Tomar wg1

Minutes of the COST 624
Meeting in Tomar
3 – 5 October
WG 1 on Plant Operation
Parallel Sessions
Jens Alex
Steven Isaacs
Ulf Jeppsson
Presentation of the Proposal of the future Action of WG 1 on

Plant Operation of the COST 624 action
The presentation of the proposal of the working programme was given by Jens Alex, Steven Isaacs and
Ulf Jeppsson. This proposal is attached in the annex of these minutes. The motivation for the proposed
programme, as described in the attached programme, was presented first by Jens Alex.
Practical Aspects
(Steven Isaacs)
The first major topic - “Practical aspects” - was explained. In addition to the attached programme the
key words “Acceptance and Understanding” and “ Meaningful Control Goals” were stressed.
Acceptance and Understanding

Of the utmost importance is the acceptance and understanding of advanced control strategies:
Acceptance: The acceptance of advanced control strategies on the part of communal leaders as well
as treatment plant personnel is of utmost importance. Regarding communal leaders, the control will
not be purchased or implemented otherwise. Regarding plant personnel, the control and automatic
measurements will not be supervised and maintained, perhaps not even used, otherwise.
Understanding: The plant personnel have to understand the controls and what they are trying to
achieve, otherwise they will make little effort to maintain the control and measurement systems.
The worst situation is to have a control system used incorrectly. This means either not used at all
(disabled) or used with wrong or poor measurements, or used with completely incorrect tuning
parameters and set points. If this occurs and continues, the respect for the control system, both at
the current plant and, through word of mouth, at neighbouring communities, will be lost. And
thereby the acceptance of the control system will decline.
Key principle: Just as in the case of automobile manufacturers, who learned that the design of a model
must also take into consideration the ease of maintenance and repair, control system design must
consider the issues of understandability and transparency in order to be accepted, understood and,
hence, well maintained and correctly implemented.
Meaningful Control Goals
Research in control system development for WWTP’s needs to consider the meaningfulness of the
control goal. Otherwise there will be a lack of acceptance for the system and a lack of willingness to
implement the system.
As an example, an obviously good control goal is the maintaining of effluent concentrations below set
limits. However, is it realistic to pursue effluent concentrations to ± a small tolerance around a set point
value? Particularly in view of all the other uncertainties in plant dynamics and measurements?
Some important criteria from a practical point of view are:

• maintaining effluent SS (turbidity) low during normal as well as storm weather;
• maintaining a stable return sludge concentration for the purpose of a stable dewatering process;
• adjusting recirculation flows, aeration intensity and aerobic/anoxic volume to meet changing
process conditions and loads;
• maintaining low effluent P, maintaining good bio-P removal, minimizing precipitant addition while
maintaining effluent P limits;
• adjusting precipitant chemical dosing so as not to overdose (economic cost and loss of needed COD)
nor underdose.
Meaningful control goals should not neglect:

• limitations in the available control handles;
• limitations in modelling accuracy;
• limitations in measurement reliability/availability;
• need for more basic but robust strategies for improvement: cost savings, hydraulic problems,
operational flexibility to changing loads.
Plant Wide Modelling

(Jens Alex)
As an important tool for the development, the evaluation of novel control and operation concepts,
simulation is widely used. One main focus of the working group actions will be the modelling on a
plant-wide level including processes for sludge treatment energy consumption/production besides the
classical activated sludge process. Details of the presentation can be found in the attached working
programme.
Benchmark
(Ulf Jeppsson)
An important ongoing action of WG 1 is the control benchmark project. The idea of this project is given
in the attached working programme. In addition the following issues were mentioned:
• summary of the present state of the benchmark action as presented in [1] and [2];
• presentation of the next steps of the benchmark, defined in a small meeting preceding the workshop
in Tomar:
- definition of a list of additional control variables (measurements) and control handles;
- preparation of a description of the properties of the newly defined measurements (delay time,
noise level, etc.).
Parallel Session No. 1 Monday, October 4th, 15.00-16.45
The aim of the discussions was to compile ideas of the participants with respect to the proposed
working programme, to discuss the proposal, to evaluate the defined goals and to plan the future actions
of the working group. There were two parallel sessions organised for each working group with a distinct
selection of participants. For this reason several topics were discussed twice.
Practical Aspects - Yes

The first topic that was discussed was to decide if and to what extent the working group should consider
practical aspects of plant operation and control. This was actually a topic for discussion, because
examples were presented, where already very basic preconditions for basic control are not available.
The level of application of control in practise depends strongly on the EU country under consideration.
Finally the result of the discussion was clearly positive, but the practice under consideration should be
the state of the art in medium and advanced (from the perspective of WWT) standard countries.
The next topic was the attempt to identify

• problems/tasks of WWTP operation,
to discuss which problems are
• addressable by control, operation (and modelling as an important methodology)
and to work out
• the need for research and tasks for WG1.
The following topics showed up during the discussion:

• sedimentation in aeration tanks, clarifier models;
• energy management, energy equalisation – minimisation;
• fault detection of sensors „reconfiguration control“, fall back control;
• state of the art in various countries: type of plants, sensor philosophy, control.
Plant Wide Modelling (for Control)

One important methodology for the development of novel control and operation concepts is the use of
simulation models. With the aim to simulate the whole treatment plant and not only the activated sludge
step in a simplified manner (usually only one line etc.) several requirements must be fulfilled.
Scalable Models
To give the opportunity to simulate complex treatment plants with the complete set of operational units,
much simpler models are needed than are usually used today. This is primarily for computational
reasons. For the correct consideration of specific problems, however, the detailed models are once again
important. Thus “scalable” models are necessary. The models should be scalable in two ways:
• scalable in time, i.e.

- different time scales: seconds, minutes->hours,days->stationary behaviour;
• scalable in complexity, i.e.

- number of fractions, number of processes considered.
One example of scalable complexity is the characterisation of organic matter by different COD fractions
as shown below.
XH
particulate
COD slowly
biodegredable
COD COD XS
soluble SA
COD readily
biodegredable
COD SF
Benchmark
The focus of the last minutes of the session were on the next steps of the benchmark project:
• additional sensors will be available -> this will allow more practical use of the benchmark;
• benchmark files will soon be readily obtainable;
• topic remains on the WG1 agenda but with varying emphasis at different meetings.
Finally, the first ideas for the programme and the location of the next working group were discussed.
This discussion continued during the second parallel session.
Parallel Session No. 2 Tuesday, October 5th, 9.00-10.15
Plant Wide Modelling
After a summary of the discussion of the first parallel session the use of simulation as a general
methodology was considered. Especially the purpose of “plant wide modelling” was put to question.
Plant wide modelling should, among other things, include:

• current situation, design of control strategies (also operation);
• examples;
• energy;
• COD adsorption;
• foaming/bulking;
• heat pumps.
Interaction
The three major topics of WG1 are strongly related to each other. The figure below gives an impression
of the interactions between the different topics of the working programme.
Requirements &
practical Aspects
Problems, Tasks
Proposals for novel

control,
„advertisement“ for Evaluation
control
plant-wide
modelling
Benchmark
Application for
Application of the models for:

• Operation, control
• Design
• Energy management
• Operation research
• Operator support (DSS)
Public Model Library
As one possible activity the idea to establish a public, unified model library for models of unit processes
of wastewater treatment plants, which are needed for the design of new control and operation schemes
was discussed.
• Establishing rules for presentations of models:

- textual description;
- written ODE’s and PDE’s;
- matrix representation;
- formal description, e.g. program code (C, FORTRAN?).
• Providing possibilities for the validation of model implementations, e.g.:

- input/output data sets.
• References to:
- cases of application of the published models;
- cases of validation of the published models.
• Guideline for modelling:

- strict mass balance principle;
- „smooth“ behaviour.
• Scalability of models (possibility to select a proper model for a given case, simple and complex
models).
• Reliability of models:
- calibration/verification methods;
- uncertainty analysis.
• Organisation and policy, e.g.:

- GNU policy.
As a result of the discussion the “Public Model Library” was supported by the participants, the use of a
web site was considered as a good technical platform for the library and a strong possibility for co-
operation with the activities of working group 2 on “Integrated Modelling” was realised. A combined
meeting of both working groups (1 & 2) was proposed. The tentative location and date is Gent, Belgium
in the fall 2001.
Interaction between WG1 and other WG’s

Finally, possible interactions between WG 1 and the other working groups were examined and
discussed.
• WG 2 on Integrated Modelling:
- model library;
- interface problems.
• WG 3 on Evaluation Tools:
- benchmark - evaluation tools;
- control evaluation.
• WG 4 on Biological Processes:
- Biosensors;
- bulking/foaming etc.
• WG 5 on Treatment Scenarios:
- new processes and units - transfer of knowledge;
- comparison of the database of unit processes (stationary behaviour) with the dynamic model
library of WG 1.
Next Meetings
• Spring, 2000, Vienna, Austria: „Plant wide modelling for WWTP control and operation“. Topics:
- guideline and framework;
- state of the art in various countries;
- typical plant configurations;
- evaluation of needs.
• Fall 2000, San Sebastian, Spain: „Anaerobic digestion as an important unit process in WWTP’s“.
Topics:
- state of the art;
- IWA working group to be invited;
- identify what local control is used.
Literature
[1] J. Alex, J.F. Beteau, J.B. Copp, C. Hellinga, U. Jeppsson, S.Marsili-Libelli, M.N. Pons, H.
Spanjers and H. Vanhooren: Benchmark for evaluating control strategies in wastewater treatment
plants. ECC’ 99, August 31- September 3, 1999, Karlsruhe, Germany.
[2] M.N. Pons, H. Spanjers and U. Jeppsson: Towards a benchmark for evaluating control strategies
in wastewater treatment plants by simulation. ESCAPE 9 (European Symposium on Computer
Aided Process Engineering - 9), May 31-June 2, 1999, Budapest, Hungary (Computer and
Chemical Engineering Supplement, pp 403-406, 1999).
Appendix
Working Programme of WG 1.

Tomar wg1

Uploaded by

Copyright:

Available Formats

Tomar wg1

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tomar wg1

Uploaded by

Copyright:

Available Formats

Minutes of the COST 624

Presentation of the Proposal of the future Action of WG 1 on

Acceptance and Understanding

Some important criteria from a practical point of view are:

Meaningful control goals should not neglect:

Plant Wide Modelling

Practical Aspects - Yes

The next topic was the attempt to identify

The following topics showed up during the discussion:

Plant Wide Modelling (for Control)

• scalable in time, i.e.

• scalable in complexity, i.e.

Plant wide modelling should, among other things, include:

Proposals for novel

Application of the models for:

• Establishing rules for presentations of models:

• Providing possibilities for the validation of model implementations, e.g.:

• Guideline for modelling:

• Organisation and policy, e.g.:

Interaction between WG1 and other WG’s

You might also like