Design and Optimization of The HVAC
Design and Optimization of The HVAC
Design and Optimization of The HVAC
Alexandre Oudet
Alexandre Oudet
-2-
Abstract
During nuclear power plants shutdown many people could be deprived of electricity and it would have a
negative impact both on the company’s image and on people activities. As a consequence, availability of
equipments in the different buildings which compose the power plant needs to be assured. HVAC system
(Heating, Ventilation and Air Conditioning) plays an important role on the reliability of these equipments
as it makes sure that ambient conditions in the buildings fit the operating temperature range of the
equipments. Consequently sizing a ventilation system is really important and it needs to be carried out
seriously. This paper introduces the methodology to size and optimize a ventilation system for nuclear
power plants’ building. This paper also develops the methodology used to size a smoke control system in a
nuclear related building. Direct application of this methodology has been realised for a specific building
which is the demineralization station of Hinkley Point C project.
Keyword: EDF, nuclear power plant, psychrometry, ventilation, demineralization, regulation, heat
science and thermodynamics, smoke control, pressurization, thermal modelling, Computational
Fluid Dynamic
Sammanfattning
Avstängda kärnkraftverk berövar många människor av elektricitet och det skulle ha en negativ inverkan
både på företagets framtoning och mänskliga aktiviteter. På grund av detta behöver tillgängligheten av
utrustningen i alla byggnaderna som kärnkraftverken består ses till. HVAC-system (Heating, Ventilation
and Air Conditioning) spelar en viktig roll när det gäller tillgänglighet av utrustning eftersom dessa
system ser till pålitligheten är på topp genom att anpassade omgivningsförhållanden till utrustningen.
Att designa ventilationssystemet rätt är därför mycket viktigt och måste göras noggrant. Denna rapport
introducerar metodologin för att designa och optimera ett ventilationssystem för en av byggnaderna i
ett kärnkraftverk. Utöver detta utvecklas och beskrivs en metodologi för att designa ett
rökkontrollssystem för en byggnad som ingår i kärnkraftverket. Dessa metodologier har implementerats
för en byggnad i en demineraliseringsstation, Hinkley Point C project.
-3-
Table of Contents
Abstract ........................................................................................................................................................................... 3
Sammafattning ............................................................................................................................................................... 3
Nomenclature ..............................................................................................................................................................10
1 Introduction ........................................................................................................................................................11
1.1 EDF (Electricité de France) / CNEPE ................................................................................................11
1.2 Hinkley Point C (HPC) nuclear power station project .......................................................................11
2 Methodology and Objectives ...........................................................................................................................12
2.1 Objectives ...................................................................................................................................................12
2.2 Methodology ..............................................................................................................................................12
2.3 Limitations .................................................................................................................................................12
3 Building description and technical background ............................................................................................13
3.1 Demineralization process ........................................................................................................................13
3.1.1 Water in a nuclear power plant ......................................................................................................13
3.1.2 Demineralized water in nuclear power plants .............................................................................15
3.1.3 How to obtain demineralized water? ............................................................................................15
3.2 Demineralization station presentation ...................................................................................................18
3.2.1 Civil works description ...................................................................................................................18
3.2.2 Building’s purpose ...........................................................................................................................18
3.2.3 Operation modes .............................................................................................................................18
4 Literature review .................................................................................................................................................19
4.1 Ventilation in industrial building ............................................................................................................19
4.1.1 Natural ventilation ...........................................................................................................................19
4.1.2 Single flow mechanical ventilation ................................................................................................20
4.1.3 Double-flow ventilation system (AHU) .......................................................................................20
4.2 Ventilation rules ........................................................................................................................................22
4.2.1 UK regulation for industrial buildings..........................................................................................22
4.2.2 Design conditions ............................................................................................................................23
4.3 Smoke and heat control ...........................................................................................................................25
4.3.1 Smoke exhaust and protection system .........................................................................................25
4.3.2 Practical application on the building.............................................................................................27
4.4 Psychrometry .............................................................................................................................................29
4.4.1 Saturation pressure of water...........................................................................................................30
4.4.2 Water content ...................................................................................................................................30
4.4.3 Dew point .........................................................................................................................................31
4.4.4 Specific enthalpy ..............................................................................................................................31
-4-
4.5 Humid air processes and air mixing .......................................................................................................31
4.5.1 Air mixing..........................................................................................................................................32
4.5.2 Representation on the psychrometric chart.................................................................................32
4.5.3 Extraction temperature for an AHU ............................................................................................33
4.5.4 Air heating .........................................................................................................................................33
4.5.5 Air cooling.........................................................................................................................................34
4.5.6 Air humidifying ................................................................................................................................38
4.5.7 Impossible air processes .................................................................................................................40
4.6 Pressure gradient and infiltration............................................................................................................40
4.6.1 Pressure gradient ..............................................................................................................................41
5 Calculation methodology review......................................................................................................................43
5.1 External conditions ...................................................................................................................................43
5.1.1 Air temperature ................................................................................................................................43
5.1.2 Ground temperature........................................................................................................................43
5.1.3 Wind ...................................................................................................................................................43
5.2 Heat gains and losses ................................................................................................................................43
5.2.1 Sensible heat gains/losses ...............................................................................................................43
5.2.2 Latent heat load ................................................................................................................................47
5.2.3 Total heat load ..................................................................................................................................48
5.3 Heat gains on supply air ...........................................................................................................................48
5.3.1 Heat gains/losses to the supply distribution duct ......................................................................48
5.3.2 Heat gain from the fan ....................................................................................................................49
5.4 Margins .......................................................................................................................................................51
5.4.1 On heat gains ....................................................................................................................................51
5.4.2 On flow rates ....................................................................................................................................52
5.5 Energy balance of a room........................................................................................................................53
5.5.1 Heat balance .....................................................................................................................................53
5.5.2 Sensible heat balance .......................................................................................................................53
5.5.3 Humidity balance .............................................................................................................................54
5.5.4 Defining the space line....................................................................................................................54
5.6 Blowing conditions ...................................................................................................................................54
5.6.1 Characteristic of the blowing air temperature difference ..........................................................55
5.6.2 Mixing rate ........................................................................................................................................55
5.6.3 Theoretical aspect and calculation ................................................................................................55
6 TH-Bât Software ................................................................................................................................................59
6.1 Input data ...................................................................................................................................................59
6.2 Zone property............................................................................................................................................60
-5-
6.3 Wall property .............................................................................................................................................60
6.4 Ventilation property..................................................................................................................................61
6.5 Opening property......................................................................................................................................61
7 Parametric study .................................................................................................................................................62
7.1 Glazed surface ...........................................................................................................................................62
7.2 HVAC system choice ...............................................................................................................................62
7.2.1 40°C outside Temperature .............................................................................................................63
7.2.2 33°C outside temperature, design without AHU........................................................................66
7.3 Scenario analysis ........................................................................................................................................67
7.3.1 Scenario 1: White painting..............................................................................................................67
7.3.2 Scenario 2: Insulation ......................................................................................................................68
7.3.3 Scenario 3: Combination of insulation and white painting .......................................................68
7.3.4 Comparison with the baseline scenario ........................................................................................69
7.4 Ventilation system control .......................................................................................................................70
7.4.1 Demineralization station HVAC system control ........................................................................70
7.4.2 HVAC system control cost ............................................................................................................73
7.5 Critical scenario analysis in chemical rooms .........................................................................................75
8 Ansys Fluent Modeling......................................................................................................................................77
8.1 Software presentation ...............................................................................................................................77
8.1.1 Meshing .............................................................................................................................................77
8.2 Turbulence modeling................................................................................................................................80
8.2.1 RANS –based turbulence model ...................................................................................................81
8.3 Methodology adopted ..............................................................................................................................81
9 Results ..................................................................................................................................................................89
9.1 HVAC system sizing.................................................................................................................................89
9.1.1 Internal Heat Gains .........................................................................................................................89
9.1.2 Blowing conditions justifications ..................................................................................................90
9.1.3 Heat gains on supply air ..................................................................................................................90
9.1.4 Final sizing ........................................................................................................................................91
9.2 Over-pressurization of the hall ...............................................................................................................95
9.3 Smoke Control System Sizing .................................................................................................................96
10 Discussion and Conclusion ........................................................................ Error! Bookmark not defined.
References.....................................................................................................................................................................99
-6-
Table of figures
Figure 1: The three water circuits of a nuclear power plant. (5) .........................................................................................13
Figure 2: Closed cooling system with its cooling tower. (6) .................................................................................................14
Figure 3: Open cooling system. (6)....................................................................................................................................14
Figure 4: Water level in two tanks separated by a semi-permeable membrane ....................................................................15
Figure 5: Osmosis principle ..............................................................................................................................................16
Figure 6: Reverse osmosis principle ...................................................................................................................................16
Figure 7: Reverse osmosis in industry ................................................................................................................................16
Figure 8: System without pressure (Water to be treated is on the left of the membrane) ......................................................17
Figure 9: System pressurized: Water is treated and flows in the right part .........................................................................17
Figure 10: Mixed beds working principle .........................................................................................................................17
Figure 11: Natural ventilation system. (11) .....................................................................................................................19
Figure 12: Single flow mechanical ventilation. (11) ...........................................................................................................20
Figure 13: Double flow ventilation system with an AHU. (11) .......................................................................................21
Figure 14: Smoke control pressurization system ................................................................................................................27
Figure 15: Psychrometric chart of humid air (22) .............................................................................................................29
Figure 16: Mixing of two air streams called “1” and “2” ................................................................................................32
Figure 17: Mixing of two humid air streams in a psychrometric chart ...............................................................................33
Figure 18: Heating of humid air on a psychrometric chart .................................................................................................34
Figure 19: Dry cooling of humid air on a psychrometric chart ............................................................................................35
Figure 20: Humid cooling of humid air on a psychrometric chart .......................................................................................36
Figure 21: Representation of sensible and latent heat during humid cooling ........................................................................38
Figure 22: humidification by water injection on a psychrometric chart ................................................................................39
Figure 23: humidification by steam injection on a psychrometric chart (dotted line: reality/ full line: theory) .......................40
Figure 24: (Left) impossible humid cooling process / (Right) impossible mixing of two air streams ....................................40
Figure 25: Drawing of a room’s energy balance .................................................................................................................53
Figure 26: Evolution of psychometry in a room, from the supply state 1 to the extraction 2 (space line) ..............................54
Figure 27: Drawing of two rooms used for Excel modelling and cardinal directions ...........................................................57
Figure 28: Wall modelling inTh-bât software ...................................................................................................................60
Figure 29: Cross section of the demineralization station with an AHU supplying air everywhere .......................................63
Figure 30: Steady state temperature of the basement (10 rooms) for 40°C outside with an AHU .....................................64
Figure 31: Cross section of the demineralization showing the two distinct parts of the building ...........................................64
Figure 32: Steady state temperature of the basement for 40°C outside without an AHU ..................................................65
Figure 33: Cross section of the demineralization station with a single flow ventilation system .............................................66
Figure 34: Temperature evolution in the main hall after a rapid increase of outside temperature from 33°C to 40°C.........66
Figure 35: Zoom in on Figure 34 ....................................................................................................................................67
Figure 36: Temperature of the hall depending on the presence of a control system ...............................................................70
Figure 37: Graphs showing the net present value of the investment for a control system ......................................................74
Figure 38: Evolution of the temperature after ACH increase in chemical rooms (summer) .................................................75
Figure 39: Evolution of the temperature after ACH increase in chemical rooms (winter) ...................................................76
Figure 40: Steady state temperature after ACH increase in chemical rooms (winter) .........................................................76
Figure 41: 2D cell types ...................................................................................................................................................77
Figure 42: 3D cell types ...................................................................................................................................................77
Figure 43: Structured grid around an airfoil .....................................................................................................................78
Figure 44: Unstructured grid around an airfoil .................................................................................................................78
Figure 45: Hybrid grid for a rotor/stator geometry ...........................................................................................................79
Figure 46: Size change between two adjacent cells (<20%) ...............................................................................................80
Figure 47: Geometry of the longest air duct in HY building (Design Modeler) ..................................................................82
Figure 48: Schematic representation of HY building’s ground floor ...................................................................................82
Figure 49: Hexahedral meshing of the air duct .................................................................................................................83
-7-
Figure 50: Tetrahedral meshing of the air duct..................................................................................................................83
Figure 51: Air duct planes (Plane n°1: green and Plane n°2: red) ...................................................................................84
Figure 52: Temperature in plane n°1 (k-omega model) .....................................................................................................85
Figure 53: Temperature in plane n°1 (k-epsilon realizable model) ....................................................................................85
Figure 54: Temperature in plane n°2 ...............................................................................................................................86
Figure 55: Walls temperature of the air duct ....................................................................................................................86
Figure 56: Temperature in plane n°1 for insulated walls ..................................................................................................87
Figure 57: Temperature in plane n°1 for an inlet temperature equal to 14.6°C ................................................................88
Figure 58: Cooling process of humid air for an external temperature of 40°C ...................................................................93
Figure 59: Psychrometric chart representing evolution of air for rooms housing workers in summer .....................................94
Figure 60: Psychrometric chart representing air humidification in winter for rooms housing employees in winter (-15°C) ....95
-8-
Table of tables
Table 1: Maximum acceptable dB in different building’s area (14) ...................................................................................23
Table 2: Recommended maximum duct velocities for low pressure ductwork. (16) ..............................................................23
Table 3: Admissible temperature range in HY building ...................................................................................................25
Table 4: Minimum ACH in HY building ......................................................................................................................25
Table 5: Recommendations for fire-fighting shafts ventilated by natural means ...................................................................26
Table 6: Electrical room mechanical smoke extraction ......................................................................................................27
Table 7: Basement mechanical smoke extraction ...............................................................................................................27
Table 8: Air leakage depending on the pressure differential applied and the leakage area (20)...........................................28
Table 9: External temperature considered for HVAC designs .........................................................................................43
Table 10: Wind speed considered for HVAC designs ......................................................................................................43
Table 11: heat transfer coefficients depending on walls and season......................................................................................44
Table 12: Values of CLTD, LM and k for concrete walls more than 300mm thick........................................................45
Table 13: Values of CLTD, LM and k for cladding ......................................................................................................45
Table 14: Values of CLTD, MSHGF, SC and CLF for a Pilkington double glazed window .......................................46
Table 15: Power from lighting (28) ..................................................................................................................................46
Table 16: Heat released by human body depending on the activity (26) .............................................................................46
Table 17: Representative rates at which heat and moisture are given off by human beings for different activities (17) ..........47
Table 18: Loss of heat and water vapor in the human body for an individual sitting in light activity (32) ..........................48
Table 19: External convective resistance for different air ducts layout ................................................................................49
Table 20: Internal convective resistance for different air speed ............................................................................................49
Table 21: Efficiency of different fans (16).........................................................................................................................50
Table 22: Motor efficiency (17) ........................................................................................................................................50
Table 23: Required mixing rate in a room (30) ...............................................................................................................55
Table 24: Thermal conductivity of materials (35) .............................................................................................................59
Table 25: Heat transfer coefficient for doors and windows (35) .........................................................................................59
Table 26: Materials’ thermal diffusivity and layer thickness..............................................................................................61
Table 27: MSGHF for different wall orientation .............................................................................................................62
Table 28: Heat gains for different windows location ..........................................................................................................62
Table 29: Results obtained with the first and the second HVAC design ...........................................................................65
Table 30: HVAC equipment required for scenario 1 .......................................................................................................68
Table 31: HVAC equipment required for scenario 2 .......................................................................................................68
Table 32: HVAC equipment required for scenario 3 .......................................................................................................68
Table 33: Comparison between the three different scenarios for 40°C and -15°C ..............................................................69
Table 34: Comparison between the three different scenarios for 2°C ..................................................................................69
Table 35: Comparison between the three different scenarios for 20.5°C .............................................................................69
Table 36: Power consumption for HVAC system with and without control by -15°C outside ...........................................70
Table 37: Average temperatures in Somerset England (36) ..............................................................................................71
Table 38: Assessment of energy gains for each month with and without a control system ....................................................72
Table 39: Energy saved each month by having a control ....................................................................................................72
Table 40: Net present value for a control system investment ..............................................................................................73
Table 41: Cell quality depending on the value of skewness ................................................................................................79
Table 42: Internal heat gains in HY building ..................................................................................................................89
Table 43: Temperature in the hall depending on supply air temperature ............................................................................90
Table 44: AHU flow rates, heating and cooling needed to be implemented in the different rooms of HY building ...............92
Table 45: List of fire doors in HY building .....................................................................................................................96
-9-
Nomenclature
EDF Electricité De France
CNEPE National Centre of Electricity Production Equipment
HPC Hinkley Point C
HVAC Heating, Ventilating and Air Conditioning
HY Name of the Demineralization station
SDA System producing water at pH7 and pH9
SER System storing and supplying water at pH9
SED System storing and supplying water at pH7
0REA System producing and supplying degassed water at pH7
DVT Name of the Ventilation system of the Demineralization station
EPR European Pressurized Water Reactor
PWR Pressurized Water Reactor
CFD Computational Fluid Dynamic
RANS Reynolds-Averaged Navier-Stokes
LES Large Eddy Simulation
DNS Direct Numerical Simulation
NPV Net Present Value
I&C Instrumentation and Control
φ Relative humidity (%)
pv Partial pressure of water vapour in air (Pa)
pv’’ Saturation pressure of water in air (Pa)
x Water content (kgwater/kgdry air)
h Enthalpy (kJ/kg)
T Temperature (K)
cp Specific heat (kJ/kg.K)
qm Mass flow rate (kg/s)
qv Volumetric flow rate (m3/s)
P Power (W)
BF Bypass Factor (%)
ξ Efficiency (%)
m Mass (kg)
p Pressure (Pa)
V Volume (m3)
v Speed (m/s)
S Surface (m²)
U Thermal transmission coefficient (W/(m².K))
CLTD Cooling Load Temperature Difference
Fo Fourier Number
α Thermal diffusivity (m²/s)
-10-
1 Introduction
At the moment EDF (Electricité de France) is working on the construction project of two EPR nuclear
reactors in Great Britain (Hinckley Point C project). They are currently carrying out pre-studies. This phase
is really important mainly for two reasons. Firstly nuclear power plants present potential risks to the
population, these risks need to be assessed during this pre-design phase so that they can be avoided during
plant operation. Secondly, EPR reactors produce an important amount of electricity and the least fault or
problem in one of the building could result in a cessation of electricity production. During the breakdown
many people could be deprived of electricity and it would have a negative impact on company’s image.
Safety and reliability are then the main concerns when it comes to nuclear buildings and it has a direct
influence on the price of such a project that is estimated to several billion Euros. Consequently modeling
and optimization studies are necessary in order reduce down to zero the risk of the project. The main goal
of EDF is then to fulfill the different requirements imposed by the regulations and tenderers while trying
to save money.
This report focuses on the HVAC system of a specific building housing the water demineralization process.
By optimizing the size of the ventilation system, installation is easier as there is more space for other
equipment such as electric cables, piping and so on. Moreover over-sizing the ventilation system is energy
consuming especially when it comes to industrial buildings used in nuclear power plants that are supposed
to be operational 24/24h and have a lifespan of sixty years. Within EDF, the CNEPE (National Centre of
Electricity Production Equipment) is among a lot of other activities in charge of the HVAC system of this
specific building.
Within the Engineering and Projects department, the CNEPE (National Centre of Electricity Production
Equipment) founded in 1955 hosts design activities related to non-nuclear buildings such as the Cold Source
and the Conventional Island of nuclear power plants for new projects and plants in operation. The main
role of this unit is to extend the operating time of nuclear power plants. It is located in France and counts
about 700 employees. The HVAC group in which this thesis was carried out is responsible for designing
HVAC system for non-nuclear buildings under CNEPE’s responsibility. (2)
EDF is currently working on a project in Somerset, South West England. It is a project that will lead to the
construction of two EPR reactors equivalent to 1600MWe each. An EPR reactor is a pressurized water
reactor (PWR) of the third generation which has been designed and developed by EDF and Areva NP. The
branch of EDF that is responsible for this project in United Kingdom is called EDF Energy. (3)
-11-
2 Methodology and Objectives
In order to carry out this project a clear methodology and precise objectives need to be set. They will be
presented in the two following paragraphs.
2.1 Objectives
The main purpose of this thesis is the design and optimization of a ventilation system for an industrial
building. The building considered is the demineralization station building also called HY building. It is a
non-nuclear safety related building which houses the SDA, SER, SED and 0REA systems and their
associated support systems:
SDA system produces water at pH7 and pH9. It gets raw water from industrial water supply system,
transform it and supply it to SER and SED systems.
SER system stores and supplies demineralized water at pH9 in normal unit operation for the plant.
SED system stores and supplies demineralized water at pH7 in normal unit operation for the plant.
0REA system collects demineralized water from SED tanks and then produces, stores and supplies
degassed demineralized water at pH7.
Within this building the HVAC system will be further analyzed. This ventilation system has many purposes
and needs to fulfill the following requirements:
Provide sufficient air renewal for personnel comfort and hygiene but also for workers’ safety in the
laboratory and chemical storage rooms.
Continuously maintain the various areas of the building at an acceptable ambient temperature and
relative humidity that is suitable for correct functioning of electrical equipment, good working
conditions and maintenance operations.
In the event of a fire, the role of DVT system is to ensure smoke control, isolate fire sectors, and
make the evacuation of the staff and the intervention of firemen easier.
2.2 Methodology
Getting familiar with the demineralization process thanks to training courses
Literature survey on ventilation system technologies, smoke control systems, UK regulations,
psychrometry and humid air.
Modeling of the building in Excel to size the ventilation
Development of a thermal model of the building on Th-bât (EDF software)
Scenario analysis in order to optimize the HVAC system of the building
Modelling on Ansys FLUENT in order to assess potential margins that would need to be taken
into account.
2.3 Limitations
Due to confidentiality issues, some documents used in references are not available for people outside EDF.
Moreover, civil works drawings, 3D model of the building and mechanical diagram of the HVAC system
cannot be displayed in a public document. Results and calculations are not fully described to be consistent
with the confidentiality policy of the company.
-12-
3 Building description and technical background
The main role of a ventilation system is to assure the good working of a process. As a consequence in order
to size a ventilation system it is primordial to understand how the demineralization process works.
The main purpose of the primary circuit is heat extraction. It is a closed pressurized water circuit at
a temperature of 320°C and 155bar. Water runs through the reactor and receives the heat generated by the
nuclear fuel fission reaction. This water heats up water from the secondary circuit via a vapor generator that
enables thermal heat transfer between these two independent circuits. To make it simple pipes from the
primary circuit heats up water from the secondary circuit by contact. (4)
The secondary circuit is used to produce steam. Through contact with thousands of U shaped tubes
composing the steam generator, water from the primary circuit transfer its heat to water flowing in the
secondary circuit. This circuit is not as pressurized as the primary circuit, that’s why water from this circuit
is converted into steam which is used to spin the turbine. It drives the generator that produces electricity.
Then steam goes back to the liquid state when it flows through the condenser. Water is sent back to the
steam generator for a new cycle. (4)
In order to condense steam and evacuate heat, the cooling circuit is made of a condenser. It is a
component composed of thousands of tubes into which cold water from the sea or the river flows. Through
contact with these tubes, steam condenses.
-13-
Two kinds of cooling system can be used in nuclear power plants. There are open cooling systems
for nuclear power plants located near the sea or close to a river with a large flow and closed cooling systems
for power plants located next to a low flow river. (6)
In a closed circuit (Figure 2) hot water from the condenser is cooled down by cold air in a large
cooling tower. A fraction of the water is vaporized and exits the cooling tower as a visible plume at the top
of the tower. The rest of the water is sent back to the condenser. With this system the water taken from the
cold source is less important (about 2 cubic meters per second in average). (6)
In an open cooling circuit (Figure 3), water is pumped directly from the sea or the river in a much
larger amount (about 50 cubic meters per second) and is released in its natural environment at a slightly
higher temperature after having circulated in the condenser. The Hinkley point C nuclear power station is
located close to the sea and will have an open cooling system.
-14-
3.1.2 Demineralized water in nuclear power plants
Demineralized water is used in primary and secondary circuit. The main purpose of demineralized water is
to prevent pipe corrosion or chemical reaction that could deteriorate the circuits and limit their performance.
Even if these circuits are closed, make-up water and draining are necessary for the proper functioning of
the nuclear power plant. As consequences make-up demineralized water needs to be provided on a daily
basis. (6)
In order to produce demineralized water three successive steps are carried out. A pretreatment by Granular
Activated Carbon, a reverse osmosis double pass and mixed bed.
The first step is the pretreatment by Granulated Activated carbon (GAC). Its main role is the protection of
reverse osmosis membranes. Even if raw water is supposed to have a good quality and a low fouling
potential, it is always recommendable to have a filtration step in order to protect reverse osmosis
membranes. (7)
The second step is the primary demineralization with two passes of reverse osmosis. Reverse osmosis is a
process that aims at producing demineralized water for various uses in industry or for private individual.
The following experience and technical considerations will lead to a better understanding of the
phenomenon of reverse osmosis. When water is poured into two tanks separated by a semi permeable
membrane, the level in each tank is shown in Figure 4.
A B
Semi-permeable
Membrane
A semi permeable membrane has the property of blocking the salts contained in water by not allowing them
to migrate through it. However, if one put water of different saline qualities on each side of the membrane,
water as a solvent will migrate from the tank with the lowest concentration (B) to the tank with the highest
concentration (A) in order to equilibrate the concentrations. A level difference will occur (Figure 5).
Consequently a pressure difference appears called “osmotic pressure” given by Van’t Hoff equation. (8)
-15-
A B
Semi-permeable
Δh
Membrane
Salt Fresh
water water
Thus, if one operate in the most saline capacity (A) a pressure greater than osmotic pressure, the
phenomenon will reverse and it will create a migration of water as a solvent without salinity (i.e. desalted
water) to the low concentration tank. It is called reverse osmosis (Figure 6). (9)
Applied
Pure water
pressure
A B
Semi-permeable
Membrane
Salt Fresh
water water
In industry a high pressure pump (from 25 bars to 80 bars) sends raw water in a compartment separated in
two by a semi permeable membrane that allows water as a solvent without salinity (osmosis purified water)
to flow through it. The salinity will increase in the upstream part of the membrane (left part of the drawing)
that’s why a loss of concentration or “deconcentration” is necessary in order to limit the saline
concentration. The principle of reverse osmosis is simplified in Figure 7: Reverse osmosis in industryFigure 7.
Semi-permeable
Membrane
Raw Water
Osmosed
Water
High pressure
pump
Deconcentration
-16-
In Figure 8 and Figure 9 are shown drawings of the demineralization process:
Semi-permeable
Membrane
Raw Water
High pressure
pump
Figure 8: System without pressure (Water to be treated is on the left of the membrane)
Semi-permeable
Membrane
Raw Water
Osmosed
High pressure Water
pump
Deconcentration
Figure 9: System pressurized: Water is treated and flows in the right part
The third and last step is the polishing step by mixed beds (Figure 10). It contains both cations and anions
resins. Basically, water flows through ion exchange resins that block anions and cations in the water by
replacing them by ions OH- and H+. System regeneration is made by extracting ions fixed by the resins and
replacing them by ion OH- and H+. This step is performed by in situ regenerated mixed beds which ensure
good reliability of operation in the case of quality variations of the water to be treated. Indeed, even in the
case of a quality degradation of the produced water by the reverse osmosis process, water quality obtained
at the output of mixed bed remains constant (there is an impact on the regeneration frequency, which can
be easily managed). It will then supply good water quality even if one of the several reverse osmosis
membranes is leaking. (10)
H+ OH-
Demineralized
Water Cation Anion Water
resins resins
Rejection Rejection
-17-
3.2 Demineralization station presentation
In this part will be presented the demineralization station civil works and purposes. It is necessary in order
to have a better representation of the building.
The value of 99.9% availability is required by the FMECA (Failure Mode, effects and criticality analysis)
study carried out during the execution phase. Consequently temperatures in the building needs to be
maintained at the desired level so that demineralized water can be produced 24/24h. However a storage
water tank will be implemented to manage unsuspected breakdown or problems.
-18-
4 Literature review
To get familiar with ventilation systems, smoke control systems and psychrometry a literature review has
been carried out on these different topics.
Air can enter a building through leakages. However, it cannot be considered as a proper ventilation system.
Indeed, the resulting air flows are completely uncontrollable and depend on the wind, parasites openings,
atmospheric pressure...
Equipments such as adjustable grilles must be arranged on the frontage for so-called "clean" rooms. Transfer
openings allow the passing of air to the so-called "wet" or "contaminated" rooms (bathroom, chemical room
...). In these locals, air is expelled through vertical ducts leading outside. (11)
Advantages:
A fully natural ventilation system requires no power consumption, the engine used to move the air being
wind pressure and temperature differences. Thus it is economic and reduces the building's impact on the
environment. In addition, natural ventilation elements generally require very little maintenance and do not
include noisy fans.
Drawbacks:
Natural ventilation depends on the natural phenomena creating the movement of air. Therefore air quality
might not be ensured in all the rooms. Indeed air change rate (ACH) can be disrupted by wind, temperature,
openings and the atmospheric pressure … Requested airflows are therefore hardly reachable especially for
rooms housing chemicals that need a huge air change rate of about 10 ACH.
-19-
4.1.2 Single flow mechanical ventilation
One talks about single-flow mechanical ventilation when either air supply or air extraction is realised thanks
to a fan. The most encountered single-flow ventilation consists in creating air circulation in the building so
that air gets in the building by rooms with low pollutants (offices) and then supply all rooms which contains
more pollutants or smell bad (sanitary room) before being extracted on the roof. To do so extraction fan
are required on the roof to suck air out of the building and supply grilles are requested on the building’s
frontages. (11)
Advantages:
Single-flow ventilation is simple and cheap. It requires only a limited space within the building as only
extraction air ducts are needed. Even though this method is cheap, airflow can be controlled thanks to
extraction fans. It is easily implementable and maintenance is almost inexistent. Moreover, balancing the
network is quite simple as operating speed of fans can be controlled.
Drawbacks:
The main drawback of this system is that really cold air in winter and really hot air in summer is supplied to
the building. As consequences, if big ACH are requested additional heating and cooling power will be huge.
Moreover a lot of energy is lost as air within the building is evacuated outside without being reused. Energy
used to heat up or cool down this air is then lost.
-20-
Figure 13: Double flow ventilation system with an AHU. (11)
Advantages:
Whatever the external conditions this system is the easiest to control. Air is treated before being supplied
to the different rooms, thus air is brought to a temperature close to the rooms’ temperatures which reduce
the lack of comfort. You can recycle air from the inside in order to heat up outside air in winter or cool it
down in summer, thus saving energy.
Drawbacks:
This system is expensive and space consuming. Many air ducts have to be installed it might be a problem if
the building is tight and cramped. It is a system really hard to balance and additional margins have to be
taken into account in order to consider the fact that balancing might not be perfect. It might lead to an over
sizing of the ventilation system.
4.1.3.2 Components
An Air Handling Unit can have many components that will depend on the temperature and air quality
requested in the building. Here is a non-exhaustive list of equipment that can be found in the Air Handling
Unit that would be implemented in the demineralization station:
An extraction fan
A supply fan
A filtration system that protect the AHU against dusts and particles harmful to its good working.
It is particularly important in our case as the nuclear power plant is really close to the sea and salt
might deteriorate our equipment. Several level of filtration can be found from low to high efficiency.
Fresh air damper
Reused air damper
A mixing box that mixes air from outside with air extracted from the rooms. Fresh air and reused
air dampers are synchronised to supply the right balance between fresh and extracted air.
Heating coil. It works either with hot water or with electricity.
Cooling coil. It works either directly with refrigerant or with cold water. Cold water flows through
a coil and cool down the air in the AHU. The cold water configuration will be chosen in order to
-21-
limit the mass of refrigerant in our system. Moreover a cold water tank is required in order to
increase the inertia of the system and increase the life span of the water chiller.
A humidifier is used to humidify the air if required. It can work either by steam injection or directly
with water.
Regulatory requirements require specific risk assessments to be carried out in respect of gaseous
contaminants that may be released into the workplace for toxic, asphyxiating, flammable and or explosive
gases, vapors or particulates.
Risks to workers must be eliminated or reduced as far as practical by removal of contaminants at the point
of release if possible using local exhaust ventilation systems or sufficient dilution ventilation where this is
not possible.
The regulations stipulate lower exposure action level of 80dB (A) and upper exposure action level of 85dB
(A), the exposure limit is 87dB (A). Whenever possible noise should be eliminated from the workplace or
reduced as far as it is reasonably practicable. In Table 1 are described the maximum admissible decibel value
in different building’s area.
-22-
Rooms Noise level
Main Control Room 45dB
Offices, rest room, laboratories 45dB
Corridors, computer rooms 60dB
Workshops 70dB
Ventilation Plant Rooms 85dB
Space containing local control with low occupancy 80dB
Table 1: Maximum acceptable dB in different building’s area (14)
Noise is generated by fans and will propagate through the ductwork in both directions to all inlets and
outlets. Additional noise can be introduced by components in the duct system and along the duct network.
Acoustic calculations should be performed when the duct design is completed to check that noise levels in
the most critical rooms (or those closest to the fan) are not exceeded.
Design guidance limits for low, medium and high velocity duct systems for specific applications / noise
constraints are published in CIBSE Guide B. (16) Maximum velocities in ducts are displayed in Table 2 in
m/s.
ACH requirements
A minimum ACH of 0.5 will be kept everywhere except in rooms housing chemicals that needs to be
supplied with an ACH equals to 10 in normal operation. ACH will be increased to 20 in chemical rooms
for accidental situations as for instance if harmful vapours are detected.
-23-
4.2.2.2 Rooms with employees
Temperatures
This building is almost autonomous as only two employees are present permanently to check the process.
The number of people present in the building can reach 20 during maintenance period but this case will be
neglected as it occurs only once to twice a year. That’s why temperatures must be maintained between a
narrow range of temperatures only for some rooms such as the laboratory and the control room in which
temperatures needs to be between 18°C and 26°C. Toilets, cloakrooms and cleaning rooms must be kept
within the temperature range 18°C/30°C. Humidity won’t be controlled except in the control room and the
laboratory. An average value equals to 45% will be considered for humidity within these rooms. All these
data have been taken from the specifications given by the company in charge of the process and equipments.
ACH requirements
For rooms such as the laboratory an ACH equals to 2 is requested due to the presence of chemicals. An
extractor hood will also be added over each bench in order to prevent fumes from spreading in the room.
For toilets, ASHRAE handbook (17) gives the value of 80m3/h exhaust ventilation. That corresponds to an
ACH equals to 2 in our situation. The same value will be taken for the cleaning room and the cloakroom.
2 people are present all the time in the control room, the reference (17) gives an air renewal rate
corresponding to 0.6L/s.m² or 8L/s per person. Taking the more disadvantageous an ACH equal to 1 is
obtained given the surface of the room.
4.2.2.5 Summary
In Table 3 is shown the admissible temperature range and the relative humidity that needs to be maintained
in the different rooms. The minimum ACH requirements is displayed in Table 4.
-24-
Temperatures
Room description Winter normal Summer normal Relative
operation operation humidity (%)
Chemical rooms 5 °C 40 °C Not controlled
Degasser room 5 °C 40 °C Not controlled
Basement 5 °C 40 °C Not controlled
Control room 18 °C 26 °C 30%-60%
Electrical room 10 °C 30 °C Not controlled
Laboratory 18 °C 26 °C 30%-60%
Cleaning room / Sanitary 18 °C 30 °C Not controlled
HVACroom
room 10 °C 35 °C Not controlled
Corridor 15°C 35°C Not controlled
Other 5 °C 40 °C Not controlled
Table 3: Admissible temperature range in HY building
ACH requirements
Room description ACH operational ACH in case of
chemicals
vapor detection
Chemical rooms 10 20
Degasser room 1 /
Basement 0.5 /
Control room 1 /
Electrical room 1 + extractor hood over each /
Laboratory 2 + extractortransformer
hood over each bench /
Cleaning room / Sanitary 2 /
room
Other 0.5 /
Table 4: Minimum ACH in HY building
-25-
4.3.1.1 Natural smoke extraction
For rooms with their ceiling in direct contact with the outside, CIBSE Guide E (19) specifies that natural
openings are enough to extract heat and smoke. The total area of the natural vents needs to be equal to 3%
of the floor surface area. Moreover, air must be supplied through an effective surface at least equivalent to
the openable vents surface area.
The second solution is more used in the nuclear field and it consists in pressurizing the firefighting shafts
(stairs plus lobby) thanks to a fan. The over-pressurization prevents the smoke from getting in this area.
Thus the area stays free of smoke and fire department personnel can approach, locate and put out a fire.
The regulation BS12101-6 (20) stipulates that a pressure differential equals at least to 50Pa (with all the
doors closed) needs to be maintained between firefighting shafts and the area in which the fire has started.
Moreover air speed must be equal to 0.75m/s through an open door to avoid smoke from getting in the
protected area. Fire cannot start in firefighting shafts as there is no risky equipment.
-26-
4.3.2 Practical application on the building
In this part smoke control systems sizing rules stated in §4.3.1 will be applied directly on the different rooms
constituting the demineralization station.
4.3.2.1 Hall level 0.00 (Process area)
The hall is made of a metal structure that can bend and collapse if the temperature rises too much. Therefore,
a smoke exhaust system must be implemented. It is realized by the mean of openable vents on the roof
equivalent to 3% of the floor surface (corresponding to 28m²). Air must be supplied through an effective
surface at least equivalent to 3% of the floor surface. In this case opening the two “equipment access” doors
located south of the building is more than enough (4.50m x 6.00 m).
4.3.2.3 Basement
The basement of this building is more than 200m² and 3m deep so according to BS 9999 a smoke exhaust
system must be implemented. It must be independent from air conditioning system, shall provide an air
change rate equal to 10, which is the usual value for this kind of system in the UK, according to CIBSE
Guide E (19). For the basement results are presented in Table 7.
P F
F Room F Room
Stairs Lobby Stairs Lobby
Air release path or
Air release path or
smoke exhaust
smoke exhaust
system of the room
system of the room
if there is one
if there is one
S
S
This system prevents smoke from entering into stairs and lobby by applying a pressure differential between
the firefighting shafts and other rooms of 50Pa doors closed and an airflow of 0.75m/s through an open
-27-
door.
In order to be sure that the overpressure will never reach in the stairwell a value too high to enable the
opening of the door, a pressure relief damper is installed in the stairwell, which opens at 80Pa. The
mechanical diagram of this pressurization system is displayed in Figure 14.
Calculation of the airflow caused by pressure differential (demonstration of the formula in §4.6.1.3):
Q = 0.83 . Ae . P1/R [1]
Where:
Q = airflow (m3/s)
Ae = area of leakage (m²)
P = Pressure differential between both sides of the opening (Pa)
R = 2 (for small openings)
For an opening, Ae is the cross-section of this opening. For a closed door, BS12101-6 (20) gives values of
leakage areas (see Table 8).
Estimation of the necessary airflow to maintain the air velocity at a value of 0.75 m/s through an open door:
Calculation of the equivalent airflow for a given air velocity through an open door:
-28-
Q = v . A [2]
With:
Q = Airflow (m3/s)
V = Air velocity (m/s)
A = Area of leakage (m²)
4.4 Psychrometry
Designing a ventilation system from scratch is quite complex, consequently a strong technical background
is necessary. In this part will be presented the most important technical aspects to consider and understand.
Psychrometry refers to thermodynamic and physical property of a gas-vapor mixture. Humid air being a
gas-vapor mixture, the study of its characteristics is called “psychrometry”. One of the devices allowing the
measurement of two intrinsic data of humid air is the psychrometer or wet-and-dry-bulb thermometers.
Indeed this device composed of two thermometers allows the measurement of:
Dry bulb temperature which is defined by the level of molecular agitation
Wet bulb temperature which is determined by measurement thanks to a scrap of fabric soaked with
water, wrapped around the measurement device and ventilated in order to cause water to evaporate.
This measurement is lower than dry bulb temperature of the air due to the evaporation of water.
These two features define precisely humid air characteristics. The diagram representing the different
characteristics of the humid air is called “psychrometric chart” (Figure 15).
In a building, temperature and humidity have to be controlled in order to insure the quality of a process and
-29-
acceptable working conditions for employees. Therefore water vapor might need to be added or removed
from air mixture and temperature can also need to be increased or decreased. (23)
With:
t = air temperature (°C)
Thanks to this definition one can define a well-known concept called relative humidity φ which can be
defined thanks to the following equation from (24):
𝒑𝒗
𝝋= [5]
𝒑′′
𝒗
With:
𝑝𝑣 = partial pressure of water vapor in air
𝑝𝑣′′ = saturation pressure of water at air temperature
With:
𝑚𝑣 = mass of water vapor in air
𝑚𝐴 = mass of dry air
𝑥 = water content (kgwater/kgdry air)
-30-
𝑝𝑣 . 𝑉. 𝑀𝑣
𝑚𝑣 =
𝑅. 𝑇
With:
𝑀𝑣 = Molecular weight of water vapor (18.016 g/mol)
𝑀𝐴 =Molecular weight of air (29g/mol)
𝑝𝐴 = pressure of dry air
𝑝𝑣 = pressure of water vapor
𝒑𝑨 = 𝒑 − 𝒑𝒗 [8]
With:
𝑝 = Total pressure of humid air
With:
ℎ𝐴 = enthalpy of dry air
𝑐𝑝𝐴 = specific heat of air (1 kJ/(kg.K))
T= temperature
ℎ𝑣 = enthalpy of water vapor
𝐿𝑣 = Latent heat of vaporization of water (2500 kJ/kg)
𝑐𝑝𝑣 = specific heat of water vapor (1.86 kJ/(kg.K))
ℎ = enthalpy of humid air (kJ/(kg.K)
𝑥 = water content (kgwater/kgdry air)
-31-
4.5.1 Air mixing
Air mixing is designed to control the adiabatic meeting of two humid air streams from different origins. It
is used to manage at best energy, depending on the ventilation needs. It is the case for the mixing of fresh
air with recycled air from a building. A mixing section is at least made of two synchronized dampers. By
knowing the characteristics of the two air streams one can get the water content, the specific enthalpy and
the temperature of the mixed stream.
With:
qm = Mass flow rate (kg/s)
h = enthalpy of humid air (kJ/(kg.K))
x = water content (kgwater/kgdry air)
T = temperature (°C)
qm2
h2
r2
T2
Figure 16: Mixing of two air streams called “1” and “2”
By applying mass balance equations and conservation of energy equation it is possible to determine the
characteristics of the mixed airflow.
𝑞𝑚𝑀 = 𝑞𝑚1 + 𝑞𝑚2
𝑞𝑚𝑀 . ℎ𝑀 = 𝑞𝑚1 . ℎ1 + 𝑞𝑚2 . ℎ2
𝒒𝒎𝟏 .𝒉𝟏 +𝒒𝒎𝟐 .𝒉𝟐
𝒉𝑴 = [11]
𝒒𝒎𝟏 +𝒒𝒎𝟐
The mixing of the two air streams can be represented by a line linking the points 1 and 2. The mixing point
“M” will be located on this line. The distance between the incoming airflow and the mixing point
corresponds to the air flow ratio. It is represented in Figure 17.
𝑞𝑚1
𝐿2 = (𝐿 + 𝐿2 )
𝑞𝑚1 + 𝑞𝑚2 1
-32-
Mixing L2
point
L1
With:
Troom = temperature in a room (°C)
qvroom = airflow in the room (m3/s)
-33-
Refrigerant heating coils:
Condensation of a refrigerant in the condenser is used to heat up humid air and increase its enthalpy.
-34-
Evaporation of a refrigerant in the evaporator is used to cool down humid air. Contrary to the cold water
heat exchanger, the temperature of this type of heat exchanger can be considered constant.
Humid cooling:
It happens when the cooling coil external temperature is lower than the dew point of humid air. Water vapor
will then condense and humid air will dehumidify. Through this process, both water content and dry bulb
temperature decrease but relative humidity φ increases.
If one consider a constant external temperature of the cooling coil, the process can be represented by a
straight line between the humid air starting characteristics and the cooling coil temperature at 100% relative
humidity. Of course the process will stop before reaching the point (Tcooling coil, φ=100%) as heat exchangers
don’t have infinite surface area (25). As a consequence the concept of bypass factor (§4.5.5.3 ) needs to be
introduced.
In real life the process path is more complicated as the cooling coil temperature is not the same everywhere.
Basically the temperature is not constant along the heat exchanger.
-35-
xin
xout
𝑻𝒘𝒂𝒕𝒆𝒓_𝒐𝒖𝒕 +𝒌.𝑻𝒘𝒂𝒕𝒆𝒓_𝒊𝒏
𝑻𝒄𝒐𝒐𝒍𝒊𝒏𝒈 𝒄𝒐𝒊𝒍 = [15]
𝟏+𝒌
With:
𝑇𝑤𝑎𝑡𝑒𝑟_𝑜𝑢𝑡 = Water out of the heat exchanger
𝑇𝑤𝑎𝑡𝑒𝑟_𝑖𝑛 = Water entering the cooling coil
𝑘 = coefficient depending on the cooling coil often estimated equal to 1 when no information concerning
the coil is available.
If k=1, 𝑇𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑐𝑜𝑖𝑙 is equal to the arithmetic mean of the inlet and outlet water temperature.
-36-
portion of untreated air and is given by the following formula:
With:
BF = Bypass Factor
𝑇𝑖𝑛 , 𝑥𝑖𝑛 , ℎ𝑖𝑛 = Temperature, water content and enthalpy of air entering the cooling coil
𝑇𝑜𝑢𝑡 , 𝑥𝑜𝑢𝑡 , ℎ𝑜𝑢𝑡 = Temperature, water content and enthalpy of air after the cooling coil
𝑇𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑐𝑜𝑖𝑙 , 𝑥𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑐𝑜𝑖𝑙 , ℎ𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑐𝑜𝑖𝑙 = Surface air temperature, water content and enthalpy of the
cooling coil
𝑇𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑐𝑜𝑖𝑙 will always be above 0°C in order to avoid the risks of frosting on the cooling coil surface
otherwise an automatic defrosting will be necessary.
Efficiency of a cooling coil can be defined as the percentage of treated air compared to the total mass of air.
It is defined by the following formula:
𝝃 = 𝟏 − 𝑩𝑭 [17]
Sensible cooling power: It is the power corresponding to the sensible heat taken from humid air to cool it
down
𝑃𝑐𝑜𝑜𝑙𝑖𝑛𝑔_𝑠𝑒𝑛𝑠𝑖𝑏𝑙𝑒 = 𝑞𝑚 (ℎ𝑜𝑢𝑡 − ℎ𝑆 )
Latent cooling power: It is the power corresponding to the latent heat removed from humid air to
dehumidify it.
𝑃𝑐𝑜𝑜𝑙𝑖𝑛𝑔_𝑠𝑒𝑛𝑠𝑖𝑏𝑙𝑒 = 𝑞𝑚 (ℎ𝑆 − ℎ𝑖𝑛 )
All these powers are negatives as they are extracted from the humid air.
-37-
Amount of water condensed on the coil:
xin
Latent
Sensible xout
Figure 21: Representation of sensible and latent heat during humid cooling
The humidifying process is ruled by the following energy conservation and water mass conservation
equation:
𝒒𝒎𝟏 . 𝒉𝟏 + 𝒒𝒎𝒘𝒂𝒕𝒆𝒓 . 𝒉𝒘𝒂𝒕𝒆𝒓 = 𝒒𝒎𝟐 . 𝒉𝟐 [19]
With:
𝑞𝑚𝑤𝑎𝑡𝑒𝑟 = Introduced mass flow of water
ℎ𝑤𝑎𝑡𝑒𝑟 = Enthalpy of added water (steam or liquid)
𝑞𝑚 = 𝑞𝑚2 = 𝑞𝑚1 , as one use the dry mass flow rate of air
-38-
∆ℎ
By combining the two equations above, = ℎ𝑤𝑎𝑡𝑒𝑟
∆𝑥
𝑘𝐽 𝑘𝐽 𝑘𝐽
Where, 𝑐𝑝,𝑤𝑎𝑡𝑒𝑟,𝑙𝑖𝑞 ≈ 4.2 , 𝐿𝑣 = 2500 and 𝑐𝑝,𝑤𝑎𝑡𝑒𝑟,𝑠𝑡𝑒𝑎𝑚 ≈ 1.8
𝑘𝑔.𝐾 𝑘𝑔 𝑘𝑔.𝐾
Sat
Out
In
-39-
Figure 23: humidification by steam injection on a psychrometric chart (dotted line: reality/ full line: theory)
In
L2
Mixing
point
Out
L1
Cooling Coil
Figure 24: (Left) impossible humid cooling process / (Right) impossible mixing of two air streams
-40-
A low over-pressurization around 4Pa is considered to be enough to minimize infiltrations from the outside.
(26)
With:
𝑝 = pressure (Pa)
𝑉 = volume (m3)
𝑛 = number of mols (mol)
T = temperature (K)
R = 8.314 (J/mol.K)
Number of mole variation (Δn) in a room at constant temperature and volume will have a direct influence
on the pressure within the room. If the number of mole varies of Δn, pressure variation will be proportional
and equal to Δp. As a consequence, in order to increase the pressure (Δp>0), the number of mole supplied
by the AHU needs to be superior to the number of mole extracted from the room. It works the other way
around if the pressure needs to be decreased (Δp<0).
From the equation above, in order to get a pressure equals to 𝑝 + ∆𝑝, the mole variation will be equal to:
(𝑝 + ∆𝑝). 𝑉 𝑝. 𝑉 ∆𝑝. 𝑉
∆𝑛 = − =
𝑅. 𝑇 𝑅. 𝑇 𝑅. 𝑇
∆𝑉 = 𝑉𝑜. ∆𝑛
∆𝑝. 𝑉. 𝑉𝑜
∆𝑛. 𝑉𝑜 = ∆𝑉 =
𝑅. 𝑇𝑜
Then,
∆𝑉 ∆𝑝. 𝑉𝑜
=
𝑉 𝑅. 𝑇𝑜
The molar volume varies with the temperature, let’s call Vo(θ) the molar volume at the temperature:
T (K) = θ(°C) + T0
-41-
𝑝0 . 𝑉0 𝑝. 𝑉0 (𝜃) 𝑉0 . 𝑇
= => 𝑉0 (𝜃) =
𝑇0 𝑇 𝑇0
1
With 𝐶 = ( )1/2
𝜉
-42-
5 Calculation methodology review
Main part in ventilation system sizing is the calculation methodology. Indeed it needs to be conducted
seriously since calculation errors might lead to equipment failure.
Winter Summer
Temperature -15°C / 100% HR 40°C / 32%HR
Table 9: External temperature considered for HVAC designs
The temperatures that are presented in Table 9 above are based on statistical calculations carried out by EDF
R&D. 40°C and -15°C corresponds to the daily (12 hours) average temperature considering a return period
of 10000 years and taking into account climate change. It means that once every 10000 years outside average
temperature will be equal to 40°C and -15°C during a period of 12 hours. These temperatures will be
considered as design basis for HVAC system according to (27).
5.1.3 Wind
Like external temperature, wind conditions are defined in (27) and detailed in Table 10:
Winter Summer
Wind speed 4m/s 1m/s
Table 10: Wind speed considered for HVAC designs
-43-
5.2.1.1 Heat transfer through walls
Heat exchanges through a wall/roof depend on its dimensions (surface area, thickness), characteristics
(material), connections (between rooms, with ground, with outside air) and position (horizontal or vertical,
exposed to solar radiation). These parameters are taken into account in the following formula:
Pwall = U . S . ΔT [25]
In HY building, all walls are either made of concrete or cladding and K can be calculated as follows:
𝟏 𝒆 𝟏 −𝟏
𝐔=( + + ) [26]
𝐡𝐢 𝝀 𝐡𝐞
Heat transfer coefficients depend on wind speed, and are calculated according to CIBSE Guide A (29):
1/he 1/hi
Vertical wall Summer 0.08 0.13
Winter 0.04 0.13
Horizontal wall Summer 0.08 0.10
(ascending flow) Winter 0.04 0.10
Horizontal wall Summer 0.08 0.17
(descending flow) Winter 0.04 0.17
Table 11: heat transfer coefficients depending on walls and season
These coefficients have been calculated considering the depth of the building’s floor.
In summer conditions, the influence of solar radiation is considered through the definition of virtual outside
temperature. This temperature has been calculated according to ASHRAE method (17), in consistency with
CIBSE Guide A (29). This method takes into consideration the solar radiation, orientation and inertia of
-44-
the building and defines a correction coefficient for the temperature of an external wall under solar radiation.
Solar radiation cause on opaque walls an increase of the external walls temperature which is higher than the
one if there was no sun. Thus calculations take into account a fictive external temperature superior to the
actual temperature outside.
CLTD = It stands for Cooling Load Temperature Difference. Its value depends on orientation, type of wall
and hour of exposure (walls are considered type A for concrete and type G for cladding, walls group depends
on inertia).
LM = correction coefficient depending on month and latitude of exposure (52° for Hinkley Point).
Tint = temperature required in the other side of the wall (40°C has been taken into account, the worst
conditions that can be seen in HY building).
Tem = mean temperature over a period of 24h (40°C for summer conditions).
k = correction factor depending on the color of the wall (here 0.83).
f = correction factor depending on the existence of an attic (here 1).
Values for CLTD, LM, k and f can be found in (17) and (28) and are presented in Table 12 and Table 13
below:
Concrete Class A
Angle Vertical Vertical Vertical Vertical Horizontal
Orientation North East South West -
CLTD 6 12 8 10 22
LM 1,1 1,6 3,3 1,6 0,5
k 0,83 0,83 0,83 0,83 1
CLTDcorr 1,993 7,388 5,479 5,728 18,6
Table 12: Values of CLTD, LM and k for concrete walls more than 300mm thick
Cladding Class G
Angle Vertical Vertical Vertical Vertical Horizontal
Orientation North East South West -
CLTD 13 17 24 31 43
LM 1,1 1,6 3,3 1,6 0,5
k 0,83 0,83 0,83 0,83 1
CLTDcorr 7,803 11,538 18,759 23,158 39,6
Table 13: Values of CLTD, LM and k for cladding
-45-
Heat gains due to direct radiations (17):
Heat gains = S . MSHGF . SC . CLF [29]
Windows
Angle Vertical Vertical Vertical Vertical
Orientation North East South West
CLTD 8 8 8 8
MSHGF 142 681 609 681
SC 0,81 0,81 0,81 0,81
CLF 0,72 0,49 0,56 0,54
Table 14: Values of CLTD, MSHGF, SC and CLF for a Pilkington double glazed window
Additional air heaters and air conditioning unit are also taken into account.
-46-
Sensible heat can be found in Table 16 taken from EN13779 (26). It depends on human activity. It is assumed
that the 2 people permanently present in the building will have a sedentary activity corresponding to a
sensible heat load equals to 75W/person.
Degree of Activity Total heat, Adult (W) Sensible heat (W) Latent heat (W)
Seated at theater 95 65 30
Seated at theater, night 105 70 35
Seated, very light work 115 70 45
Moderately active office work 130 75 55
Standing, light work: walking 130 75 55
Walking, Standing 145 75 70
Sedentary work 160 75 80
Light bench work 220 80 140
Moderate dancing 250 80 160
Walking 5km/h, light machine work 295 90 185
Bowling 425 110 255
Heavy work 425 170 255
Heavy machine work 470 185 285
Athletics 525 210 315
Table 17: Representative rates at which heat and moisture are given off by human beings for different activities (17)
-47-
Table 18: Loss of heat and water vapor in the human body for an individual sitting in light activity (32)
It only concerns the supply metal ductwork. To be precise, it only concerns, for a given room, the part of
the metal ductwork located outside of the room. It means that the supply air, before entering the room it is
supposed to cool, will thermally exchange with the rooms that it is passing through. Given that the
difference between the temperature of supply air and the temperatures of the crossed rooms is significant
(around 15-20K in average), then heat transfers will be important.
The impact of this design tolerance is the over-cooling of the rooms crossed by supply ductworks (i.e. close
to the supply shaft), and the under-cooling of the rooms located at the end the distribution network (i.e. far
from the supply shaft). Even if the overall energy balance is correct, it actually results in a destabilization of
-48-
the cooling distribution, and thus the calculated temperatures will be exceeded in some rooms (the further
ones). (28)
The case of heating is neglected as heat gains from the fan compensate heat losses while air is crossing the
rooms. Whereas for the cooling case heat gains from the fan and from the rooms crosses are added, it is a
lot more restrictive.
The rate of heat transfer Φ (W) between the room at temperature T room and the temperature of the air in
the duct Tduct, with external surface area S (m²) can be written:
Φ = U. S. (Tduct − Troom )Pwall = U . S . ΔT [25]
Where U (W/(m².K)) is the thermal coefficient for the duct (circular or rectangular)
Calculation of thermal transmittance for rectangular ducts (28):
o Internal surface resistance (convective) = 1/hi
o The thermal resistance constituted by successive layers of thickness e with conductivity λ
o Internal surface resistance (convective) = 1/he
1𝑈=1ℎ𝑖+𝑒λ+1ℎ𝑒𝐔=(𝟏𝐡𝐢+𝒆𝝀+𝟏𝐡𝐞)−𝟏 [26
Calculation of thermal transmittance for circular ducts (28):
The principle is the same as for rectangular ducts except that the thermal resistance of the duct
changes.
𝟏 𝟏 𝟏 𝑫 𝟏
= + (∑ 𝐥𝐨𝐠( 𝑬 )) + [31]
𝑼 𝒉𝒊 𝛌 𝑫𝒊 𝒉𝒆
Information from reference (33), are gathered in Table 19 and Table 20 below:
Duct Vertical Duct horizontal (heat Duct horizontal (heat flow Circular
arrangement Duct flow up) down) duct
The influence of heat transfer between air in ducts and rooms will be studied further in the report thanks
to ANSYS Fluent which is a computational fluid dynamics software. After the simulation, it will be possible
to assess the impact of this heat exchange between the room and air within the air duct.
-49-
The absorbed fan power is realized in the airstream as a temperature rise across the fan and additionally as
temperature rise due to frictional interaction throughout the duct network.
For fan arrangements where the motor is outside the airstream, the heat gain to be taken into account is the
one of the fan power (16):
𝑸.∆𝑷
𝑷𝒇𝒂𝒏 = [32]
𝜼𝑭
Where:
𝑃𝑓𝑎𝑛 = fan power (kW)
Q = volumetric airflow (m3/s)
∆𝑃 = Fan differential pressure (kPa)
𝜂𝐹 = Fan efficiency
However where the fan motor is located in the airstream the motor losses must also be taken into account,
the heat gain to be taken into account is the absorbed motor power (16):
𝑸.∆𝑷
𝑷𝑴𝒐𝒕𝒐𝒓 = [33]
𝜼𝑭 .𝜼𝑴 .𝜼𝑫
Where:
𝑃𝑀𝑜𝑡𝑜𝑟 = Motor absorbed power (kW)
𝜂𝑀 = Motor efficiency
𝜂𝐷 = Drive efficiency
Fan efficiency depends on fan selected and are presented in Table 21 below:
Drive efficiencies are considered to be around 97% for belt driven fans and 100% for direct driven fans.
Motor efficiency can be found in ASHRAE (17) and are presented in the Table 22 below:
Minimum Nominal Full Load Efficiency (%) for Motors Manufactured after December 2010
Number of poles 2 4 6
Synchronous speed (RPM) 3600 1800 1200
Motor (kW)
0.8 77 85.5 82.5
1.1 84 86.5 86.5
1.5 85.5 86.5 87.5
2.2 85.5 89.5 88.5
3.7 86.5 89.5 89.5
5.6 88.5 91.0 90.2
7.5 89.5 91.7 91.7
11.1 90.2 93.0 91.7
14.9 91.0 93.0 92.4
18.7 91.7 93.6 93.0
22.4 91.7 94.1 93.6
29.8 92.4 94.1 94.1
37.3 93.0 94.5 94.1
44.8 93.6 95.0 94.5
56.0 93.6 95.0 94.5
Table 22: Motor efficiency (17)
-50-
The temperature difference can then be calculated by the following formula:
∆𝒑
∆𝑻 = [34]
𝜼𝑭 .𝜼𝑴 .𝜼𝑫 .𝒄𝒑𝒂𝒊𝒓 .𝝆
With:
𝝆 = air density (kg/m3)
∆𝑃 = Fan differential pressure (kPa)
𝜂𝑀 = Motor efficiency
𝜂𝐷 = Drive efficiency
𝜂𝐹 = Fan efficiency
∆𝑇 = Temperature variation (K)
5.4 Margins
The objective of this section is to identify and propose compensations to the design tolerances that appear
along the design processes. Without allowance, a design tolerance may pose a threat to several activities,
such as safety justifications, sizing of the systems in interface, contract management, layout activities or
electrical supply sizing. In this part margins that have been taken into account in the design are presented.
The document used for reference in this part is the allowance strategy report (34).
A design tolerance is defined as a characteristic which alters the precision or the correctness of the activities
performed in the design process. It mainly comes from uncertainties in input data and from imprecision
intrinsic to the establishment of models for calculations.
Because of the risk it brings along, the design tolerance needs to be compensated by an allowance. The
design tolerance and the associated allowance can concern different parameters such as: heat load, air flow
rate, cooling power, electrical power or mechanical dimension. (34)
-51-
5.4.2 On flow rates
Due to flow rate balancing
Even if commissioning activities happen at the end of the design stage, they have to be anticipated in the
first design stage. Indeed, balancing a HVAC system is a very tricky activity, with results that cannot be as
perfect as the theoretical calculations.
Actually the thermal calculation results in a minimum supply flow rate, associated to a supply temperature,
to be set for each room. The setting is performed by an action on the pressure loss, through setting a manual
damper when looking at the flow rate measurement in the associated line. This setting is not precise and
moving the damper does not provide a linear response in terms of flow rate. Another issue is that modifying
one line unbalances all others. Finally, the final flow rate cannot be lower than the one calculated, as it is a
minimal requirement. In addition, the flow rate measurement chain has its own uncertainties.
So in fact, two Design Tolerances can be identified as part of the flow rate balancing activity during
commissioning activities. The first one is linked with the flow rate measurement uncertainties. The second
one is linked to the impossibility of achieving a perfect balancing.
It is to be noted that most of the air-conditioning systems are equipped with supply and exhaust ducts. Thus
the balancing is necessary for supply and exhaust ductworks, as flow rate in a given room represents the
pressure balance in supply and exhaust lines. (34)
First Design tolerance
It can be considered as the flow rate measurement uncertainties that is actually a combination of
uncertainties that depend on measurement parameters in reference to the standard depending on the type
of measurement mean used (Hot wire anemometer, Pitot tube …), the I&C system settings and the entire
chain measurement.
This flow rate uncertainty margin is taken equal to 10% and is due to the flow rate measurement uncertainty
using measurement devices as for instance a Pitot tube or an axial anemometer. (34)
Second design tolerance
It represents the difficulty of the ductwork balancing activity due to the technical difficulty of the task.
According to the HVAC tenderer, leeway value can’t be decreased below +5% on the global supply flow
rate. (34)
Total margin
The total margin due to flow rate balancing is equal to 1.05x1.10 = 1.15 = 1+15%
Duct leakages
Ductworks that are used for air distribution in HVAC systems are not airtight as there is a light over-pressure
in supply ducts due to the fan. This non-air tightness means that there is a loss of flow rate between the
supply fan and the rooms to be ventilated. For air-conditioning purposes, this loss is prejudicial as it means
that the calculation conditions are not met. This is a design tolerance. Only the part of the ductwork outside
of the ventilated room is concerned by the design tolerance. A leakage inside the room has no negative
effect.
To cope with this Design Tolerance, allowance acting on the global supply flow rate is defined. This
allowance is a margin, which value represents the average leakages in the system ductwork. The value is
quite difficult to assess though. It has been considered as 1% in previous designs and this value will be
maintained in this thesis. (28)
-52-
5.5 Energy balance of a room
Doing the energy balance of a room where temperature and humidity are constant means that:
Power supplied to the local is equal to the power lost by the local
Humidity supplied to the local is equal to humidity lost due to condensation and extraction
of humid air.
On Figure 25, air is supplied at a state 1 and extracted at a state 2.
With:
𝑃1 = 𝑞𝑚1 . ℎ1
𝑃2 = 𝑞𝑚2 . ℎ2
∆𝑞
𝑞𝑚2 = 𝑞𝑚1 − . 𝑞𝑚1 − 𝑞𝑣𝑖𝑛𝑓
𝑞
Then,
𝑷𝒕𝒐𝒕𝒂𝒍 +𝒒𝒗𝒊𝒏𝒇 .𝒉𝟐
𝒒𝒎𝟏 = ∆𝒒 [35]
((𝟏− ).𝒉𝟐 −𝒉𝟏 )
𝒒
-53-
5.5.3 Humidity balance
In a similar way humidity equation can be written, 𝑀1 + 𝑀 = 𝑀2
With:
𝑀1 = 𝑞𝑚1 . 𝑥1
𝑀2 = 𝑞𝑚2 . 𝑥2
The parameter γ can then be defined by neglecting 𝑞𝑣𝑖𝑛𝑓 and is equal to:
∆𝑞
𝑃𝑡𝑜𝑡𝑎𝑙 (1 − 𝑞 ) . ℎ2 − ℎ1
𝛾= ≈
𝑀 ∆𝑞
(1 − 𝑞 ) . 𝑥2 − 𝑥1
It represents the slope of the space line (see Figure 26); it is an intrinsic characteristic of the room.
Δh
Δx
Figure 26: Evolution of psychometry in a room, from the supply state 1 to the extraction 2 (space line)
-54-
5.6.1 Characteristic of the blowing air temperature difference
It represents the temperature difference between the air supplied to the room and the ambient temperature
in the room.
∆𝑇 = 𝑇𝑟𝑜𝑜𝑚 − 𝑇𝑠𝑢𝑝𝑝𝑙𝑦
It can be either positive or negative. Usual values to be taken are consistent with reference (30):
Cold blowing: +5𝐾 < ∆𝑇 < +12𝐾
Hot blowing: −20𝐾 < ∆𝑇 < −5𝐾
5.6.2 Mixing rate
It is written τ (see Table 23) and is equal to:
qv
τ=
V
With:
qv = Blowing volumetric rate (m3/s)
V = Room’s volume (m3)
τ = Mixing rate (Vol/h)
-55-
Comparison between the airflow required in winter, summer and the ACH. For each room the
maximum of these three airflows is selected in order to get a first guess of the airflow that needs to
be supplied.
-56-
5.6.3.2 Second approach
In order to go a bit further, optimize the system and check the results, exact temperatures in each room have been calculated for steady state under extreme external
temperatures in winter (-15°C) and in summer (40°C). Here are the different stages of this second approach:
Writing of the thermal equation for each room in the building, giving us a system of 25 (number of rooms) equations and 25 unknown data (temperature in
each room).
Conversion of these 25 equations into a matrix system: A*T = B
A is a 25 rows, 25 columns matrix
T is the vector containing the 25 temperatures
B is a constant vector depending on each room’s supplied airflow and constant temperatures.
Inversion of the matrix A in order to get the exact temperatures T, T=A-1*B
Verification of the temperature in each room and optimization of the DVT system by varying the airflow.
The second approach is realized thanks to a matrix system. In order to explain calculations easily a simplified model with only two rooms in contact has been chosen:
C
N
q2 q1
Tsupply W E
T1 Tsupply
T2
Tout
P1 P2
S F
Tfloor
Figure 27: Drawing of two rooms used for Excel modelling and cardinal directions
In order to keep the model simple, only two rooms are considered in the model and walls’ temperature is considered to be independent from the orientation (see
Figure 27). Only the floor has a different temperature. Real calculations have been realized considering different temperatures depending on orientation of the wall.
T1 = Room 1 temperature Xf = Parameter X for floor
P1 = Internal gains within room 1 Xc= Parameter X for ceiling
T2 = Room 2 temperature Xn= Parameter X for North wall
P2 = Internal gains within room 2 Xs= Parameter X for South wall
Tsupply=Temperature of incoming air Xw= Parameter X for West wall
q1= airflow supplied in room 1 Xe= Parameter X for East wall
q2= airflow supplied in room 2
Tout= Outside Temperature
Tfloor=Floor temperature
S = Surface
U=overall heat transfer coefficient
∆𝑞
Uf1 ∗ Sf1 ∗ (Tf − T1) + (Uc1 ∗ Sc1 + Un1 ∗ Sn1 + Us1 ∗ Ss1 + Uw1 ∗ Sw1) ∗ (Tout − T1) + Ue1 ∗ Se1 ∗ (T2 − T1) + P1 = q1 ∗ ρ ∗ cp ∗ ((1 − ) . 𝑇1 − 𝑇𝑠𝑢𝑝𝑝𝑙𝑦 ) − 𝑞𝑣𝑖𝑛𝑓 . 𝑐𝑝. 𝑇𝑠𝑢𝑝𝑝𝑙𝑦
𝑞
∆𝑞
Uf2′′′′ ∗ Sf2 ∗ (Tf − T2) + (Uc2 ∗ Sc2 + Un2 ∗ Sn2 + Us2 ∗ Ss2 + Ue2 ∗ Se2) ∗ (Tout − T2) + Uw2 ∗ Sw2 ∗ (T2 − T1) + P2 = q2 ∗ ρ ∗ cp ∗ ((1 − ) . 𝑇2 − 𝑇𝑠𝑢𝑝𝑝𝑙𝑦 ) − 𝑞𝑣𝑖𝑛𝑓 . 𝑐𝑝2 . 𝑇𝑠𝑢𝑝𝑝𝑙𝑦
𝑞
∆𝑞
Then the associated matrix A is identified for rooms that are not over-pressurized ( =0 and 𝑞𝑣𝑖𝑛𝑓 =0):
𝑞
−(Uf1 ∗ Sf1 + Uc1 ∗ Sc1 + Un1 ∗ Sn1 + Us1 ∗ Ss1 + Uw1 ∗ Sw1 + Ue1 ∗ Se1) − q1 ∗ ρ ∗ cp Ue1 ∗ Se1= Uw2 ∗ Sw2
Ue1 ∗ Se1= Uw2 ∗ Sw2 −(Uf2 ∗ Sf2 + Uc2 ∗ Sc2 + Un2 ∗ Sn2 + Us2 ∗ Ss2 + Uw2 ∗ Sw2 + Ue2 ∗ Se2) − q2 ∗ ρ ∗ cp
Eventually,
𝑇1 −Tf ∗ Uf1 ∗ Sf1 − (Uc1 ∗ Sc1 + Un1 ∗ Sn1 + Us1 ∗ Ss1 + Uw1 ∗ Sw1) ∗ Tout − P1 − q1 ∗ ρ ∗ cp ∗ Tblow
𝐴. =
𝑇2 −Tf2 ∗ Uf2 ∗ Sf2 − Tout ∗ (Uc2 ∗ Sc2 + Un2 ∗ Sn2 + Us2 ∗ Ss2 + Ue2 ∗ Se2) − P2 − q2 ∗ ρ ∗ cp ∗ Tblow
By inverting the matrix A, T1 and T2 can be found depending on q1 and q2. Then the relevance of airflows found during the first approach can be checked and
improved. Optimization of the design is possible by changing airflows and checks their influence on rooms’ temperature. Same reasoning process is used for the hall
which is pressurized except that Δq/q ≠0 and qvinf≠0.
-58-
6 TH-Bât Software
Thermal calculations in buildings are made with ThBat software. This software was developed under the
supervision of EDF Research & Development, specifically to meet the challenges of the Great ‘’Hot’’
project for extreme external temperature in summer.
ThBat is a code dedicated to thermo-aeraulic calculations in buildings for transient regime (permanent
regime appears as a limit case of the transitional regime). It is based on a modeling nodal principle: each
node is called air zone. For each air zone, representing a room or group of rooms, the software solves a
system of differential equations representing mass balance, momentum and energy. It therefore does not
take into account the rooms’ geometry (apart from the heat exchange surfaces through the walls, with
adjacent areas) and assumes the temperature of a thermal zone homogeneous, which leads up to the concept
of ambient temperature. Gases are considered as ideal gases. Heat transfer through the walls follows the
heat diffusion law 1D.
The rooms are represented by homogeneous air zones. For each air zone one associate:
- Thin walls (doors, windows) or thick (walls, partitions).
- Internal gains (regulated or not).
- Ventilations that can be regulated depending on the temperature.
- Vertical or horizontal openings
Elements whose temperatures do not vary during the calculations such as the deep ground are represented
as adjacent zones (ZA).
The following values also need to be provided and input in the software:
Outside temperature (°C)
Supply temperature in the rooms
Simulation time (s)
The time interval of the calculation (s)
Adjacent zones that are thermodynamic sources at constant temperature and humidity.
6.2 Zone property
A zone is an object where the evolution of temperatures will be calculated during the simulation. Concretely,
zones represent rooms and are defined by the following characteristics:
Its volume (m3)
Air density and specific heat within the room (kg/m3 and J/kg.K). If the fluid in the zone is air, the
density will be calculated automatically as a function of the temperature, pressure and humidity of
the air.
A total thermal power dissipated within the zone (W). This power will be chosen constant at its
maximum in our case as calculations will be made in the worst case scenario.
Its height under the ceiling (m)
In addition to calculated zones, the outside and adjacent zones are considered as zones with constant
temperature. The adjacent zones and the outside will be considered as “air zone”. It means that density is a
function of the temperature, pressure and humidity.
Each wall layer represents a part of the wall. A layer has to be defined by:
A material (thermal conductivity, specific heat and density)
A thickness (m)
hleft T1 T2 T3 T4 T5 T6 hright
-60-
The thickness of each layer must be sagely chosen depending on the nature of the material it is made of and
the timescale of the studied phenomena. The decisive criteria that will be considered to choose a layer
thickness depend mainly on the thermal diffusivity of the material and the speed of the studied phenomena.
For example if studied temperatures are in the range of the minute, the layers will have to be thinner than
for variations in the order of the hour of the day. Indeed if layers are too thick, some quick thermal
phenomena could be hidden. If the layers are too thin, the calculation time might be too long. Results for
concrete and cladding are displayed in Table 26.
A good solution for most of the cases is to take a Fourier number for each layer around 3 or 4.
𝝀
𝛂= [39]
𝝆∗𝒄𝒑
𝛂 .𝒕
𝑭𝒐 = [40]
𝒍²
-61-
7 Parametric study
Ventilation systems are energy consuming especially in nuclear buildings because they work 24/24h. It is
then necessary to carry out a parametric study in order to optimize the system and thus save money and
energy. In this part the impacts of building civil works and ventilation design on energy consumption of will
be studied.
It is logical that windows should be implemented on the North surface as the MSGHF is much smaller
during summer months. In winter it is not supposed to have any influence as heat gains due to solar
radiations are neglected in order to size the HVAC system in the worst case scenario.
In Table 28 are presented the results obtained with the building model realized on Excel for a glazed surface
equivalent to 20% of the total wall surface, an outside temperature of 40°C and an inner temperature
assumed equal to 40°C (worst acceptable temperature in the hall).
Windows location North South West
Heat gains through walls in the main hall (W) 21800W 36500W 35525W
Comparison compared to the minimum value +0% +67.5% +62.8%
Table 28: Heat gains for different windows location
The location of glazed surfaces has an important impact on thermal heat gains through walls in the hall.
After discussion with engineers responsible for thermal design of nuclear based buildings at EDF, it is more
reasonable to consider windows on the North wall of the building. As a consequence, all the parts that
follow have been made with the assumption that the building will have a glazed surface corresponding to
20% of the total North wall surface. The value of 20% is consistent with UK building regulation. (35)
-62-
be interesting if it shows that for a rapid increase of outside temperature to 40°C, it takes long enough to
exceed the maximum admissible temperature.
AHU
Figure 29: Cross section of the demineralization station with an AHU supplying air everywhere
As shown in Figure 29, one Air Handling Unit supplies air to the whole building. It means that air sent to
the building is treated by the AHU before.
The calculations for this design have been realized both with Excel and with Th-bât software. Air supplied
to the building by the Air Handling Unit has a temperature of 15°C. Figure 30Figure 31 shows the results
obtained on Th-bât for rooms located in the basement.
The maximum admissible temperature for all these rooms which are located in the basement is 40°C. On
this graph which plots the room temperature as a function of the time it is clear that temperature is not
going over 30°C when the steady state is reached. It is 10°C lower than the maximum admissible
temperature. Using an AHU to insure air renewal in the basement is pointless. Indeed, it has a negative
impact on the energy use and the size of the Air Handling Unit.
-63-
31
29 Soda/Acid/Morpholine Room
27 Bisulphite room
Temperature (°C)
Acid/Base room
25
Lobby West level -6,8
23 Stairs West
21 Degasser room
Lobby South level -6,8
19
Fire Fighting Valve room
17
Stairs South East 1
15 Main hall basement
0 5 10 15 20 25
Time (days)
Figure 30: Steady state temperature of the basement (10 rooms) for 40°C outside with an AHU
Discussion:
This design is really energy consuming in summer. Indeed, the basement is underground and naturally
cooled down by the chill ground and deep ground. Using an AHU in order to supply air to the basement is
not the best solution as heat transfers through the basement walls totally compensate heat gains from the
basement. As a consequence, this design is clearly not optimized as it would result in an over-sizing of the
ventilation system.
7.2.1.2 Second design: AHU supplying only rooms above level 0.00
Air supplied by the AHU
AHU
Figure 31: Cross section of the demineralization showing the two distinct parts of the building
-64-
As shown in Figure 31, air supplied to the basement comes directly from outside whereas air that is sent to
rooms above level 0.00 is conditioned by an AHU.
The calculations for this design have been realized both with Excel and with Th-bât software. Air supplied
to the building by the Air Handling Unit has a temperature of 15°C. However conditioned air will only
supply rooms located above ground level. The basement will be supplied with air from outside in order to
take advantage of chill ground to cool down warm air from outside. Below are shown graphs realized with
Th-bât and then exported on Excel.
37
35 Soda/Acid/Morpholine Room
33 Bisulphite room
Temperature (°C)
31 Acid/Base room
29 Lobby West level -6,8
27 Stairs West
25 Degasser room
Figure 32: Steady state temperature of the basement for 40°C outside without an AHU
Discussion:
With this design, the steady state temperature in the basement gets closer to the maximum admissible value
of 40°C without going over it. A simple extraction fan will be required to ventilate the basement. No air
conditioning by an AHU is necessary.
The size of the AHU can be decreased by almost 40% with the second design. Moreover energy saving with
the second design in summer is important; there is about 44% difference between the two designs.
Installation costs for the first design are much more important as really long supply and extract air ducts are
needed in order to link the basement to the AHU. Additional costs of a bigger AHU, cooling coil and water
chiller are also quite important. The second design is put aside for the rest of the study as it is not relevant
being more expensive and more energy consuming.
-65-
7.2.2 33°C outside temperature, design without AHU
Some nuclear buildings are designed for 33°C as outside temperature but only if it is possible to provide
enough cooling with a single flow mechanical ventilation system (27). Consequently, no AHU will be used
but only several extraction fans located on the building’s roof (see Figure 33). It is only possible because the
maximum admissible temperature in most of the rooms is bigger than outside air temperature. This design
results in a cost reduction at many levels (HVAC, civil works...). Indeed total length of air ducts is drastically
reduced, AHU and water chillers disappear. In order to validate this design, the influence of a quick
temperature rise from 33°C to 40°C will be studied.
Figure 33: Cross section of the demineralization station with a single flow ventilation system
The calculations for this design have been realized both with Excel and with Th-bât software. Air supplied
to the building has a temperature equal to the outside temperature. Below are shown graphs realized on Th-
bât and then exported on Excel.
Only the rooms with sensitive equipment which are not cooled down by additional air conditioning unit
will be plotted on Figure 34 and Figure 35:
45
Temperature (°C)
40
35
25
20
0 10 20 30 40 50
Time (days)
Figure 34: Temperature evolution in the main hall after a rapid increase of outside temperature from 33°C to 40°C
-66-
46
45
44
Temperature (°C)
43
42
41 Main hall level 0
40
39
38
37
-10 -5 0 5 10
Time (hours)
The first step is a 33°C outside temperature condition until the steady state is reached (horizontal
asymptote). The second step consists in a rapid increase of the temperature to 40°C. These two graphs
above plot the temperature evolution in the hall housing the demineralization process after a temperature
rise from 33°C to 40°C.
From §4.2.2.1 membranes in the reverse osmosis skids can withstand air temperature up to 43°C. Then the
design will be considered as relevant if the temperature in the hall does not exceed 43°C during six hours
for a constant outside temperature equal to 40°C. (28)
Figure 34 and Figure 35 show that it is impossible to design a reliable ventilation system without an AHU for
33°C outside. Indeed the temperature within the hall reaches 44°C in less than two hours. Thus it is not
reasonable because it represents risks towards the demineralization process which are not consistent with
the FMECA study. This solution will be abandoned for the rest of the study due to the risks it represents
toward the process.
This coefficient has been taken equal to 0.83 for the baseline scenario. For walls painted in white, the value
of this coefficient decreases to 0.65
-67-
Values for k can be found in (17) and (28).
In Table 30 are shown the results obtained with Th-bât and Excel:
Main purpose of the AHU is to ensure a minimum air change rate but the power supplied by the AHU is
often not enough to maintain air temperature within the requested range in most of the rooms. Thus,
Padditional heating and Padditional cooling refer to air conditioning units and convectors that are added in order to
compensate for the lack of power supplied by the AHU.
It has a direct influence on the thermal heat transfer coefficient U of the walls §5.2.1.1. In Table 31 are
displayed the results obtained with Th-bât and Excel:
-68-
7.3.4 Comparison with the baseline scenario
In Table 33, are summarized the differences in AHU airflows and power required in order to maintain
ambient conditions within the rooms at a reasonable level for extreme external temperature.
Baseline scenario Scenario 1 Scenario 2 Scenario 3
AHU airflow 22050 m3/h -4,8 % -14,3 % -17,4 %
Pcooling coil 330 kW -4,8% -14,3 % -17,4%
Pheating coil 115 kW -7,8 % -11,3 % -12,1 %
Padditional cooling -31 kW -16,1% -16,1 % -19,3 %
Padditional heating 111,5 kW +3.1 % -18,4 % -18,4%
Table 33: Comparison between the three different scenarios for 40°C and -15°C
In order to assess what is the influence of the different scenario on the power consumption, it is relevant
to consider reasonable external temperature and not extreme external temperatures which are used to design
the ventilation system. The temperature range that will be studied in this part is the range 2°C/20.5°C which
correspond to the average coldest month and average warmest month temperature in Somerset England
according to (36).
2°C Baseline Scenario 1 Scenario 2 Scenario 3
Pheating coil scenario
57kW 54kW 51kW 50kW
Padditional heating 19,5kW 19,5kW 15kW 14,5kW
Total 76,5kW 73,5kW 70,5kW 64,5kW
Table 34: Comparison between the three different scenarios for 2°C
Influence of these parameters on the power consumption of the system is important for extreme external
temperature (see Table 33) but it can be overlooked for temperatures between 2°C and 20.5°C (see Table 34
and Table 35). Furthermore after discussion with one of the person in charge of civil works in the company,
it is not reasonable to consider these scenarios for the following reasons:
- Concrete walls must be checked for cracks at a regular frequency. Consequently, additional external
insulation is not a feasible solution as it would hide the possible cracks in the walls.
- Painting could be a solution but the power plant is close to the sea and salt from seawater will
damage the painting really quickly. Then they would have to redo it every ten to fifteen years which
is not worth it considering the energy gains.
- The price of the two previous solutions is really important and it would not compensate the energy
gains.
A further analysis month by month could have been carried out as in §7.4 but considering the three previous
statements written before, it has been abandoned.
The baseline scenario will be maintained as the more feasible scenario and used in the following part of this
paper.
-69-
7.4 Ventilation system control
In nuclear power plants HVAC system needs to be kept as simple as possible. Indeed, robustness will be
favored over energy efficiency. As a consequence, in most of the buildings in a nuclear power plant there is
no control. It means that the Air Handling Unit is designed for summer condition which is the most
restrictive season and same airflows are considered for winter conditions.
15
14
13
Temperature (°C)
12
11
10 Main hall level 0 without
control
9
8 Main hall level 0 with control
7
6
5
0 5 10 15 20 25
Time (days)
Figure 36: Temperature of the hall depending on the presence of a control system
HVAC control system that is going to be treated in this part is really simple in order to stick to the nuclear
building ventilation sizing rules. It consists in a two speed fan both for supply and extraction (each speed
corresponding to one season either winter or summer), motorized dampers that will balance the network
when the speed changes and one captor that will monitor the outside temperature. If the outside temperature
reaches a certain threshold the fan speed will change and the dampers will be activated.
Airflows have been calculated for summer as it is the most restrictive season. For winter conditions the
minimum air renewal will be insured in order to reduce the heating coil power and AHU’s fan power.
Additional heating elements such as radiator or air heater will be preferred. The outside temperature at
which the fan speed changes is equal to 19°C according to the calculation realized. In order to avoid fan
speed changes all the time when temperature is around 19°C, fan speed will change when temperature
reaches 19°C and will change again at 18°C. This way one has a hysteresis cycle.
Winter (-15°C) / Without control Winter (-15°C) / With control
AHU airflow 22050 m3/h 13069 m3/h
Pheating coil 115kW 98kW
Pheating 111,5kW 116kW
Pmotor fan 7,5kW (495Pa) 1,8kW (200Pa)
Ptotal 234kW 216kW
Table 36: Power consumption for HVAC system with and without control by -15°C outside
-70-
By knowing the length and the cross section of the air ducts it is possible to estimate the pressure losses in
the circuit for both scenarios (with and without regulation). The formula used to calculate the motor
absorbed power can be found in section §5.3.2.
𝑄. ∆𝑃
𝑃𝑀𝑜𝑡𝑜𝑟 = [33]
𝜂𝐹 . 𝜂𝑀 . 𝜂𝐷
In order to assess the benefits of having a control system, a deeper analysis month by month needs to be
carried out. The daily average temperatures are summarized in Table 37.
The outside temperature rarely exceeds 19°C (fan speed switching temperature) so a control system is
necessary otherwise the HVAC system will be oversized.
In order to assess the real benefits of a control system it is necessary to calculate the power gains for normal
weather condition and not for extreme conditions as it was the case for sizing the HVAC system. Thus for
each month the energy consumption of the HVAC system will be assessed with and without the control
system (at normal average temperature).
-71-
With Control Without Control
Padditional heating 11kW 11kW
May Pheating coil 0kW 0kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Padditional heating 6kW 6kW
June Pheating coil 0kW 0kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Padditional heating 3,5kW 3,5kW
July Pheating coil 0kW 0kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Padditional heating 3,5kW 3,5kW
August Pheating coil 0kW 0kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Padditional heating 6kW 6kW
September Pheating coil 0kW 0kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Padditional heating 11kW 11kW
October Pheating coil 0kW 0kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Padditional heating 11kW 11kW
November Pheating coil 25kW 34kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Padditional heating 13kW 13kW
December Pheating coil 31kW 41kW
Pmotor fan 1,8kW (200Pa) 7,5kW (495Pa)
Table 38: Assessment of energy gains for each month with and without a control system
Table 39 is a summary of the energy saved each month by having a control system. First column has been
calculated by calculating the difference between Padditional heating Pheating coil and Pmotor fan (from Table 38) for
HVAC system without control and for HVAC system with control.
Month Power saved Energy saved per month
January 16,7kW 12,4MWh
February 16,7kW 11,2MWh
March 15,2kW 11,3MWh
April 5,7kW 4,1MWh
May 5,7kW 4,2MWh
June 5,7kW 4,1MWh
July 5,7kW 4,2MWh
August 5,7kW 4,2MWh
September 5,7kW 4,1MWh
October 5,7kW 4,2MWh
November 14,7kW 10,6MWh
December 15,7kW 11,7MWh
Energy saved in a year 86,4MWh
Table 39: Energy saved each month by having a control
-72-
7.4.2 HVAC system control cost
Having a control system with a two-speed fan involves an additional cost. An assessment of the financial
viability of this investment needs to be carried out. For EDF as an electricity producer, energy saved can be
sold to the consumers because it won’t be used to power HVAC equipment, one can then calculate the net
present value of the investment over a period of 60 years (average lifespan of a nuclear power plant).
𝑹𝒕
𝑵𝑷𝑽 = ∑𝑵
𝒕=𝟎 − 𝑰 [41]
(𝟏+𝒊)𝒕
With:
The calculation has been made considering a selling price equal to 110 €/MWh and a discount rate equal to
10% according to EDF company.
The initial investment for 2 two-speed fans (1 AHU extraction fan and 1 AHU supply fan) with their control
system is about equal to 50 000 Euros according to the constructor documentation. Additional maintenance
cost is considered to be 2% of the additional initial investment. Consequently every year EDF can save up
to 8500 €. Table 40 gives the cash flow and the net present value every year.
-73-
Net present value of a control investment
20000
10000
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Net present value
-10000
-20000
-30000
-40000
-50000
-60000
Figure 37: Graphs showing the net present value of the investment for a control system
Discussion:
As shown on Figure 37 above, implementing a control is interesting after a period of 10 years. This
investment would then be economically viable considering the plant in operation for sixty years.
-74-
7.5 Critical scenario analysis in chemical rooms
In the basement, some rooms are used for chemicals storage. In §4.2.2.5, it is stated that ACH requirements
needs to be risen to 20 if chemicals vapors are detected within these rooms. An assessment of the
temperature evolution in the basement when the ACH is increased needs to be carried out both for summer
and for winter.
The following results have been obtained on Th-bât and then exported on Excel to plot graphs. On the two
graphs below, the first part (t<0) corresponds to the normal ACH requirement equivalent to ten volumes
per hour. After chemical vapors detection at t=0, ACH is increased to twenty. The influence of this brutal
rise on the temperature is studied.
Summer case:
For summer, results are presented on Figure 38. It is obvious that it is not a problem if ACH rises as
temperature won’t exceed the maximum admissible temperature of 40°C.
37
36
35 Soda/Acid/Morpholine Room
Bisulphite room
Temperature (°C)
34 Acid/Base room
Lobby West level -6,8
Stairs West
33
Degasser room
Lobby South level -6,8
30
-50 0 50 100 150
Time (hours)
Figure 38: Evolution of the temperature after ACH increase in chemical rooms (summer)
Winter case:
For winter case temperature must not drop below 0°C as it would cause water to freeze. Temperature in
the basement drop from 5,3°C to 3,8°C in less than 1h30min. After one day the temperature reaches 3°C
in the basement. According to the tenderer of the demineralization process if toxic vapors are detected in
the basement, actions will be taken to fix the problem in one day. As a consequence, increasing the ACH in
chemical rooms presents no risk even if temperature drops below the minimum admissible temperature
(Tmin=5°C).
-75-
9
8
Soda/Acid/Morpholine Room
7 Bisulphite room
Temperature (°C)
Acid/Base room
6
Lobby West level -6,8
Stairs West
5
Degasser room
4 Lobby South level -6,8
Fire Fighting Valve room
3
Stairs South East 1
Figure 39: Evolution of the temperature after ACH increase in chemical rooms (winter)
However vapor detection system might trigger but signal transmission to the control room might fail and
workers would not notice it. The steady state temperature reached in the basement after an ACH rise is
presented on Figure 39 and Figure 40. It will be equal to 1.6°C. The temperature won’t drop below 0 and it
won’t harm the process.
10
9
Soda/Acid/Morpholine Room
8
Bisulphite room
7
Temperature (°C)
Acid/Base room
6
Lobby West level -6,8
5
Stairs West
4
Degasser room
3
Lobby South level -6,8
2
Fire Fighting Valve room
1
Stairs South East 1
0
-40 10 60 110 160 210 Main hall basement
Time (hours)
Figure 40: Steady state temperature after ACH increase in chemical rooms (winter)
Discussion:
ACH increase due to harmful vapor detection will change the final temperature within the different rooms
but it doesn’t have a direct influence on the process.
-76-
8 Ansys Fluent Modeling
Ansys Fluent has been used in order to assess the impact of heat gains to the supply distribution duct due
to the temperature difference between the room and air inside the duct as presented in §5.3.1. The software
that has been used in this part is an academic version (downloaded from KTH webpage) of the actual
software. The simulation process follows four distinct and successive steps:
Building of CAD model
Meshing
Application of boundary conditions
Computational Analysis and Visualization
8.1.1 Meshing
Meshing is the spatial discretization of a continuous medium or a discrete representation of the geometry
involved in a problem. The purpose of meshing is to simplify a system with a model representing the system
in its environment. Mesh generation is a really important step as it has an impact on the solution accuracy
and the rate of convergence. (37)
-77-
8.1.1.2 Grid Classification
Three different types of grid can be used in order to mesh an object. They are structured grid (Figure 43),
unstructured grid (Figure 44) and hybrid grid (Figure 45). Following information has been taken from
reference (38).
A structured grid is composed of elements which are orthogonal in i, j space (2D) or i, j, k space (3D). It
has many advantages as equations are easily discretized. It gives also a faster convergence with fewer
iterations, a better accuracy and a higher resolution than for unstructured grids but it is difficult to apply
this type of grid to complex geometries. This type of grid usually uses quadrilateral (2D) and hexahedra
(3D).
Unstructured grid is composed of cells that are arranged in an arbitrary fashion. Basically, there is no
regularity to the mesh. The main advantages of this type of mesh are the fast generation of meshes for
complex geometry and its capacity to concentrate easily the meshing where needed in the geometry.
This type of grid usually uses triangles (2D) and tetrahedra (3D).
-78-
An hybrid grid is a combination of structured and unstructured grids. Parts of the geometry that are complex
will be meshed with unstructured grids and regular geometry parts are meshed with structured grids. In 2D,
hybrid mesh is composed of both triangles and quadrilateral. In 3D, hybrid mesh is made of tetrahedral and
hexahedra.
In this project due to the simplicity of the geometry, a structured grid will be used to mesh the air duct.
In order to obtain good results, skewness needs to be minimized (see Table 41) and its value should not
exceed;
- 0.85 for Hex and quad cells
- 0.85 for triangular cells
- 0.9 for tetrahedral cells
Skewness can be calculated by two methods, one is based on equilateral volume and applies only to triangles
and tetrahedral. The other one is based on the deviation from a normalized equilateral angle and it applies
to all the shape of cells.
o Equilateral volume skewness:
𝑂𝑝𝑡𝑖𝑚𝑎𝑙 𝑐𝑒𝑙𝑙 𝑠𝑖𝑧𝑒 − 𝑐𝑒𝑙𝑙 𝑠𝑖𝑧𝑒
𝑆𝑘𝑒𝑤𝑛𝑒𝑠𝑠 = [41]
𝑂𝑝𝑡𝑖𝑚𝑎𝑙 𝑐𝑒𝑙𝑙 𝑠𝑖𝑧𝑒
o Skewness based on the deviation from normalized equilateral angle:
-79-
𝜃𝑚𝑎𝑥 − 𝜃𝑒 𝜃𝑒 − 𝜃𝑚𝑖𝑛
𝑆𝑘𝑒𝑤𝑛𝑒𝑠𝑠 = max [ ; ]
180 − 𝜃𝑒 𝜃𝑒
With:
𝜃𝑚𝑎𝑥 = largest angle in face or cell
𝜃𝑚𝑖𝑛 = smallest angle in face or cell
𝜃𝑒 = angle for equilateral face or cell, (60° for triangle and 90° for square)
Aspect ratio:
Aspect ratio represents the ratio between the longest edge and the shortest edge of a cell. The best achievable
value is equal to 1 for an equilateral triangle or a square.
Smoothness:
The change in the size of cells in a grid must be as smooth as possible. It means that the size of neighbor
cells should not vary more than 20%.
In order to be useful a turbulence model needs to fulfill the following requirements, it must be reliable,
simple, economical to run and have a wide range of applicability (39). Thus the simulation realized in this
report has been carried out using the RANS model.
-80-
8.2.1 RANS –based turbulence model
RANS turbulence can be classified in terms of number of transport equation solved simultaneously in
addition to the RANS equation:
Zero equation or algebraic model (Cebeci-Smith, Baldwin Lomax, Johnson King …)
One equation models (Spalart-Allmaras, Baldwin-Barth …)
Two equation models (k-ε, k-ω…)
Zero equation models are often considered too simple to be applied to actual physic problem or general
situation. It is being ignored in CFD software nowadays. One equation models are weak to analyze and
calculate complex internal flows with strong curvature. Then, these RANS – based turbulence models won’t
be used in this work.
Two equation models such as k-epsilon model and k-omega model are the most common models used in
industry to solve the majority of engineering problems. It solves two additional transport equations in
addition to the RANS equation. Most of the time, it solves the turbulent kinetic energy equation “k”. The
second equation that is solved depends on the type of model that is used. For k-epsilon model the turbulent
dissipation rate “ε” is added. For k-omega model the turbulent dissipation rate equation is replaced by the
specific dissipation “ω” (38) (39).
8.2.1.1 k – ε model
K-epsilon model is a really interesting model as it is relatively easy to implement and provides accurate
results for many flows. However its performance is limited for highly curved streamlines, swirling and
rotating flows. Some variations of the k-epsilon model exist and give improved predictions for some flows.
For instance the Renormalization Group Method k-ε or Realizable k-ε improve the standard k-ε model
performances for more complex flow, high streamline curvature flow, wall heat, mass transfer and swirling
flow. This model works only for fully turbulent flow and it will tend to create some turbulence when there
is none. It will increase the heat transfer between the air within the air duct and the outside.
8.2.1.2 k – ω model
K-omega model is similar to k-epsilon model however it solves for the specific dissipation and not the
turbulent dissipation. The k-omega model can be very useful in cases where k-epsilon model fails and can’t
be used as for instance for internal flows, flows with strong curvatures. However calculations take longer
and convergence is a bit more laborious than for k-epsilon model.
-81-
Direction of the flow
Figure 47: Geometry of the longest air duct in HY building (Design Modeler)
The first outlet encountered in the direction of the flow is supplying air to a first set of rooms and the last
five are providing air to the main hall where water is demineralized. Air temperature within the duct and
walls’ temperature needs to be measured for the following reasons:
If the temperature of the walls is inferior to the dew point temperature of humid air of air at 40°C
and 32% relative humidity, then it will be necessary to insulate the air duct to prevent water from
condensing and dripping in the rooms.
Average air temperature in the duct just before it is supplied to the main hall (after outlet 1) needs
to be measured in order to know if additional margins on the supply temperature need to be taken
into account.
6 5 4 3 2 1
Hall
-82-
Figure 49: Hexahedral meshing of the air duct
With ANSYS academic version, the number of cells is limited. The size of the cells has then been limited
by the software capacity. Size has been chosen so that number of cells used to mesh the geometry is as close
as possible to the maximum numbers of cells authorized by the software. In this way, the maximum possible
precision will be obtained.
For shapes such as rectangle or square ducts, the hydraulic diameter is then defined by:
-83-
4𝐴
𝐷𝐻 = [43]
𝑃
P is the wetted perimeter. That means the perimeter of all channel walls in contact with the flow. In this
case it means the total perimeter of the duct.
For this simulation, 𝑅𝑒 = 4,1. 105 > 4000. Then the airflow is turbulent. Moreover, Mach number M<0,3
as a consequence air can be assimilated to an incompressible fluid.
For turbulent flows two main models can be used, k-ε model and k-ω model. Calculations will be realized
with k-omega model that performs better for wall bounded flow with curvatures. However it will be
compared to realizable k-epsilon model which is an improvement of standard k-epsilon model
Second step is the boundary conditions implementation. Inlet has been chosen as a mass flow inlet, outlets
have been set as pressure outlets with a target mass flow. The air duct is placed in an atmosphere at 40°C
and 32% humidity that is identical to actual temperature in the building.
Step 4: Calculation and Visualization of the results
Screenshots that will be presented in the report displays temperature variation of two different cross sections
of the air duct which are shown on Figure 51 below.
Figure 51: Air duct planes (Plane n°1: green and Plane n°2: red)
Results obtained with both meshing are almost the same at +/- 0.1°C. On Figure 52 and Figure 53 are
represented the results obtained after 500 iterations for k-omega model and k-epsilon realizable model. The
air duct is not insulated and made of 2mm thick steel. On these screenshots it is obvious that temperature
varies between the edge and the center of the ducts. Close to the wall temperature is at its maximum and in
the middle temperature is at its lowest. Such a temperature difference can be explained by the fact that
airflow within an air duct has a velocity close to zero near the walls. Velocity keeps increasing by getting
closer to the center of the duct. This difference of velocity leads to a heterogeneous temperature distribution
as shown in Figure 52 and Figure 53.
-84-
Figure 52: Temperature in plane n°1 (k-omega model)
Another temperature gradient appears while moving along the air duct. The further from the inlet it is, the
bigger the average temperature of a cross section is. It can be explained by the fact that air within the duct
has more time to exchange heat with the room. Energy = Power x Time, consequently more energy is
exchanged when the distance travelled by air increases resulting in a bigger average temperature within the
air duct. This temperature gradient appears clearly in Figure 54.
-85-
Direction of the flow
For both models, results are really close which gives robustness to the modeling. Average temperature in
plane n°1 has been calculated thanks to the function calculator in ANSYS Fluent. For k-epsilon model,
average temperature is equal to 16.50°C. For k-omega model, average temperature in plane n°1 is 16.55°C.
In ventilation calculations realized, supplied air temperature is supposed to be equal to 15°C so the target
temperature is exceeded by 1.5°C.
Minimum wall temperature (displayed in Figure 55) has been calculated thanks to the function calculator tool
in ANSYS Fluent. It stipulates that it is equal to 18.7°C. Consequently condensation might occur at some
-86-
spots along the air duct considering that 18.7°C is inferior to the dew point temperature of air at 40°C and
32% relative humidity equal to 20.20°C.
In order to avoid condensation and decrease the temperature of air in plane n°1, two solutions are possible:
The first one consists in decreasing the temperature of air at the inlet so that it reaches 15°C before
entering in the hall but it would lead in an energy consumption rise and wouldn’t solve the problem
of condensation as minimum wall temperature would remain below 22.20°C.
The second solution consists in insulating the air duct so that its thermal resistance increases and it
gets harder to exchange heat with the room. Consequently temperature increase problem and
condensation problem would be solved.
In Figure 56 are presented the results for an air duct insulated with 4cm of glass wool covered with a thin
aluminum layer. Thermal conductivity of such a material is equal to 0.04 W/(m.K).
One can easily notice that temperature is more homogeneous, there are fewer differences between inner
wall temperature and air temperature within the duct. Function calculator tool in ANSYS Fluent stipulates
that average temperature of air in plane n°1 is equal to 15.4°C.
As a consequence, energy can be saved this way as temperature increase is limited, moreover condensation
problems are avoided. However temperature rise from 15°C to 15.4°C cannot be neglected. Thus a margin
equal to 0.4°C will have to be considered in calculations so that air entering the main hall has a temperature
equal to 15°C in order to be consistent with the calculations carried out. Figure 57 shows the results
considering an inlet air temperature equal to 14.6°C.
-87-
Figure 57: Temperature in plane n°1 for an inlet temperature equal to 14.6°C
Average temperature in plane n°1 considering that air is supplied at a temperature of 14.6°C is equal to
14.95°C. This value is really close to 15°C which is the air supply temperature used for the HVAC system
sizing.
Step 5: Discussion
Heat transfers due to temperature difference between air in ducts and rooms cannot be overlooked and
must be considered even in the case where ducts are insulated. After having carried out this thermal study,
two design choices appear. Either a margin needs to be taken into account on the inlet temperature in order
to maintain the air supply temperature at 15°C or the inlet temperature is maintained at 15°C and air supply
temperature needs to be changed. In the calculations presented in the following paragraph, the system has
been sized for a supply temperature equal to 15°C and considering a margin equal to 0.4°C on the inlet
temperature (14.6°C).
-88-
9 Results
In this part are presented the results about ventilation system and smoke control sizing.
For winter conditions, no internal heat gains are considered except the ones provided by local heating
equipment as the worst case scenario with all equipments shut down is considered.
Latent heat gains are neglected as only two people will be permanently present in the building. Moreover
building is pressurized and basement’s walls are made of concrete resulting in negligible infiltrations. The
process of demineralization is completely closed thus there is no water emission by the process.
-89-
9.1.2 Blowing conditions justifications
Summer:
The supply temperature must be as small as possible. Indeed by maximizing the temperature difference
between supplied air and the maximum admissible temperature, airflows will be smaller. The smaller the
airflows, the smaller the air ducts, the AHU and fan energy consumptions are. But it cannot be too low for
comfort and thermodynamic reasons (relative humidity of supplied air, φ cannot be lower than 90%).
15°C is the best compromise as φ=82% and ΔT=26°C -15°C=11°C where 26°C is the maximum admissible
temperatures for rooms with workers inside. It satisfies the condition from §5.6.1.
Winter:
The limiting parameter in winter is the temperature of the hall. Indeed, the hall doesn’t need that much
heating due to its good insulation. If the blowing temperature is too high, then the heating coil will be
oversized and energy will be wasted.
Blowing temperature Hall temperature (without control) Hall temperature (with control)
For winter conditions, blowing temperature will be chosen equal to 15°C to limit the energy loss in the
heating coil for heating the hall. It cannot be lower than 15°C as otherwise it would be uncomfortable for
workers.
Taking the specific heat capacity for air as 1.02 kJ/kgK and air density as 1,2kg/m3, then:
Δ T= 4,253/(1.02 x 1.2 x 6.125) = 0,56°C
Conclusion:
Total temperature rise of air in air ducts is assumed to be equal to 1°C. As a consequence, in order to provide
air at 15°C, air must exit the cooling coil at a temperature of 14°C.
-90-
9.1.4 Final sizing
One standalone Air Handling Unit (AHU) of 22050m3/h will be implemented in the HVAC room of HY
building. It will be designed to supply air at 15°C for extreme summer conditions (40°C) and 15°C during
extreme winter conditions (-15°C).
A minimum supply of air is necessary for the removal of odors, carbon dioxide and any other contaminants
produced by human occupation.
The maximum volumetric flow for AHUs has been determined for the summer case. In some cases, heating
and cooling power supplied by the AHU is not enough to maintain the different rooms at the requested
temperature. Therefore additional cooling and heating have to be provided locally.
In order to optimize the ventilation system and save energy during plant operation, the maximum flow that
needs to be supplied by the AHU has to be minimized. To do so, the building has been divided in two
distinct parts as explained in §7.2.1.2. The first one is composed of all the rooms located underground. The
underground part will be supplied with air coming directly from the outside thanks to fans. Indeed due to
the temperature of the ground and deep ground this part doesn’t need cooling in summer. For winter
conditions, additional heating has been added mainly in the hall in order to maintain the required
temperatures.
The AHU will recycle air from the inside (except from the toilets and the laboratory) in winter in order to
save some energy. However for summer conditions it is not worth it to do so because recycled air from the
building is hotter than outside air (especially due to the fact that transformers release a large amount of
heat).
Results are presented in Table 1Table 44 below.
-91-
Airflow Heating Cooling
Level Room
(m3/h) (W) (W)
4.60 Cleaning room 235.7 2000 0
4.60 Lobby 5 13 0 0
8.00 Lobby 6 29.8 500 0
TOTAL 22050 111500 -31000
Table 44: AHU flow rates, heating and cooling needed to be implemented in the different rooms of HY building
Rows colored in red correspond to rooms supplied with air coming from the AHU. The airflow that needs
to be provided by the AHU is equal to 22050m3/h for blowing temperatures equal to 15°C both for summer
and winter.
(*) This symbol represents reversible heat pumps (providing both cooling and heating)
𝑇𝑤𝑎𝑡𝑒𝑟_𝑜𝑢𝑡 + 𝑘. 𝑇𝑤𝑎𝑡𝑒𝑟_𝑖𝑛
𝑇𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑐𝑜𝑖𝑙 =
1+𝑘
With:
𝑇𝑤𝑎𝑡𝑒𝑟_𝑜𝑢𝑡 = Water out of the heat exchanger
𝑇𝑤𝑎𝑡𝑒𝑟_𝑖𝑛 = Water entering the cooling coil
𝑘 = Coefficient depending on the cooling coil often estimated equal to 1 when no information is available
about the coil.
If k=1, 𝑇𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑐𝑜𝑖𝑙 is equal to the arithmetic mean of the inlet and outlet water temperature
The AHU supplies 22050m3/h and needs to cool down outside air having a density of 1.2 kg/m3 from 40°C
to 14°C. By using the psychometric chart shown in Figure 58:
Pcooling = 330kW
35,6 − 23,8
𝐵𝐹 = = 20,7%
80,8 − 23,8
ξ = 1 – BF = 79,3%
According to §4.5.5.3 average value for the bypass factor is about 20% which is really close to the value
obtained.
-92-
h = 80,8
h = 35,6
h = 23,8
Figure 58: Cooling process of humid air for an external temperature of 40°C
∆𝑥 = 100 gwater/h
450 𝑘𝐽
𝛾= . 3600 = 16 200
0.1 𝑘𝑔𝑤𝑎𝑡𝑒𝑟
Let’s check if the humidity got for 26°C is within the range 30-60%.
-93-
It means that for ∆ℎ = 40 𝑘𝐽/𝑘𝑔𝑤𝑎𝑡𝑒𝑟 then ∆𝑥 = 2,47 . 10−3 𝑘𝑔𝑒𝑎𝑢 /𝑘𝑔𝑔𝑎𝑠
ΔH=40 kJ/kg
ΔX=0.00247
Φ=40%
26°C
Figure 59: Psychrometric chart representing evolution of air for rooms housing workers in summer
Relative humidity within the rooms is about 40% which is consistent with UK regulation.
o Winter case:
In winter, air is really dry. Additional humidifiers need to be added in rooms where workers spend most
their time according to (12). In order to stay in the most restrictive scenario, no water emission from workers
is considered in winter. Then ∆𝑥 = 0 and the space line is horizontal according to §5.5.4. (see Figure 60)
-94-
h = 34
h = 17,5
0,0075
0,001
Fresh air Air mixed Extracted air Supply air 22
Figure 60: Psychrometric chart representing air humidification in winter for rooms housing employees in winter (-15°C)
From §4.5.6.2 and considering that air density is equal to 1,2 kg/m3:
191 + 229
𝑃ℎ𝑢𝑚𝑖𝑑𝑖𝑓𝑖𝑒𝑟 = 𝑚.̇ 𝛥ℎ = . 1,2. (34 − 17,5) = 2,5 𝑘𝑊
3600
-95-
Here:
∆𝑝 = 4Pa
𝑉𝑜 = 22.41 l/mol
𝑇0 =273.15 K
𝜃 = 40°C (Worst case scenario in the calculation)
R=8.31 SI
∆𝑞 𝑚3
= 4,5% => ∆𝑞 = 0,045 . 𝑞 = 0,045. 13015 = 585
𝑞 ℎ
Calculation of leakages through doors is given by formula in §4.6.1.3:
2
𝑞𝑣𝑖𝑛𝑓 = 0.62 . 𝑆 . ( )0.5 . ∆𝑝0.5 ≈ 0.83 . 𝑆. ∆𝑝0.5
𝜌
Here:
Number of doors = 2
Door size = 4,5m x 6m
Gap = 4mm
∆𝑝 = 4Pa
𝑚3 𝑚3
𝑞𝑣𝑖𝑛𝑓 = 0,83 . (4 . 10−3 ). 4,5 . 40,5 = 0,030 = 108
𝑠 ℎ
Conclusion:
𝑚3
∆𝑞𝑡𝑜𝑡𝑎𝑙 = 108 + 585 = 693
ℎ
Extracted airflow needs to be decreased by 534 m3/h so qextracted = 12357 m3/h
-96-
From §4.3.1.3, margins to be considered are in consistency with (40):
o For airflow through a closed door : +50%
o For airflow through an open door : +15%
Airflow required for maintaining an air speed of 0.75m/s through an open door with a surface area
equivalent to 2.2m² is equal to:
o Q = (2.2 x 0.75)*1.15 = 1.90 m3/s (for both stairs with margin)
Airflow required for maintaining a pressure difference through closed doors is equal to:
o Q = (0.06 + 0.12 + 0.06)*1.5 = 0.3 m3/s (for stairs 1)
o Q = (0.06 x 6 + 0.12)*1.5 = 0.72m3/s (for stairs 2+3)
For both stairs the required airflow to have 0.75m/s is much higher than the one necessary to get the
pressure differential of 50Pa. As a consequence fans will be sized with an airflow equal to 1.90m3/s
-97-
10 Discussion and Conclusion
The main purpose of this work is the sizing of a robust and efficient ventilation system for a nuclear
related building for HPC project. Literature about HVAC on nuclear building is really simplistic and needs
to be adapt for non-safety classified buildings such as the demineralization station. The main question that
needs to be answered is how to maintain the ambient conditions within a given range of temperature and
relative humidity while assuring the minimum ACH requirements. In order to do so several ventilation
systems have been studied going from natural ventilation systems to double flow ventilation systems. The
most efficient system is halfway between a single flow and a double flow ventilation system. Indeed
investment costs and operating costs are much smaller than for a double flow ventilation system and much
more robust than natural or single flow systems. Before this work, the system was supposed to be a 100%
double flow ventilation system but thanks to the studies carried out in this paper, previous thoughts have
been abandoned for a more energy efficient and cheaper system. Smoke control system has been size
according to regulations taken from CIBSE Guide E (19) and no infringements to these rules is allowed as
it is a matter of safety for employees. Usually, nuclear buildings are deprived of ventilation control system
as robustness is preferred over energy efficiency. However for a non-safety classified building which is
halfway between an industrial and a nuclear building, it makes sense to consider a simple ventilation control
in order to save energy and decrease operating costs. It is something new that hasn’t already been done on
this kind of buildings but the results presented in this thesis show that it is something doable that could be
interesting to consider.
However there are some limitations. Indeed, the model that has been built in this thesis work only for
a given set of input data, if the nuclear safety commission change these data then the calculations will have
to be redone. But the work carried out still provides a good basis for EDF as ideas and some parts of this
design can be used on similar buildings for other projects around the world.
HVAC system is a primordial system for nuclear associated buildings as it insures the quality of a
process. The work presented in this thesis describes the first step of the HVAC system development which
is the system sizing. It has impacts on civil works and installation since shaft needs to be allocated for air
ducts and room needs to be saved for bulky equipment such as extraction fans, water chillers and Air
Handling Unit. Ventilation affects also electrical engineering as power consumption of HVAC equipment
is important and cannot be overlooked. HVAC system sizing is then an important step since it impacts
many downstream activities.
The first part of the design is to set input data for the design such as external temperature, relative
humidity, wind speed and solar radiation in order to carry out thermal calculations. Internal admissible range
of temperatures, relative humidity and minimum ACH in each room also need to be defined. Calculated
heat transfers from previous data will be added to internal heat gains from equipment, lighting and
employees in order to assess the required power (either positive or negative) to maintain the different rooms
at the requested temperature previously stated.
A ventilation system is expensive especially in the nuclear field as some equipment need to be safety
classified. As a consequence, by reducing the number of HVAC equipment investment costs will decrease.
Moreover electricity is produced 24/24h for about 60 years consequently by optimizing the ventilation
system energy and money can be saved. It represents a benefit for the company since which earns more
money and for the environment as energy is saved.
Robustness of the system is one of the complicated parts as additional margins needs to be considered
in order to ensure the reliability of the system. Indeed, if margins are taken too small, it will result in a risk
of failure of HVAC system. If margins are taken too big, it will lead to an over-sizing of the ventilation
system. However it is always preferable to have a system that is over-sized rather than under-sized.
-98-
References
1. Company, EDF. [Online] https://www.edf.fr/.
2. Company, EDF; Au cœur de l'énergie nucléaire. 2015.
3. Hinckley point C, securing the UK's energy future. [Online] EDF energy UK, 2015.
https://www.edfenergy.com/energy/nuclear-new-build-projects/hinkley-point-c.
4. Institute, Nuclear Energy. [Online] http://www.nei.org/.
5. lenergeek. [Online] 2015.
http://lenergeek.com/wp-content/uploads/2012/08/circuits_eau_reacteur_nucl%C3%A9aire.png.
6. Company, EDF. L'utilisation de l'eau dans les centrales nucléaires. 2013.
7. DeSilva, Frank. Water treatment guide. [Online] 2000.
http://www.watertreatmentguide.com/activated_carbon_filtration.htm.
8. Maurel, Alain. Osmose inverse et ultrafiltration, Technologie et applications. s.l. : Techniques de
l'ingénieur.
9. Lenntech. Reverse osmosis. [Online] http://www.lenntech.com/library/reverse-osmosis/whatisro.htm.
10. Distilled,deionized and demineralized water. [Online]
http://www.lenntech.com/applications/process/demineralised/deionised-demineralised-water.htm.
11. La ventilation hygiénique. Energie plus, Efficacité énergétique des bâtiments tertiaires. [Online] 2015.
http://www.energieplus-lesite.be/.
12. UK Workplace (Health, Safety and Welfare) Regulations.
13. UK Control of Substances Hazardous to Health Regulations.
14. UK Control of Noise at Work Regulations.
15. UK Environmental Noise Regulations.
16. CIBSE Guide B; Heating, Ventilating and Air Conditioning.
17. ASHRAE Handbook. 2013.
18. Standards, BSI British. BS9999, Code of practice for fire safety in the design, management and use of
buildings. 2008.
19. CIBSE Guide E, Fire Safety Engineering.
20. BS EN 12101-6; Smoke and heat control systems. 2005.
21. EDF. GFPA, Global FIre Preliminary Assessment for HY building.
22. Tim Dwyer. CIBSE Journal. [Online] http://www.cibsejournal.com/cpd/modules/2009-08/.
23. Duminil, Maxime. Air Humide. Techniques de l'ingénieur. 1986.
24. H.Havtun, P. Bohdanowicz, J.Claesson, E.Granryd and G.Vicatos. Sustainable Energy Utilisation,
Chapter 6 "Psychrometrics".
25. Bensafi, Ahmed. Air humide - Traitement et conditionnement de l’air. Techniques de l'ingénieur. 2007.
26. EN13779 Ventilation for non-residential buildings - Performance requirements for ventilation and
room-conditioning.
27. ECUK121009, "Minute of meeting – HVAC input data & hypothesis".
-99-
28. ECEF040074 B, “Engineering rules ENG 2.83 (MD2) : Ventilation System functional
requirements,preliminary sizing & material selection”.
29. CIBSE Guide A; Environmental Design.
30. RT2000; Thermal regulation.
31. AFNOR. Thermal performance of buildings, Heat transfer via the ground. 2008.
32. Le Recknagel; Manuel pratique du génie climatique – 2ème édition – PYC Edition.
33. Porcher, G. Bases du calcul des installations de climatisation – 4ème édition.
34. EDF. Design tolerances allowance strategy for HVAC systems design.
35. UK building regulation.
36. Weather statisitics for Bath and North East Somerset. [Online]
http://www.yr.no/place/United_Kingdom/England/Bath_and_North_East_Somerset/statistics.html.
37. R.Fahkrai. Introduction to computational methods in energy technology. s.l. : KTH Royal Institute of
Technology.
38. CFD Online. [Online] http://www.cfd-online.com.
39. ANSYS Fluent 12.0 User's Guide. 2009.
40. BSEN12101-6:2005, “Smoke and heat control system – Part 6: Specification for pressure differential
systems – Kits.".
-100-