Waste - To - Energy - Concepts For Efficient Power Plants

Download as pdf or txt
Download as pdf or txt
You are on page 1of 150
At a glance
Powered by AI
The text discusses processes involved in converting municipal solid waste to energy and discusses problems like corrosion that affect plant efficiency. It also explores innovative cycles combining waste-to-energy plants with gas turbines to further increase efficiency.

The main sections discussed are the delivery and storage section, energy recovery section, and combustion chamber.

Grate incinerators like reverse reciprocating grate and push forward grate are some common types of incinerators mentioned.

Waste-to-Energy

Lisa Branchini

Waste-to-Energy
Advanced Cycles and New Design Concepts
for Efficient Power Plants

1  3
Lisa Branchini
CIRI- Energy and Environment
University of Bologna
Bologna
Italy

ISBN 978-3-319-13607-3    ISBN 978-3-319-13608-0 (e-Book)


DOI 10.1007/978-3-319-13608-0

Library of Congress Control Number: 2015933516

Springer Cham Heidelberg New York Dordrecht London


© Springer International Publishing Switzerland 2015
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recita-
tion, broadcasting, reproduction on microfilms or in any other physical way, and transmission or in-
formation storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publica-
tion does not imply, even in the absence of a specific statement, that such names are exempt from the
relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)


To my parents,
who always gave me the chance to choose.
“Nothing in life is to be feared, it is only to
be understood.”
Marie Sklodowska Curie

v
Preface

Who Should Read This Text?

This text is intended for students and engineers interested in the design of classi-
cal and novel waste-to-energy (WTE) systems. It focuses on: (i) the description of
processes involved in the conversion of municipal solid waste (MSW) into energy;
(ii) key problems concerning corrosion, which negatively affect WTE steam cycle
design parameters; (iii) possibilities to increase WTE conversion efficiency and
performance through innovative cycles based on integration with gas turbines (GT).
This text has evolved from 3 years of PhD research studies carried out at the
DIN—Department of Industrial Engineering, Alma Mater Studiorum—Università
di Bologna.
The target readers are graduate students, masters students, or PhD students with
a thorough understanding of basic thermodynamic processes in energy systems, and
engineers working in the field of energy systems and MSW conversion. Common
prerequisites are knowledge of general mechanical engineering topics such as ther-
modynamics, fluid machines, energy systems, etc.

Why Has This Text Been Written?

Several countries have waste management systems with significant improvement


boundaries to reach worldwide state-of-the-art engineering, health, and safety stan-
dards. An increase in environmental and health concerns, combined with new possi-
bilities to exploit these ordinary life by-products as a valuable energy resource, has
led to the exploration of alternative methods for waste disposal. In this context, the
energy conversion of MSW in WTE power plants is one of the principal means of
integrated waste management; its potential is increasing throughout Europe, both in
terms of plant number and capacity, furthered by legislative directives. The aim of
the new Waste Framework Directive 2008/98/EC is to drive WTE facilities to maxi-
mize energy recovery and utilization of waste heat, in order to substitute energy

vii
viii Preface

produced with conventional fossil fuel-fired power plants. This calls for novel ap-
proaches to maximize the conversion of MSW into energy. In particular, the idea of
an integrated configuration made up of a WTE and a GT derives from the desire to
eliminate, or at least mitigate, limitations affecting the WTE conversion processes
bounding the thermodynamic efficiency of the cycle. The aim of this book is to
study and propose innovative integrated WTE cycles. It investigates, from a ther-
modynamic point of view, integrated WTE–GT system sharing, particularly the
steam cycle. The book also proposes and analyzes novel approaches to maximize
the conversion of MSW into energy, and discusses, in its final chapters, the difficult
task of optimizing conversion efficiency for multi-fuel (MF) energy systems.

What Can Be Learned from This Text?

The primary emphasis explains the basic concepts related to WTE conversion
processes, highlighting the most relevant aspects limiting the thermodynamic ef-
ficiency of a WTE power plant. In particular, a preliminary thermodynamic analysis
is carried out to quantify the influence of the main steam cycle parameters and plant
configurations for WTE conversion efficiency. The book’s core lesson describes
and discusses the integrated WTE–GT concept. Two basic types of hybrid dual-
fuel combined cycle arrangements are explained: (i) steam/waterside integration
and (ii) windbox repowering. Focusing on steam/waterside integration, various lay-
outs are proposed; for each analyzed configuration, the optimum WTE–GT plant to
maximize steam generation and power output, and to minimize discharged outlet
temperatures is identified.
Issues of power output allocation and conversion efficiency for an MF energy
system, receiving different type of input fuels and generating a single useful output,
are discussed. Based on several performance indexes, specifically introduced to
evaluate an MF energy system, different approaches are compared in order to evalu-
ate the WTE–GT integrated system performance.
Finally, a specific case study of a medium-sized WTE facility integrated with
different market-available GT units is presented; for each considered, GT units and
integrated plant configurations, as well as detailed modifications to the WTE origi-
nal steam cycle and the resultant achieved performance gains are discussed.

Acknowledgments

A number of people have directly or indirectly contributed to this text. I would


like to thank, in particular, my advisors, Prof. Michele Bianchi and Prof. Antonio
Peretto, for pushing me to keep learning. Sincerest thanks are due to my colleagues,
Francesco and Andrea, for their ongoing support, and advice since I was a student.
I would like to thank Springer, the publisher of the book, and my editors, Tiffany
Preface ix

Gasbarrini and Rebecca Hytowitz for giving me this opportunity and guiding me
through every step of the way from the beginning. Finally, a special thanks goes to
my friend and colleague Elisa for her continuous support and friendship over all
these years.

Bologna, October 2014 Lisa Branchini

Book organization

The manuscript is divided into four main parts.


Part I: Presents the outline of the scenarios of interest and describes the funda-
mental key aspects concerning the waste-to-energy conversion processes. Chap. 2
presents an overview of municipal solid waste production and disposal for both
the European and the Italian markets. Chapter 3 introduces and describes the basic
concepts related to waste-to-energy conversion processes, highlighting the most rel-
evant aspects circumscribing the thermodynamic efficiency of a WTE power plant.
Part II: In Chap. 4, a thermodynamic analysis is carried out in order to quantify
the influence of steam cycle parameters and plant configurations on WTE ther-
modynamic efficiency. The aim of this preliminary analysis is to understand and
compare possibilities and benefits of a thermodynamic cycle upgrade for a WTE
facility.
Part III: Explains and discusses the hybrid combined cycle (HCC) concept.
Two basic types of hybrid dual-fuel combined cycle arrangements are detailed in
Chap. 5: (i) steam/waterside-integrated HCC and (ii) windbox repowering.
Steam/waterside integration represents a cost-effective method to increase the
electric efficiency of a WTE power plant without the necessity of new expensive
materials to prevent corrosion risks. Indeed, a further superheating of the steam is
performed in a separated heat recovery steam generator (HRSG), by means of a
high-grade fuel, that does not cause corrosion problems in the superheater banks.
Windbox repowering, however, due to a considerable increase in combustion gas
volumetric flow rate, requires a significant redesign of combustion, convective, and
gas cleaning sections.
With reference to WTE–GT steam/waterside integration, the logic governing
plant match in terms of steam production as the function of thermal power is inves-
tigated and explained. The thermodynamic analysis carried out in Chap. 6 assesses
and defines, for given layout and operative conditions, the optimum WTE–GT plant
match to maximize steam generation and power output, and to minimize discharged
outlet temperatures. Moreover, several proposed WTE–GT steam/waterside-
integrated layouts, with reference to one pressure level heat recovery boiler con-
figuration, are presented and detailed.
Part IV: In Chap. 7, open research issues to evaluate the performance and the
efficiency conversion of multi-fuel (MF) energy systems are discussed. The dif-
ficulty in defining a performance index capable of quantifying the efficiency of
x Preface

the integrated system, compared to separate generation, lies in assigning the extra
power generated as a consequence of systems integration. Thus, in Chap. 7, sev-
eral performance indexes are proposed and described. Finally, in Chap. 8, detailed
modifications to the WTE cycle and the resulting enhancement of its performance
are presented for two different WTE–GT integrated cases. Numerical results for a
representative WTE plant integrated with different commercial GT units are shown
and discussed. The performance indexes and output allocation approaches defined
in Chap. 7 are applied to quantify, for each case, integrated system performance.
Contents

Part I  WTE State-of-the-Art

1 Introduction �����������������������������������������������������������������������������������������������    3
References����������������������������������������������������������������������������������������������������    5

2 Municipal Waste Overview ����������������������������������������������������������������������    7


2.1 Municipal Solid Waste Definition and Management
System Hierarchy������������������������������������������������������������������������    7
2.2 Overview of Waste Production and Disposal for European Countries   9
2.2.1 Overview of Municipal Solid Waste Production and
Disposal in Italy����������������������������������������������������������������������  12
2.3 Municipal Solid Waste Landfill Average Costs����������������������������������  13
References����������������������������������������������������������������������������������������������������  16

3 Waste-to-Energy ����������������������������������������������������������������������������������������   19
3.1 Basics of a WTE Power Plant�������������������������������������������������������������  19
3.1.1 Waste Delivery and Storage Section���������������������������������������  20
3.1.2 The Combustion Section���������������������������������������������������������  21
3.1.3 The Energy Recovery Section�������������������������������������������������  26
3.1.3.1 Corrosion Protection��������������������������������������������������  28
3.2 WTE Plant Distribution in the European Scenario�����������������������������  30
3.2.1 WTE Plant Efficiency in a Representative
National Scenario�������������������������������������������������������������  31
3.3 EU Regulation Framework Oriented to WTE Efficiency�������������������  33
References����������������������������������������������������������������������������������������������������  36

Part II  WTE Thermodynamic Analysis

4 Waste-to-Energy Steam Cycle������������������������������������������������������������������   39


4.1 Steam Cycle State-of-the-Art Parameters and Layout������������������������  39
4.2 Steam Cycle Upgrade: Effects on Cycle Efficiency���������������������������  44
4.3 New Designs for High-Efficiency WTE Plant������������������������������������  49
References����������������������������������������������������������������������������������������������������  53
xi
xii Contents

Part III  WTE Advanced Cycles

5 Waste-to-Energy and Gas Turbine: Hybrid Combined


Cycle Concept �����������������������������������������������������������������������������������������    57
5.1 The HCC Concept�����������������������������������������������������������������������������    57
5.1.1 WTE–GT Steam/Waterside Integration��������������������������������    59
5.1.2 WTE–GT Windbox Integration��������������������������������������������    61
5.2 State-of-the-Art of Integrated WTE–GT�������������������������������������������    63
5.3 Existing WTE–GT Integrated Power Plants�������������������������������������    64
5.3.1 Zabalgarbi WTE–GT Power Plant: The SENER Solution���    65
5.3.2 Moerdijk WTE–GT Power Plant: The Dutch Solution���������    67
5.3.3 Takahama WTE–GT Power Plant: The Japanese Solution���    68
References��������������������������������������������������������������������������������������������������    69

6 WTE–GT Steam/Waterside Integration: Thermodynamic


Analysis on One Pressure Level��������������������������������������������������������������    71
6.1 Thermodynamic Analysis of Steam Production�������������������������������    71
6.1.1 Influence of Evaporative Pressure and GT Outlet
Temperature on Steam Production����������������������������������������    76
6.2 Numerical Results�����������������������������������������������������������������������������    78
6.2.1 Optimum Plant Match in Terms of Electric Power Ratio�����    80
6.2.2 Traditional WTE Versus Integrated Plant: Steam
Turbine Capacity���������������������������������������������������������������    81
6.3 Conclusion����������������������������������������������������������������������������������������    83
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG��������    83
6.5 Comparative Results of WTE–GT One-Pressure-Level
Integrated Layouts����������������������������������������������������������������������������  104
References��������������������������������������������������������������������������������������������������  109

Part IV  Performance and Efficiency Conversion Issues

7 Performance Indexes and Output Allocation for Multi-fuel


Energy Systems����������������������������������������������������������������������������������������   113
7.1 Context����������������������������������������������������������������������������������������������  113
7.2 Performance Evaluation of an MF Energy System���������������������������  115
7.2.1 MF Energy System Arrangement������������������������������������������  115
7.2.2 Indexes for MF Energy System Performance Evaluation�����  116
7.2.2.1 First Law Efficiency�������������������������������������������������  116
7.2.2.2 Electric Equivalent Efficiency���������������������������������  116
7.2.2.3 Relative SI����������������������������������������������������������������  118
7.2.2.4 MF SI�����������������������������������������������������������������������  119
7.2.3 Useful Output Allocation to Each ith Fuel����������������������������  121
7.2.3.1 Allocation Approach #1�������������������������������������������  121
7.2.3.2 Allocation Approach #2�������������������������������������������  121
Contents xiii

7.3 Application Example: Two-fuel Co-combustion Power Plant����������  122


7.4 Conclusions���������������������������������������������������������������������������������������  125
References��������������������������������������������������������������������������������������������������  126

8 Specific Application Cases with GT Commercial Units ����������������������   127


8.1 Midsize WTE Reference Steam Cycle���������������������������������������������  127
8.2 WTE Integration with GT Units: Investigated Layout
Cases and Results������������������������������������������������������������������������������  130
8.2.1 GT Unit Selection�����������������������������������������������������������������  132
8.2.2 WTE–GT Integrated Plant Numerical Results����������������������  133
8.3 Conclusion����������������������������������������������������������������������������������������  138
References��������������������������������������������������������������������������������������������������  139

Index ���������������������������������������������������������������������������������������������������������������  141


List of Abbreviations

BC Bottomer cycle
C Compressor
CC Combined cycle
CHP Combined heat and power
DEA Deaerator
ECO Economizer
EVA Evaporator
EU European Union
FGR Flue gas recirculation
GT Gas turbine
HCC Hybrid combined cycle
HE general heat exchanger
HP High pressure
HRSG Heat recovery steam generator
IBC Inverted Brayton cycle
ICE Internal combustion engine
IP Intermediate pressure
LHV Lower heating value
LP Low pressure
MF Multi-fuel
MSW Municipal solid waste
ORC Organic Rankine cycle
PL Pressure level
RH Reheater
SF Single fuel
SH Superheater
ST Steam turbine
T Turbine
TC Topping cycle
WFD Waste framework directive
WTE Waste-to-energy

xv
xvi List of Abbreviations

Symbols

c specific heat [kJ/kgK]


E electric energy output [MWh]
F fuel input power [MW]
fb equivalent factor accounting for energy losses [–]
h specific enthalpy [kJ/kg]
Iimp Imported Enery [MWh]
m mass flow rate [kg/s]
p pressure [bar]
P power output [MW]
Q heat [MW]
SI Synergy Index [–]
T temperature [°C]
U total useful electric energy output [MWh]

Greek Symbols

ε electric equivalent conversion factor [–]


η efficiency [–]
∆ Difference
ξ HRSG effectiveness [–]
x fuel input weighting factor [–]

Subscripts and Superscripts

I first law
boil Waste-to-Energy boiler
C Carnot
EE Electric Equivalent
EXH exhausted
ev evaporation
i generic i-th input
in input
it isentropic efficiency
j generic j-th input
lat latent heat
max maximum
nu non useful heat
List of Abbreviations xvii

O Outlet
Prod produced
rec recoverable
ref reference
s steam
sc sub-cooling
TOT Total
W Waste
Part I
WTE State-of-the-Art
Chapter 1
Introduction

In recent years, the increased focus on energy resources and environment has
changed perceptions about waste. The growth in economic activities, standard of
living, and population has led to a sharp increase in the quantity of generated waste.
Waste poses a highly complex and heterogeneous environmental problem since all
human activities, inevitably, result in generation of waste due to imperfect utiliza-
tion of energy and resources.
Currently, refuse disposal methods are determined by cost and their effect on the
environment. Intensification of environmental and health concerns, combined with
the possibility of exploiting day-to-day by-products as valuable energy resources,
has led to the exploration of alternative methods for waste disposal. Energy conver-
sion of municipal solid waste (MSW) in waste-to-energy (WTE) power plants is
one of the principal means to integrated waste management; its potential is increas-
ing throughout Europe, both in terms of plant number and capacity and is furthered
by legislative directives.
Waste Framework Directive (WFD) 2008/98/EC [1] sets the basic concepts and
definitions related to waste management and lays down waste management prin-
ciples, introducing the so-called “waste hierarchy” in which waste prevention is
still considered the main goal followed by reuse and recycle. Nevertheless, growing
interest in the WTE conversion process is evident.
WTE plants have been established worldwide as a sustainable method for the
disposal of residual waste.
Big investments were made in new WTE facilities and recycling activities in
many European Union (EU) countries at the beginning of this century. There are
enough incineration and recycling capacities in Central Europe to prevent landfill-
ing of organic waste. Since huge amounts of greenhouse gas (GHG) emissions have
been prevented by WTE facilities as compared to landfills, the unused capacities
of WTE installations should be used to recover energy from waste in EU Member
States where waste disposal is predominantly through landfills.

© Springer International Publishing Switzerland 2015 3


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_1
4 1 Introduction

The possibility of generating energy in the form of heat, electricity, or combined


heat and power (CHP) became an attractive way of dealing with waste. Although
various WTE technologies such as waste incineration, anaerobic digestion, gasifica-
tion, etc. exist, mass incineration is still the most common.
Incineration plants, which were a viable option earlier, are the predecessors of
the present day’s WTE plant. Heat recovery mechanisms were added to the incin-
erators in the form of waste heat boilers and later in the form of low pressure and
temperature steam boilers. These incinerators then evolved into waterwall boilers
with an integrated furnace.
At present, direct combustion over a moving grate with the generation of super-
heated steam feeding a steam turbine in a Hirn cycle is the dominant technology
used to recover energy from MSW. The amount of energy recovered from the MSW
combustion can vary significantly with the characteristics of the MSW fed into the
boiler (composition, mass flow rate, and heating value), the combustion technology,
the configuration and features of the recovery boiler (adiabatic or integrated), and
the characteristics of the thermodynamic cycle. Owing to the heterogeneous nature
of waste, there are some differences between a conventional fossil fuel power plant
and a WTE plant with respect to the chemical-to-electrical energy conversion pro-
cess. The thermodynamic efficiency of a WTE power plant is constrained, mainly,
by the following aspects [2]:
• Maximum temperature of the steam cycle is limited by corrosion problems,
mainly affecting the high-temperature sections.
• The steam-superheated temperature makes it necessary to keep the evaporative
pressure low to avoid build-up of liquid at the steam turbine outlet.
• Typically modest power output and steam mass flow rate of a WTE power plant
imply low steam turbine isentropic efficiency.
Consequently, the thermodynamic efficiency of WTE power plants typically ranges
from 25 to 30 %.
The new WFD 2008/98/EC lays down the criteria for producing energy effi-
ciently from waste (the so-called “R1 formula”) to evaluate recovery status of the
plant. Although the energy efficiency criteria can only be regarded as a starting
point for all the Member States, they are a benchmark for quality standards in waste
recovery. The energy recovery status provides incentives for future investments in
WTE plants, especially those located close to customers. The aim of the Directive
is to spur WTE facilities to maximize energy recovery and waste heat utilization so
they can eventually substitute conventional fossil fuel-fired power plants.
There is substantial scope for improvement in WTE systems. This calls for novel
approaches to maximize the conversion of MSW into energy.
This text investigates, from a thermodynamic point of view, the scope for per-
formance improvement in WTE plants, starting with the use of state-of-the-art tech-
nologies. Both repowering and advanced WTE options are analyzed.
References 5

In particular, the idea of an integrated configuration made up of a WTE and a


gas turbine (GT) derives from the desire to eliminate, or at least mitigate, the limita-
tions of the WTE conversion processes bounding the thermodynamic efficiency of
the cycle.
The study on the integrated WTE–GT system, that is, sharing (i) the steam cycle,
(ii) the flue gas paths, or (iii) combining both the mechanisms, is the book’s core.
The thermodynamic analysis investigates and defines the logic governing plants
and it should match in terms of steam production and steam turbine power output as
a function of the thermal power generated. The book presents detailed discussions
on the modifications to a typical WTE steam cycle arrangement in order to evaluate
the resulting performance enhancement. Numerical results of a conventional refer-
ence WTE plant integrated with different GT commercial units are discussed.
This book aims to provide guidelines for designing advanced WTE solutions for
future implementation in new and/or existing plants of industrial partners and other
similar industrial arrangements.
Another purpose of the book is to discuss meaningful performance indicators of
systems such as multi-fuel (MF) energy systems. Performance indexes, specifically
introduced to assess the performance of the proposed integrated configuration and
to determine the power output of each input fuel, have been presented and applied
to the plant under consideration.
The framework of this research integrates a collaborative project involving Alma
Mater Studiorum—Università di Bologna—DIN and a leading Italian multiutility
company operating, particularly, in the field of design and management of WTE
facilities. This joint research project was aimed at assessing the possibility of inte-
grating new and existing WTE power plants with GT units.

References

1. Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008 on
waste. Official Journal of the European Union OJ L 312/3, 22.11.2008
2. Bianchi M, Branchini L, De Pascale A, Falchetti M, Fiore P (2014) Advanced waste-to-energy
steam cycles. Energy Procedia 45:1205–1214. doi:10.1016/j.egypro.2014.01.12
Chapter 2
Municipal Waste Overview

Improved waste management is an essential component in order to make countries


more resource efficient. This important target is driving European environmental
policies and legislations towards a more efficient waste management system.
In this chapter, the “waste management hierarchy,” proposed by the 2008/98/EC
European Union Directive, is described. Moreover, in order to provide a general
overview on municipal solid waste (MSW), this chapter focuses on waste generation
and disposal for both the European and the Italian markets. Information and data
concerning European Union (EU) countries’ progress in enhancing recycling rates
will be discussed, with a focus on on related policies such as landfill average costs
and taxes.

2.1 Municipal Solid Waste Definition and Management


System Hierarchy

The definition of “municipal waste” used all around the EU countries varies,
­reflecting different waste management operations. In order to collect and com-
pare yearly reporting data of waste generation and management from European
countries, “municipal waste” is defined as follows [1]: “Municipal waste is mainly
­produced by households, though similar wastes from sources such as commerce, of-
fices and public institutions are included. The amount of municipal waste generated
consists of waste collected by or on behalf of municipal authorities and disposed of
through the waste management system.”
In this context, municipal waste is understood as waste collected by or on behalf
of municipalities.
In the EU’s Landfill Directive, MSW is defined as [2] “waste from households,
as well as other waste which, because of its nature or composition, is similar to
waste from households.”

© Springer International Publishing Switzerland 2015 7


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_2
8 2  Municipal Waste Overview

Fig. 2.1   Waste management hierarchy as suggested by 2008/98/EC Directive

Thus, the definition refers to “household waste and similar waste.” The typical
MSW composition contains food and kitchen waste, green waste, glass, paper, plas-
tics, as well as other recyclable materials, etc.
The revised 2008/98/EC Directive also sets the basic concepts and definitions
related to waste management and lays down waste management principles such
as the “waste hierarchy” [3]. The EU’s approach to waste management is based on
three fundamental principles: waste prevention, recycling and reuse, and improving
final disposal and monitoring. Based on the EU’s approach to waste management,
the best and most economical way of dealing with waste is to minimize its produc-
tion; if waste cannot be prevented, as many of the materials as possible should be
recovered, preferably by recycling. Where possible, waste that cannot be recycled
or reused should be safely incinerated, recovering the energy released with waste
combustion and leaving landfill as the last option for waste disposal.
Figure 2.1 schematically shows, through an inverted pyramid, the waste manage-
ment hierarchy suggested by European Commission’s directives.
European environmental policies and national efforts must be in the direction
of shifting waste management up the waste hierarchy—reducing waste disposal
(i.e., landfilling) while focusing on waste prevention, reuse, recycling, and energy
recovery.
Development and progress on waste management for each one of the 32 Euro-
pean Environment Agency (EEA) member countries1 will be discussed in the next
paragraphs, reviewing the latest available data.

1 
The EU-27, Iceland, Norway, Switzerland, Turkey, and Croatia.
2.2  Overview of Waste Production and Disposal for European Countries  9

2.2 Overview of Waste Production and Disposal


for European Countries

The socioeconomic growth of the most industrialized countries has involved a pro-
gressive increase of MSW production. In the EU-27, the average MSW production
was about half a metric ton per person in 2010 [4]; hence, issues related to the dis-
posal of MSW turn out to be very important. Figure 2.2 shows the amount of MSW
produced per person for the 32 EEA members in 2001 and 2010.
A comparison of yearly MSW generation data shows that most of the countries
(21 of 32) generated more municipal waste per capita in 2010 than in 2001, while
11 reduced municipal waste per capita.
Despite the problem that the MSW definition differs between countries (which
means that data should be used with caution), a very wide range of waste generation
can be observed.
The EU-27 average value equaled 502 kg/p.p. in 2010; countries such as Cy-
prus or Switzerland considerably exceed the EU average value, whereas Eastern

800

2001
700 2010
MSW generated per capita [kg/p.p.y]

600

500

400

300

200

100

0
Slovenia

Slovakia
Lithuania
United Kingdom
Iceland*

Croatia**
Italy
Netherlands

Finland

Latvia
Romania
Czech Republic

Estonia
Switzerland

Ireland

Turkey
Germany

Bulgaria
Luxembourg

Belgium

Poland
Portugal
Denmark

Norway

Hungary
Malta
Austria

Spain
Cyprus

Greece
France

Sweden
EU-27

Fig. 2.2   Municipal waste generated per capita in 2001 and 2010 [4]. (Iceland: 2008 data used for
2010; Croatia: 2004 data used for 2001)
10 2  Municipal Waste Overview

UHF\FOLQJ
LQFLQHUDWLRQ
ODQGILOOLQJ
 QRLQIRDERXWWUHDWPHQW




06:PLOOLRQWRQV










         

Fig. 2.3   Development of municipal waste management in 32 European countries, from 2001 to
2010 [4]

­ uropean countries have lower values. Italy, with a MSW production of about
E
530 kg/p.p./year, lies just above the EU-27 average value.
Figure 2.3 shows the MSW total generation and final treatments for the 32 EU
countries from 2001 to 2010. The data highlight that European countries are climb-
ing up the waste hierarchy for municipal waste management, thereby implementing
one or more key principles of the Waste Framework Directive (such as recycling
and incineration) while reducing the use of landfills.
Figure 2.3, however, highlights that more than half of the countries still land-
fill over 50 % of their municipal waste. The main reason is that landfill is still the
cheapest and simplest way, among all waste-processing technologies, to deal with
waste disposal in most of the countries.
In Fig. 2.4, percentages of MSW final treatments in the EU-27 countries are re-
ported in detail. Only few exceptions (such as the Netherlands, Denmark, and Swe-
den) have a high level of alternatives for final treatment disposal. The percentage
of incinerated waste ranges from zero (e.g., East European countries) to over 50 %
(e.g., Denmark) with an average value equal to about 18 % for 27 EU countries in
2.2  Overview of Waste Production and Disposal for European Countries  11

recycling and composing landfilling


incineration
100

80
% of waste final treatement

60

40

20

0
Slovenia
Lithuania

United Kingdom
Slovakia

Iceland
Italy

Finland
Netherlands
Ireland

Latvia

Romania
Czech Republic

Estonia

Turkey
Switzerland
Belgium
Bulgaria

Poland
Portugal

Croatia
Denmark

Hungary

Norway
Luxembourg
Malta
Austria
Spain

Cyprus
Greece
France

Sweden
Germany

Fig. 2.4   Final treatment percentage allocation for 27 European countries in 2010 [4]. (No data is
available for Croatia)

2010. In general, during the period from 2001 to 2010, the increase in recycling and
incineration of MSW has led to a decrease in landfill waste (see Fig. 2.3).
Figure 2.5 compares MSW recycling rates (including also composting and diges-
tion of organic waste) for EEA countries in 2001 and 2010. In the analyzed period,
although percentage rates show considerable differences in performance between
nations, 22 countries have increased their recycling performance between five and
ten percentage points. This clearly indicates a significant improvement in recycling
performance. Although, five countries (Switzerland, the Netherlands, Belgium,
Austria, and Germany) have already achieved the 50 % recycling target imposed
by the European 20-20-20 targets, the majority of countries will need to make an
extraordinary effort in order to achieve this goal.
12 2  Municipal Waste Overview



06:UHF\FOOLQJUDWHV>@  






:DVWH)UDPHZRUN'LUHFWLYH
 WDUJHWZLWKLQ










,WDO\
+XQJDU\

1RUZD\
7XUNH\

,FHODQG

,UHODQG
&\SUXV

/X[HPERXUJ
%XOJDULD

/DWYLD
/LWKXDQLD

*UHHFH

)UDQFH

%HOJLXP
3RUWXJDO

8QLWHG.LQJGRP
'HQPDUN

1HWKHUODQGV
)LQODQG

6ZHGHQ
&URDWLD

0DOWD

$XVWULD
5RPDQLD

6ORYDNLD

&]HFK5HSXEOLF

(VWRQLD

6ORYHQLD
3RODQG

6SDLQ

6ZLW]HUODQG

*HUPDQ\

Fig. 2.5   MSW recycled for 32 European countries in 2001 and 2010 [4]. WFD Waste Framework
Directive

2.2.1 Overview of Municipal Solid Waste Production


and Disposal in Italy

Figures 2.6 and 2.7 present the data of MSW total generation from 1998 to 2013
[5] and of MSW generated per capita in 2009 and 2013, focusing on the Italian
national and regional markets. According to the report of Istituto Superiore per
la Protezione e la Ricerca Ambientale (ISPRA), there are remarkable differences
in per capita production across regions. In 2009, central regions had the highest
per capita production of waste, equal to about 604 kg/p.p.y, whereas lower values
were recorded for southern regions, namely, 493 kg/p.p.y. Emilia Romagna (with
666 kg/p.p.y) had the highest MSW production per habitant, followed by Toscana
(with 663 kg/p.p.y). In 2013, the total generation in Italy was about 29,595 metric
tons of municipal waste.
2.3  Average Costs of Municipal Solid Waste Landfill 13

Fig. 2.6   MSW produced in Italy from 1998 to 2013 [5]

In 2009, landfills received around 45 % of the total MSW managed, as shown in
Fig. 2.8. Comparing the regional data for waste disposal in 2009, Fig. 2.8 highlights
an inhomogeneous situation between regions—northern regions show reduced use
of landfill (Lombardia records the lowest use representing 7 % of its total produc-
tion). An exception, in the southern regions, is represented by Sardegna where re-
gional legislation is driving towards recycle and reuse of waste.

2.3 Average Costs of Municipal Solid Waste Landfill

Cost of various waste disposal methods varies depending on the technology ad-
opted and on the country’s specific policy measures (such as taxes). As a rule of
thumb, incineration costs are twice the costs of landfill. A recent study [6] quanti-
fies incineration costs between 100 and 250 €/t of waste, whereas costs for landfill
have a range down to 20 €/t of waste. Thus, the main reason why landfill is still the
predominant treatment option for most of the EU countries is primarily economic.
Taxes and charges tend to drive the economics of recycling and composting.
14 2  Municipal Waste Overview

700
2009

600 2013
MSW generated per capita [kg/p.p.y]

500

400

300

200

100

0
Friuli Venezia Giulia
Trentino Alto Adige

Lazio
Piemonte

Abruzzo
Lombardia

Veneto

Toscana
Liguria
Emilia Romagna

Sicilia
Marche

Basilicata

Sardegna
Valle d'Aosta

Molise
Umbria

Puglia

Calabria
Campania

Fig. 2.7   MSW produced per capita in Italian regions in 2009 and 2013 [5]

The most relevant case is Denmark, where cost for landfilling both household
waste and hazardous2 waste drastically increased in the past 17 years [7]. Table 2.1
collects available cost data for some of the EU countries.
A correlation between recycling rates (see Fig. 2.5) and landfill costs is evident.
Nations which have sharp policy measures, such as additional landfill taxes or eco-
nomic support to build up recycling infrastructures, are expected to achieve high
recycling rates. As shown in Table 2.1, there is a general trend in increasing costs
for waste disposal throughout almost all countries considered. Figure 2.9 shows
gate fee and landfill taxes for MSW landfilling in EU Member States. EU countries
appear much more likely to meet a 50 % recycling target once landfill charges (or
the cost of the cheapest disposal option) approach EUR 100 per metric ton [8].

Waste derived from common manufacturing and industrial processes, from specific industries
2 

and from commercial chemical products, etc. are classified as hazardous.


2.3  Average Costs of Municipal Solid Waste Landfill 15

120
2009
2013
100
MSW landfilled rates [%]

80

60

40

20

Trentino Alto Adige


Friuli Venezia Giulia
Emilia Romagna
Sicilia

Basilicata
Italy

Campania

Lombardia
Calabria

Piemonte
Veneto

Valle d’Aosta
Toscana
Puglia

Molise

Liguria
Marche
Umbria

North
Lazio
South

Center
Sardegna

Abruzzo

Fig. 2.8   Percentage of landfill disposal of MSW generation in Italian regions in 2009 and 2013 [5]

Table 2.1   Estimated cost range for landfilling of waste (excluding any landfill tax, if relevant) [7]
Country Latest data Cost range for MSW [€/t] % change
Austria 1999 50–150 –
Belgium-Flemish 2003 116  + 5 (in 2 years)
Denmark 2004 110  + 57 to + 340 (in 17 years)
Finland 2003 30–121 –
Germany 2005 123 (average from 12 sites) –
Greece 2005 8–35  + 75 (in 6 years)
Ireland 2005 120–240  + 52 (in 4 years)
Italy 2003 90–110 –
Luxembourg 2003 50 –
Netherlands 2002 58 − 8 (in 4 years)
Portugal 2004 26 –
Spain 2004 12 –
Sweden 2004 70–90 –
UK 2003 21 –
16 2  Municipal Waste Overview

/DQGILOOJDWHIHH
/DQGILOOWD[HV







/DQGILOOLQJFKDUJH¼W










%HOJLXP:DO

/DWYLD
/LWXDQLDK

3RUWXJDO
&]HFK5HSXEOLF
%HOJLXP)O

%XOJDULD

1HWKHUODQGV
(VWRQLD

*UHHFH

0DOWD

5RPDQLD

6ORYHQLD
6ORYDNLD

8QLWHG.LQJGRP
'HQPDUN

)LQODQG

,UHODQG
,WDO\
*HUPDQ\

/X[HPERXUJ
$XVWULD

3RODQG
)UDQFH

+XQJDU\

6YHGHQ
&\SUXV

6SDLQ

Fig. 2.9   Typical charges (gate fee and taxes) for MSW landfilling in EU Member States and
regions [4, 8]

References

1. Eurostat (2012) Reference Metadata in Euro SDMX Metadata Structure (ESMS). http://epp.
eurostat.ec.europa.eu/cache/ITY_SDDS/EN/env_wasmun_esms.htm. Accessed 11 Jan 2013
2. EU (1999) Council Directive 1999/31/EC of 26 April 1999 on the landfill of waste, Official
Journal of the European Union OJ L 182, 16.7.1999, pp 1–19
3. Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008 on
waste. Official Journal of the European Union OJ L 312/3, 22.11.2008
4. European Environment Agency EEA, Report No 2/2013, Managing municipal solid waste—a
review of achievements in 32 European countries. http://www.eea.europa.eu/data-and-maps/
figures/municipal-waste-generated-per-capita. Accessed 1 April 2014
5. ISPRA, Istituto Superiore per la Protezione e la Ricerca Ambientale, Edizione (2014) Rapporto
Rifiuti Urbani, ISBN: 978-88-448-0665-1. http://www.isprambiente.gov.it/files/pubblicazioni/
rapporti/RapportoRifiutiUrbani2014_web.pdf. Accessed 19 Sept 2014
References 17

6. Göerner K (2003) Waste Incineration European State of the Art and New Developments. IFRF
Combust J Article Number: 200303. ISSN 1562-479X
7. Report on Implementation of the Landfill Directive in the 15 Member States of the European
Union; October 2005, European Commission. http://ec.europa.eu/environment/waste/pdf/re-
port_a2.pdf. Accessed 7 March 20134
8. Bio Intelligence Service (2012) Use of economic instruments and waste management per-
formances, Contract ENV.G.4/FRA/2008/0112, European Commission (DG ENV). http://
ec.europa.eu/environment/waste/use.htm. Accessed March 2014
Chapter 3
Waste-to-Energy

This chapter introduces and describes the basic concepts related to the waste-to-
energy (WTE) conversion processes, highlighting the most relevant aspects that
limit the thermodynamic efficiency of a WTE power plant.
An overview of the distribution of WTE plants around Europe, electrical and
thermal energy generated, and average plant capacity is provided at the beginning
of the chapter. Basic sections and related operations of a WTE facility are described
along with typical operative steam cycle parameters. The chapter’s final section is
dedicated to the definition of the “energy recovery status” (R1 formula) according
to the 2008/98/EC Waste Framework Directive (WFD). The thresholds to achieve
the energy recovery status are discussed according to WTE plants’ operative data.

3.1 Basics of a WTE Power Plant

Waste incineration became established in many European countries at the end of


the nineteenth century as a device to minimize waste’s volume and to make it hy-
gienic. Since then, as environmental awareness has increased, WTE technology
has continuously developed and improved, increasing the importance of efficient
energy generation. WTE power plants, in many European countries, have become
important tools to generate heat and electricity, replacing the energy produced by
conventional fossil fuel power plants.
Various WTE technologies, such as waste incineration, anaerobic digestion,
and gasification, exist today; despite that mass burn incineration is still the most
­common.
The dominant technology for energy recovery from municipal solid waste
(MSW) is direct combustion over a moving grate with the generation of superheat-
ed steam feeding a steam turbine in a Hirn cycle. The amount of energy recovered
from the MSW combustion can vary significantly with the characteristics of MSW
fed into the boiler (composition; mass flow rate; and lower heating value, LHV),
the combustion technology, the configuration and features of the recovery boiler
(adiabatic or integrated), and the characteristics of the thermodynamic cycle.

© Springer International Publishing Switzerland 2015 19


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_3
20 3 Waste-to-Energy

Fig. 3.1   Schematic of a WTE power plant highlighting four macro sections

A simplified and schematic representation of a WTE power plant is shown in


Fig. 3.1 highlighting four macro processes and related plant sections:
• Waste delivery and storage section (bunker)
• Waste combustion section (furnace)
• Energy recovery and conversion section
• Flue gas cleaning section
Table  3.1 schematically presents the purpose of the main components of a WTE
power plant.

3.1.1 Waste Delivery and Storage Section

Trucks, trains, or containers arrive at the delivery area in order to dump the waste
into the bunker, usually after visual control and weighing. Enclosure of the delivery
area can be an effective means to avoid odor, noise, and emission problems. The
bunker is usually a waterproof concrete bed. The waste is piled and mixed in the
bunker using cranes equipped with grapples. The mixing of the waste helps achieve
a balanced heat value, size, structure, composition, etc., of the material dumped into
the combustion section. In order to avoid excessive dust development and gas for-
mation (e.g., methane) from spontaneous chemical reactions between the disposals,
as well as the accumulation of odor and dust emissions, waste has to be permanently
moved, mixed, and turned by crane operators. Moreover, depending on the calorific
value of the waste as well as the layout and the concept of the plant, a fraction of
total combustion air (usually the under-grate air) is extracted from the bunker. The
3.1 Basics of a WTE Power Plant 21

Table 3.1   Purpose of the main components of a waste incineration plant [1]
Objective Responsibility of
Delivery and storage of waste Bunker
Mixing of waste
Feeding of waste to combustion section
Destruction of organic substances Furnace
Evaporation of volatile heavy metals and
inorganic salts
Production of potentially exploitable slag
Volume reduction of residues
Evaporation of water Energy conversion section
Recovery of useful energy
Removal and concentration of volatile heavy Flue gas cleaning
metals and inorganic matter into solid residues
Minimizing emission to all media

bunker usually has a storage capacity of several days (commonly 3–5 days) of the
plan operational throughput; thus, its depth can reach a few dozen meters.
Proper waste feeding from bunker to grate means dosing the right quantity of
fuel for steady combustion. Therefore, waste feeding is continuously adjusted as a
consequence of achieved combustion temperature and energy production.

3.1.2 The Combustion Section

Basically, waste incineration is the oxidation of the combustible materials contained


in the waste. Waste is generally a highly heterogeneous material, consisting essen-
tially of organic substances, minerals, metals, and water. During incineration, flue
gases containing most of the available fuel energy as heat are created. Burning of the
organic fuel substances occurs once they have reached the necessary ignition tem-
perature and come in contact with oxygen. The combustion process takes place in the
gas phase in fractions of seconds and simultaneously releases energy whenever waste
calorific value and oxygen supply is sufficient. This can lead to a thermal reaction
chain and self-supporting combustion, that is, there is no need to add other fuels.
The main phases of the waste incineration process are [1, 2]:
• Drying and degassing: When volatile content is evolved (e.g., hydrocarbons and
water) at temperatures generally between 100 and 300 °C. The drying and degas-
sing processes do not require any oxidizing agent and are only dependent on the
supplied heat.
• Pyrolysis and gasification: Pyrolysis is the further decomposition of organic
substances in the absence of an oxidizing agent at approximately 250–700 °C.
Gasification of carbonaceous residues is the reaction occurring between water
22 3 Waste-to-Energy

vapors and CO2 at temperatures of typically 500 –1000 °C. Thus, solid organic


matter is transferred to the gaseous phase. In addition to heat, water, steam, and
oxygen support this reaction.
• Oxidation: When the combustible gases created in the previous stages are oxi-
dized at flue gas temperatures generally between 800 and 1450 °C.
During waste combustion, air is typically added in two main stages: Primary air is
generally taken from the waste bunker and supplied through the grate layer into the
fuel bed, and secondary combustion air is supplied at high speeds over the grate
layer. Additional air (tertiary air) and flue gas recirculation (FGR) can also be used,
in particular with the aim to increase turbulence and decrease pollutant emissions.
The primary air is usually preheated either by using combustion grate cooling
and/or steam supplied by steam turbine bleed.
Secondary air is blown into the incineration chamber at high speeds via, for
example, injection lances. This is carried out to assure complete incineration, and it
is also responsible for the intensive mixing of flue gases and the prevention of the
free passage of unburned gas streams. Preheating of primary combustion air can
have a positive influence on overall energy efficiency in case of electricity produc-
tion. Moreover, combustion air preheating is particularly beneficial for assisting the
combustion of high moisture content waste.
When designing a waste boiler, the design metric tons per day of waste to be
combusted and the typical range of heating value that is expected for the refuse in
that location need to be known. The boiler is typically designed for the maximum
metric ton per day input and the maximum refuse heating value.
Figure  3.2 shows a typical “capacity diagram” for a waste incineration plant
where the operating area is highlighted as a function of waste mass flow rate for
different waste LHV. The two variables, waste mass flow rate and LHV, identify on
the vertical axis the thermal input in the system. From Fig. 3.2, the WTE operating
area is defined by:
• OP1: Maximum firing rate (100 % capacity) at design point waste input (design
LHV, design waste throughput)
• OP2: Design LHV waste input at minimum boiler load (e.g., 60 % capacity;
design LHV, but minimum waste throughput)
• OP3: Maximum boiler load with highest LHV (means reduced waste through-
put)
• OP4: Minimum boiler load with maximum LHV
• OP5: Reduced boiler load with minimum LHV (means maximum waste through-
put)
• OP6: Reduced boiler load with minimum LHV (means intermediate waste
throughput)
Due to the extremely heterogeneous nature of MSW, in contrast to fossil fuel, the
operating condition of a WTE plant is continuously adjusted, as much as possi-
ble, around the design point (OP1 in Fig. 3.2) in order to maintain nominal plant
conditions.
3.1 Basics of a WTE Power Plant 23

Fig. 3.2   Typical furnace capacity diagram of a waste incineration plant showing the design and
off-design operating points

In the normal operating area, two different zones can be detected: a zone where,
due to low LVH value, preheating of combustion air is necessary and an overloaded
zone where WTE operation can be allowed only for a limited amount of time.
Grate incinerators are widely applied for the incineration of mixed municipal
waste. In Europe approximately 90 % of installations treating MSW use grates [1].
Grate incinerators usually have the following components: waste feeder, in-
cineration grate, bottom ash discharger, incineration air duct system, incineration
chamber, and auxiliary burners.
Depending on waste LHV, grate cooling can be accomplished with air or a liquid
cooling medium (usually water). The flow of the cooling medium is from colder
zones to progressively hotter ones in order to maximize the heat transfer. The heat
absorbed by the cooling medium may be transferred for use in the process or for
external supply. Water cooling is most often applied when waste LHV is high (e.g.,
> 12–15 MJ/kg of MSW).
Moreover, the use of water for grate cooling may allow grate metal temperature
and local combustion temperature to be controlled with greater independence from
the primary air supply. A better control of grate temperature can allow combus-
tion of higher calorific value waste without the normally increased operational
and maintenance problems. A typical waste feeder and grate device is shown in
Fig. 3.3.
24 3 Waste-to-Energy

Fig. 3.3   Schematic of waste feeder, grate device, and combustion air system

Waste residence time over the grates is typically not more than 60 min. Different
grate systems can be distinguished by the way the waste is conveyed through the
different zones in the combustion chamber. There are essentially two main recipro-
cating grate variations, namely, continuous or discontinuous feeders:
• Reverse reciprocating grate: The grate bars oscillate back and forth in the re-
verse direction of the flow of the waste, consisting of fixed and moving grate
steps.
• Push-forward grate (Fig. 3.4): The grate bars consist of many steps oscillating
horizontally, pushing waste in the direction of the ash discharge section.
Combustion takes place above the grate in the incineration chamber (Fig. 3.5). As
a whole, the incineration chamber typically consists of a grate situated at the bot-
tom, cooled and uncooled walls on the furnace sides, and a ceiling or boiler surface
heater at the top [1]. As municipal waste generally has a highly volatile content, the
volatile gases are driven off, and only a small part of the actual incineration takes
place on or near the grate. The following requirements influence the design of the
incineration chamber:
• Form and size of the incineration grate: The size of the grate determines the size
of the cross-section of the incineration chamber.
• Vortex and homogeneity of flue gas flow: Complete mixing of the flue gases is
essential for good flue gas incineration.
3.1 Basics of a WTE Power Plant 25

Fig. 3.4   Forward-moving


reciprocating grate

Waste feeding

Sewage sludge
feeder Secondary Secondary
air
air

Feeder table

Flue gas

Waste

Drying

Degassing Primary
air
Slag discharge
Exhaust
Incineration

Fig. 3.5   Schematic of an incineration chamber


26 3 Waste-to-Energy

• Sufficient residence time for the flue gases in the hot furnace.
• Sufficient reaction time at high temperatures must be assured for complete incin-
eration.
• Partial cooling of the flue gases in order to avoid fusion of hot fly ash at the
boiler; the flue gas temperature must not exceed an upper limit at the incineration
chamber exit.
Modern standards require flue gases to be exposed to a temperature of minimum
850 °C for a minimum residence time of  2 s after the last secondary air injection [1].
Therefore, the furnace must have a post-combustion chamber above the grate. The
final burnout of the flue gas takes place in that chamber, and the required amount of
secondary combustion air is added in a way that ensures maximum turbulence. To
ensure complete combustion, a certain amount of excess combustion air is needed.
The excess air, representing a loss of energy and hence a loss of thermal efficiency,
can be reduced by using a water-cooled grate and/or by FGR. A fraction (approxi-
mately 10–20 % by volume [1]) of the (usually cleaned) flue gases is recirculated,
normally after pre-dusting, to replace a fraction of secondary air feeds in the com-
bustion chamber. This technique is reported to reduce heat losses with the flue gas
and to increase the process energy efficiency by around 0.75–2 %. Additional ben-
efits of primary NOX reduction are also reported [1].

3.1.3 The Energy Recovery Section

Boilers for WTE power plants are typically water tube boilers, and most often they
have four passes: three vertical radiation passes and a convective pass. The first
radiation pass is integrated in the furnace (integrated boiler) as the post-combustion
chamber. The convection pass, in which the evaporators, superheaters, and econo-
mizers are located, may be vertical or horizontal.
Thus, as schematically shown in Fig. 3.6, it is possible to differentiate three dif-
ferent heat surface areas:
• Economizer (or feedwater-preheating heat exchanger (HE); ECO in Fig. 3.6): In
this area, the boiler feedwater is heated by flue gases to a temperature close to
the boiling point.
• Evaporator (EVA in Fig. 3.6): In this area, the water coming out from the econo-
mizer is heated until it reaches the saturated steam temperature (designed as a
bundled heating surface, envelopment wall of the incineration chamber).
• Superheater (SH in Fig. 3.6): In this area, the saturated steam coming out from
the evaporator is superheated to the maximum temperature. Spray coolers and
surface coolers are used in circulation boilers in order to maintain the exact re-
quired steam temperature. Their function is to balance the fluctuations of the
steam temperature, which are the consequence of load fluctuations, changes in
waste quality, surplus air, as well as contamination of the heat surfaces (clean or
dirty surface).
3.1 Basics of a WTE Power Plant 27

Fig. 3.6   Typical heat surface


areas in a steam generator.
SH superheating, EVA evapo-
ration, ECO economizing

When designing a boiler for waste firing, the most important aspect to take into
account is the special risk of corrosion. Corrosion problems, as described in de-
tail in the following section, limit the steam parameters to a maximum of around
450 –500 °C and 40–60 bar, respectively, for steam-superheated temperature and
pressure. In general, at the entrance of the convection section (located after the third
pass), flue gas has a temperature lower than 700–650 °C.
Good quality of the boiler and the adoption of a specific protective layer must be
expected when increased steam parameters are used. Thus, a compromise between
high steam cycle parameters and increased corrosion problems is required in order
not to penalize waste conversion efficiency into energy and, at the same time, not to
increase capital and maintenance plant costs.
Different boiler concepts can be used in waste incineration plants [1]:
• Horizontal boilers
• Vertical boilers
• Combination of vertical and horizontal boilers
Horizontal arrangement boilers are characterized by the fact that the flue gas in the
convective heating surfaces travels horizontally. By contrast, in vertical arrange-
ment boilers, the flue gas in the convective heating surfaces travels in the vertical
direction. The main advantages of the horizontal arrangement are: (i) easier clean-
ing of the heating surface, (ii) reduced risk of tube bundle blocking due to particles
28 3 Waste-to-Energy

Fig. 3.7   Typical corrosion diagram

and dusts,1 (iii) the possibility to place outside flue gas passing supports for the
heating surface. But the major drawback of the horizontal configuration is that it
needs huge space. Compactness is, however, the main advantage of vertical con-
figurations, thus optimizing performance per metric tons of steel in the boiler.

3.1.3.1 Corrosion Protection

Modern WTE plants have significantly improved and are far superior compared
with the older polluting incinerators. However, the variability in MSW LHV and
its relatively high content of chlorine contribute to a highly corrosive atmosphere
that shortens the life of the heat exchanger tubes [3]. The incineration chamber, the
waterwalls of the first passes, and the superheaters are boiler components most af-
fected by corrosion. A typical corrosion diagram, derived from practical experience,
is shown in Fig. 3.7 [4]. The diagram, where horizontal and vertical axis report flue
gas and waterwall surface temperature, respectively, highlights two different zones
characterized by different corrosion rates. As highlighted in the figure, flue gas and
waterwall tube temperature are the fundamental variables to determine corrosion
risk.

1 
The high proportion of ash in flue gas causes a risk of a correspondingly high contamination
of the heat transfer surfaces. This leads to a decline in heat transfer and therefore a performance
loss. Thus, heat transfer surface cleaning plays an important role. This cleaning can be accom-
plished manually or automatically with lances (compressed air or water jet), with agitators, with
soot blowers using steam, with a hail of pellets (sometimes shot cleaning), with sound and shock
waves, or with tank cleaning devices.
3.1 Basics of a WTE Power Plant 29

The most important factors affecting corrosion probability inside a WTE boiler
are [3]:
• Metal surface temperature. High temperature of the metal surface due to high
radiation fluxes and/or inadequate heat transfer rate to the steam flow inside
the tube results in the melting of deposits and acceleration of the corrosion
rate. In general, the metal temperatures of waterwall and superheater tubes are
maintained at temperatures below 300  and 450 °C, respectively. However, as
mentioned earlier, operation at higher superheater temperatures increases the
thermal efficiency of the steam turbine.
• Gas temperature. The temperature of the combustion gases can affect the deposi-
tion rates and also the composition of the deposit, and thus accelerate corrosion.
The temperature gradient between gas temperature and metal surface tempera-
ture is a driving force for the condensation of vaporized species, such as metal
chlorides, on the cooled surface.
• Temperature fluctuation. The heterogeneous MSW composition and the cor-
responding fluctuation in LHV with time results in significant fluctuations of
the gas temperature within the combustion chamber. Experimental studies have
confirmed that the corrosion rate increases several times because of wide tem-
perature fluctuation.
• Characteristics of molten salt deposits. The presence of chlorides, sulphides,
alkaline, and heavy metal components in deposits affects both the chemical and
physical properties of deposits, such as gas permeability. Deposit composition
is affected by waste feed composition and the gas–metal temperature gradient.
Corrosion rate also increases with an increase in thickness of deposits.
A detailed description of corrosion mechanisms inside a WTE facility lies beyond
the scope of the present work (additional information can be found in [3–9]).
Metal tube corrosion is the major operating problem as it results in downtime
and periodic shutdowns in WTE plants and accounts for a significant fraction of
the total operating costs of WTE plants. Moreover, as introduced in Sect. 3.1.3,
the increase in steam cycle parameters to achieve high plant efficiency causes heat
transfer surfaces to be exposed to a severe corrosion atmosphere.
Besides corrosion problems, another negative aspect of a WTE plant is erosion,
that is, the abrasion of surface material through vertical wear-and-tear, which is
primarily caused by the ash particles present in flue gas. Erosion appears mostly in
the area of gas redirection. Tube wear is caused by a combination of corrosion and
abrasion.
Countermeasures, only helping to reduce corrosion damage to an acceptable
level, can be divided into primary and secondary measures.
Primary measures seek to eliminate corrosion by influencing the process condi-
tions in the boiler [7], mainly minimizing gas temperature fluctuations and enhanc-
ing mixing of gas through gas recirculation and optimal boiler system design.
Secondary measures involve the use of protective layers to extend the lifetime of
the boiler tubes. To prevent corrosive attacks, heating surfaces in the radiant section
and superheaters are, typically, protected by resistant refractory materials and/or
30 3 Waste-to-Energy

140

Number of operative plants [-]


120

100

80

60

40

20

Portugal

Switzerland
Czech Republic

United States
Netherlands
Belgium

United Kingdom
Ireland
Italy
Finland

Denmark
Denmark
Austria

France

Hungary

Spain
Norway

Sweden
Fig. 3.8   Number of operative incinerator plants in EU-17 and the USA in 2011

welded high-alloy materials. Cladding consists of overlaying a layer of Inconel 625


on tubes which provides excellent corrosion resistance [8, 9]. The most important
drawback in the use of Inconel 625 is its extremely high cost, which is only partially
compensated by absent corrosion costs.2
A Waste-to-Energy Research and Technology Council (WTERT) study [10]
shows that the nonscheduled downtime due to corrosion typically ranges from 0
to 20 days per year and that the yearly maintenance cost per boiler unit due to cor-
rosion ranges from US$ 18,000 to 1,200,000 or from US$ 0.23 to 8.17 per ton of
MSW combusted.

3.2 WTE Plant Distribution in the European Scenario

The International Solid Waste Association (ISWA) report [11] examined the situa-
tion in 18 countries (17 European countries and the USA) in 2011. Figure 3.8 shows
the results of the study: The total number of operative WTE power plants in the
European Union (EU)-17 was 455, while there were 86 operative WTE plants in
the USA. The waste treatment capacity of each installation varies greatly across Eu-
rope: WTE plant capacity for each country is shown in Fig. 3.9. The average EU-17

The application of Inconel 625 on superheater tubes is more complicated because the perfor-
2 

mance of the cladding depends on the metal temperature reached during operation [8, 9].
3.2 WTE Plant Distribution in the European Scenario 31

80

Average plant capacity [t/h] 70

60

50

40

30

20

10

United States
France

Italy
Austria

Portugal
Denmark

Denmark

Spain

EU-17
Ireland

Norway
Belgium

Finland

Hungary

Sweden
Netherlands

Switzerland
United Kingdom
Czech Republic

Fig. 3.9   Average incinerator plant capacity (t/h) in EU-17 and the USA in 2011

capacity value is 32 t/h; although in some countries very large plants are common,
in others small plants are characteristic.
A recent study by the Confederation of European Waste-to-Energy Plants
(CEWEP) [12] forecasts a significant increasing trend for incineration with en-
ergy recovery in Europe: The MSW treatment capacity of European plants will
grow from 64 million metric tons in 2006 to over 100 million metric tons in 2020
(+ 59 %), with a marked increase in the recovery of energy, both in the form of heat
and electricity.

3.2.1 WTE Plant Efficiency in a Representative National


Scenario

Focusing on Italy, as representative case, latest recorded data (2010) [13] show 50
operating incinerators with a total capacity equal to 7.12 Mt/year, corresponding to
2925 MWt thermal capacity and to 782 MW electric power output. Energy recovery
is obtained in almost all plants (49 out of 513): All plants produce electricity, while
only 11 plants are combined heat and power (CHP) systems. Figure 3.10 shows op-
erative incinerators in Italy from 1995 to 2010, and Fig. 3.11 shows the distribution
of WTE plant nominal gross electric size and thermal capacity. It is interesting to

3 
51 incinerators are installed out of which 50 are operative plants.
32 3 Waste-to-Energy

With energy recovery


Total (with and without energy recovery)
60 No energy recovery

50
Number of operative plants [-]

40

30

20

10

0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

Fig. 3.10   Operative incinerators in Italy with and without energy recovery,1995–2010



 
*URVVHOHFWULFSRZHU >0:@








η 





   
7KHUPDOFDSDFLW\>0: @
W

Fig. 3.11   WTE electric power output and input thermal capacity in the Italian scenario

note that 34 out of 50 plants have a capacity equal to or below 50 MWt (mainly old
generation), while only few new plants exceed 100 MWt. Figure 3.11 also shows
different values of gross electric efficiency, depending on the technology level, size,
and age of the plant. In most cases, it is between 18 and 25 %, while only in a few
3.3 EU Regulation Framework Oriented to WTE Efficiency 33


(OHFWULFDOHQHUJ\

7KHUPDOHQHUJ\

*HQHUDWHGHQHUJ\>*:K@














  
  




  
  
 
 

Fig. 3.12   Electrical and thermal energy generated in WTE power plants in Italy, 2004–2013

cases the gross efficiency is above 30 %. A detailed description of the state-of-the-
art WTE steam cycle arrangement and parameters is carried out in Chap. 4.
Figure 3.12 shows the total amount of electrical and thermal energy generated by
WTE power plants in Italy between 2004 and 2013 [13]. A growing trend of gener-
ated energy is evident, with an increase of more than 1888 GWh and 1913 GWht for
electrical and thermal energy, respectively. A further significant increase in capacity
is expected in the Italian scenario by the end of 2014.

3.3 EU Regulation Framework Oriented to WTE


Efficiency

The 2008/98/EC WFD promotes production of energy from MSW, introducing an


energy efficiency criterion (the so-called R1 formula [14, 15]) for WTE plants to
obtain the “recovery” or “waste disposal” status under certain conditions. The aim
of the WFD is to drive WTE facilities to maximize energy recovery and utilization
of waste heat in order to substitute energy produced with conventional fossil fuel
power plants.
The R1 ratio is a modified efficiency index, including in its expression all energy
input/output of an WTE plant, that is, electricity and thermal production, losses, input
waste, imported electricity, and additional fuels. The R1 formula [15] is defined as:

Eprod − ( Ef + I imp )
R1 = , (3.1)
f B ⋅ ( Ef + Ew )
34 3 Waste-to-Energy

where:
• Eprod is the annual energy produced as heat or electricity. It is calculated with
energy in the form of electricity being multiplied by 2.6 and heat produced for
commercial use multiplied by 1.1. So, Eprod can be regarded as an “equivalent
primary energy”: the energy input, with conventional fossil fuels, to generate the
same amount of electricity and heat as the WTE power plants. It is important to
understand that the equivalence values are not exact coefficients or conversion
factors. They provide an estimation of the energy required to produce the same
amount of energy output externally.
• Ef is the annual energy input to the system from fuels contributing to the produc-
tion of steam (e.g., natural gas used, if necessary, as an additional fuel to increase
post-combustion temperature).
• Ew is the annual energy contained in the waste, calculated using the net calorific
value (LHV) of the waste.
• Iimp is the annual energy imported, excluding Ef and Ew.
• fB is a factor accounting for energy losses due to bottom ash and radiation
(­assumed equal to 0.97).
A minimum value of R1 is required for WTE facilities to achieve the status
(R1 = 0.60 for existing plants, R1 = 0.65 for new plants). It poses no problem for the
few large, new and most advanced WTE plants to achieve the “energy recovery”
status; by contrast, for the diffused, small and old facilities, only the less perform-
ing “waste disposal” status can be achieved. The reason is linked to the limited
thermodynamic conversion efficiency of typical WTE power plants. As described in
Sect. 3.1.3 [16, 17], low steam cycle parameters and small capacity of WTE facili-
ties negatively affect the efficiency of MSW energy conversion.
Figures 3.13 and 3.14 show the maximum, minimum, and average R1 values for
existing European WTE plants [12] and the thresholds for energy recovery status.
In particular, Fig. 3.13 shows the effects of the plant capacity on the achieved R1
values; Fig. 3.14, however, shows a comparison of the achieved R1 values by WTE
plants producing only electricity, or providing only heat, or by CHP plants. On
average, R1 is higher for WTE plants with highest capacity (> 250 Mt/year) which
work in CHP mode.
The described scenario justifies a design trend towards large WTE plants with
a CHP arrangement in order to achieve the energy recovery status. Moreover, even
if R1 threshold values can be satisfied, this does not imply that high conversion ef-
ficiency is achieved in conventional WTE plants. Indeed, the current technological
limits of conventional WTE plants, based on a superheated steam thermodynamic
cycle, still penalize the electric energy efficiency of existing WTE plants.
Improved electric energy efficiency of MSW-fired steam units can be achieved
only by tolerating higher corrosive rates for increased superheating temperatures,
and consequently higher maintenance costs [16].
Thus, especially for small, inefficient but diffused WTE plants, new strategies
to maximize the MSW energy conversion are currently under investigation. This
calls for novel approaches and possibilities. In particular, the idea of an integrated
3.3 EU Regulation Framework Oriented to WTE Efficiency 35



5>@
0D[
0LQ
 $YHUDJH
7KUHVKROGVIRU
HQHUJ\UHFRYHU\
 VWDWXV





1HZ
([LVWLQJ 






&DSDFLW\ &DSDFLW\!W\ &DSDFLW\
W\ DQG !W\
W\

Fig. 3.13   Maximum, minimum, and average R1 values for existing WTE power plants as func-
tions of plant size


5>@

0D[
 0LQ 7KUHVKROGVIRU
$YHUDJH HQHUJ\UHFRYHU\
VWDWXV






1HZ
([LVWLQJ 






(OHFWULFLW\ +HDW &+3

Fig. 3.14   Maximum, minimum, and average R1 values for existing WTE power plants generating
only electricity, only heat, or working in the combined heat and power (CHP) mode
36 3 Waste-to-Energy

configuration made up of a WTE and a gas turbine (GT) originates from the desire
to eliminate or, at least mitigate, limitations of the thermodynamic cycle affecting
the WTE conversion process.

References

  1. European Commission (2006) Integrated Pollution Prevention and Control Reference Docu-
ment on the Best Available Techniques for Waste Incineration
  2. Babcock & Wilcox volund brochure, 21’ Century Advanced Concept for Waste-Fired Power
Plants. http://www.volund.dk. Accessed March 2012
  3. Lee SH, Themells NJ, Castaldi MJ (2006) High temperature corrosion phenomena in waste to
energy boilers. J Therm Spray Technol 16(1):104–110. doi:10.1007/511666-006-9005-4
  4. Carminati A (2003) Caldaie alimentate con RSU Accorgimenti per ridurre la corrosione, La
Termotecnica – Novembre 2003. (In Italian)
  5. CESI—Ricerca di Sistema (2005) Indagine problematiche di affidabilità degli impianti di
termovalorizzazione, A5049283
  6. Rademarkers P, Hesseling W, Wetering J (2002) Review on corrosion in waste incinerators,
and possible effect of bromine. TNO Industrial Technology
  7. Albina DO (2005) Theory and experience on corrosion of waterwall and superheater tubes of
waste-to-energy facilities. Master Thesis, Columbia University
  8. Kawahara Y (2002) High temperature corrosion mechanisms and effect of alloying elements
for materials used in waste incineration environment. Corros Sci 44(2):223
  9. Wilson A, Forsberg U, Noble J (1997) Experience of composite tubes in municipal waste
incinerators. Corrosion 97:153
10. Zwahr H (2003) Ways to improve the efficiency of waste to energy plants for the production
of electricity, heat and reusable materials. Proceedings of the 11th North America Waste to
Energy Conference, Florida
11. International Solid Waste Association ISWA (2012) Waste to energy state of the art report-
statistics 6th Edition. http://www.iswa.org/fileadmin/galleries/Publications/ISWA_Reports/
ISWA_Report_2012.pdf. Accessed March 2014
12. Reimann DO, CEWEP Energy Report II (status 2004–2007) Results of specific data for en-
ergy R1 plant efficiency factor and Net calorific value of 231 European Waste to Energy
plants
13. ISPRA Istituto Superiore per la Protezione e la Ricerca Ambientale Edizione (2014) Rapporto
Rifiuti Urbani, ISBN: 978-88-448-0665-1. http://www.isprambiente.gov.it/files/pubblicazio-
ni/rapporti/RapportoRifiutiUrbani2014_web.pdf. Accessed 19 Sept 2014
14. Bianchi M, Branchini L (2012) Advanced waste-to-energy power plant: integration with gas
turbine. International Conference on Applied Energy ICAE2012, July 5–8, 2012, Suzhou,
China, Paper ID: ICAE2012-A10330
15. Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008
on waste. Official Journal of the European Union OJ L 312/3,22.11.2008
16. Bianchi M, Branchini L, Melino F, Peretto A (2010) Cycle upgrade for Waste-to-energy pow-
er plants, ASME-ATI-UIT 2010 Conference on Thermal and Environmental Issues in Energy
Systems 16–19 May, 2010, Sorrento, Italy
17. Bianchi M, Branchini L, De Pascale A, Falchetti M, Fiore P (2014) Advanced waste-to-ener-
gy steam cycles. Energy Procedia 45:1205–1214. doi:10.1016/j.egypro.2014.01.126
Part II
WTE Thermodynamic Analysis
Chapter 4
Waste-to-Energy Steam Cycle

This chapter focuses on state-of-the-art waste-to-energy (WTE) steam cycle arrange-


ment and possibilities to improve thermodynamic efficiency, upgrading both steam
cycle parameters and cycle layout.
Based on a collection of WTE facilities data, a “state-of-the-art” WTE power
plant is identified: Typical values of steam cycle pressures and temperatures are dis-
cussed with reference to the corrosion problem and characteristic plant size. Start-
ing from the “state-of-the-art” WTE power plant, a sensitivity analysis is carried
out to analyze the effects of the most important steam cycle parameters and con-
figuration, improving the thermodynamic efficiency of the WTE facility. An initial
assessment of the importance of the main steam cycle parameters must be outlined
as a preliminary study in order to understand the possibilities and the benefits of a
thermodynamic cycle upgrade for a WTE power plant.

4.1 Steam Cycle State-of-the-Art Parameters and Layout

The amount of energy recovered from waste combustion can vary significantly with
the characteristics of municipal solid waste (MSW) fed into the boiler (composition;
mass flow rate; and lower heating value, LHV), the combustion technology, the
configuration and features of the recovery boiler (adiabatic or integrated), and the
characteristics of the thermodynamic cycle.
For boiler normal operative conditions (excluding start-up and shutdown), as
explained in Chap. 3, the kinetics of the corrosion process is influenced by the wall
heat exchanger and flue gas temperatures. Thus, the characteristics of the produced
steam, pressure, and temperature play a fundamental role in corrosion.
All of these factors expose the boiler structure to a variety of critical events; it is
therefore necessary to protect the heat exchanger section with refractory materials,
reducing the heat flux and limiting the flue gas temperatures, and to plan a specific
heat exchanger section organization.

© Springer International Publishing Switzerland 2015 39


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_4
40 4  Waste-to-Energy Steam Cycle

The abovementioned problems concerning the WTE conversion limit the maxi-
mum superheated steam temperature. Aggressive compound condensation is anoth-
er limiting factor. In order to avoid it, high outlet temperatures must be carried out.
These limits reduce the thermal energy input with waste actually convertible into
useful energy, constraining the steam cycle efficiency.
Corrosion problems bind the inlet turbine temperature to be between 370 and
450 °C. Low superheating temperatures also imply moderate evaporation pressure
in order to limit the liquid fraction at the steam turbine (ST) outlet, typically be-
tween 40 and 50 bar. Besides, higher values of the evaporation pressure (more than
50 bar) make a protection of the combustion chamber with a noble metal layer to
avoid corrosion problems necessary.
It has to be pointed out that the possibility to increase the evaporation pressure
and steam-superheated temperature is mostly limited by economical and plant com-
plexity aspects.
Typical flue gas outlet temperatures range from 180 to 250 °C, significantly
higher than those typical for fossil fuel power plants. Values equal to or below
160 °C can only be achieved in advanced plants.
The typical size of an incineration plant is smaller than that of a conventional
fossil fuel plant. The electric power generated from the combustion of MSW (usu-
ally between a few megawatts up to a maximum of 70–80 MW) is one or two orders
less than that of a conventional power plant. Moreover, the layout of the plant must
be as simple as possible to reduce the investment costs.
Surface condenser with air-cooling towers or air cooler condenser is generally
used for steam condensation, implying high condensation pressure (between 0.2
and 0.1 bar), while 0.05 bar is the value for conventional power plants using surface
condensers with open cycles.
The net electric efficiency is also negatively influenced by the high air excess
necessary for the MSW combustion and, as a consequence, by the high volume
of combustion products. Thus, considerable amount of exhaust flue gases are dis-
charged, raising the auxiliary power consumptions and the discharged heat.
Due to these reasons, the conversion efficiency of the steam cycle for a WTE
power plant hardly exceeds 30 %, while the net average value of electric efficiency
is around 25 %.
Figure 4.1 shows the distribution of Italian WTE facilities in terms of electric
power output as a function of waste thermal input and steam cycle parameters
(steam maximum pressure and temperature).
As stated by this study, evaporative pressure values range between 10 and 75 bar;
55 % of the cases (27 plants) are within a range of 20 and 40 bar, with a peak around
40 bar. The operating temperature is typically 400 °C. Brescia WTE power plants
achieved the best performance with 75 bar and 450 °C of evaporation pressure and
steam maximum temperature, respectively [1–3].
The average values of plant capacity and steam parameters, highlighted in
Fig.  4.1, suggest that a WTE representative “average” facility can be identified.
This representative “average” WTE is characterized by medium size (average input
capacity equal to 55 MWt and gross electric power output of 15.7 MW), moderate
4.1 Steam Cycle State-of-the-Art Parameters and Layout 41

 ƌĞƐĐŝĂ

ĐĞƌƌĂ

*URVVHOHFWULFSRZHU>0:@





DŝůĂŶŽ

WĂƌŽŶĂ

 η сϭϱй
Ğ


DYHUDJHYDOXHV
0: 0:
W

   
7KHUPDOFDSDFLW\>0: @
a W



6XSHUKHDWHGVWHDPWHPSHUDWXUH>ƒ&@





 DYHUDJHYDOXHV
EDUƒ&






          

b 6XSHUKHDWHGVWHDPSUHVVXUH>EDU@

Fig. 4.1   Italian WTE gross electric power output and input thermal capacity (a) and steam cycle
parameters (b)

steam cycle parameters (ST inlet pressure of 43 bar and temperature of 390 °C), and
typically steam extraction from the turbine and regeneration.
Opposed to state-of-the-art “average” midsize WTE (Fig. 4.1), “high-performing”
WTE power plants can be identified, characterized by high steam cycle parameters
42 4  Waste-to-Energy Steam Cycle

Fig. 4.2   Schematic layout of a midsize reference WTE power plant. HE heat exchanger, ST steam
turbine, DEA deaerator, EVA evaporator, SH superheater, ECO economizer

(ST inlet pressure higher than 80 bar and temperature higher than 450 °C) and high
plant capacity (over 100 MWt). This high-performing steam cycle arrangement is
typical of modern and high-size WTE power plants where significant investment
costs have been accepted.
A superheated steam cycle without reheating and with steam extraction is the
typical layout arrangement of a midsize WTE power plant. A schematic layout of
state-of-the-art cycle arrangement is presented in Fig. 4.2. The steam generator is a
natural circulation type, integrated with the combustion chamber. It comprises sev-
eral radiation channels with vertical flue gas flow and a convection section contain-
ing the superheaters (SH), evaporators (EVA), and economizers (ECOs). As already
detailed in Chap. 3, the configuration of the convective section can be either verti-
cal, horizontal, or combining both types.
Flue gas produced by combustion of waste passes through the steam generator,
where superheated steam required to be fed to the turbogenerator for the generation
of electricity is produced. The SHs are tagged SH1, SH2, and SH3, following the
steam flow. The layout of SH banks, following the flue gas flow, is usually designed
to limit corrosion from chlorine at high temperature; thus, high-temperature flue gas
comes in contact first with SH1, disposed in concurrent steam flow. The SH1 parallel
flow configuration is necessary to guarantee the lowest metal temperature because
here the coolest steam conditions are exposed to the hottest gas temperature. Never-
theless, to obtain satisfactory refuse boiler SH performance, the use of Inconel tube
material in the highest tube metal temperature sections is required.
4.1 Steam Cycle State-of-the-Art Parameters and Layout 43

While the second and third SH encountered by the flue gas are designed with
countercurrent steam flow, carbon steel is commonly used in SH sections with low-
er metal temperatures [4].
The upper furnace, where EVA sections are placed, must be properly sized in
order to provide adequate heat transfer surface to reduce the flue gas temperature
entering the SH to an acceptable level (see also Sect. 3.1.3).
Superheated steam, generated inside the convective section of the WTE boiler
in SH bank, is sent to the ST, connected to a synchronous generator for the produc-
tion of electricity ECO. ECOs are located at the end of the WTE boiler convective
section, while the EVA is split in different heat exchanger banks, located in both the
radiating and the convective sections (EVA1 and EVA2) of the boiler.
ECOs can be either vertical longflow or horizontal crossflow.
An ST bleed is used to feed the deaerator (DEA) and to complete the primary
air preheating before combustion. Air preheating is necessary for two reasons: (i) to
help refuse-drying and ignition processes and (ii) to increase the thermal efficiency
increasing the reactant’s enthalpy. Primary (or under-grate) combustion air is typi-
cally preheated up to 120–150 °C. To achieve this temperature level, in addition to
grate cooling exploitation an ST bleed feeding a heat exchanger is necessary. Sec-
ondary (or over-fire) air is generally preheated taking advantage just of grate cool-
ing up to about 50°C.
Under-grate air systems are typically designed for 70 % of the total air with an
expected normal operation of 60 % [4]. Because waste contains a high percent-
age of volatiles, a significant portion of the total combustion air should enter the
furnace secondarily, or as over-fire air (about 40 % in normal operation), through
the furnace wall. The basic function of secondary air is to complete combustion,
providing both the necessary oxygen and the adequate turbulence necessary to mix
combustion gases.
Water cooling with a closed loop is used for the combustion grate; after passing
through the grate, hot water is used to preheat the primary and secondary inlet air
streams. Thus, removed heat through grate cooling is exploited for combustion air
preheating.
A fraction of exhaust gas (typically 15 % by volume) is recirculated inside the
WTE boiler in the post-combustion section in order to reduce nitrogen oxide (NOx)
emissions.
Minimum oxygen content, %O2, in exhaust gases from WTE boilers of 6 % was
required by earlier legislation but was removed from the most recent European
Commission (EC) directives on incineration [5]. Despite that, the typical operative
value of oxygen content in dry exhaust gases ranges between 6 and 9 %. Lower O2
content may increase carbon monoxide emission and may increase corrosion risk
[5].
Minimum exhaust gas outlet temperature is mostly determined in order to avoid
acid dew point corrosion. Minimum possible dry gas temperature is 130–140 °C us-
ing normal boiler construction material. When temperature falls below the acid dew
point, wet chemical corrosions appear on cold surfaces. This damage can be avoided
by raising the temperature or by selecting an appropriate material. Typical gas outlet
temperature ranges between 150 and 190 °C for modern WTE facilities [5].
44 4  Waste-to-Energy Steam Cycle

A second ST bleed (at lower pressure), if present, can be used in the feedwater
heat exchanger (after extraction pump) to preheat water coming out from the con-
denser before entering the DEA. The presence of an additional ST bleed for feedwa-
ter preheating can be found in higher-capacity WTE, where additional complexity
and associated costs can be more easily accepted.

4.2 Steam Cycle Upgrade: Effects on Cycle Efficiency

An initial assessment of the importance of the main steam cycle parameters must
be outlined as a preliminary study in order to understand the possibilities and the
benefits of a thermodynamic cycle upgrade for a WTE power plant. The selection of
steam cycle parameters has a considerable impact on energy efficiency providing a
great opportunity for an increased use of energy out of waste. Despite that, achiev-
able conversion efficiency gain must be always weighed with increased investment
cost that can occur.
A sensitivity analysis has been carried out to analyze the effect of the most im-
portant steam cycle parameters and configuration to improve the performance of a
WTE plant. The parameters considered in this study are:
• Condenser pressure, pk
• Evaporative pressure, pev
• Steam-superheated temperature, TSH
• ST isentropic efficiency, ηit
• Exhaust gases temperature, TO, WTE
• Oxygen content in dry exhaust gases, %O2
• Primary and secondary air temperature, Tair
• Cycle configuration in terms of regeneration
The influence of each key parameter has been studied considering a fixed capacity
of the WTE power plant: MSW composition, mass flow rate, and LHV of waste fed
into the integrated boiler are supposed to be constant.
If the input of thermal power with waste is constant, percentage increments in
power output are achievable due to an upgrade of steam cycle parameters or the
configuration and agree with increments in the electric efficiency value.
Thermodynamic and parametric analysis was carried out by the use of Gate Cy-
cleTM [6], a commercial software from General Electric for plant design and simu-
lation. The software, solving mass and energy balance using a lumped model ap-
proach, allows to evaluate inlet and outlet conditions of each system’s components
and to predict performance of the plant.
The influence of each key parameter has been studied individually or in relation
to others.
Figure  4.3 shows the influence of evaporative pressure on electric efficiency
for two different values of superheated steam temperature. Percentage increases
achieved on efficiency are compared with a fixed increase in evaporative pressure
4.2 Steam Cycle Upgrade: Effects on Cycle Efficiency 45

4.5
∆ η [%]

T SH = 400 °C
e

4.0
T = 500 °C
SH

3.5

3.0

2.5

2.0

1.5

1.0

0.5

p [bar]
p [bar]

Fig. 4.3   Percentage increments on electric efficiency as a function of evaporative pressure for
different values of steam-superheated temperature

value equal to 10 bar, that is, at 400 °C, passing from (pev1) 30 bar to (pev2) 40 bar;
the increase in efficiency is equal to 4.2 %. This figure clearly highlights that higher
benefits can be found for lower values of pev, passing from 4 % (from 30 to 40 bar)
to about 1 % (from 90 to 100 bar). High values of TSH slightly narrow the realized
increase (Fig. 4.3).
Figure 4.4 shows the effect of condenser pressure on electric efficiency for differ-
ent values of evaporative pressure. Percentage increments on efficiency are shown
for a fixed decrease (equal to 25 mbar) in condenser pressure value. The possibility
to reduce the pk has considerable effects on the steam cycle efficiency. It should be
noted that lower pk values (i.e., 0.05 bar) can be achieved only by the use of surface
condenser with an open cycle; this possibility is strictly limited by the presence of
a natural water source close to the WTE power plant. Figure 4.4 clearly highlights
that higher benefits can be found for lower values of pk passing from more than 3 %
(from 50 to 75 mbar) to about 1 % (from 1.75 to 2 mbar). Percentage gains increase
if the evaporative pressure is lower (Fig. 4.4).
The increase in superheated steam temperature (TSH) has an obviously positive
effect on the efficiency of the steam cycle. The electric efficiency trend is shown in
Fig. 4.5 for a fixed steam-superheated temperature increment, equal to 10 °C. The
slope of the curve is slightly influenced by the actual value of TSH (as evaporative
pressure is equal to 100 bar); on the contrary, for a lower pev value, the percentage
increment on efficiency increases with the increase of the actual value of TSH.
46 4  Waste-to-Energy Steam Cycle

S
η

S >EDU@
S >EDU@

Fig. 4.4   Percentage increments on electric efficiency as a function of condenser pressure for dif-
ferent values of evaporative pressure

0.5
∆ η [%]

pev = 50 bar
e

p = 100 bar
dev

0.4

0.3

0.2

0.1

0.0 3
6 0 370 8
3 0 390 0
4 0 410 420 430 4 0 450 6
4 0 470 480 490 5 0
0

T
T

Fig. 4.5   Percentage increments on electric efficiency as a function of steam-superheated tempera-


ture for different values of evaporative pressure
4.2 Steam Cycle Upgrade: Effects on Cycle Efficiency 47

P
η

η
η

Fig. 4.6   Percentage increments on electric efficiency as a function of ST isentropic efficiency for
different values of evaporative pressure

High values of ST isentropic efficiency (ηit) are also related to the size of the ST
and, as a consequence, to the size of the plant. Percentage increments on electric ef-
ficiency are shown in Fig. 4.6 for a fixed ηit increment set equal to 0.01. Figure 4.6
points out that increments are not influenced by the evaporative pressure value.
In Fig. 4.7, percentage increments gained for a fixed decrease in TO, WTE equal
to 10 °C are shown. Increments are almost constant, not dependent on evaporative
pressure and steam-superheated temperature. It should be noted that, in the con-
sidered range, the lowest value (160 °C) is typical of advanced WTE power plants.
Even a reduction of oxygen content (vol%O2) in the dry exhaust gases at WTE
boiler outlet causes a reduction in the amount of heat lost to the stack. Benefits on
efficiency are shown in Fig. 4.8 for two different values of evaporative pressure.
Figure 4.8 emphasizes that the percentage gain for a fixed percentage point of de-
crease is a function of the actual value of %O2 but is not dependent on the pev value.
The influence of secondary air temperature on electric efficiency has been also
analyzed. A percentage increase in electric efficiency of more than 1.6 % has been
found by preheating secondary combustion air from ambient temperature to 150 °C.
By combining both secondary air and feedwater preheating, a gain of about
5.5 % can be achieved on WTE electric efficiency.
The sensitivity analysis results for each analyzed steam cycle parameters are
summarized in Table 4.1. Of course, cycle improvements can be also combined
together to obtain the corresponding electric efficiency increase.
48 4  Waste-to-Energy Steam Cycle

P T
η

P T

T
T

Fig. 4.7   Percentage increments on electric efficiency increments as a function of exhaust gas
outlet temperature for different values of evaporative pressure

p
η

Fig. 4.8   Percentage increments on electric efficiency increments as a function of volume oxygen
content in exhaust dry gases for different values of evaporative pressure
4.3 New Designs for High-Efficiency WTE Plant 49

Table 4.1   Steam cycle improvements: effect on WTE efficiency


Steam cycle operative Assumed variation (increase Percentage increment gained
parameter or decrease) on WTE electric efficiency
(%)
Evaporative pressure, pev + 10  bar 1–4
Steam-superheated tempera- + 10 °C 0.35
ture, TSH
Condenser pressure, pk − 0.025 bar 1–3
ST isoentropic efficiency, ηit + 0.01 1.25
Exhaust gases outlet tempera- − 10 °C 0.95
ture, TO, WTE
Oxygen content in dry − 1 % 0.8–1.3
exhaust gases, %O2
WTE waste-to-energy, ST steam turbine

It has to be pointed out that some strategies, such as increasing evaporative pres-
sure and steam-superheated temperature or decreasing O2 content and exhaust gas
temperature, require specific solutions to protect the integrity of the waste-fired
boiler. Increased corrosion risks will follow both plant complexity and investment
cost raises.
Along with upgrading efficiency of key steam cycle parameters, increase can
also be achieved with cycle improvements. In fact, the steam cycle layout has a
significant impact on WTE plant’s performance. To better analyze the influence
of a regenerative steam cycle on electric efficiency, the influence of regeneration
for feedwater preheating has been investigated; results are shown in Fig. 4.9. An
increase equal to about 5 % on efficiency can be obtained with preheating the water
before entering the DEA. With an additional ST bleed, preheating the water before
the ECO inlet will further raise the efficiency to over 22 %. As shown in Fig. 4.9,
increments are quite independent of evaporative pressure values.

4.3 New Designs for High-Efficiency WTE Plant

Concerning high-efficiency WTE plants, some interesting cases must be discussed.


New WTE concepts are still under investigation aiming at increasing plant effi-
ciency. Proposed and already constructed plant featuring new design concepts are
located in:
• Amsterdam—the Netherlands: high-efficiency WTE plant
• Reno—Nord Denmark: SteamBoost concept
• Germany—Karlsruhe Institute of Technology research: bypass of fuel gas con-
cept
50 4  Waste-to-Energy Steam Cycle

Fig. 4.9   Percentage increments on electric efficiency for regenerative cycles for different evapo-
rative pressure values

The first highly efficient WTE plant, installed in Amsterdam, uses higher steam
temperature/pressure together with an intermediate superheating using steam from
the boiler drum and achieves a net electrical efficiency of 30 % [7]. The concept
applied in Amsterdam involves, basically, the reheating of a part of the steam out of
the high-pressure turbine using a steam-steam shell-and-tube heat exchanger to ob-
tain a net electrical efficiency of 30 % and availability of over 90 % by using Inconel
625 as boiler cladding. Thus, a more classical approach is applied to WTE in order
to increase efficiency as widely diffused in classical fossil-fueled steam cycles.
In Fig. 4.10, a schematic of the WTE concept with steam intermediate reheating
is presented. As illustrated in Fig. 4.10, an intermediate reheater (RH), as additional
component, is present where the saturated steam from the boiler drum is used to
reheat the steam that comes out from the high-pressure ST.
In this plant, the steam pressure is increased from the typical 40 bar to 130 bar.
The temperature of the live steam at the outlet of the SH is 440 °C, while the tem-
perature of the reheated steam is 320 °C. The main advantage of this configuration
is its high energy efficiency due to the high pressure and the reheating combined
with modest SH temperature of 440 °C.
4.3 New Designs for High-Efficiency WTE Plant 51

Fig. 4.10   Schematic of the Amsterdam WTE concept of steam intermediate reheating. FGR flue
gas recirculation, EVA evaporator, SH superheater, ECO economizer, ST steam turbine, HPT high-
pressure turbine, LPT low-pressure turbine

Combustion takes place at an excess air ratio equal to 1.4 in order to reduce flue
gas losses. A portion of the flue gas is recirculated back into the lower part of the
boiler, which reduces the temperature and improves the mixing in the post-com-
bustion zone. In this plant, a condensing pressure of 0.03 bar is used (considerably
below the typical WTE plant condensing pressure value).
Different concepts to further enhance the WTE plant efficiency of plants focus
on flue gas and the fuel gas splitting concept, proposed by Babcock & Wilcox and
Karlsruhe Institute of Technology Research Center, respectively.
More in detail, the Babcock & Wilcox Company has developed a new approach
to improve the net electrical efficiency of WTE plants receiving a world patent [8].
The basic idea of the proposed concept (see Fig. 4.11) is to divide the flue gas from
the grate into two fractions, one fraction of the flue gas having a highly corrosive
content of chlorine and another fraction having a low chlorine concentration since
the corrosibility of the flue gas varies significantly over the grate length. The low
corrosive part of the flue gas may be directed to a separate SH section to increase
the steam temperature, thereby boosting the electrical efficiency of the plant [8].
When using an extra SH with the flue gas split concept, the steam data are increased
to 50 bar/500 °C, thereby increasing the efficiency by 3 % [9,10]. This method is
applied in Denmark, in the Reno-Nord WTE CHP plant.
The Karlsruhe Institute of Technology Research Center has proposed a new tech-
nology for WTE plants based on the fuel gas splitting concept. In this configuration,
schematically presented in Fig. 4.12, the temperature of the steam leaving the main
52 4  Waste-to-Energy Steam Cycle

Fig. 4.11   Schematic of the Reno WTE concept of the steam boost concept. FGR flue gas recircu-
lation, EVA evaporator, SH superheater, ECO economizer, ST steam turbine

boiler at 120 bar/400 °C is further increased up to 540 °C by burning the extracted


fuel in a small combustion chamber and utilizing the generated heat for improved
superheating of the pre-generated steam.
Combustion of MSW takes place in the main furnace, but a fraction of the formed
fuel gas is extracted before entering the flue gas burnout zone and goes to a second-
ary combustion chamber, where high steam parameters are reached.
Therefore, bypass of a part of the fuel gas is achieved; this bypass system has
the following components [7]: heat exchanger, fuel gas cleaning system, secondary
combustion chamber, and SH.
As shown in Fig. 4.12, at the temperature of 900 °C, a part of the fuel gas is
extracted from the furnace and subsequently cooled down to about 400 °C using a
heat exchanger to preheat the air. In fact, the cooled gas has to be cleaned prior to
combustion in the secondary combustion chamber in order to operate steam super-
heating at high temperatures avoiding corrosion risks.
At this temperature, alkali and heavy metal chlorides present in the fuel gas are
transferred from the gaseous to the solid state; injections of alkaline compounds to
the fuel gas allow efficient absorption of HCl and, to a lesser extent, of H2S. Sub-
sequently, the cleaned fuel gas is fed into the secondary combustion chamber of the
bypass system. Here it is burnt with low air excess atmosphere. From low corrosive
flue gas, generated heat is used in the superheater of the bypass system for enhanced
temperature of the preheated steam streaming out of the furnace. It must be pointed
References 53

Fig. 4.12   Schematic of the Germany-Karlsruhe WTE concept of bypass of fuel gas [7]. FGR flue
gas recirculation, EVA evaporator, SH superheater, ECO economizer, ST steam turbine

out that the external burnout of the fuel gas and the recirculation of the bypass flue
gases into the main furnace allow the reduction of the excess air in the main com-
bustion chamber by reducing the secondary air supply. As a benefit, the heat loss in
the exhaust gas decreases, improving energy efficiency.
In this configuration, the temperature of the steam leaving the main boiler at
120 bar and 400 °C is further increased up to 540 °C. Compared with the typical
WTE plant steam parameters, that is, 40 bar and 400 °C, the efficiency of gross
power generation shows an increase of approximately 4 % points [7].

References

1. ENEA- FederAmbiente (2012) Rapporto sul recupero energetico da rifiuti urbani in Italia, 3°
edizione, Marzo 2012 (in Italian).
2. Bianchi M, Branchini L, Melino F, Peretto A (2010) Cycle upgrade for waste-to-energy power
plants, ASME-ATI-UIT, 2010, Conference on thermal and environmental issues in energy sys-
tems 16–19 May, 2010, Sorrento, Italy
3. Branchini L (2012) Advanced waste-to-energy cycle, PhD Thesis
4. Stultz SC, Kitto JB (ed) (1992) Steam its generation and use, 40th edition. The Babcock &
Wilcox Company, Barberton, Ohio U.S. A., ISBN 0-9634570-0-4
5. European Commission (2006) Integrated pollution prevention and control reference document
on the best available techniques for waste incineration, August 2006
54 4  Waste-to-Energy Steam Cycle

  6. Gate CycleTM by General Electric Energy (o J) http://site.ge-energy.com/prod_serv/products/


oc/en/opt_diagsw/gatecycle1.htm
  7. Bogale W, Viganò F (2013) A preliminary comparative performance evaluation of highly
efficient Waste to Energy plants, ATI 2013–68th conference of the Italian thermal ma-
chines engineering association. Energy Procedia 45(2014):1315–1324. doi:10.1016/j.egy-
pro.2014.01.137
  8. Ralf K (2008) Innovative concepts of high-efficiency EfW plants, 16th annual North Ameri-
can waste-to-energy conference, May 19–21, Philadelphia, Pennsylvania, USA
  9. Madsen O (2008) High electrical efficiency by dividing the combustion products”, 16th an-
nual North American waste-to-energy conference, May 19–20, Philadelphia, Pennsylvania,
USA
10. Madsen O (2007) Next generation of waste fired power plants, NAWTEC 15, Miami, USA
Part III
WTE Advanced Cycles
Chapter 5
Waste-to-Energy and Gas Turbine: Hybrid
Combined Cycle Concept

This chapter focuses on the hybrid combined cycle (HCC) concept. The HCC, based
on thermal integration between a topper cycle (TC) and a bottomer cycle (BC),
denotes specifically “dual-fuel” combined power cycles. The possibility to use dif-
ferent fuels for the TC and BC is one of the advantages of the HCC. This chapter
describes in detail two basic types of hybrid dual-fuel combined cycle (CC) arrange-
ments applied to waste-to-energy (WTE) power plants using as TC a gas turbine
(GT), a steam/waterside-integrated HCC, and windbox repowering.
Furthermore, a literature review on HCC applied to WTE and GT is discussed,
and three operative WTE power plants based on HCC distributed worldwide are
described together with their technical data.

5.1 The HCC Concept

To introduce the concept of a waste-to-energy–gas turbine (WTE–GT) integrated


power plant, a general overview on HCC is presented in this section.
Although standardized terminology to address the thermal connection between dif-
ferent systems does not exist, CC is the term commonly used for it. The CC concept is
not new and widely spread: It dates back to as long back as the idea for a conventional
gas-steam CC, based on GT as topper unit and a steam cycle as bottomer.
The term topper cycle (TC) addresses the power cycle of any heat engine that ac-
cepts thermal energy at a high temperature level and whose remaining exhaust heat
is exploited in another cycle at a lower temperature level. Typical examples, widely
spread today as TC, are GTs and internal combustion engines (ICE). TC systems
typically utilize high-grade gaseous or liquid fuels.
The term bottomer cycle (BC) refers to any power cycle whose heat supply
comes in the form of rejected heat from another power cycle. The BC itself rejects
heat at the lowest possible temperature level. A typical example of a BC is the
well-known Rankine cycle, working with water/steam or any other two-phase fluid

© Springer International Publishing Switzerland 2015 57


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_5
58 5  Waste-to-Energy and Gas Turbine: Hybrid Combined Cycle Concept

(such as organic Rankine cycle). Power units of this type can have their own fuel
input while serving as BC to a topping engine.
The term hybrid cycle denotes specifically “dual-fuel” combined power cycles
where in different fuels are used for the TC and BC [1]. One of the main advantages
of the HCC is the possibility to utilize low-grade fuels (solid fuels, such as mu-
nicipal solid waste (MSW) or biomass) in the BC, together with exploiting the full
potential of high-grade fuels (gaseous or liquid) in the TC.
The BC has then its own individual combustion chamber in which the bottoming
fuel is fired. Two basic types of hybrid dual-fuel CC arrangements are possible (see
Fig. 5.1) depending on the TC purpose:
• Windbox repowering, where the TC exhaust, with or without precooling, is sup-
plied to the bottoming boiler and used as combustion air for firing the BC fuel
• Steam/waterside integration, where thermal energy from the TC exhaust is uti-
lized for feedwater preheating and/or steam superheating and/or additional steam
generation parallel to the BC

Fig. 5.1   Schematic classification of combined cycles and TC–BC integration options. CC
combined cycle
5.1 The HCC Concept 59

Windbox repowering cycles can be further divided in two main categories: hot
windbox and cold windbox. In the hot windbox type, the GT exhaust is fed directly
into the BC boiler. In the cold windbox type, the GT exhaust is first cooled down
to a lower temperature level (by various options, for example, supplying heat for
parallel steam generation or feedwater preheating), after which it is fed into the BC
boiler. Cold windbox arrangements actually allow for features typical to both the
windbox repowering and steam/waterside-integrated hybrid cycles.
As will be discussed in detail, steam/waterside integration represents a cost-effec-
tive method to increase the electric efficiency in comparison with a reference WTE
plant without new, expensive materials; indeed, further superheating of the steam is
performed in a separate heat recovery steam generator (HRSG) by means of a high-
grade fuel, which cannot cause corrosion problems in the superheater (SH) bank. By
contrast, both cold and hot windbox repowering arrangements, due to a considerable
increase in terms of combustion gas volumetric flow rate, require a significant rede-
sign of the original WTE combustion, convective, and gas cleaning sections.

5.1.1 WTE–GT Steam/Waterside Integration

Focusing on steam/waterside integration, the WTE plant basically acts as an ad-


ditional source of saturated steam and, if convenient, of hot water for the CC. All
the equipment for power production is concentrated in the “island” comprising the
CC. The waste treatment island comprises the equipment for waste handling, waste
combustion, and flue gas treatment. Figure 5.2 schematically shows a possible
WTE–GT steam/waterside-integrated power plant layout.
Being almost exempt from corrosion, the HRSG can take steam to temperatures
typically reached in fossil fuel power plants (520–560 °C). In turn, higher steam
temperatures allow higher evaporation pressures without undue liquid fractions in
the last stages of the steam turbine (ST). Also, generation of saturated steam outside
the HRSG (in the WTE boiler) gives a more favorable temperature profile (smaller
ΔT between gas and steam or water) with lower irreversibility and higher efficiency.
The flue gas treatment section and thus pollutant concentrations at the stack can be
identical to those of conventional WTE plants; this also applies to emissions specific
to the amount of MSW because the integration causes no dilution of the combustion
products. No dilution also means no penalties on the cost for flue gas treatment.
Increasing steam temperature externally in the Rankine cycle will improve the ef-
ficiency of electricity generation for MSW-fired power plants. This can be achieved
by MSW and natural gas HCC that involve two different thermodynamic cycles
with two types of fuel. In the CC, the TC consists of a GT, while the BC, driven
by low-quality MSW fuel, is a steam cycle. In a dual-fuel CC system, there must
be a well-designed thermal link between the TC and the bottomer steam cycle. The
integration has to provide thermodynamic and operating advantages for both the TC
and the BC. Generally, steam superheating by turbine exhaust heat is viable. This
arrangement can substantially increase the efficiency of MSW energy conversion
while avoiding the described corrosion problems (see Chap. 3).
60 5  Waste-to-Energy and Gas Turbine: Hybrid Combined Cycle Concept

([KDXVW JDVHV

&RQGHQVHG
ZDWHU
:DVWH :7(
$LU ERLOHU

6DWXUDWHG
VWHDP
67

([KDXVW JDVHV
6XSHUKHDWHG
+56* VWHDP

*7
1DWXUDO JDV
&&

& 7

$LU

Fig. 5.2   Schematic of the steam/waterside integration between a waste-to-energy (WTE) and a gas
turbine (GT). C compressor, ST steam turbine, HRSG heat recovery steam generator, CC combined
cycle, T turbine

The potential advantages of a WTE–GT plant match compared with a stand-


alone WTE plant can be summarized as follows [2]:
• Increase in the maximum temperature of the steam-cycle-transferring SH from
the WTE to the HRSG, the most problematic component for what concerns high-
temperature corrosion.
• Reduction of the HRSG irreversibility caused by high mean temperature differenc-
es, moving the generation of saturated steam from the HRSG to the WTE section.
• Increase in ST efficiency due to its increased size: The ST serving both the WTE
section and the HRSG is larger than that typically used in conventional WTE plants.
Moreover, from an economic point of view, the integrated system leads to further
benefits such as the reduction of:
• Maintenance costs of the WTE section, due to elimination of the SH heat ex-
changer
• Capital costs since a significant number of equipment and services are shared in
the integrated plant
5.1 The HCC Concept 61

5.1.2 WTE–GT Windbox Integration

Windbox repowering is accomplished by installing a GT to provide extra power


and direct its exhaust into the original boiler’s windbox with or without precooling.
Figure 5.3 shows a schematic WTE–GT hot windbox-integrated layout, while in
Fig. 5.4 a schematic of cold windbox integration is presented.
GTs can be used for retrofitting existing steam plants (hot windbox repowering).
GTs are operated with large amounts of excess air. Thus, the exhaust of GTs con-
tains a high concentration of oxygen (14–16  vol%), and the exhaust temperature is
also high (480–600 °C). This makes a GT well suited for integration with an MSW
boiler in the CC [3]. Moreover, modern GTs are generally optimized with respect to
maximum power density (output per unit air flow) rather than efficiency. This coin-
cidentally meets well with the requirements of optimum efficiency of the CC plants.
The hot windbox repowering has the highest degree of technical complexity of
all the combustion-turbine-based repowering options. The air heaters may need to

([KDXVW JDVHV

&RQGHQVHG
ZDWHU
:DVWH
:7(
$LU ERLOHU

6XSHUKHDWHG 67
VWHDP

([KDXVW JDVHV

*7
1DWXUDO JDV
&&

& 7

$LU

Fig. 5.3   Schematic of hot windbox integration between a waste-to-energy (WTE) and a gas turbine
(GT). C compressor, ST steam turbine, CC combined cycle, T turbine
62 5  Waste-to-Energy and Gas Turbine: Hybrid Combined Cycle Concept

Fig. 5.4   Schematic of cold windbox integration between a waste-to-energy (WTE) and a gas
turbine (GT). C compressor, ST steam turbine, ST steam turbine, CC combined cycle, T turbine

be modified based on the revised air and gas flows, and the ductwork must be
upgraded to accommodate the higher temperature and larger volume of air. The
furnace burners must be modified or replaced because of the lower oxygen content
of the flow from the combustion turbine exhaust. Furthermore, the lower oxygen
content of the combustion air will change the heat release profile in the furnace, and
some derating of the boiler or a redesign of the convective parts of the furnace may
be necessary. Other necessary modifications can include bypass ducts for admitting
variable amounts of combustion turbine exhaust, a steam air heater to allow inde-
pendent operation of the existing boiler when the combustion turbine is not avail-
able, an induced draft fan to reduce the back pressure on the combustion turbine,
and a combustion turbine bypass stack for unit start-up [4].
A variant of the hot windbox repowering approach includes a heat exchanger to
reduce the temperature of the combustion turbine exhaust and, for example, pro-
duce additional steam. The possibility to share both gas and steam/water paths be-
tween cycles is referred to as “cold windbox repowering,” presented in Fig. 5.4.
5.2 State-of-the-Art of Integrated WTE–GT 63

Advantages of hot and cold windbox WTE–GT integration can be summarized


as follows:
• Elimination of WTE combustion air preheating due to the higher temperature of
the GT exhaust gas.
• Reduction of the total exhaust gas mass flow rate and thermal power discharged
in comparison with WTE and GT stand-alone systems.
• Reduction of the environmental impact compared with separate systems: GT
exhaust gas would be subject to the same WTE exhaust cleaning treatments.
• Reduction of the water fraction in WTE exhaust gases that lower the acid dew
point allowing a decrease in WTE minimum allowed temperature: The increase
in mass flow rate in respect to fresh combustion air, for the same water content
in the waste, decreases the water fraction in the WTE exhaust.
Performance of windbox integration can be further improved if external superheat-
ing is applied in the cold windbox configuration. In this case, the integrated cycle
can have a twofold advantage: The steam SH located in the GT exhaust duct is not
exposed to the corrosive gases and can thus achieve the same level as that in the
SH, but it has a simpler design and a much smaller surface area than a heat recovery
boiler.

5.2 State-of-the-Art of Integrated WTE–GT

HCC with MSW as BC fuel has become a very attractive topic for scientific and
industrial research during the past 15 years [5–15] due to the growing interest in (i)
energy utilization from waste and (ii) improvement in the electrical efficiency of
WTE with cost-effective methods.
As already detailed in Chap. 4, improved electrical efficiency from MSW-fired
steam units can be achieved by tolerating higher corrosive rates for increased super-
heat temperatures and consequently higher maintenance costs. A more cost-effective
method for increasing the electrical efficiency without new materials or expensive
investments is the possibility to superheat the steam in a separate heat exchanger
by combusting cleaner fuels. The incorporation of the MSW boiler as a BC into an
HCC, where the topping exhaust provides superheating, can substantially improve
the MSW conversion efficiency into electrical energy using only conventional tech-
nology. This is relevant to any steam cycle with low steam parameters.
Several authors investigated possible arrangements of HCC plants with WTE.
First, Eber et al. [5] proposed external superheating of steam, coming from an MSW
boiler, in a simple natural gas burner. The study, where natural gas had been limited
to 25 % of the total energy input, highlighted the thermodynamic and economic
feasibility of this solution.
The idea of integrating WTE with a topping GT was put forward by Lowry
and Martin [6] and Wiekmeijer [7]. Lowry and Martin evaluated a simplified ar-
rangement in which the GT exhaust gas superheats the MSW-generated steam.
64 5  Waste-to-Energy and Gas Turbine: Hybrid Combined Cycle Concept

­ iekmeijer’s study focused on a more complicated steam/waterside integration


W
­arrangement with an economizer (ECO) and a final SH in an HRSG [7].
Terasawa and Ogura [8] mentioned the HCC alternative for rationalization of
MSW incineration practices. Ito et al. [9] evaluated the economic and energy char-
acteristics of an MSW boiler integrated with a topping GT in combined heat and
power (CHP) mode. Sue [10] suggested the application of a steam-injected GT as
TC in an MSW-fired power unit. Otoma et al. [11] performed a life cycle analysis
(LCA) for WTE with low steam parameters and an electric efficiency of 15 %. Then
they evaluated options for topping this base-case boiler with a GT. Holmgren [12]
focused on a thermodynamic and sensitivity analysis of the overall performance of
three different HCC configurations of a WTE–GT in CHP mode. Results highlight-
ed the advantages of the HCC in comparison with separate plants (existing MSW-
fired WTE, CHP plant, and gas turbine combined cycle (GTCC) plant). Korobitsyn
et al. [13] examined different WTE–GT HCC configurations, featuring final steam
superheating by the GT exhaust gas. Their results showed an increase in the HCC
electric efficiency compared with the average efficiency of the individual single-
fuel plants (one simple MSW-fired steam cycle with efficiency close to 25 % and
one GTCC). Moreover, this study suggested definitions to evaluate within the HCC
the separated contribution of MSW to the electric efficiency. Finally, Udomsri et al.
[14] investigated the economic performance of a WTE–GT, showing that HCC is a
viable solution in comparison with conventional WTE.
Despite this large literature, the actual spread of integrated WTE–GT plants
is really poor [15], confirming that this is a solution which is still new and under
development. To the author’s knowledge, only two WTE–GT plants are located in
the European Union (EU; Zabalgarbi, Spain [16–18], and Moerdijk, the Netherlands
[19]) and one in Japan (Takahama [20]). Only the first one has been designed as a new
WTE–GT integrated plant while the others are retrofits of WTE. The limited spread
of the integration concept can be mainly linked with the lack of regulations concern-
ing multiple-fuel and renewable incentive allocation, rather than technical issues.
The most promising short-term HCC solution not requiring a complete redesign
of the MSW boiler is the steam/waterside integration investigated in the following
section. One important aim of this study not covered in the above cited references is
to find out the optimal GT size for the proposed WTE–GT integrated configurations.

5.3 Existing WTE–GT Integrated Power Plants

The spread of integrated WTE–GT power plants is really poor, which confirms
that this solution is new and still under research. There are two operative WTE–GT
power plants in Europe, namely in Spain and in the Netherlands. Another opera-
tive plant is located in Japan. All the existing and operating WTE-GT power plants
feature the steam/waterside-integrated configuration. However, it must be pointed
out that only the Spanish WTE–GT integrated power plant was originally projected
with the idea of an integrated power plant while the Dutch and the Japanese WTE
power plants have been repowered by means of a GT as TC.
5.3 Existing WTE–GT Integrated Power Plants 65

5.3.1 Zabalgarbi WTE–GT Power Plant: The SENER Solution

The integrated power plant of Zabalgarbi was built by Constructions Industrielles


de la Méditerranée (CNIM), an expert in MSW with energy recovery construction
and operation and SENER Ingeniería y Sistemas, S.A., an engineering company for
the design and development of the project in Bilbao, Spain [16, 17]. Plant construc-
tion started in September 2001 and was completed after approximately 36 months.
The start-up of the plant was in 2004.
The WTE–GT power plant is based on steam/waterside integration. It utilizes
the SENER-2 high-efficiency thermodynamic energy cycle [17] designed and pat-
ented by SENER to minimize corrosion problems which are normal in conventional
plants. As stated in the plant summary report, the aim of the project was to promote
a new concept for generation of electricity from MSW to allow efficient power gen-
eration, efficient waste disposal, and a low environmental impact.
The process and a schematic layout of the plant are shown in Fig. 5.5, and
technical data are summarized in Table 5.1.

Fig. 5.5   Zabalgarbi integrated WTE–GT schematic layout. C compressor, CC combined cycle,
T turbine, SH superheater, ST steam turbine, RH reheater, EVA evaporator, HP high pressure,
IP intermediate pressure, ECO economizer, LP low pressure, FGR flue gas recirculation, DEA
deaerator, HE general heat exchanger
66 5  Waste-to-Energy and Gas Turbine: Hybrid Combined Cycle Concept

Table 5.1   Zabalgarbi’s technical data [16]


WTE section
Type of waste MSW and assimilated
Number of lines 1
Consumption of waste 33.08 t/h
Waste average lower heating value (LHV) 8000 kJ/kg
Productive capacity 240,000 t/year
Thermal power from waste 73.52 MW
Steam
Saturated steam mass flow rate 100 t/h
Steam evaporative pressure (HP and LP) 100 and 25 bar
Steam-superheated and reheated temperature 540 °C
Natural gas
Consumption of natural gas (GT + HRB) 13,870 N m3/h
LHV of natural gas 38,992 kJ/N m3
Thermal power introduced with natural gas 150 MW
Energy
Gross electric power 99,176 MW
Internal consumption (with externals) 4,745 MW h (5,200 MW h)
Net power (grid connection) ST read 54,320 MW
Net power to the grid 94,431 MW
Net power (grid connection) GT read 40,111 MW
Net electricity generation 730,000 MW h/year
Efficiency and savings
Total electrical gross efficiency 44.31 %
Overall net plant efficiency 42 %
Savings on conventional primary energy 47 %
(thermal plant CC with natural gas comparison)
CO2 emission avoided 300,000 t/year
WTE waste-to-energy, LHV lower heating value, HP high pressure, LP low pressure, GT gas tur-
bine, HRB hybrid burner ring, ST steam turbine, CC combined cycle, MSW municipal solid waste

Main components in the system are: an MSW furnace with boiler for the genera-
tion of saturated steam, a GT generator, an HRSG working at 100 bar (equipped
with auxiliary burners), an ST generator, a main condenser and an auxiliary one,
and a flue gas cleaning system.
The saturated steam goes out of the main boiler (located in the WTE section)
at 350 °C and 100 bar. To protect the boiler against corrosion, all exposed furnace
tubes are coated with Inconel 625.
Saturated steam gets in the HRSG where it is heated up to 540 °C with heat
coming from the exhaust of the GT and auxiliary burners. Superheated steam gen-
erated at 540 °C and 100 bar goes into the high-pressure ST to be expanded. Steam
5.3 Existing WTE–GT Integrated Power Plants 67

comes out of the high-pressure ST at a lower pressure (25 bar), goes again to the
HRSG to be reheated at the same temperature (540 °C), and finally goes to the
low-pressure ST. After condensation and before entering the furnace’s boiler, the
water is preheated with a low-pressure ST extraction.
The GT generator chosen was a General Electric LM6000 of 43 MW aver-
age gross capacity fueled with natural gas and equipped with an intake air cool-
ing system (compression chiller). The GT exhaust temperature of the LM6000 is
455 °C; thus, to generate superheated steam at 540 °C, auxiliary burners (using natu-
ral gas and fresh air) are necessary, which work at temperatures around 650–700 °C.
LM6000 was chosen considering that, at the time of the project and authoriza-
tions, the plant was considered in a special Spanish regime (renewables and cogen-
eration), where the maximum size of the integrated plant should have not exceeded
100 MW; thus, considering WTE and consequently ST average capacity, the in-
stalled GT could not exceed 45 MW.
The effect of eliminating the superheated exchangers from the WTE section has
been clearly stated: In conventional plants, maintenance (exchange of tubes) must
be done every 1–2 years; in Zabalgarbi, the first rows of tubes were exchanged in
the 6th year of operation.
Due to system integration, the power plant has a high flexibility. The following
operations modes are feasible:
• Normal operation
• CC operation
• Fresh air mode
• Incineration and GT through bypass stack
• Incineration only
• Shutdown
Based on Zabalgarbi’s operational and economic success, the Vizcaya Regional
Government has focused its 2016 Integral Waste Management Plan on doubling
the first plant’s capacity. Investment costs for the WTE–GT integrated power plant
have been estimated at about 136 million euros.

5.3.2 Moerdijk WTE–GT Power Plant: The Dutch Solution

The history of the WTE–GT power plant of Moerdijk, located in the south of the
Netherlands, is closely related to the nearest MSW incinerator (property of the
AZN company) and a CC plant (property of EPZ, the electricity-producing com-
pany for the Southern Netherlands). The two systems have been commissioned in
the same period (1996 for the incinerator and 1997 for the CC power plant). Thus,
the proximity and their contemporary construction are the main reasons for the
idea of the integrated plant. For this reason, the CC has all the components and the
design of an autonomous system. Figure 5.6 schematically presents the layout of
the integrated plant, showing a complex three-pressure-level CC with reheating.
68 5  Waste-to-Energy and Gas Turbine: Hybrid Combined Cycle Concept

Fig. 5.6   Moerdijk integrated WTE–gas turbine (GT) schematic layout

The HRSG is used to superheat the steam coming from the furnace boiler to avoid
corrosion problems in the WTE section. STs serve both the CC and WTE sections.
The steam coming from the WTE boiler (at 100 bar) is mixed with the steam gen-
erated in the HRSG section. The superheated steam at 520 °C is then fed into the
high-pressure ST. Middle-pressure and low-pressure steam are generated only in
the HRSG section.
The incinerator has a capacity of about 80 t/h of waste with a lower heating value
(LHV) of about 10,450 kJ/kg. Thus, the power introduced with waste is 232 MWt.
The CC is equipped with three GTs, each with an electric power of 60 MW. The
total power of the integrated system is about 330 MW with an increase, with refer-
ence to separate production, of 18 MW.

5.3.3 Takahama WTE–GT Power Plant: The Japanese Solution

Near the city of Takahama in Japan, an MSW incineration plant was built in 1988,
which is able to process 18 t/h of waste with an average calorific value of about
8400 kJ/kg. The primary objective of the incinerator is the disposal of waste, while
References 69

Fig. 5.7   Takahama integrated WTE–gas turbine (GT) schematic layout. C compressor, CC com-
bined cycle, T turbine, SH superheater, ST steam turbine, EVA evaporator, ECO economizer, HRSG
heat recovery steam generator

energy recovery was limited to the generation of only 1.3 MW, essentially equal to
the auxiliary and internal consumptions of the plant.
Eight years later the idea of increasing the electricity production of the plant
came up by integrating the WTE section with a small GT and a heat recovery boiler,
where saturated steam generated in the incinerator at a temperature of 255 °C and
a pressure of 20 bar could be superheated. In 1996 the integrated plant was real-
ized: a 15-MW GT with a heat recovery boiler, a 10-MW ST, and the incinerator.
The layout of the Takahama integrated power plant is shown in Fig. 5.7. The figure
shows that water out of the ECO section is divided into two streams; thus, saturated
steam (at 20 bar) is generated both in the WTE and HRSG section; after mixing,
the steam is superheated in the heat recovery boiler at a temperature of 400 °C. The
integrated plant has a total capacity of 25 MW.

References

1. Petrov MP, Martin AR, Hunyadi L (2002) Hybrid dual-fuel combined cycles: general perfor-
mance analysis, ASME International Joint Power Generation Conference, Phoenix AZ, USA,
June 2002
2. Consonni S (2000) Combined cycles for high performance, low cost, low environmental
impact waste-to energy. Systems, Proceedings of ASME TURBOEXPO, May 8–11, 2000,
Munich Germany
70 5  Waste-to-Energy and Gas Turbine: Hybrid Combined Cycle Concept

 3. De Carli (2004) “Fattibilità tecnico – economica, inclusi gli aspetti autorizzativi,
dell’integrazione inceneritore – ciclo combinato, con riferimento sia all’utilizzo di turbine a
gas che di cicli a vapore”. Technical report, CESI, Ricerca di Sistema (In italian)
  4. William C, Stenzel PE, Dale M, Sopocy PE, Stanley E Pace, repowering existing fossil steam
plants, SEPRIL. http://coal2nuclear.com/MSR%20-%20Repowering%20Existing%20Fos-
sil%20Steam%20Plants%20-%20SEPRIL%20.pdf
  5. Eber SM, Kettler DJ, Worsley CR, Fedele G (1989) Gas fired superheaters in a resource
recovery installation. Proceedings of the American Power Conference, vol 51, Chicago, IL,
April, pp 858–864
  6. Lowry WL, Martin CA (1990) Economics of combined cycle waste-to-energy. Proceedings of
the International Joint Power Generation Conference, Boston MA, USA, October, pp 31–37
  7. Wiekmeijer IT (1990) Improvements in incinerators by means of gas turbine based cogen
systems, ASME Paper 90-GT-180, Presented at the International Aeroengine Congress and
Exposition, Brussels, Belgium, June
  8. Terasawa H, Ogura M (1993) System to rationalize energy utilization at waste incineration
plants. Proceedings of JSME—ASME International Conference on Power Engineering,
Tokyo, September, vol 2, pp 591–594
  9. Ito K, Yokoyama R, Shimoda M (1996) Optimal planning of a super waste incineration co-
generation plant. ASME Paper 96-GT-384, Proceedings of the International Gas Turbine and
Aeroengine Congress and Exposition, Birmingham, UK, June
10. Sue M (1996) Performance characteristics of waste-to-energy system utilising steam-inject-
ed gas turbine. Transactions of the Japan Society of Mechanical Engineers, Part B, vol 62,
Nr.597, May, pp 2013–2020
11. Otoma S, Mori Y, Terazono A, Aso T, Sameshima R (1997) Estimation of energy recovery
and reduction of CO2 emissions in municipal solid waste power generation. Resour Conserv
Recycl 20(2), 95–117
12. Holmgren R Hybridkraftverk för avfallseldning, M. Sc. Thesis, report Nr. 1998:228 CIV,
ISSN: 1402-1617, ISRN: LTU-EX–98/228-SE, Division of Energy Technology, Luleå Uni-
versity of Technology, Luleå, Sweden, August 1998
13. Korobitsyn MA, Jellema P, Hirs GG (1999) Possibilities for gas turbine and waste incinerator
integration. Energy 24(9):783–793
14. Udomsri S, Martin AR, Fransson TH.(2010) Economic assessment and energy model sce-
narios of municipal solid waste incineration and gas turbine hybrid dual-fueled cycles in
Thailand. Waste Manag 30:1414–1422
15. Petrov MP, Hunyadi L (2002) Municipal solid waste boiler and gas turbine hybrid combined
cycles performance analysis. 1st International Conference on Sustainable Energy Technolo-
gies 12–14 June, Porto, Portugal paper n. EES6
16. CNIM and SENER (2005) Bizkaia waste to energy plant project, summary report. Contract
No BM 80/96 ES-PO
17. SENER Ingeniería y Sistemas S.A., High efficiency technologies applied for waste to energy
(WTE)
18. AZN and EPZ, Combination of a waste incineration plant and a combined cycle power plant,
project No BM/349/93/NL
19. Japanese National Team (1997) First Japanese waste incinerator and combined cycle power
plant repowered with a gas turbine. Caddet Energy Effciency, Waste and Biogas, Newsletter
No 2, 1997
Chapter 6
WTE–GT Steam/Waterside Integration:
Thermodynamic Analysis on One Pressure
Level

This chapter focuses on waste-to-energy (WTE) and gas turbine (GT) integrated con-
figurations concerning one-pressure-level heat recovery steam generator (HRSG).
The thermodynamic and parametric analysis of steam/waterside-integrated WTE–
GT power plant has been carried out, first of all, with the aim to investigate the logic
governing plants and that it should match in terms of steam production as a function
of the thermal power generated. Steam generation, optimum plant match condition,
inlet and outlet conditions of heat exchangers, etc., as a consequence of system in-
tegration are analyzed and explained. A sensitivity analysis, varying with evapora-
tive pressure and HRSG inlet conditions, has also been presented to investigate the
influence of operative parameters on steam mass flow rate.
Starting with a simple integrated plant layout, several configurations have been
proposed and analyzed. Positive aspects and limitations of each layout have been
investigated and discussed.
The following paragraphs assess and define, for a given layout and operative
conditions, the optimum WTE–GT plant match in terms of system input of thermal
power to maximize steam generation and plant performance and to minimize dis-
charged outlet temperature.

6.1 Thermodynamic Analysis of Steam Production

Focusing on a waste-to-energy–gas turbine (WTE–GT) steam/waterside-integrated


power plant, the simplest layout with one-pressure-level HRSG is shown in Fig. 6.1
(named Layout_1). This configuration can be regarded as a starting point to investi-
gate the logic concerning steam production in an integrated plant.
In the proposed layout, while saturated steam gets produced in the waste boiler,
water gets preheated and steam superheated in the HRSG economizer (ECO) and
superheater (SH), respectively; one bleed from the steam turbine (ST) is present to
preheat the primary air necessary for waste combustion and to feed the deaerator
(DEA).

© Springer International Publishing Switzerland 2015 71


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_6
72 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.1   WTE–GT integrated power plant layout (Layout_1). C compressor, CC combined cycle,
T turbine, SH superheater, ST steam turbine, EVA evaporator, ECO economizer, HRSG heat recov-
ery steam generator

As we focus on steam production, we find that steam generation is affected by:


• The thermal power discharged from the GT entering the HRSG section:

 (
Q EXH = m EXH c p ,EXH TO,GT − Tref ,) (6.1)

where TO,GT is the GT outlet temperature, Tref is the reference temperature assumed
to be equal to 15 °C, and c p ,EXH is the specific heat of the exhaust gases, a function
of TO,GT.
• The input power introduced with waste in the WTE boiler:
FW = m W LHVW , (6.2)

where m W and lower heating value ( LHVW ) are waste mass flow rate and lower
heating value, respectively.
6.1 Thermodynamic Analysis of Steam Production 73

Consequently, energy balance allows us to evaluate the superheated steam pro-


duced in the HRSG (m s,HRSG) and the saturated steam generated by the WTE boiler
(m s,WTE) with the following equations:
 ξQEXH
m s,HRSG = , (6.3)
∆ hSH + ∆ h ECO

 ηboil FW
m s,WTE = , (6.4)
∆ h lat + ∆ hsc

where:
• ξ is the HRSG effectiveness, which mainly depends on the HRSG inlet and
outlet temperatures, TO,GT and TO,HRSG.
• ηboil is the WTE boiler efficiency, which is mainly due to outlet temperature of
WTE gases, TO,WTE, and the ratio of air to fuel waste mass flow rates.
• ∆ hSH and ∆ h ECO are steam- and water-specific enthalpy, which increases in the
SH and in the ECO, respectively.
• ∆ h lat + ∆ h sc is the specific enthalpy rise in the evaporator (EVA), which is the
sum of latent heat and subcooling enthalpy. Once selected, the evaporative pres-
sure value ( pev ), ∆ h lat is constant, while ∆ hsc is equal to the difference between
evaporative and ECO outlet temperature. Thus, it can be equal to or higher than a
minimum value imposed to avoid water evaporation in the ECO section depend-
ing on the Q EXH value, which will be detailed later.
The adopted layout proposes to have the same value of steam mass flow rate in the
HRSG and in the WTE sections. Consequently, by equating Eqs. (6.3) and (6.4) the
following must be satisfied:

Q EXH ηboil ∆ h SH + ∆ h ECO
= . (6.5)
FW ξ ∆ h lat + ∆ h sc

Figure 6.3 shows how Eq. (6.5) could be solved qualitatively to calculate steam pro-
duction as a function of GT-discharged thermal power (Q EXH ), assuming the values
TO,GT , p ev , and FW remain constant.
In particular, m s,max, calculated from Eq. (6.4), represents the upper limit of
steam production, regardless of the HRSG size, determined only by the thermal
power introduced with waste, FW .
The main design point (point B in Fig. 6.2) can be identified, corresponding to
*
optimum GT-discharged thermal power, Q EXH .
For GT-discharged thermal power higher than Q*EXH (point B) and moving from
B to A, the following takes place:
• HRSG outlet temperature (TO,HRSG) increases (decreasing HRSG effectiveness,
ξ ); thus, it overcomes its minimum value (Fig. 6.3a).
74 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.2   Steam to waste mass ͘


flow rate as a function of ŵ
gas turbine (GT)- discharged Ɛ & сĐŽŶƐƚ
t
thermal power ͘ Ɖ сĐŽŶƐƚ͕d сĐŽŶƐƚ
Ğǀ K͕'d
ŵ
Ɛ͕ŵĂdž

  


'dhEZ^/ 'dKsZ^/

Y
Ύ Y
y, y,

• No change in subcooling temperature difference, ∆Tsc (Fig. 6.4b), which keeps


its minimum value.
• Superheated steam temperature, TSH (Fig. 6.3c), remains constant and at its
maximum value if we assume a minimum approach difference between T O,GT
and TSH .
A reduction of Q EXH under Q*EXH can be managed with two possible strategies:
1. Dotted line B → C: To keep the steam mass flow rate at its maximum value
m s,max , a reduction in TSH is required. In this case, the thermal power discharged
from the GT is not enough to superheat, at maximum level, the steam that the
WTE boiler can produce.
2. Continuous line B → D: TSH can be kept constant so that the efficiency of the
steam cycle is not compromised, thus reducing the steam mass flow rate and
increasing ∆Tsc .
In this case, the difference between the EVA and ECO outlet temperatures is higher
than the minimum value allowed; this increase in the subcooling temperature is
equivalent to the introduction of an ECO section in the WTE boiler.1
Therefore, point B represents the best choice (maximum TSH , minimum ∆Tsc,
and TO,HRSG ) for the assumed layout.

1 
If an ECO section was placed into the WTE boiler, point D will show an increase in steam mass
flow rate, due to a lower temperature water entering the WTE boiler, causing a further reduction
of TO, WTE.
6.1 Thermodynamic Analysis of Steam Production 75

Fig. 6.3   TO,HRSG (a), ∆Tsc

O, HRSG
p = const, T = const a
(b), and TSH (c) realized as ev O,GT
F = const
functions of GT-discharged A

T
W

thermal power

C B
D

∆T
sc b

C ∆T B A
sc,min

T c
SH

D B A

* Q
Q EXH
EXH

It is possible to operate the plants by keeping the thermodynamic parameters at


their optimum values (maximum TSH and minimum TO,HRSG, line B–D), but, in this
case, steam mass flow rate is lower than its maximum value. Thus, the WTE section
turns out to be undersized.
76 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.4   T–Q diagram for d


the HRSG section of the inte- ΛKƉƟŵƵŵYΎy,ͬ&t
grated plant. SH superheater,
ECO economizer ;ƉŽŝŶƚͿ

džŚĂƵƐƚŐĂƐĞƐ
^,

^ƚĞĂŵͬǁĂƚĞƌ

K

Y Y Y
^, K

However, it is not convenient to operate by keeping the steam mass flow rate
*
m s,max at the maximum, but with Q EXH far from  Q EXH (line A–B or line B–C);
this means that one has to operate with low HRSG effectiveness (A–B) or with low
thermodynamic steam cycle efficiency (B–C).
This analysis highlights that, for a given evaporative pressure and GT outlet
temperature, an optimum plant match, in terms of thermal power ratio, Q *EXH FW,
is found.
Beyond this value, an increase in HRSG thermal power input does not lead to
additional benefits in terms of generated steam mass flow or HRSG effectiveness.
Figure 6.4 shows the T–Q diagram for the HRSG section, corresponding to the
optimum value (point B, Q *EXH FW ). As previously mentioned, one of the main
advantages of a WTE–GT integrated plant is the reduction of the HRSG irrevers-
ibility caused by high mean temperature differences. As stressed in the figure, re-
moving the EVA from the HRSG reduces the distance between exhaust gases and
steam/water lines, with respect to a typical one-pressure-level HRSG T–Q diagram.
Finally, for the investigated layout, it must be pointed out that TO,WTE strongly
depends on evaporative temperature; thus, the choice of evaporative pressure main-
ly affects ηboil; the higher the pev , the lower is the WTE boiler efficiency.

6.1.1 Influence of Evaporative Pressure and GT Outlet


Temperature on Steam Production

In order to investigate the influence of steam cycle parameters on optimum WTE–


GT plant match, a parametric analysis has been performed by varying steam evapo-
rative pressure and GT outlet temperature.
6.1 Thermodynamic Analysis of Steam Production 77

Figures 6.5 and 6.6 show, qualitatively, the influence of three different evapo-
rative pressures ( pev < pev ′′ ) and TO,GT values (TO,GT1 < TO,GT2 < TO,GT3) on
′ < pev
steam mass flow rate production, respectively. Focusing on Fig. 6.5, an increase
in the evaporative pressure value and a decrease in ∆ h lat can lead to an increase in
steam mass flow rate as expressed in Eq. (6.4).
If the thermal power introduced in the GT exhaust is not enough to bring about
maximum steam flow rate (points B, Bʹ, and Bʹʹ), the amount of produced steam
is lower than the maximum value, and its variation is not dependent on pev . By

Fig. 6.5   Steam mass flow ͘


rate as a function of GT- ŵ d сĐŽŶƐƚ;d сĐŽŶƐƚͿ
discharged thermal power for Ɛ K͕'d ^,
ΖΖ ƉΖΖ
different evaporative pressure & сĐŽŶƐƚ Ğǀ
t
values

Ζ ƉΖ
Ğǀ

 Ɖ
Ğǀ

Y
Ύ
;Y
Ύ
ͿΖ ;Y
Ύ
ͿΖΖ Y
y, y, y, y,

Fig. 6.6   Steam mass flow


rate as function of GT-

ŵ
discharged thermal power Ɛ Ɖ сĐŽŶƐƚ
Ğǀ
for different GT outlet &tсĐŽŶƐƚ
temperature

  
Ϯ ϯ

d
K͕'dϭ

d d
K͕'dϮ K͕'dϯ

Ύ Ύ Ύ
Y
y,
;Y Ϳ ;Y
y, Ϯ y, ϯ
Ϳ Y
y,
78 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

*
increasing QEXH over QEXH , a constant trend is achieved for every evaporative
pressure value. Moreover, by increasing the evaporative pressure, an optimum con-
dition (B, Bʹ, Bʹʹ, respectively, for pev , p′ev , and p′′
ev ) is achieved, which is char-
*
acterized by a greater optimum thermal power ratio, that is, the QEXH / FW value.
The influence of TO,GT (and consequently TSH ) is shown in Fig. 6.6; for QEXH
below the optimum point, an increase in GT outlet temperature (and on steam-
superheated one) decreases the steam mass flow rate. Once the optimum point is
reached (B, B2, B3, respectively, for TO,GT1 , TO,GT2, and TO,GT3), steam production
is the only function of the input waste thermal power and is not affected by TO,GT .

6.2 Numerical Results

The main results of the study are collected and shown in Fig. 6.7. A constant waste
composition and LHVW value have been assumed, as reported in Table 6.1. In Ta-
ble 6.2, additional assumptions used for numerical analysis are listed.
The grid of Fig. 6.7 represents optimum plant match (point B) for different evap-
orative pressure and GT outlet temperature values.

Fig. 6.7   Grid representing ͘


ŵ ϲ͘ϰ
optimum plant match for a Ɛ
given evaporative pressure ͘ ΀Ͳ΁ ϴϬďĂƌ
ŵ
and GT outlet temperature as t
ϲ͘ϯ
a function of thermal power
input ratio
ϲ͘Ϯ
ϳϬďĂƌ

ϲ͘ϭ d сϰϮϬΣ ϰϱϬ ϱϬϬ ϱϱϬ ϱϴϬΣ


K͕'d

ϲ͘Ϭ
ϲϬďĂƌ

ϱ͘ϵ

ϱϬďĂƌ
ϱ͘ϴ

ϱ͘ϳ
Ɖ сϰϬďĂƌ
Ğǀ

ϱ͘ϲ
Ϭ͘ϵϬ

Ϭ͘ϵϱ

ϭ͘ϬϬ
Ϭ͘ϳϬ

Ϭ͘ϳϱ

Ϭ͘ϴϬ

ϭ͘ϮϬ
Ϭ͘ϴϱ

ϭ͘Ϭϱ

ϭ͘ϭϬ

ϭ͘ϭϱ

YΎ ͬ&t΀Ͳ΁
y,
6.2 Numerical Results 79

Table 6.1   Municipal solid waste (MSW) composition and lower heating value (LHV) assumed
for the analysis
Proximate Humidity 23.0 % of volatile C 52.5
analysis (% Ashes 15.0 matter H 7.5
by weight)
Volatile matter 62.0 O 38.5
N 1.3
S 0.2
LHVW = 11.85 MJ/kgW (2.83 kcal/kgW)

Table 6.2   Main assumptions


ST extraction pressure for primary air heating and deaerator (bar) 3.5
ST isentropic efficiency (−) 0.85
ST outlet quality (−) 0.84 ÷ 0.95
Condenser pressure (bar) 0.1
Deaerator working pressure (bar) 3
Oxygen in exhaust dry gases (vol. %) 7
WTE exhaust gas recirculation (vol. %) 15
WTE exhaust gas recirculation temperature (°C) 150
Primary air temperature (°C) 130
Secondary air temperature (°C) 50
ΔT pinch point (°C) 10
ΔT approach (°C) 20
ΔTsc (°C) 10
TO, WTE (°C) ≥ 160
TO, HRSG (°C) ≥ 110
ST steam turbine, WTE waste-to-energy

Intersecting the line at constant pressure with the line at constant temperature,
the steam mass flow rate and the optimum ratio of HRSG to WTE thermal power
input is obtained.
For example, at evaporative pressure equal to 60 bar and a GT outlet tempera-
ture of 500 °C (corresponding to 480 °C of superheated steam temperature), about
6 kg/s of steam mass flow rate is generated for every kg/s of waste. Moreover, the
optimum plant match corresponding to Q *EXH FW = 0.93 is found.
As highlighted in Fig. 6.7, high evaporative pressure and superheated steam tem-
perature means high Q *EXH FW . Thus, the higher the steam cycle parameters, the
higher must be the GT-discharged thermal power per unit of thermal power intro-
duced with waste.
80 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.8   GT power output as ϮϬϬ

W ΀Dt ΁
Ğ
a function of GT-discharged
heat for commercial GT units

'd
ϭϱϬ

W
'd
ϭϬϬ сϬ͘ϳϱ
Y
y,

ϱϬ W
'd
сϬ͘ϯϬ
Y
y,

Ϭ
Ϭ ϱϬ ϭϬϬ ϭϱϬ ϮϬϬ
Y ΀Dt΁
y,

6.2.1 Optimum Plant Match in Terms of Electric Power Ratio

Optimum plant match can be expressed in terms of electric power output rather than
the thermal power input.
In order to evaluate the GT electric size ( PGT ) that must be chosen to realize the
integrated power plant of Fig. 6.1, the following equation can be formulated:

 PGT P Q F
= GT · EXH · W , (6.6)
PWTE QEXH FW PWTE

where PWTE is the electric capacity of a traditional WTE power plant fed with the
same waste input as that of the integrated system.
Taking into account that
• the electric capacity of a GT is a fraction of the discharged thermal power QEXH
ranging from 0.30 to 0.75 depending on the GT characteristics (see Fig. 6.8) and
that
• for a traditional WTE power plant, the efficiency PWTE FW typically ranges
from 0.25 to 0.30.
From Eq. (6.6) it can be concluded that optimum matching requires one to select
a GT with an electric power output ranging from the same WTE size (minimum
PGT QEXH and maximum PWTE FW ) up to three times of the WTE original plant
(maximum PGT QEXH and minimum PWTE FW).
To keep the GT size at its minimum, the PGT QEXH ratio has to be the low-
est. Since a relationship between PGT QEXH and the GT efficiency ( ηGT ) can be
identified as
6.2 Numerical Results 81

Fig. 6.9   Grid representing ͘


optimum plant match as a ŵ ϲ͘ϰ
function of output power ͘ Ɛ ΀Ͳ΁ W W
ϴϬďĂƌ
ratio for WTE efficiency ŵ 'd
сϬ͘ϯϬ
td
сϬ͘ϯϬ
t Y
of 30 % and a ratio of GT y,
&
t

electric power output to the ϲ͘Ϯ


discharged heat, which is ϳϬďĂƌ
equal to 0.3 d сϰϮϬ ϰϱϬ ϱϬϬϱϱϬ ϱϴϬΣ
K͕'d

ϲ͘Ϭ ϲϬďĂƌ

ϱϬďĂƌ
ϱ͘ϴ

Ɖ сϰϬďĂƌ
Ğǀ

ϱ͘ϲ

ϭ͘ϬϬ

ϭ͘ϮϬ
Ϭ͘ϳϬ

Ϭ͘ϳϱ

Ϭ͘ϴϬ

Ϭ͘ϴϱ

Ϭ͘ϵϬ

Ϭ͘ϵϱ

ϭ͘Ϭϱ

ϭ͘ϭϬ

ϭ͘ϭϱ
W ͬW ΀Ͳ΁
'd td

 1
ηGT ≈ , (6.7)
 Q 
1 +  EXH 
 PGT 

it follows that the GT with low efficiency is the best choice for this purpose.
Figure 6.9 shows, as an example, the steam-to-waste mass flow rate ratio as a
function of electric power output PGT PWTE . The figure refers to WTE efficiency
of 30 % and a ratio of GT electric power output and discharged heat equal to 0.30.
Thus, optimum conditions expressed in terms of the thermal power ratio (see
 P Q* 
Fig. 6.7) agree with those found in terms of power output ratio  GT = EXH  .
 PWTE FW 
However, Fig. 6.10 shows that optimum values in terms of electric power ratio are
about three times those found in Fig. 6.9, assuming WTE efficiency of 25 % and a
ratio of GT electric power output to discharged heat equal to 0.75.

6.2.2 Traditional WTE Versus Integrated Plant: Steam Turbine


Capacity

The analysis compared the steam mass flow rate produced by an integrated plant
with that generated by a traditional WTE fed with the same amount of waste power
input and highlighted how a considerable increase can be achieved.
82 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.10   Grid represent- ͘


ing optimum plant match as ŵ ϲ͘ϰ W
͘ ΀Ͳ΁
Ɛ W ϴϬďĂƌ
a function of output power 'd
сϬ͘ϳϱ td сϬ͘Ϯϱ
ratio for WTE efficiency of ŵ Y &
t y, t
25 % and a ratio of GT power
ϲ͘Ϯ
output to discharged heat ϳϬďĂƌ
equal to 0.75 d сϰϮϬ ϰϱϬ ϱϬϬ ϱϱϬ ϱϴϬΣ
K͕'d

ϲ͘Ϭ ϲϬďĂƌ

ϱ͘ϴ ϱϬďĂƌ

Ɖ сϰϬďĂƌ
Ğǀ

ϱ͘ϲ
Ϯ͘ϬϬ

Ϯ͘ϮϬ

Ϯ͘ϰϬ

Ϯ͘ϲϬ

Ϯ͘ϴϬ

ϯ͘ϬϬ

ϯ͘ϮϬ

ϯ͘ϰϬ

ϯ͘ϲϬ
W ͬW ΀Ͳ΁
'd td

Fig. 6.11   Increase in steam ∆W ϳϬ


^d
turbine power output as ΀й΁
a function of evaporative W
td ϲϱ ϱϯϬΣ
pressure and GT outlet
temperature
ϲϬ
ϰϴϬΣ
ϱϱ
ϰϯϬΣ
ϱϬ

d сϰϬϬΣ
ϰϱ ^,

ϰϬ

ϯϱ
ϰϬ ϰϱ ϱϬ ϱϱ ϲϬ ϲϱ ϳϬ ϳϱ ϴϬ
Ɖ ΀ďĂƌ΁
Ğǀ

As a consequence, the capacity of the ST in the WTE–GT integrated system


will be higher than the one employed in a WTE stand-alone power plant. It can be
expressed as
 PST = PWTE + ∆ PST , (6.8)

where ∆PST is proportional to the steam mass flow increase.


Figure 6.11 shows the percentage increment of ST power output with respect to a
WTE stand-alone power plant and highlights that it ranges, for the considered cases,
6.4  WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 83

from 35 to 70 %. It reaches its highest value for the maximum evaporative pressure
and steam-superheated temperature.

6.3 Concluding Remarks on Thermodynamic Analysis

The study suggests that an optimum WTE–GT plant match in terms of the system’s
thermal power input must be pursued to maximize steam generation and steam-
superheated temperature and to minimize exhaust gas temperature.
Deviation from optimum conditions can be resolved through the following steps:
(i) oversize the GT without additional benefits in terms of generated steam mass
flow rate or the effectiveness of HRSG, (ii) depress the WTE section, decreasing the
amount of generated steam, or (iii) work with low steam cycle thermodynamic effi-
ciency. Thus, the thermodynamic and parametric analysis provide useful guidelines
in selecting an optimum GT size to match WTE–GT maximum performance. The
study highlights that the higher the steam cycle parameters (evaporative pressure
and steam-superheated temperature), the higher must be the GT-discharged thermal
power with respect to the thermal power introduced with waste.
A correspondence between optimum thermal and electric power ratio can be
achieved only if the GT efficiency is low, otherwise, optimum values in terms of
electric power ratio can be up to three times that found corresponding to input ther-
mal power.
Comparing steam mass flow rate produced in the WTE–GT integrated system
with a WTE stand-alone, we find significant increases. Moreover, percentage gains
in steam mass flow rate increase the increasing evaporative pressure and steam-su-
perheated temperature. Increments in ST power output proportional to steam mass
flow rate increase, and new ST capacity, for the integrated plant, has been derived:
Power output increase ranges from 35 to 70 %, reaching its highest values for the
maximum evaporative pressure and superheated temperature.

6.4 WTE–GT Proposed Layouts for a One-Pressure-Level


HRSG

In this paragraph, WTE–GT layouts for a one-pressure-level HRSG are proposed


and investigated. Simulations have been performed using Gatecycle software. In-
tegrated system layouts are optimized, case after case, in order to exploit both the
thermal power input with the waste and the GT-discharged thermal power. For each
one of the proposed configurations, steam mass flow generation is analyzed and
discussed as a function of GT-discharged thermal power, and the optimum plant
match condition is identified. Based on common assumptions, the numerical results
for each one of the WTE–GT integrated plant layouts are highlighted.
Municipal solid waste (MSW) composition and, as a consequence, waste LHV
assumed for the analysis have been taken according to Table 6.1.
84 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Table 6.3   Assumption of the Molecular weight 28.6344


composition of GT exhaust
gases [2] Molecular fraction composition
Oxygen, O2 0.1531
Nitrogen, N2 0.7610
Water, H2O 0.0510
Carbon monoxide, CO 0.0000
Carbon dioxide, CO2 0.0258
Methane, CH4 0.0000
Hydrogen, H2 0.0000
Argon, Ar 0.0091
Carbonyl sulphide, COS 0.0000
Hydrogen sulphide, H2S 0.0000
Sulphur dioxide, SO2 0.0000
Specific enthalpy value
hEXH (at 500 °C) (kJ/kg) 519.78
hrif (at 15 °C) (kJ/kg) − 57.04

Table 6.4   WTE outlet pev (bar) TO, WTE (°C)


temperature as a function of
evaporative pressure 40 260
50 274
80 305

Being interested in evaluating the optimum plant match condition in terms of


thermal power discharged by a GT exhaust, we have assumed a typical GT exhaust
gas composition according to [2] and as summarized in Table 6.3. The reference
composition entering the HRSG section, shown in Table 6.4, refers to an air-to-fuel
(CH4) mass flow rate ratio equal to 68.94. Taking into account the stoichiometric
ratio of combustion of air to methane (17.23), the excess air in the composition of
exhaust gases turns out to be equal to 4, in line with typical GT combustion values.
The remaining steam cycle common assumptions are reported in Table 6.2.
The following aspects have been considered in the analysis of factors that affect
waste combustion:
• The total air required for the waste boiler is set to obtain a prescribed oxygen
concentration in dry exhaust gases; based on the oxygen content in the exhaust
gas, the program automatically calculates the excess of air with respect to a stoi-
chiometric combustion. For the assumed value of oxygen content in dry exhaust
gases, the combustion air-to-waste mass flow ratio turns out to be equal to 6.55
(thus, the excess air fraction is equal to 0.63).
• Moreover, in the program, the primary air is specified in relation to the mass
flow rate of the fuel, that is, as a mass fraction (mass flow of primary air/mass
flow of fuel) assumed equal to 4.1; thus, 63 % of the total amount of combustion
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 85

air is identified as primary air, while the remaining part is secondary combustion
air.
• Primary and secondary combustion air are supposed to be preheated up to a tem-
perature of 50 °C by exploiting grate cooling.
• The integrated boiler (representing three vertical radiation passes, see Chap. 3,
Sect. 3.1.3) has been modeled taking into account the contribution of both the
thermal radiation and convective heat transfer; the overall heat transferred to
EVAs and the waterwalls bordering the combustion chamber are mainly due to
thermal radiation (about 87 % of the total heat exchanged).
• The assumed MSW composition, the calculated combustion air mass flow, the
assumed exhaust flue gas recirculation (FGR), and combustion parameters lead
to an adiabatic combustion temperature equal to about 1185 °C, which is in line
with typical values for a WTE integrated boiler.
• A convective heat exchanger section (convective pass, see Chap. 3, Sect. 3.1.3)
has also been modeled. The temperature at the convective section inlet has been
fixed according to a typical WTE value, equal to 650 °C (Chap. 3, Sect. 3.1.3).
• The effectiveness of the heat exchanger has been kept at an upper limit to restrict
heat exchanger surfaces.
Numerical results for WTE–GT integrated layout (Layout_1 in Fig. 6.1) are shown
in Figs. 6.12 and 6.13 for T–Q diagram for the HRSG section corresponding to the


7>ƒ&@
DWRSWLPXP%
S  EDU
HY
([KDXVWJDV
7 ƒ&
 2*7

6+



6WHDPZDWHU



(&2



       


4 4
6+ (&2 4>N:@

Fig. 6.12   T–Q diagram for HRSG section for Layout_1. ECO economizer
86 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

 
P >NJV@
V
)  0:
:

 7 ƒ& 7 ƒ&
2*7 6+
S EDU
HY

 %




/D\RXWB






       

4
(;+
4 >0:@
(;+

Fig. 6.13   Steam mass flow rate for Layout_1 as a function of GT-discharged thermal power

optimum point and steam mass flow rate generated as a function of GT-discharged
thermal power, respectively.
Numerical results confirm the thermodynamic analysis described in the previ-
ous paragraph: The steam mass flow rate increases linearly, reaching the maximum
value corresponding to the optimum plant match (point B), above which a constant
trend is reached. Optimum plant match condition, for the considered layout, turns
out to be equal to about 87.7 MW corresponding to maximum steam mass flow
rate of 49.12 kg/s. The lowest considered value (corresponding to a steam mass
flow rate equal to about 44 kg/s) agrees with the highest effectiveness assumed for
SH heat exchanger. By contrast, there is no upper limit, but the constant trend has
been interrupted in proximity to point B because, as previously described, it is not
*
convenient to operate with values higher than QEXH .
For the investigated layout, it must be noticed that TO,WTE strongly depends on
the evaporative temperature; having assumed only the presence of evaporative heat
exchanger in the WTE integrated boiler, the exhaust gas temperature is a func-
tion only of the evaporative pressure and the pinch point assumed, as reported in
Table 6.4.
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 87

Table 6.5   Layout_1 main Layout_1


results
FW (MW) 100
pev (bar) 50
TO, GT (°C) 500
TSH (°C) 480
m EXH (kg/s) 152
*
QEXH (MW) 87.68

* 0.88
QEXH
Fw
(−)

m s, max (kg/s) 49.12


TO, HRSG (°C) 143
TO, WTE (°C) 274
PST (MW) 45.76

The main results of the proposed layout, corresponding to optimum plant match
*
condition (point B), are summarized in Table 6.5. The optimum QEXH FW corre-
*
sponding to an evaporative pressure of 50 bar is below the unit ( QEXH FW equal to
0.88). This means that the thermal power introduced with waste is greater than that
introduced with the GT exhaust gases. Thus, the biggest supply in terms of thermal
power introduced is assigned to the WTE section.
The main negative aspect of the proposed layout is the high WTE exhaust gas
temperature: A relatively high amount of heat is discharged from the WTE boiler
( TO, WTE higher than 270 °C), which is not completely used at the minimum tempera-
ture allowed (160 °C).
Layout_2 and Layout_2bis  The next proposed layout (Layout_2) is schematically
shown in Fig. 6.14. As can be seen in the figure, one ECO section is present in the
WTE and one in the HRSG section. Water out of the DEA is first fed to the ECO1,
which is placed in the WTE convective pass, where part of the economization is
performed; economization is completed in the ECO2, inside the HRSG section.
The idea to share water economization comes from results of the Layout_1 that
tried to exploit the thermal power available in the boiler, minimizing the WTE out-
let temperature. Compared with the previous layout, there was considerable reduc-
*
tion of QEXH corresponding to the optimum point ( point Bʹʹ, Fig. 6.15). Optimum
( )
* ′
plant match condition is found corresponding to QEXH and equal to 81.9 MW.
*
A decrease in QEXH equal to 6.6 % of Layout_1 was obtained, while the maximum
amount of steam produced remains the same. Focusing on steam production, for
a fixed evaporative pressure and steam-superheated temperature, Layout_2 shows
an increase in steam mass flow rate compared to Layout_1 for QEXH lower than
the optimum value (see Fig. 6.15). The increase in steam mass flow can be ex-
plained considering the presence of the first ECO section in the WTE—allowing an
88 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.14   Schematic of Layout_2 waste-to-energy (WTE)–GT integrated plant. C compressor, CC


combined cycle, T turbine, SH superheater, ST steam turbine, ECO economizer

increase in water inlet temperature at ECO2, for a fixed value of QEXH , increases the
amount of generated steam. However, once the optimum condition is reached, the
two layouts reach the same maximum value, since the amount of generated steam is
determined only by the thermal power available with waste, FW.
However, corresponding to an optimum value, an increase in HRSG outlet tem-
perature is noticed when compared to Layout_1. In this configuration, water enters
ECO2 at a higher temperature. The T–Q diagram for the HRSG section of Layout_2
is shown in Fig. 6.16. About 71 % of the thermal power introduced in HRSG is used,
with 60 % used in the SH section.
Figure 6.17 shows an alternative possibility to share water economization be-
tween the two subsystems, namely, Layout_2bis. In the considered configuration,
water coming out of the DEA, is first fed to the ECO2 (placed in the HRSG), where
part of the economization is performed which is completed in the ECO1 inside the
WTE section.
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 89

 
P >NJV@
V
)  0:
:

 7 ƒ& 7 ƒ&
2*7 6+
S EDU
HY

 %

 /D\RXWB


/D\RXWB






       
4


4
(;+ (;+

4 >0:@
(;+

Fig. 6.15   Steam mass flow rate for Layout_1 and Layout_2 as a function of GT-discharged ther-
mal power

The T–Q diagram relative to Layout_2bis is shown in Fig. 6.18. In this case, with
respect to Layout_2, a lower HRSG-discharged temperature is obtained. Compar-
ing the steam mass flow rate production across the proposed layouts (Fig. 6.19), it
can be observed that the optimum condition of Layout_2bis (Bʹ) is within point Bʹʹ
(optimum value for Layout_2) and B (optimum value for Layout_1). Meanwhile,
for all the proposed configurations, the maximum amount of generated steam mass
flow rate remains the same.
The main results of Layout_2 and Layout_2bis are summarized in Table 6.6. It
can be noticed that corresponding to the optimum conditions (point Bʹ in Fig. 6.19),
Layout_2 reaches the minimum-allowed WTE exhaust gas temperature, while the
HRSG outlet temperature turns out to be considerably higher than the minimum
value allowed.
Layout_3 In the proposed layout, shown in Fig. 6.20, a parallel configuration
between the ECO sections is evaluated. Water, coming out of the DEA, is divided
into two streams: A fraction ( f) goes into ECO2, the ECO section inside HRSG,
90 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …


7>ƒ&@
DWRSWLPXP%

S  EDU
HY
([KDXVWJDV
7 ƒ&
 2*7

6+


6WHDPZDWHU


(&2



       


4 4 4>N:@
6+ (&2

Fig. 6.16   Layout_2 T–Q diagram for the HRSG section. SH superheater, ECO economizer

while the remaining part is sent to ECO1, inside WTE. Before entering the WTE
boiler, a mixer, combining both streams, is present.
Water mass flow rate splitting is adjusted to have similar mixer inlet tempera-
tures (or ECO outlet temperatures) in trying to maximize the exploitation of the
available thermal power, minimizing exhaust gas temperatures.
Corresponding to the optimum condition, for an evaporative pressure equal to
50 bar and a GT outlet temperature of 500 °C, the following water mass flow rate
splitting has been found: 70 % of the total water mass flow is sent to ECO2 inside
*
HRSG. Corresponding to the optimum QEXH , the assumed water mass flow rate
splitting allows one to reach the same minimum water subcooling temperature dif-
ference of both streams. Below the optimum value (point Bʹʹʹ, Fig. 6.21), due to a
decrease in QEXH , a slight decrease in f must occur, increasing the water mass flow
rate economized in the WTE section.
As highlighted in Fig. 6.21, the proposed configuration allows a considerable
* *
decrease in QEXH (about 11 % decrease in Layout_1). Moreover, for QEXH < QEXH ,
Layout_3 achieves the best performance in terms of the generated steam mass flow
rate.
Focusing on the T–Q diagram for Layout_3 (Fig. 6.22), a reduction in the dis-
tance between the exhaust gas and steam/water lines with respect to previous layouts
can be observed; moreover, water mass flow rate splitting has been also optimized
in order to obtain a parallelism between exhaust gas and water lines.
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 91

Fig. 6.17   Schematic of Layout_2bis waste-to-energy (WTE)–GT integrated plant. C compressor,


CC combined cycle, T turbine, SH superheater, ST steam turbine, ECO economizer

The main results of Layout_3 are summarized in Table 6.7. It can be noticed that
corresponding to optimum conditions, the WTE exhaust gas temperature is close to
its minimum value ( TO, WTE equals about 168 °C).
Layout_4  Starting with Layout_2, the possibility of further exploiting the HRSG-
discharged thermal power has been evaluated in the next proposed configuration,
namely, Layout_4. Starting with the results of Layout_2, in terms of HRSG outlet
temperature, the possibility to preheat waste combustion air has been investigated
in this layout. At the exit of the ECO section, a heat exchanger has been introduced
fed by HRSG-discharged thermal power. This solution allows eliminating the ST
bleed necessary for air preheating.
Layout_4 refers to the preheating of only the primary combustion air up to the set
temperature of 150 °C, while the temperature of the secondary combustion air is set
at 50 °C. Preheating of both primary and secondary combustion air has been investi-
92 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …


7>ƒ&@
#RSWLPXP%

S  EDU
HY
([KDXVWJDV
7 ƒ&
 2*7

6+


6WHDPZDWHU



(&2


       


4 4
6+ (&2 4>N:@

Fig. 6.18   Layout_2bis T–Q diagram for the HRSG section. SH superheater, ECO economizer

gated (namely, Layout_4_tot_air), where both temperatures (primary and secondary


combustion air) have been set equal to 150 °C.
Focusing on Fig. 6.23, an increase in the maximum amount of steam mass flow
rate has been observed in both cases, equal to about 1 and 3.5 %, respectively, for
Layout_4 and Layout_4_tot_air. For both layouts, an increase in steam mass flow
rate (points BIV) is achieved with respect to Layout_1 and Layout_2. This can be
explained considering the increase in combustion air temperature which, for the
considered layouts, reaches 150 °C with an increase equal to 20 °C on previous con-
figurations (see Table 6.2) and increases the reactants’ enthalpy, therefore increas-
ing the combustion temperature.
As shown in Table 6.8, the increase in maximum steam mass flow rate and the
elimination of ST bleed for air preheating allow an increase in ST power output.
Layout_5  In the proposed layout (namely, Layout_5, shown in Fig. 6.24) the pos-
sibility of introducing an integral DEA inside the HRSG section has been analyzed,
thus eliminating the need for a ST bleed to feed it. Water coming out of the con-
denser is sent to a pre-ECO placed in the HRSG, feeding the integral DEA (EVA
DEA). Water mass flow rate out of the deaerator is then divided into two streams to
be economized in parallel. Concerning water splitting between ECO1 and ECO2,
similar considerations as those for Layout_3 were maintained.
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 93

 
P >NJV@
V
)  0:
:

 7 ƒ& 7 ƒ&
2*7 6+
S EDU
HY

 %

/D\RXWB %
/D\RXWBELV



/D\RXWB






       

4 >0:@
(;+

Fig. 6.19   Steam mass flow rate for Layout_1, Layout_2, and Layout_2bis as functions of GT-
discharged thermal power

Table 6.6   Layout_2 and Layout_2 Layout_2bis


Layout_2bis results
FW (MW) 100 100
pev (bar) 50 50
TO, GT (°C) 500 500
TSH (°C) 480 480
m EXH (kg/s) 142 145
Q *EXH (MW) 81.91 83.64
Q *EXH 0.82 0.84
FW
(−)
m s, max (kg/s) 49.12 49.12
TO, HRSG (°C) 181 141
TO, WTE (°C) 160 246
PST (MW) 45.76 45.76
94 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.20   Schematic of Layout_3 waste-to-energy (WTE)–GT integrated plant. C compressor, CC


combined cycle, T turbine, SH superheater, ST steam turbine, ECO economizer

In Fig. 6.25, the steam mass flow rate produced in Layout_5 as a function of GT-
discharged thermal power is shown. Comparing results of Layout_5 with other con-
*
figurations, an increase in the optimum value of Q EXH (point BV) is found. More-
*
over, for Q EXH < Q EXH , Layout_5 has the lowest steam mass flow rate production.
When the integrated system is operating below the optimum condition, water split-
ting between ECO2 and ECO1 deviates from the condition of point BV (70 % of the
total water mass flow to ECO2). A decrease in Q EXH decreases the water mass flow
rate that ECO2 can manage. Corresponding to the lowest considered value of Q EXH,
only 28 % of the total mass flow rate is fed to ECO2.
Focusing on the T–Q diagram in Fig. 6.26, the proposed configuration corre-
sponding to the optimum plant match condition (point BV) allows one to achieve the
lowest HRSG outlet temperature ( TO, HRSG is about 112 °C). Besides, the presence of
the integral DEA inside HRSG increases the distance between the exhaust gases and
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 95

 
P >NJV@ )  0:
V
:
7 ƒ& 7 ƒ&
2*7 6+

S EDU
HY

 %

%
/D\RXWB


/D\RXWB


 /D\RXWB




       

4 >0:@
(;+

Fig. 6.21   Steam mass flow rate for Layout_1, Layout_2, and Layout_3 as functions of GT-dis-
charged thermal power

steam/water lines. As shown in Table 6.9, the elimination of the ST bleed to feed the
DEA allows an increase in ST power output.
Layout_6 In Layout_6, as shown in Fig. 6.27, there is a ST bleed to preheat
feedwater (heat exchanger) before it enters the ECO. Out of the DEA, water econo-
mization is first performed in WTE (ECO1), then inside the HRSG section (ECO2).
The ST bleed for water preheating is set at 1.1 bar. This value is chosen so that
the ST power output is not compromised and the T–Q diagram for HRSG is im-
proved. Even in this layout, an integral DEA (EVA DEA) is present.
Starting with a DEA pressure of 2 bar, the working pressure has been changed
to improve the HRSG T–Q profile (minimize the distance between exhaust gas and
steam/water lines and decrease HRSG outlet temperature); an optimized pressure
value of 3.5 bar has been found. The configuration obtained with the new value of
the DEA pressure refers to Layout_6bis. T–Q diagrams relative to Layout_6 and
Layout_6bis are shown in Figs. 6.28 and 6.29, respectively.
*
As shown in Fig. 6.30, the improved layout leads to a decrease in QEXH optimum
value (point BVI bis < BVI) and to an increase in generated steam mass flow for
Q EXH < Q *EXH (Table 6.10).
96 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …


7>ƒ&@
DWRSWLPXP%

S  EDU
HY
([KDXVWJDV
7 ƒ&
 2*7

6+


6WHDPZDWHU



(&2



       


4 4 4>N:@
6+ (&2

Fig. 6.22   Layout_3 T–Q diagram for HRSG section. SH superheater, ECO economizer

Table 6.7   Layout_3 results Layout_3


FW (MW) 100
pev (bar) 50
TO, GT (°C) 500
TSH (°C) 480

m EXH (kg/s) 128


73.83
Q *EXH (MW)
Q *EXH 0.74
FW
(−)
m s, max (kg/s) 49.12
TO, HRSG (°C) 140
TO, WTE (°C) 168
PST (MW) 45.76
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 97

 
P >NJV@ )  0:
V
: /D\RXWBBWRWBDLU
7 ƒ& 7 ƒ&
2*7 6+

S EDU ,9
HY %

 %
,9 /D\RXWB
%

/D\RXWB %

%




/D\RXWB

 /D\RXWB




       

4 >0:@
(;+

Fig. 6.23   Steam mass flow rate for different layouts as functions of GT-discharged thermal power

Table 6.8   Layout_4 and Layout_4 Layout_4_tot_air


Layout_4_tot_air results
FW (MW) 100 100
pev (bar) 50 50
TO, GT (°C) 500 500
TSH (°C) 480 480
m EXH (kg/s) 144 148
83.06 85.37
Q *EXH (MW)
Q *EXH 0.83 0.85
FW (−)

m s, max (kg/s) 49.57 50.84

TO, HRSG (°C) 158 144


TO, WTE (°C) 160 160
PST (MW) 46.69 47.88
98 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

1DWXUDO JDV
&&

& 7

$LU
4(;+ 72*7

6+

67

:7(
ERLOHU (&2  0L[HU
6HFRQGDU\
): I
DLU

ZDVWH

(9$ '($ 6SOLWWHU


3ULPDU\
(&2  DLU
([KDXVW JDV

(&2
72:7(

([KDXVW JDV
72+56*
([KDXVW JDV I

Fig. 6.24   Schematic of Layout_5 waste-to-energy (WTE)–GT integrated plant. C compressor, CC


combined cycle, T turbine, SH superheater, ST steam turbine, ECO economizer, EVA evaporator,
DEA deaerator

Layout_7  In the last proposed layout, Layout_7, Fig. 6.31, for a one-pressure-level
integrated power plant, the possibility of generating a fraction of the total saturated
steam in the HRSG section is investigated. Thus, a more complex configuration is
proposed and investigated. The proposed layout can be interesting, in particular,
considering the transient behavior of the system, LHV deviation from the design
value, and the different system start-up times.
In Layout_7, an evaporative heat exchanger (EVA2) is placed in the heat re-
covery boiler unit; thus, a fraction of the total saturated steam is generated here.
Water coming out of the condenser, before entering the deaerator, is divided into
two streams: One part ( h) is preheated in the ECO section inside HRSG, while the
remaining ( 1-h) is sent to a heat exchanger fed by an ST bleed. Out of the DEA,
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 99

 
P >NJV@ )  0:
V
:
7 ƒ& 7 ƒ&
2*7 6+

S EDU
HY

 9
%

% %
/D\RXWB


/D\RXWB


/D\RXWB

/D\RXWB




       

4 >0:@
(;+

Fig. 6.25   Steam mass flow rate for Layout_1, Layout_2, Layout_3, and Layout_5 as functions of
GT-discharged thermal power

water economization is performed in parallel: one part ( f  ) is sent to the WTE ECO
1, while the other one ( 1-f  ) is sent to another ECO section inside HRSG (ECO
2); economized streams are mixed and then split again before they enter the EVA
sections. Splitting of water entering the ECO sections is adjusted to exploit all the
available heat inside the HRSG. Before entering the superheated heat exchanger,
saturated steam streams are mixed together.
Contrary to the previously analyzed configurations, for Layout_7 there is no
optimal condition in terms of Q EXH; indeed, by increasing the value of Q EXH the
total steam mass flow rate, the sum of steam generated in the WTE boiler and
in the HRSG section, always increases. However, a minimum value of GT-dis-
charged thermal power is required to have steam generation inside the HRSG
section. Figure 6.32 shows the steam mass flow rate generated in the WTE and
HRSG sections plus the total amount as function Q EXH ; while in Fig. 6.33, the total
amount of steam mass flow rate generated for Layout_7 is presented as a function
100 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …


7>ƒ&@
9
DWRSWLPXP%
S  EDU
HY
([KDXVWJDV
7 ƒ&
 2*7

6+



6WHDPZDWHU



(&2



'(* (&2

        


4 4 4 4>N:@
6+ (&2 (&2

4
'(*

Fig. 6.26   T–Q diagram for the HRSG section for Layout_5. SH superheater, ECO economizer

Table 6.9   Layout_5 results Layout_5


FW (MW) 100
pev (bar) 50
TO, GT (°C) 500
TSH (°C) 480

m EXH (kg/s) 155

89.41
Q *EXH (MW)

Q *EXH
0.89
FW (−)

m s, max (kg/s) 49.12

TO, HRSG (°C) 112


TO, WTE (°C) 168
PST (MW) 47.97
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 101

Fig. 6.27   Schematic of Layout_6 waste-to-energy (WTE)–GT integrated plant. C compressor, CC


combined cycle, T turbine, SH superheater, ST steam turbine, ECO economizer, EVA evaporator,
DEA deaerator, HE general heat exchanger

of GT-discharged thermal power. As highlighted in both figures, an increase in


Q EXH increases m s,HRSG ; thus increasing the total amount (m s,TOT ) of the generated
steam.
The proposed configuration allows one to achieve the minimum HRSG out-
let temperature (see the T–Q diagram of Fig. 6.34). Thus, with the above con-
figuration, the GT-discharged thermal power is completely exploited. Moreover, as
suggested in Table 6.11, the WTE outlet temperature reaches the minimum value
allowed. The T–Q diagram shown in Fig. 6.34 and the numerical results reported in
Table 6.11 refer to Q EXH equal to 89.4 MW; since, for Layout_7, it is not possible
to identify an optimum plant match condition, the highest value of GT-discharged
thermal power previously found (see Layout_5 optimum condition) has been taken
as a reference.
102 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …


7>ƒ&@
9,
DWRSWLPXP%
S  EDU
HY
([KDXVWJDV
7 ƒ&
 2*7

6+



6WHDPZDWHU



(&2



'(*
(&2
       
4 4 4 4>N:@
6+ (&2 (&2

4
'(*

Fig. 6.28   T–Q diagram for the HRSG section in Layout_6. SH superheater, ECO economizer


7>ƒ&@
9,
DWRSWLPXP%
ELV
3  EDU
([KDXVWJDV HY

 7 ƒ&
2*7

6+



6WHDPZDWHU



(&2



(&2
'(*
       
4 4 4 4>N:@
6+ (&2 (&2

4
'(*

Fig. 6.29   T–Q diagram for the HRSG section in Layout_6bis. SH superheater, ECO economizer
6.4 WTE–GT Proposed Layouts for a One-Pressure-Level HRSG 103

 
P >NJV@ )  0:
V
:
7 ƒ& 7 ƒ&
2*7 6+

S EDU
HY

 9, 9,
%

% ELV % %
/D\RXWB


/D\RXWBELV

/D\RXWB

 /D\RXWB




       

4 >0:@
(;+

Fig. 6.30   Steam mass flow rate for Layout_1, Layout_3, Layout_6, and Layout_6bis as functions
of GT-discharged thermal power

Table 6.10   Layout_6 and Layout_6 Layout_6bis


Layout_6bis results
FW (MW) 100 100
pev (bar) 50 50
TO, GT (°C) 500 500
TSH (°C) 480 480

m EXH (kg/s) 142 137


81.91 79.02
Q *EXH (MW)
Q *EXH
0.82 0.79
FW (−)

m s, max (kg/s) 49.12 49.12

TO, HRSG (°C) 143 130


TO, WTE (°C) 160 160
PST (MW) 46.72 46.72
104 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

Fig. 6.31   Schematic of Layout_7 WTE–GT integrated plant. C compressor, CC combined cycle,
T turbine, SH superheater, ST steam turbine, ECO economizer, EVA evaporator, DEA deaerator

6.5 Comparative Results of WTE–GT One-Pressure-Level


Integrated Layouts

The results of all the proposed one-pressure-level WTE–GT integrated layouts are
summarized in Figs. 6.35, 6.36, and 6.37, showing the maximum amount of steam
mass flow rate, the ST power output, and the ratio of steam to waste mass flow rate,
respectively, as a function of GT-discharged thermal power in correspondence to
optimum plant match conditions (points B) identified in the previous section.
As previously discussed, for the proposed configurations, Layout_3 reaches the
minimum value of Q *EXH. While Layouts 1, 2, 2bis, 5, 6, and 6bis attain the same
maximum steam mass flow rate generation, Layouts 4, 4_tot_air, and 7 lead to an
increase in the maximum amount of steam that a plant could produce.
As far as the ST power capacity is concerned, Layout_7 gives the best perfor-
mance, followed by Layout_5 and Layout_4_tot_air.
6.5 Comparative Results of WTE–GT One-Pressure-Level Integrated Layouts 105

P >NJV@   

P
)  0:

V+56*
:
 L ƒ& 7 ƒ&
V

2*7 6+

>NJV@

S EDU
HY


 

P P
V727
V:7(








P
V+56*



 
      
4 >0:@
(;+

Fig. 6.32   Steam mass flow rate for Layout_7 as a function of GT-discharged thermal power

 
P >NJV@
V )  0:
:
7 ƒ& 7 ƒ&
2*7 6+

S EDU
HY



%


/D\RXWB


 /D\RXWB




       

4 >0:@
(;+

Fig. 6.33   Steam mass flow rate for Layout_1 and Layout_7 as a function of GT-discharged ther-
mal power
106 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …


7>ƒ&@
#4  0:
(;+
([KDXVWJDV
S EDU
HY

 7 ƒ&
2*7

6+


6WHDPZDWHU

(9$



(&2


'(*

(&2

        


4 4 4 4 4 4>N:@
(9$ (&2 '(* (&2
6+

Fig. 6.34   T–Q diagram of HRSG section for Layout_7. SH superheater, ECO economizer, EVA
evaporator

Table 6.11   Layout_7 results Layout_7


FW (MW) 100
pev (bar) 50
TO, GT (°C) 500
TSH (°C) 480

m EXH (kg/s) 155

89.41
Q EXH (MW)

Q EXH
0.89
FW (−)

m s, max (kg/s) 50.70

TO, HRSG (°C) 112


TO, WTE (°C) 160
PST (MW) 48.95
6.5 Comparative Results of WTE–GT One-Pressure-Level Integrated Layouts 107



P >NJV@ )  0:
:
V 
S EDU
HY
/D\RXWBBWRWBDLU
 7  ƒ&
2*7

 /D\RXWB
/D\RXWB

/D\RXWB


/D\RXWBELV /D\RXWB
 /D\RXWB
/D\RXWBELV
 /D\RXWB
DQG
/D\RXWB



























4 >0:@
(;+

Fig. 6.35   Maximum steam mass flow rate (optimum condition, points B) as a function of GT-
discharged thermal power


3 >0:@ )  0:
67 :

S EDU
HY /D\RXWB
 7  ƒ&
2*7


/D\RXWBBWRWBDLU

/D\RXWB

/D\RXWB /D\RXWB
 /D\RXWBELV



 /D\RXWB
/D\RXWB
 /D\RXWBELV /D\RXWB




























4 >0:@
(;+

Fig. 6.36   Steam turbine power output as a function of GT-discharged thermal power
108 6  WTE–GT Steam/Waterside Integration: Thermodynamic Analysis …

 
P
 V >@  S EDU
HY
P 7  ƒ&
: 2*7 /D\RXWBBWRWBDLU


 /D\RXWB




/D\RXWB


 /D\RXWB /D\RXWBELV /D\RXWB

 /D\RXWB
 /D\RXWBELV
 /D\RXWB
DQG
 /D\RXWB



































4
(;+ >@
)
:

Fig. 6.37   Steam-to-waste mass flow rate ratio as a function of thermal power input ratio

The analysis shows that the steam mass flow rate produced by the integrated
plant, for all the proposed layouts, is higher than that generated with a tradition-
al WTE power plant fed with the same amount of waste. Figure 6.38 shows an
increment in steam mass flow rate and ST power output for each investigated
layout compared to a traditional WTE fed with the same thermal power input
( FW = 100 MW) and working at the same evaporative pressure and steam-super-
heated temperature equal to 50 bar and 480 °C, respectively. The steam mass flow
rate production of a stand-alone WTE would be equal to about 32.09 kg/s, while
the ST capacity would be equal to 29.7 MW; thus, increments range, for the con-
sidered cases, from 53 to 58 % for the steam mass flow and from 54 to 65 % for
the ST capacity.
Numerical results in terms of integrated system efficiency and power output
allocation to each input fuel will be discussed, in detail, in the next chapter.
References 109

0DVVIORZUDWH
67SRZHURXWSXW

/D\RXWB

/D\RXWBELV

/D\RXWB

/D\RXWB

/D\RXWBBWRWBDLU

/D\RXWB

/D\RXWB

/D\RXWBELV

/D\RXWB

/D\RXWB

          
LQFUHPHQW

Fig. 6.38   Increments in steam mass flow rate with respect to a traditional stand-alone WTE power
plant. ST steam turbine

References

1. Gate CycleTM by General Electric Energy. http://site.ge-energy.com/prod_serv/products/oc/


en/opt_diagsw/gatecycle1.htm
2. Lozza G Turbine a gas e cicli combinati, Progetto Leonardo, 2° Edizione (In Italian)
Part IV
Performance and Efficiency
Conversion Issues
Chapter 7
Performance Indexes and Output Allocation
for Multi-fuel Energy Systems

In this chapter, the concept of conversion efficiency by exploring multi-fuel (MF)


energy systems is discussed. An MF system can be defined as a system with various
fuel energy inputs and useful product outputs. The difficulty in defining a perfor-
mance index for MF systems lies in quantifying the contribution of each input fuel
to the total output energy.
This chapter intends to make a few proposals and start a discussion that would
help assess the MF system performance.
The conventional first law efficiency, normally used to assess the performance
of a single-fuel (SF) system, can be applied, but it provides incomplete information
about an MF system. The electric equivalent efficiency and the relative and overall
MF synergy index (SI) concepts are introduced as the most significant performance
indexes; these new performance indexes aim to evaluate the MF performance com-
pared with reference SF scenarios. Proposed performance indexes are analyzed, dis-
cussed, and compared from a general point of view, identifying aspects of relation
and the main differences. The introduced conversion efficiency indexes are then
applied to a specific co-combustion power plant case. The influence of the reference
scenarios, as it is highlighted in the chapter, is of great importance. The evaluation
of the proposed indexes is of relevance for the formulation of a specific and unified
theory about MF power systems, such as a waste-to-energy–gas turbine (WTE–GT)
integrated power plant.

7.1 Context

A trend towards energy supply diversification, based on an energy infrastructure


for which the feedstock supply no longer depends on a single fuel or network, is
needed to prevent reliability reductions in the provided output energy that may
occur in a market-oriented system. In addition, supply diversification provides
more degrees of freedom in selecting the supply source while also taking into ac-
count the availability of storage from various sources [1]. The diversification of

© Springer International Publishing Switzerland 2015 113


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_7
114 7  Performance Indexes and Output Allocation for Multi-fuel Energy Systems

energy feedstock can also provide, as a consequence, the push towards MF en-
ergy systems, using different fuels in the same power plant. Another reason that
makes the study particularly relevant is the increasing depletion of primary fos-
sil fuel resources. Due to their depletion and the contextual increase of renewable
sources [2, 3], MF energy systems are becoming attractive, and their penetration
is increasing in the energy market [4, 5], especially within external combustion
energy systems [6–9]. For example, MF energy systems based on co-firing of a
fossil fuel together with a limited fraction of renewable or waste-derived fuel can
be a strategy to use conventional power plants by introducing only modifications in
the combustion section and/or limited challenges in the boiler redesign [10]. More-
over, MF conditions can also be realized in renewable-energy-based [11] or waste-
to-energy (WTE) [12–14] power plants, which nevertheless need conventional fuels
in more or less limited periods of their operation.
In SF energy systems, where input fuel energy is converted into useful output
energy, the evaluation of conversion efficiency is a simple issue resulting from the
comparison between the generated output and the introduced input. By-products
of the conversion process, such as discharged heat, are disregarded and/or treated
as waste. For MF energy systems, a standardized performance index that is able
to evaluate both the overall conversion efficiency of the input into the total output
and the relative influence on performance of the single input is still under research.
The most problematic aspect concerning the MF system conversion efficiency is
that different fuels with different characteristics and value all contribute to the gen-
eration of the overall system energy output. Therefore, it is hard to measure each
individual input contribution to the global output. The problem of fuel valorization
can become very useful from a practical point of view, for example when one of the
input fuels of an MF power plant is renewable and the other is not. Therefore, it is
important to understand which fraction of the output can be recognized as “clean,”
and subsequently opening it to incentives or special tariff regimes. Another problem
can arise when the integration between two or more fuels (e.g., one renewable and
another natural gas) is accomplished in a plant that, due to the MF arrangement,
cannot achieve the combined-cycle-state-of-the-art efficiency values close to 60 %.
In this case, the question that can be raised is: “Is the integrated MF arrangement
energetically convenient?” Generally, from a theoretical point of view, it should be
useful to identify comparison indexes, representing “best” or “reference” or at least
theoretical limiting efficiency values in order to stimulate the technical develop-
ment towards significant improvements in MF plants. This chapter puts forward
some new definitions and discusses the conversion efficiency of an MF energy sys-
tem from different points of view. Absolute and comparative performance indexes
are presented to evaluate and compare conversion efficiency of an MF system with
respect to SF systems fed with the same amount of input energy. Eventually, exam-
ples of MF systems are given to highlight their potential advantages in comparison
with separate SF systems. Moreover, the analyzed cases show how the selection of
the reference SF scenario can affect the convenience of the MF solutions.
7.2 Performance Evaluation of an MF Energy System 115

7.2 Performance Evaluation of an MF Energy System

The complexity of the issues related to MF energy systems calls for synthetic indi-
cators to characterize and define the performance of the system. In particular, the
proposed approach considers the energy system as a black box. Using this black-
box approach, it is possible to build an equivalent performance model on the basis
of only input–output energy flows, without considering the detailed representation
of the internal components.
For notation purposes, the various energy entries are calculated over a given time
span (for instance, hourly, daily, annual) depending on the purpose of the study.
Moreover, this discussion is restricted to energy systems in which flows are steady.

7.2.1 MF Energy System Arrangement

A general MF energy system is shown in Fig. 7.1 as a black box, fed by n fuel in-
puts ( Fi is the ith associated energy), which produce a useful electric energy output
( E), while rejecting a non-useful heat ( Qnu ). Moreover, within the framework of a
general-purpose approach, heat exchanges between the MF system and the exter-
nal ambient or with other systems are taken into account in this study. Indeed, as
schematically shown in Fig. 7.1, both input heat and output heat are considered. In
particular, the system can receive additional input heat ( Qin ), and it can deliver use-
ful recoverable heat (Qrec ). The control surface, surrounding the MF energy system,
can be applied to single components constituting the power plant or the whole sys-
tem, to evaluate, respectively, the subsystem and/or the overall energy system con-
version efficiency. The energy system input Fi can be multiple and from different
types of sources such as (i) primary fuels, directly available in nature (e.g., natural
gas, coal, oil, biomass, etc.) and (ii) non-primary fuels (energetically expensive and

Fig. 7.1   Basic multi-fuel


(MF) energy system
116 7  Performance Indexes and Output Allocation for Multi-fuel Energy Systems

not directly available in nature, e.g., H2, syngas, methanol, etc.). In more detail, in
the case of primary fuels, the expression of Fi is the following:

Fi = mi ·LHVi , (7.1)

where, for each input source, mi represents the fuel mass flow and LHVi represents
the lower heating value per unit of mass flow at the ambient temperature (T0 ) at
which the ith reactant enters. For non-primary fuel input cases, the term Fi repre-
sents the primary fuel energy ideally necessary to produce the actual input. In the
case of high-temperature input streams, the relating enthalpy content is given by
the term Qin .

7.2.2 Indexes for MF Energy System Performance Evaluation

In order to evaluate the efficiency of an MF energy system, different performance


indicators can be used, as described below.

7.2.2.1 First Law Efficiency

As a result of the laws of thermodynamics, by comparing useful output energy and


input energy, the following index can be introduced:
First law efficiency:

E + Qrec .  
ηI = n (7.2)
∑ Fi + Qin
i =1

According to this definition, the useful output heat has the same importance as the
electric output. This definition can be misleading as it does not consider the actual
value of heat at different temperature levels.

7.2.2.2 Electric Equivalent Efficiency

In order to consider the quality of the useful input and output heat, including their
contribution to the products, the electric equivalent of heat ε must be considered.
Thus, in the proposed electric equivalent efficiency definition, the useful heat con-
tribution is expressed as +(ε rec · Qrec ) , while, with the same methodology, the inlet
heat contribution can be expressed as −(ε in · Qin ) and can be considered a negative
electric equivalent term; ε rec and ε in are the electric equivalent conversion fac-
tors of the heat terms, which depend on the temperature of the heat-exchanging
7.2 Performance Evaluation of an MF Energy System 117

Fig. 7.2   Electric equivalent


of heat, ε, as a function of the ε
temperature level, TQ, of the
heat source for different heat
recovery systems. CC2PL
combined cycles with two-
pressure-level arrangement,
CC1PL combined cycles with
one-pressure-level arrange-
ment, ORC organic Rankine
cycles, IBC inverted Brayton
cycles

process. Therefore, the MF performance can be evaluated according to the follow-


ing definition:
Electric equivalent efficiency:

E + ε rec · Qrec − ε in · Qin U
ηEE = n
= n
(7.3)
∑ Fi ∑ Fi
i =1 i =1

where U represents, with a compact notation, the total useful electric output, thus
including the equivalent heat contributions. Different possibilities can be chosen
to quantify the value of ε for each heat contribution, considering that the upper
theoretical limit to ε is represented by the Carnot efficiency (ηC ). Figure 7.2 shows
the trend of ε versus temperature level TQ of the heat source for different exist-
ing and innovative energy conversion systems, namely, combined cycles with two-
pressure-level arrangement (CC 2 PL), combined cycles with one-pressure-level
arrangement (CC 1 P L), organic Rankine cycles (ORC) with octamethyltrisilox-
ane (MDM) as the working fluid [7], Stirling engines, and inverted Brayton cycles
(IBC) [6], where ηClm is the efficiency of an ideal reversible recuperation cycle with
limited capacity of the heat source.1

An isentropic expansion and an isothermal compression is made up of an isobaric heat absorption


1 

during the process of heat exchange. The hot source temperature decreases from TQ to T0.
118 7  Performance Indexes and Output Allocation for Multi-fuel Energy Systems

Fig. 7.3   a Multi-fuel


(MF) energy system versus
b single-fuel (SF) energy
system

7.2.2.3 Relative SI

A different approach to measure the MF system performance is to quantify im-


provements obtained with the MF integrated system in comparison with separate
exploitation of the various fuel inputs. Thus, a comparative energy conversion sce-
nario must be taken into account. This approach is clearly described in Fig. 7.3,
where the MF energy system (Fig. 7.3a) and the comparative scenario (Fig. 7.3b)
are presented.
The improvement in the integrated MF configuration, in comparison with the
separate and SF ones, is evaluated as being equal to all of the ith fuel energy input.
A comparative index, named relative synergy of the ith source SIi, is introduced
here. This index tries to express the benefit of the ith fuel in the MF integrated plant
by excluding the effects of all the other fuels and using the reference scenario to
measure the useful output. In order to quantify the contribution of the ith source, the
idea is to subtract from the total MF output (U ) the output of all the reference SF
n
systems (USF, j in Fig. 7.3) except the ith is ∑ U SF, j . This quantity allows one to
j =1, j ≠ i
calculate SIi.
Relative SI:
n

U− ∑ U SF, j
(7.4)
j =1, j ≠ i
SIi =
Fi
7.2 Performance Evaluation of an MF Energy System 119

To evaluate the output produced by an SF system USF, j, it is necessary to identify


the specific SF reference system and in particular the characteristics of conversion
efficiency. Thus, the generic USF, j can be expressed as

U SF , j = η SF, j ·F j , (7.5)

where ηSF, j represents the generic jth SF system conversion efficiency. Therefore,
Eq. (7.4) can be rewritten as
n

U− ∑ (ηSF, j ·F j )
j =1, j ≠ i
SIi = . (7.6)
Fi

The selection of an appropriate value of each ηSF, j can severely affect the advantages
or disadvantages of the integrated system, as shown by a numerical example. Several
possibilities of estimating the reference efficiency can be considered. For example,
the “Best Available Technologies (BAT) Reference Document” (BREF) [15] can be
used. For some of the most common fuels, the reference efficiencies are reported in
[15] (Table 7.1) as a function of the combustion technology or the plant type.

7.2.2.4 MF SI

To evaluate the overall MF system performance and to give information about the
synergy effect of all the input fuels in comparison with the SF systems of Fig. 7.3,
the following index is also introduced:
MF SI:
n

U − ∑ U SF,i
i =1
SI = n
. (7.7)
∑USF,i
i =1

Evidently, when SI > 0 , the integrated MF energy system is better than the sum of
the SF ones; the synergy effect due to the configuration of the plant and the con-
textual fuel exploitation increases the useful output in comparison with the overall
output of the SF reference scenario. Otherwise, when SI ≤ 0 , the integrated con-
figuration is equal to or worse than the collection of separate and SF systems.
The SI evaluates the overall benefit of the exploitation of all the input streams
inside the integrated system differently from the relative SI, even though it is dif-
ficult to highlight and quantify the contribution of each SF source in terms of use-
ful energy output. Nevertheless, it is possible to correlate both SI and ηEE to the
ensemble of SIi, according to the following equations:
120 7  Performance Indexes and Output Allocation for Multi-fuel Energy Systems

Table 7.1   Efficiency associated with BAT for different fuels as a function of the combustion
technique [15]
Fuel Combustion technique ηSF (%)
New plants Existing plants
Coal Pulverized combustion 43–47 35–40
Fluidized bed > 41
Pressurized fluidized bed > 42
combustion
Biomass Grate firing 20 –
Spreader-stoker > 23
Fluidized bed combustion > 28–30
Lignite Pulverized combustion 42–45 35–40
Fluidized bed > 40
Pressurized fluidized bed > 42
combustion
Peat Fluidized bed combustion > 28–30 –
Plant type ηSF (%)
New plants Existing plants
Gas turbine 36–40 32–35
Gas engine 38–45 –
Gas-fired boiler 40–42 38–40
CC with or without supplementary firing 54–58 50–54
(HRSG) for electricity generation
CC combined cycle, HRSG heat recovery steam generator

 n 
 ∑ SIi ·Fi 
1  i =1
(7.8)
SI = − 1
n n 
 ∑ηSF,i ·Fi 
 i =1 

1 n n 
 η EE =  ∑ SI i ·xi + ( n − 1) ∑η SF ,i ·xi  (7.9)
n  i =1 ,
i =1 

n
where xi = Fi / ∑ Fi represents a fuel input weighting factor of the ith relative SI.
i =1 n
Due to the SIi definition, when SI > 0 (i.e., U > ∑ U SF,i ) the following inequal-
i =1
ity occurs:
n n
∑ SIi Fi > U > ∑USF,i ,
(7.10)
i =1 i =1
7.2 Performance Evaluation of an MF Energy System 121

which means that, in this case, the sum of the energies associated with the various
SIi is higher than the sum of the output energies of the SF reference scenario.

7.2.3 Useful Output Allocation to Each ith Fuel

Besides the issue of conversion efficiency assessment of an MF energy system, there


is another aspect with practical implications to be solved: What is the contribution
(U i ) of each fuel input to the total output (U ) ? How can it be properly quantified?
Different approaches can be used to allocate output energy to different fuels.

7.2.3.1 Allocation Approach #1

The first proposed approach for output allocation to fuel input consists in using the
following equation:
Fi
Ui = U· = ηEE ·Fi .
 n (7.11)
∑ Fi
i

Thus, according to this approach, each U i is simply proportional to the ith fuel
input. As shown in the numerical example, this approach can be misleading when
input fuels have different characteristics and value because, using Eq. (7.11), all
fuels have the same weighting factor (namely, ηEE ).

7.2.3.2 Allocation Approach #2

An alternative evaluation method is proposed here to account for the actual conver-
sion efficiency gain of each fuel in comparison with the SF reference scenario; the
new allocation formula is
SIi ⋅ Fi .  
Ui = U ⋅
 n (7.12)
∑ SIi ⋅ Fi
i

According to this definition, each fuel input is weighted via the corresponding SIi
term. In this way, the input fuel with higher conversion efficiency in comparison
with the SF reference scenario (i.e., higher SIi) can be identified with making a
higher contribution to the global output. Thus, this method privileges the fuel which
is better exploited in the MF system in comparison with the SF scenario, whereas
Approach #1 does not include such reference to the SF scenario.
122 7  Performance Indexes and Output Allocation for Multi-fuel Energy Systems

Fig. 7.4   Two-fuel co-


combustion power plant (a)
versus separate combustion
energy systems (b)

The various efficiency definitions and allocation approaches can be applied to


obtain useful information about the system convenience under different operating
conditions or to compare different plant design arrangements. An application is pre-
sented in the following section.

7.3 Application Example: Two-fuel Co-combustion Power


Plant

A specific case in which the above general methodology for MF systems perfor-
mance evaluation can be applied is the simple case of co-combustion of two differ-
ent fuels in a generic power plant. The scheme of the system is shown in Fig. 7.4a,
where F1 and F2 are the fuel input energy values. In this case, the reference sepa-
rate energy systems, exploiting the same two input fuels, are considered, as shown
in Fig. 7.4b.
In particular, the co-combustion power plant produces the useful electric output
E, while the two separate reference SF systems produce, respectively, ESF,1 and
ESF,2 . In this co-combustion case, the first law efficiency (also coincident with ηEE )
according to Eq. (7.2) is
 E .
ηI =  (7.13)
F1 + F2

Evaluated in comparison with the reference scenario, the MF SI according to


Eq. (7.7) is
 E − ( ESF,1 + ESF,2 ) .
SI = (7.14)
( ESF,1 + ESF,2 )

Finally, the relative SI referred to the two separate fuel inputs according to Eq. (7.6)
are
7.3 Application Example: Two-fuel Co-combustion Power Plant 123

 E − ESF,2 E − F2 ·ηSF,2
SI1 = = (7.15)
F1 F1

E − ESF,1 E − F1·ηSF,1
(7.16)
SI 2 = = ,
F2 F2

where ηSF,1 and ηSF,2 are the efficiencies of the two reference SF separate plants.
In order to obtain a global energy benefit from the co-combustion, the condition
SI > 0 leads to
 E > ( ESF,1 + ESF,2 ).  (7.17)

This condition can be rearranged as a function of the co-combustion first law


­efficiency and of the relative SI of the two reference SF systems:

 F1 (η I − ηSF,1 ) + F2 (η I − ηSF,2 ) > 0, (7.18)

which is, first of all, verified when ηI > ηSF,1 and ηI > ηSF,2 ; otherwise, for
ηSF,2 > ηSF,1, the condition (7.18) can be also verified if the following relation is
satisfied:

 F2 ηI − ηSF ,1 . (7.19)
<
F1 ηSF ,2 − ηI

This equation leads to a simple link between the co-combustion fuel ratio F2 / F1,
the SF reference efficiencies, and a minimum first law conversion efficiency value
ηI _ min , which is required to have energy convenience (i.e., SI ≥ 0). This link is
shown for two different reference SF efficiency cases (Fig. 7.5) corresponding to
typical biomass and coal reference plants, with high- and low combustion technol-
ogy levels, respectively, according to the values reported in Table 7.1, where fuel F1
represents biomass and F2 represents coal. The figure clearly shows that the conve-
nience of the co-combustion plant is linked to the choice of the reference scenarios.
The graph shows that to have convenience (i.e., a synergic effect) the first law ef-
ficiency of the co-combustion MF system must be higher than the weighted average
of the two SF systems’ efficiency, using the input fuel energy terms as weights.
Finally, in this co-combustion case the global convenience condition, that is
SI > 0 can be linked to the relative SI of each single fuel as follows:

 SI > 0 ⇒ ( E − ESF,1 > ESF,2 ) / F2 ⇒ SI 2 > ηSF,2 (7.20)

 SI > 0 ⇒ ( E − ESF,2 > ESF,1 ) / F1 ⇒ SI1 > ηSF,1.  (7.21)


124 7  Performance Indexes and Output Allocation for Multi-fuel Energy Systems


η
,BPLQ η6)  η 6) 

η6)   η 6) 

η 6)

0)FRQYHQLHQW

 % $
0)QRWFRQYHQLHQW
& IRUKLJKFRPE
 WHFKOHYHO

η
6)
 0)QRWFRQYHQLHQW


     
)  )  ) ) ) 
    

Fig. 7.5   Minimum total efficiency to have SI = 0 for a biomass–coal typical co-combustion power
plant versus F1 / ( F1 + F2 ) . MF multi-fuel

These two conditions basically affirm that in the co-combustion plant both fuels are
exploited better than in the two reference SF plants. Thus, the condition of relative-
SI for each fuel entering the MF system higher than the corresponding SF system
efficiency is equivalent to global MF convenience in terms of total output energy.
Table  7.2 shows numerical data of three co-combustion cases (corresponding
points A, B, and C shown in Fig. 7.5) of biomass with coal used in the same power
plant, with given size and first law efficiency; two MF cases (A and B) show syn-
ergy convenience and one (C) is not convenient. Due to inconvenience of case C
(SI < 0 ), the evaluation of relative SIs and output allocation Approach #2 does not
give meaningful results in this case.
The convenience of co-combustion is strictly related to the ratio between the two
fuels and the different reference conversion efficiencies. The contribution of each
input fuel to the total output is calculated according to the two above-mentioned
approaches. In particular, using Approaches #1 and #2, the following expressions
are obtained:
Approach #1:

(7.22) F1 F2
E1 = E · ; E2 = E ·
F1 + F2 F1 + F2

Approach #2:

 F1·SI1 F2 ·SI 2 .  (7.23)


E1 = E · ; E2 = E ·
F1·SI1 + F2 ·SI 2 F1·SI1 + F2 ·SI 2
7.4 Conclusions 125

Table 7.2   Example of MF co-combustion performance


MF size and E (MW) 100
efficiency 33.3 %
ηI

SF reference ηSF,1 20 % (biomass)


efficiency
ηSF,2 35 % (coal)

Co-combustion Cases A B C
MF performance F1 (MW) 100 50 10
data
F2 (MW) 200 250 290
SI > 0 OK OK NO
SI1 0.30 0.25 –
SI2 0.40 0.36 –
Approach #1
E1 (MW) 33.3 16.7 3.3
E2 (MW) 66.7 83.3 96.7
Approach #2
E1 (MW) 27.3 12.2 –
E2 (MW) 72.7 87.8 –
MF multi-fuel, SF single-fuel

The obtained results show that by adopting Approach #2, the relative contribution
to the total output of fuel F1 (biomass, which is characterized by a lower SF conver-
sion efficiency) decreases in comparison with Approach #1.

7.4 Conclusions

In this chapter, a few proposals have been made to evaluate the performance of an
MF system and to quantify the contribution of each input fuel to the system output.
In particular, the concept of relative SI expresses the benefit of each input fuel to
the MF system by excluding the effects of all other fuels in comparison with a
reference scenario. On the other hand, the global MF SI evaluates the overall per-
formance, giving information about the synergy effect of all the input fuels. The
proposed indexes allow one (i) to investigate the best way to measure the MF inte-
grated system conversion efficiency; (ii) to measure the benefit of the integration
in comparison with the SF scenario; (iii) to establish a criterion for the selection of
the best configuration of integrated MF systems, for example, when various inte-
grated plant layouts have to be compared. Moreover, two different approaches have
been discussed to allocate the useful output to each input fuel. The introduced effi-
ciency indexes have been discussed and applied to a specific co-combustion power
126 7  Performance Indexes and Output Allocation for Multi-fuel Energy Systems

plant case. It has been highlighted that the convenience of the co-combustion case
is linked to the choice of the reference scenario. Besides co-combustion, the intro-
duced performance assessment methodologies can also be applied in the case of
topper–bottomer-integrated hybrid MF energy systems.

References

  1. IEA World Energy Outlook (2010) http://www.worldenergyoutlook.org/media/weo2010.pdf.


Accessed June 2013
  2. Branchini L, Perez-Blanco H (2012) Handling wind variability using gas turbines. Proceed-
ings ASME Turbo Expo. 6. pp 727–734. doi:10.1115/GT2012-68045
  3. Branchini L, Perez-Blanco H (2012) Computing gas turbine fuel consumption to firm up
wind power. Proceedings of ASME Turbo Expo 2012, Vol 6. pp 735–741; doi:10.1115/
GT2012-68046.
  4. Al-Mansoura F, Zuwala J (2010) An evaluation of biomass co-firing in Europe. Biomass
Bioenergy 34(5):620–629
 5.  Savolainen K (2003) Co-firing of biomass in coal-fired utility boilers. Appl Energy
74(3–4):369–381
  6. Bianchi M, De Pascale A (2011) Bottoming cycles for electric energy generation: parametric
investigation of available and innovative solutions for the exploitation of low and medium
temperature heat sources. Appl Energy 88(5):1500–1509
  7. Branchini L, De Pascale A, Peretto A (2013) Systematic comparison of ORC configura-
tions by means of comprehensive performance indexes. Appl Therm Eng 61(2):129–140.
doi:10.1016/j.applthermaleng.2013.07.039
  8. Bianchi M, Branchini L (2012) Advanced waste-to-energy power plant: integration with gas
turbine. International Conference on Applied Energy, ICAE 2012, July 5–8 2012, Suzhou,
China, Paper ID: ICAE2012–A10330
  9. Campana F, Bianchi M, Branchini L, De Pascale A, Peretto A, Baresi M, Rossetti N, Fermi A,
Vescovo R (2013) ORC waste heat recovery in European energy intensive industries: energy
and GHG savings. Energy Convers Manag 76:244–252
10. Tillman DA (2000) Biomass cofiring: the technology, the experience, the combustion conse-
quences. Biomass Bioenergy 19(6):365–384
11. Donatini F, Zamparelli C, Maccari A, Vignolini M (2007) High efficiency integration of ther-
modynamic solar plant with natural gas combined cycle. International Conference on Clean
Electrical Power, ICCEP ’07
12. CNIM and SENER (2005) Bizkaia waste to energy plant project, summary report, Contract
No BM 80/96 ES-PO
13. Unda JI (2009) Steps forward in energy upgrading of waste to energy plants (WTE), Dubai,
May 27th 2009 Middle East Waste Summit
14. Bianchi M, Branchini L, De Pascale A, Falchetti M, Fiore P (2013) Advanced waste-to-ener-
gy steam cycles. 68th Conference of the Italian Thermal Machines Engineering Association,
ATI2013. To be published in Energy Procedia
15. European Commission (2006) Integrated pollution prevention and control reference docu-
ment on the best available techniques for waste incineration, August 2006
Chapter 8
Specific Application Cases with GT
Commercial Units

In this final chapter, detailed application cases of waste-to-energy (WTE) integrated


with different commercial gas turbine (GT) units are shown and discussed. Detailed
modifications of the WTE cycle and the resulting enhancement of its performance
are presented. The analysis shows that combining WTE and GTs provides power
output increase of up to 80 % and a steam turbine (ST) mass flow increase of up to
50 %, compared with a state-of-the-art midsize reference WTE with an efficiency
of 27 %. A first law efficiency increase in the range of 8–15 percentage points
can occur, depending on the GT model and layout selection. Results of waste-to-
energy–gas turbine (WTE–GT)-specific application cases suggest issues and useful
guidelines to (i) create new advanced WTE–GT integrated power plants or to (ii)
repower existing low-performing WTE power plants to increase waste conversion
into energy.

8.1 Midsize WTE Reference Steam Cycle

In order to perform a numerical investigation on the potential of a WTE–GT inte-


grated power plant, a reference stand-alone WTE has been defined. This standard
state-of-the-art plant represents the starting point for the implementation of the pro-
posed integration with GT. The following aspects have been taken into account
and fixed in order to characterize the reference WTE: (i) a given size in terms of
municipal solid waste (MSW) input capacity, (ii) a given thermodynamic arrange-
ment in terms of layout, and (iii) a standard level of steam cycle parameters of the
WTE. The stand-alone WTE size and performance data are collected in Table 8.1. A
schematic layout of the reference WTE steam cycle is shown in Fig. 8.1. Assumed
steam cycle thermodynamic parameters are summarized in Table 8.1.
The assumed capacity, equal to 400 t/day of input MSW, is typical of a medi-
um-sized WTE plant, and it is in line with the average values of data presented in
Chap. 4 (see Fig. 4.1).

© Springer International Publishing Switzerland 2015 127


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0_8
128 8  Specific Application Cases with GT Commercial Units

Table 8.1   Main data of the reference stand-alone WTE plant


Reference WTE size and capacity data
Daily capacity [t/d] 400
Annual capacity [t/y] 135,000
MSW mean LHV [MJ/kg] 11.9
Thermal capacity [MWt] 55
Number of lines 2
Combustion system Grate
Gross output power [MW] 15.7
Net electric power output [MW] 14.9
Net electric efficiency [%] 27.3
WTE waste-to-energy, MSW municipal solid waste, LHV lower heating value

A superheated steam cycle without reheating and with steam extractions has been
considered. This adopted cycle layout arrangement, as already detailed in previous
chapters, is also quite common for medium-sized WTE power plants [1].
As shown in Fig. 8.1, the superheated steam generated inside the WTE boiler,
coming from the Superheater (SH) bank located in the convective section of the
boiler, is sent to the ST that features two ST bleeds: One bleed (at higher pressure)
is used to feed the deaerator (DEA) and to complete the primary air preheating
before combustion; the second bleed (at lower pressure) is used in the feedwater
heat exchanger (HE). Water cooling, in closed loop, is used for combustion grate.
After passing through the grate, the hot water is used to preheat the primary and
secondary inlet air streams. The economizer (ECO) is located at the end of the WTE

67

(9$
(9$

6+ +LJK SUHVVXUH
67 EOHHG

(9$
&RQGHQVHU
:DVWH

(&2 /RZ SUHVVXUH


67 EOHHG

7R IOXH +,JK SUHVVXUH


&
JDV FOHDQLQJ 67 EOHHG
EOHHG
FRQGHQVHU
6HFRQGDU\ 72:7(
DLU
*UDWH +LJK SUHVVXUH
FRROLQJ 67 EOHHG
+(
'($
3ULPDU\
& FRPEXVWLRQ
DLU
7R '($

Fig. 8.1   Schematic of the reference stand-alone WTE plant. EVA evaporator, SH superheater,
ECO economizer, DEA deaerator, ST steam turbine, HE general heat exchanger
8.1 Midsize WTE Reference Steam Cycle 129

Table 8.2   Steam cycle thermodynamic parameters of the reference stand-alone WTE plant
Steam turbine data
Steam mass flow rate [kg/s] 18.9
Turbine inlet temperature [° C] 390
Turbine inlet pressure [bar] 43
Turbine isentropic efficiency [-] 0.9
ST high pressure bleed [bar] 3.5
ST low pressure bleed [bar] 1.35
Condenser pressure [bar] 0.1
WTE boiler data
Evaporation pressure [bar] 49
Steam-superheated temperature [° C] 395
Primary air inlet temperature [° C] 120
Secondary air inlet temperature [° C] 50
Oxygen vol. dry fraction in exhaust gases [%] 7.5
FW outlet temperature [° C] 100
DEA pressure [bar] 3
Exhaust gas temperature [° C] 164
Recirculated exhaust gas rate [%] 15
Recirculated exhaust gas temperature [° C] 150
Steam cycle auxiliary consumption [MW] 0.9
Pressure losses in each heat exchanger [%] 2

boiler convective section, while the evaporator (EVA) is split in different HE banks,
located in both the radiating and the convective sections of the boiler.
The steam cycle thermodynamic arrangement, the fluid properties, and the over-
all performance of this reference WTE plant have been fully reproduced, simulating
it by means of a commercial software [2] based on a lumped parameter-modeling
approach, which is used in the conventional power plant industry for complex en-
ergy system characterization and performance prediction. The software basically
solves mass and energy balance equations in steady state conditions for each com-
ponent (HEs, expanders, compressors, pumps, etc.), providing flows, thermody-
namic states, exchanged heat and power between the system components.
Thermodynamic input for the WTE simulation, collected in Table 8.2, has been
selected considering statistical data distribution of steam cycle parameters for exist-
ing WTE plants. Figure 4.1 in Chap. 4 shows superheated steam temperature and
pressure of available plants, with reference to the Italian scenario. Data shown in
this figure, collected in [3], have been averaged here to identify the main character-
istics of the reference WTE power plant at the end of 2010 (see Table 8.1). Other
input data required to simulate the WTE steam cycle, for example, steam pressure
losses and heat losses in the boiler drum and distribution pipes, have been assumed
in line with typical WTE steam cycle power plants.
130 8  Specific Application Cases with GT Commercial Units

8.2 WTE Integration with GT Units: Investigated Layout


Cases and Results

The first proposed WTE steam/waterside-integration layout is shown in Fig. 8.2.


The WTE–GT integrated layout is an improvement of the integrated scheme (Lay-
out_1) presented in Chap 6 (see Fig. 6.1). A one-pressure-level (1PL) arrangement
with heat recovery steam generator (HRSG) has been considered. In this layout
(named Case A), the MSW boiler has the task of producing saturated steam, while
water preheating (in the ECO) and steam superheating (in the SH) are incorporated
into the HRSG. Indeed, in this proposed arrangement, in order to eliminate corro-
sion problems from the original SH of the WTE, the steam superheating process is
moved inside the HRSG. Inside the WTE convective section, a small dryer bank
replaces the original SH of Fig. 8.2. The dryer is used only to slightly increase the
saturated steam temperature (of about 5 °C) to avoid steam condensation in the dis-
tribution pipes towards SH.
A preliminary parametric analysis on this integrated WTE–GT base configura-
tion was carried out in Sect. 6.1 of Chap. 6. The additional steam mass flow rate
generated in the MSW boiler caused by the integration was highlighted. This in-
vestigation in the past showed a link between the steam mass flow ( ms) and the GT
size. Results of the study suggested that two key factors of the WTE–GT hybrid
combined cycle (HCC) optimal design are the GT-discharged thermal power ( QEXH)
and the MSW input power ( FW). An optimum condition in terms of ratio of QEXH
to FW had been identified to maximize steam generation and steam-superheated
temperature and to minimize exhaust gas temperature. This optimum depends on
the steam cycle pressure and SH temperature.

EĂƚƵƌĂů 'ĂƐ

&&

& 7

72*7
67
(9$ 6+
(9$

'5<(5

+LJK SUHVVXUH
67 EOHHG
(9$ +56*
:DVWH
&RQGHQVHU

(&2

& 7R IOXH
7R VWDFN /RZ SUHVVXUH
JDV FOHDQLQJ
72+56* 67 EOHHG

6HFRQGDU\ 72:7(
DLU
*UDWH +,JK SUHVVXUH
FRROLQJ 67 EOHHG +,JK SUHVVXUH
67 EOHHG
EOHHG FRQGHQVHU
3ULPDU\
& FRPEXVWLRQ
DLU
7R '($ ):
'($

Fig. 8.2   WTE–GT integrated layout Case A. EVA evaporator, DEA deaerator, SH superheater,
HRSG heat recovery steam generator, ECO economizer, ST steam turbine
8.2 WTE Integration with GT Units: Investigated Layout Cases and Results 131

Fig. 8.3   ms versus QEXH /FW for three steam cycle p–T sets—Case A and Case B layouts; com-
mercial gas turbine (GT) units are shown

Figure 8.3 shows steam production versus QEXH /FW for the WTE–GT plant fea-
turing the above-described Case A layout (continuous line). This line is obtained,
according to the analysis performed in [4] and already explained in Sect. 6.1 of
Chap. 6, by simulating the GT exhaust as a variable heat source and for fixed steam
cycle parameters. The p–T values at the ST inlet are initially set equal to pST = 80 bar,
TST = 480 ° C; these values are compatible with conventional GT exhaust temperature
values (typically around 500–550 ° C), considering a minimum approach of 20 °C.
A maximum value of the generated ms can be identified; beyond this value, an
increase in the HRSG thermal input ( QEXH) does not lead to additional benefits, as
the GT would be thermally oversized, causing a higher HRSG outlet gas tempera-
ture, that is, the GT-discharged heat is not fully exploited (see [4]). The maximum
calculated ms values of the WTE–GT plant are, in this p–T set case, 150 % higher
than values of the steam mass flow in the reference WTE.
In order to further improve the WTE–GT integrated thermodynamic perfor-
mance, an advanced configuration  (Case B), shown in Fig. 8.4, has been also con-
sidered. Even in this second application case, the assumed integrated layout comes
from an improvement of Layout_3 (see Fig. 6.20). The possibility of sharing the
ECO section between WTE boiler and HRSG is proposed with a parallel arrange-
ment. The water flow downstream DEA is split into two streams: a fraction ( f) goes
132 8  Specific Application Cases with GT Commercial Units

EĂƚƵƌĂů 'ĂƐ

&&

& 7

72*7
67
(9$ 6+
(9$

'5<(5

+LJK SUHVVXUH
67 EOHHG
(9$ +56*
:DVWH 0,;(5
&RQGHQVHU
(&2 
(&2 
63/,77(5

7R IOXH
& 7R VWDFN /RZ SUHVVXUH
JDV FOHDQLQJ
72+56* 67 EOHHG

6HFRQGDU\ 72:7(
DLU
*UDWH +LJK SUHVVXUH
FRROLQJ 67 EOHHG +,JK SUHVVXUH
67 EOHHG
3ULPDU\ EOHHG FRQGHQVHU
& FRPEXVWLRQ
DLU
7R '($ ):
'($

Fig. 8.4   WTE–GT integrated layout with parallel economizers—Case B. EVA evaporator, DEA
deaerator, ECO economizer, SH superheater, HRSG heat recovery steam generator, ST steam
turbine

into ECO2, that is, the ECO section inside the HRSG, while the remaining part (1-f)
is sent to ECO1, inside the WTE boiler. A mixer is introduced before the EVA, to
mix the two ECOs, outlet streams. The split factor f has been set in order to equalize
the mixer inlets’ temperatures.
The reduction in the amount of heat necessary for water economization inside
the HRSG, due to the parallel ECO configuration of Case B, leads to a decrease
in the optimal QEXH/FW value, as shown in Fig. 8.3 (dotted line). However, the
maximum generated ms is the same for both layouts, for a given p–T set. Thus, this
investigation suggests that Case B can be more convenient, producing the same
amount of steam of Case A, but using a smaller GT.

8.2.1 GT Unit Selection

In order to evaluate, more accurately, the effect of specific GT machines on the


WTE–GT integrated plant performance, a new numerical investigation has been
performed by introducing a model-library commercial GT machine in the full
WTE–GT thermodynamic model [2], with specified nominal data of the exhaust
mass flow and temperature.
Different commercial units have been selected in this study, namely, GE
5251M (GT0), Rolls Royce RB211 (GT1), GE LM2500 PJ (GT2), and GE
LM2500 PH (GT3). The design data of the GTs at ISO full-load condition are
reported in ­Table  8.3 [2], while the corresponding points are shown in Fig. 8.3.
The GT-discharged thermal power is also given in Table 8.3.
8.2 WTE Integration with GT Units: Investigated Layout Cases and Results 133

Table 8.3   Nameplate data of selected gas turbine (GT) units at ISO conditions
GT0 GT1 GT2 GT3
5251M RB211 LM2500 PJ LM2500 PH
PGT [MW] 17.8 27.1 22.6 20.8
TO, GT [°C] 510 506 544 533
Air flow [kg/s] 96.0 90.6 64.7 63.6
Exhaust flow 97.5 92.3 68.3 64.2
[kg/s]
ηGT [%] 24.9 35.2 35.7 35.7
FNG [MWt] 71.3 77.3 63.4 58.4
QEXH [MWt] 57.5 54.1 43.2 39.8
QEXH/FW [-] 1.04 0.98 0.79 0.72

As shown in Fig. 8.3, the GT1 unit has been selected close to the optimum QEXH
value (i.e., min QEXH granting max ms) for Case A. In a similar way, GT2 is close
to the optimum point for Case B. Finally, GT3 is characterized by an undersized
thermal power, and GT0 is oversized for both layouts. As shown in Table 8.3, the
electric efficiency of the GTs, ranging between 25 and 36 %, is a key factor that
strongly affects the value of QEXH. Indeed, GT0 is the smallest electric size unit, but
it results in oversized QEXH because of the lowest ηFT value.

8.2.2 WTE–GT Integrated Plant Numerical Results

Numerical results of the WTE steam cycle integration with GT units are shown in
Table 8.4, for both Case A and Case B layouts. A performance assessment has been
carried out, depending on the selected GT unit and WTE–GT integrated configura-
tion. In particular, due to the multi-fuel (MF) configuration of the investigated HCC
plants, as already detailed in Chap. 7, the conversion efficiency quantification is not
a simple task [5, 6]. For this reason, different key performance indicators have been
taken into account to evaluate the performance of the integrated system, namely:
1. Steam cycle mass flow rate, ms
2. HRSG effectiveness, ( TO, GT −TO, HRSG)/( TO, GT −Tamb)
3. ST power increase versus WTE stand-alone
4. First law efficiency, ( PWTE + GT)/( FNG + FW)
5. Waste synergy index, ( PWTE + GT −0.41∙FNG)/FW
6. Natural gas (NG) synergy index, ( PWTE + GT −PWTE)/FNG
7. MF synergy index, ( PWTE + GT −PWTE −PGT)/( PWTE + PGT)
Among the performance indicators listed above, the first law efficiency (4) is a
basic well-known index, which considers both fuels (natural gas (NG) and MSW)
of equal importance. Using a different approach, the extra power, generated as a
consequence of WTE–GT integration, can be assigned to the NG or MSW input
Table 8.4  Key thermodynamic variables and performance results of WTE–GTs integrated systems
134

WTE GT0 GT1 GT2 GT3


stand- GE 5251M RB 211 LM2500 PJ LM2500 PH
alone Case A Case B Case A Case B Case A Case B Case Case A Case B Case
B-opt B-opt
ST inlet temperature (TST) [°C] 390 480 480 480 480 480 480 514 480 480 490
ST inlet pressure (pST) [bar] 43 80 80 80 80 80 80 80 80 80 80
Steam-superheated temperature at SH outlet 395 485 485 485 485 485 485 519 485 485 495
[°C]
Evaporation pressure at ECO outlet [bar] 49 85 85 85 85 85 85 85 85 85 85
TO, GT in integrated configuration [°C] – 511 511 507 507 545 545 545 534 534 534
Approach temperature difference [°C] – 26 26 22 22 60 60 26 49 49 39
Mass flow splitting factor (f) [–] – – 0.7 – 0.7 – 0.7 0.67 – 0.67 0.67
Steam mass flow rate ms [kg/s] 18.9 28.4 28.4 28.4 28.4 26.3 28.4 27.8 25.4 27.7 27.4
Subcooling temperature difference [°C] 50 10 10 10 10 35 10 16 47 17 20
HRSG outlet temperature (TO, HRSG) [°C] – 162 227 139 208 138 141 138 138 138 138
HRSG effectiveness – 0.704 0.572 0.748 0.608 0.758 0.762 0.768 0.763 0.763 0.763
(TO, GT−TO, HRSG)/(TO, GT−Tamb) [–]
WTE outlet temperature [°C] 164 314 182 314 182 314 182 172 314 173 174
High-pressure ST bleed mass flow rate [kg/s] 1.81 2.43 2.41 2.43 2.41 2.28 2.41 2.33 2.22 2.37 2.34
Low-pressure ST bleed mass flow rate [kg/s] 1.68 2.56 2.54 2.56 2.54 2.35 2.54 2.45 2.26 2.48 2.44
Regeneration ratio 0.41 0.34 0.34 0.34 0.34 0.34 0.34 0.34 0.34 0.34 0.34
WTE–GT integrated gross output power – 46.4 46.4 55.9 55.9 49.2 51.4 52.2 46.4 48.8 48.9
[MW]
WTE–GT integrated net output power – 44.9 44.9 54.2 54.2 47.7 49.7 50.6 44.9 47.2 47.3
PWTE + GT [MW]
8  Specific Application Cases with GT Commercial Units
Table 8.4  (continued)
WTE GT0 GT1 GT2 GT3
stand- GE 5251M RB 211 LM2500 PJ LM2500 PH
alone Case A Case B Case A Case B Case A Case B Case Case A Case B Case
B-opt B-opt
GT output power [MW] – 17.7 17.7 27.0 27.0 22.6 22.6 22.6 20.7 20.7 20.7
ST net power [MW] 14.9 27.2 27.2 27.2 27.2 25.1 27.2 28.0 24.2 26.4 26.6
ST power increase versus WTE stand-alone 0 12.2 12.2 12.2 12.2 10.2 12.2 13.1 9.3 11.5 11.7
[MW]
First law efficiency 27.3 35.5 35.5 41.0 41.0 40.3 42.2 42.7 39.6 41.6 41.7
(PWTE + GT)/(FNG + FW)[%]
Natural gas synergy index – 0.420 0.420 0.509 0.509 0.517 0.549 0.562 0.514 0.553 0.555
(PWTE + GT−PWTE)/FNG [–]
MF synergy index – 0.373 0.373 0.290 0.290 0.272 0.327 0.348 0.257 0.322 0.326
(PWTE + GT-PWTE-PGT)/(PWTE + PGT) [–]
Waste synergy index – 0.284 0.284 0.410 0.410 0.394 0.432 0.447 0.381 0.423 0.425
(PWTE + GT−0.41∙FNG)/FW [–]
R1 0.726 1.099 1.099 1.113 1.103 1.053 1.156 1.195 1.023 1.130 1.140
(EWTE·2.6-EEXH·0.21)/(0.97·EW) [–]
WTE waste-to-energy, ST steam turbine, SH superheater, ECO economizer, HRSG heat recovery steam generator, WTE–GT waste-to-energy gas turbine, GT
gas turbine, MF multi-fuel
8.2 WTE Integration with GT Units: Investigated Layout Cases and Results
135
136 8  Specific Application Cases with GT Commercial Units

(see Chap 7, Sect. 7.2.2). If the benefit is assigned to NG, the NG synergy index (6)
can be used to measure the WTE–GT incremental performance, and it can be com-
pared to the GT unit efficiency. On the other hand, the waste synergy index (5) can
be used to measure the WTE–GT incremental performance compared to the WTE
stand-alone. In this case, a reference constant conversion efficiency equal to 41 %
of a reference 1PL combined cycle (CC) [7, 8] has been taken into account. Finally,
the MF synergy index (7) can be used to measure the actual WTE–GT marginal per-
formance in comparison with the separated reference WTE and GT power plants.
According with the investigation, the following performance results can be high-
lighted:
Steam mass flow rate increases significantly in all the layout configurations and
GT unit cases, ranging from 34  up to 50 % of the original WTE stand-alone value
(18.9 kg/s).
For fixed steam cycle p–T set ( pST = 80 bar, TST = 480 °C), the points of maximum
steam mass flow rate lead to maximum values of ST power increase versus WTE
stand-alone. The values of extracted steam mass flows are also reported in Table 8.4
along with the calculated values of regeneration ratio.1
In order to additionally improve the WTE–GT integrated plant design, as the
GT2 and GT3 approach temperature difference is significantly higher than typical
values, a further optimization has been carried out for Case B, increasing the steam-
superheated temperature at the SH outlet (taking into account a minimum allowed
approach equal to 20 °C). In these steam-cycle-enhanced conditions ( Case B-opt),
the steam mass flow rate slightly decreases, but all the other performance indicators
increase, demonstrating the possibility of optimizing the thermodynamic design of
the steam cycle for GTs with a high value of TO, GT. On the contrary, for both GT0
and GT1, TO, GT does not allow a further increase in the steam-superheated tempera-
ture; indeed, the approach temperature difference is already close to minimum typi-
cal values (it is equal to 26 °C and 22 °C for GT0 and GT1, respectively).
Nevertheless, all the investigated WTE–GT integrated configurations provide
a gain in power output in comparison with the separate stand-alone WTE and GT
systems. The ST power increase versus WTE stand-alone ranges between 9 and
12 MW, with a gain of 60–80 %. The achieved PWTE + GT values are 45–54 MW, that
is, about three times more than the reference WTE.
The obtained values of the first law efficiency are in the range from 35 to 43
%, depending on the case, with a maximum for GT2 Case B-opt. In all WTE–GT
cases, the first law efficiency is higher than the conversion efficiency of both the
stand-alone reference WTE and the used GT unit. This is due to the synergic effect
derived by integration. It follows that, for all configurations, positive values of MF
synergy index are obtained.

The regeneration ratio is calculated as the ratio of the enthalpy rise in the feedwater HEs to the
1 

enthalpy increase in the ECO section without feedwater heating.


8.2 WTE Integration with GT Units: Investigated Layout Cases and Results 137

The NG synergy index is relevant (reaching values up to about 0.55 in the best
case), showing that if the benefit of integration is attributed to NG, the resulting
performance index is higher than the efficiency of GT units and of reference 1PL
CC (with assumed mean efficiency equal to 41 % [7, 8]).
Also, the calculated waste synergy index values are positive and Case B layout
provides higher values for GT2 and GT3. The values of this index are higher than
the WTE reference efficiency, if Case B with GT2 is considered.
The highest HRSG effectiveness values (and lowest HRSG outlet temperature)
are achieved in Case A, with most of the examined GT units. On the other hand,
Case B provides the minimum WTE exhaust temperature values, allowing for better
exploitation of the MSW energy content.
Finally, as a minor result, the decrease in the GT’s power output in the integrated
configurations, compared to the nominal value reported in Table 8.3, is due to the
presence of the HRSG’s additional pressure losses, causing an increase in GT back-
pressure and output power.
The application of the “R1 formula,” as defined in [9], in the case of integrated
WTE–GT configurations, appears questionable. The difficulty in defining a perfor-
mance index for a MF energy system lies in quantifying the contribution of each
input fuel to the total output energy and till date the problem is open. For the sake of
completeness, results for a “modified” R1 formula have been included in Table 8.4,
for each integrated configuration, assuming (i) a plant capacity factor equal to 0.9
(7800 h/year operating at full load), (ii) only the ST electrical output from the WTE
section as the annual energy produced, and (iii) the electric equivalent of the GT’s
discharged energy recovered in the HRSG section as WTE annual “imported” en-
ergy.2 With this adapted definition, the R1 value increases in comparison with the
reference case, and it will allow all the WTE–GT integrated configurations to be
classified as recovery facilities. This result suggests that the benefit of GT integra-
tion could be useful also in cases of existing, old, and low-performance WTE plants
that are not satisfying the R1 threshold in the stand-alone configuration.
A final global comparison of all the analyzed cases shows that the best-perform-
ing configuration in terms of almost all the indexes is WTE-GT2 Case B (and in
particular the B-opt optimized steam cycle settings), confirming that a guideline
for the WTE–GT design optimization is to select the GT unit providing the mini-
mum QEXH but, at the same time, allowing for the achievement of the maximum ms
(Fig. 8.3).
The case of an existing WTE repowered with GT will call for additional and
specific evaluations; nevertheless, some general considerations can be made:
In order to accommodate the steam mass flow rate and the increase of the steam
cycle parameters (temperature and pressure) increase due to the GT integration, a
replacement of the original ST, condenser, and generator is necessary.

A factor equal to 0.21 has been assumed as the electric equivalent of GT-discharged heat, in line
2 

with typical 1PL CC values [7–8].


138 8  Specific Application Cases with GT Commercial Units

In order to avoid original steam cycle component replacement, only an increase


of up to 10 % of the nominal electric power can be supported by the system working
continuously in normal operating condition. One major design challenge, when re-
powering an existing plant, is to match the steam production capability of the topper
cycle and the HRSG with the steam needs of the existing ST. Thus, the GT’s optimal
size, in case of a limited ST power output increase, is expected to be significantly
reduced compared to the selected commercial units of Table 8.3.
As a general remark, for repowering an existing WTE plant, GT the optimal size
of the GT must be chosen strictly connected to the bottomer cycle size, in order to
achieve optimum plant match condition.
Despite the increased costs in comparison with the reference WTE due to ST en-
largement, GT installation, and operational costs, economic benefits occur through
both for the additional energy production incomes and the removed corrosion prob-
lems in the WTE convective SH section. These economic aspects have not been
quantified in this study, but they are in accordance with the field data from the only
existing WTE–GT plant in Europe [10–12]. The planned maintenance intervals for
corrosion problems can be reduced to a third in comparison with stand-alone WTE.
Finally, the resulting cost of electricity generated by the integrated plant is main-
ly linked to the method used to allocate the power output to each input fuel [5].
The first law efficiency can be used to allocate the generated power output to waste
and NG in a proportional way. For example, in GT2 Case B-opt, PNG = [PWTE + GT/
( FNG + FW)]∙FNG = 27.1  MW and PW = [PWTE + GT/( FNG + FW)]∙FW = 25.1  MW. Thus, a
fraction of the ST power output increase is allocated to NG contribution according
to this method. The result of electricity allocation affects, as a consequence, the cost
of electricity.

8.3 Conclusion

A detailed numerical analysis of the achievable performance by a WTE power plant


that is integrated with different existing GT units has been presented. The resulting
performance enhancement compared with a reference stand-alone WTE has been
quantified by considering two different layout configurations of the plant, both with
1PL arrangement of the HRSG.
Combining WTE with GTs in a hybrid CC provides an increase in the ST power
output of up to 80 %. This is a significant increase compared with typical mid size
stand-alone WTE power plants. The steam cycle mass flow rate generation is in-
creased by up to 50 %.
Numerical results of WTE repowering show that the best GT size to match maxi-
mum performance (maximum steam generation, ST power output, and minimum
exhaust gas temperature) varies depending on the layout configuration and the spe-
cific selected GT unit.
References 139

This study shows that by integrating a WTE with a commercial GT, a substantial
increase in the first law efficiency occurs, compared with the separated reference
WTE and GT. Moreover, considerable synergic effect is shown by our numerical
evaluation of a midsize conventional WTE.
Due to the high exhaust temperature values of commercial GT units, an increase
in the steam cycle parameters (i.e., steam-superheated temperature and evaporation
pressure) compared with the p–T set of the reference WTE must be pursued in the
design of the WTE–GT integrated plant. Consequently, significant power output
increase can be achieved.
Finally, this study provides useful guidelines and performance benchmarks for
the optimal design of future advanced WTE power plants integrated with GTs.

References

  1. Martin GmbH Waste-to-Energy Plant KHKW Zürich-Hagenholz. Switzerland. http://www.


martingmbh.de/
  2. Thermoflex software version 22 by Thermoflow inc. (2012)
  3. ENEA-Federambiente (2012) Report on energy recovery from municipal solid waste in Italy,
3rd edn. (In Italian)
  4. Bianchi M, Branchini L (2012) Advanced waste-to-energy power plant: integration with gas
turbine. International conference applied energy, July 5–8, 2012, Suzhou, China Paper ID:
ICAE2012–A10330
  5. Bianchi M, Branchini L, De Pascale A (2014) Performance indexes and output allocation for
multi-fuel energy systems. Energy Proc 45:32–41. doi:10.1016/j.egypro.2014.01.005
  6. Bianchi M, Branchini L, De Pascale A, Falchetti M, Fiore P (2014) Advanced waste-to-
energy steam cycles. Energy Proc 45:1205–1214. doi:10.1016/j.egypro.2014.01.126
 7. Gas Turbine World (2004–2005) GTW handbook, vol 24. Pequot, Southport, Pequot
Publication
 8. Bhargava RK, Bianchi M, Campanari S, De Pascale A, di Montenegro GN, Peretto AA
(2010) Parametric thermodynamic evaluation of high performance gas turbine based power
cycles. J Eng Gas Turbines Power 132(2). Article no. 022001. doi:10.1115/1.3155782
  9. Directive (22.11.2008) 2008/98/EC of the European Parliament and of the Council of 19
November 2008 on waste. Official Journal European Union OJ L 312/3, 22.11.2008
10. CNIM and SENER (2005) Bizkaia waste to energy plant project, summary report. Contract
No. BM 80/96 ES-PO
11. SENER Ingeniería y Sistemas S.A. High efficiency technologies applied for waste to energy
(WTE)
12. AZN and EPZ Combination of a waste incineration plant and a combined cycle power plant.
Project no. BM/349/93/NL
Index

A section, 26–28
Application cases, 127 status, 4, 19, 34
Average values, 40, 127 European environment agency (EEA), 8, 9, 11
European legislation, 7
B European union (EU)
Boiler, 4, 22, 26, 87 MSW landfilling in, 14
horizontal boilers regulation framework, 33
advantages of, 27 Existing integrated power plant, 64
vertical boilers, 27
combination of vertical and horizontal G
boilers, 27 Gas turbine (GT), 5, 36, 57, 64, 71
Bottomer cycle (BC), 57, 138 selection, 132
bottoming, 58 Generation per capita, 9, 12
boiler, 59
H
C Heat surface areas
Co-combustion power plant, 113, 122 economizing, 26
Cold windbox integration, 61 evaporation, 26
Combustion superheating, 26
chamber, 24, 26, 29, 42, 52, 53, 58, 85 High efficiency waste-to-energy plant, 30, 49
section, 20–26, 43, 114 Amsterdam—The Netherlands: High
Commercial units, 127, 132, 138 efficiency WTE plant, 49
Conversion efficiency, 34, 40, 44, 63, Reno—Nord Denmark: SteamBoost
113–115, 121, 123, 136 concept, 49
Corrosion protection, 28–30 Germany—Karlsruhe Institute of
metal tube corrosion, 29 Technology research: bypass of fuel
gas concept, 49
D Hirn cycle, 4, 19
Delivery and storage section, 20, 21 Holmgren, R., 64
Disposal status, 33 Hot windbox integration, 61
Hybrid combined cycle (HCC)
E concept of, 57
Eber, S.M., 63
Efficiency evaluation, 116 I
Electric efficiency, 45, 47, 49, 59, 64 Incinerators
Energy feedstock, 114 grate incinerators, 23
Energy recovery reverse reciprocating grate, 24
push forward grate, 24

© Springer International Publishing Switzerland 2015 141


L. Branchini, Waste-to-Energy, DOI 10.1007/978-3-319-13608-0
142 Index

Integrated R
layouts, 104 Rankine cycle, 57, 59
plant numerical results, 133
plant operative parameters, 136, 137 S
system Sections description, 20
performance indicators of, 5 Sensitivity analysis, 39, 44, 47, 64, 71
Ito, K., 64 Separate generation, 116
Single-fuel (SF) energy systems, 113
K State of art, 39, 41, 42, 63, 114
Korobitsyn, M.A., 64 Steam cycle
effects on cycle efficiency, 44–49
L efficiency, 40, 45, 76
Landfill costs, 14 layout, 49
Lowry, W.L., 63 operative parameters, 71
parameters upgrade, 49
Steam
M
generation, 58, 59, 71, 83, 99, 130, 138
Martin, C.A., 63
production
MF synergy index, 113, 133, 136
thermodynamic analysis on, 71
MF energy systems
Steam mass flow increase, 82
arrangement, 115, 116
Steam/waterside integration, 58–60, 63,
conversion efficiency assessment of, 121
65, 71–109, 130, see also WTE-
indexes for, 116
GTpower plant
performance evaluation of, 115
Sue, M., 64
Moerdijk plant, 67, 68
Municipal solid waste (MSW), 3, 7, 8
and disposal in Italy, 12 T
landfill costs, 14 Takahama plant, 64, 68
Municipal waste Thermodynamic analysis, 5, 86
definition of, 7 on steam production, 72, 76–78, 87
Terasawa, H., 64
Topping cycle (TC), 57–59, 64
N
Two-fuel co-combustion power plant,
New designs, 49
122–125
O
U
Ogura, M., 64
Udomsri, S., 64
One pressure level HRSG, 71, 76, 83–104
Optimum plant match, 78, 80–85, 87, 94, 101,
104 W
Otoma, S., 64 Waste delivery
Output allocation, 113, 121 and storage section, 20
Waste disposal, 3, 8, 10, 13, 14, 33, 65
Waste framework directive (WFD) 2008/98/
P
EC, 3, 4
Parametric analysis, 44, 71, 76, 130
Waste generation, 7, 9
Performance indexes, 5, 113–125
Waste hierarchy, 3, 8, 10
Plants
Waste incineration, 4, 19, 21–23, 27
capacity, 19
capacity diagram, 23
statistics, 129
phases of, 21
Power output
drying and degassing, 21
evaluation, 137, 138
oxidation, 22
increase, 83, 127, 138
pyrolysis and gasification, 21
Primary combustion, 22
Index 143

Waste management, 3, 8, 10 Wiekmeijer, I.T., 63


systems, 7 Windbox
Waste production, 9–12 integration, 61–63
Waste recycled, 8 repowering, 59, 61, 62
Waste-to-energy (WTE), 19 WTE–GT power plant, 67
conversion, 3, 5,35, 40 of Moerdijk, 67
EU regulation framework, for efficiency, of Takahama, 68
33–36 of Zabalgarbi, 64–67
-GT windbox integration, 61–63 WTE–GT integrated layouts, 104–108
power plant, 3–5, 19–35, 42, 44–47, 57,
64, 80, 108, 117, 128, 139 Z
steam cycle, 5, 31, 129, 133 Zabalgarbi plant, 64–67
Waste-to-Energy Research and
TechnologyCouncil (WTERT)
study, 30

You might also like