Telemac project

1
Telemac project
Kick-off Meeting
INRIA, Sophia Antipolis, 13-14 September
Workpackage 2
2
Telemac project
Workpackage 2
Workpackage 2.1 Sensor specifications

• Objectives (4.6 MM)
• Specify technical requirements used sensors
  (precision, reliability, drift)
• Overview of the state-of-the-art of practically
  usable sensors

Partners APPLITEK  1.8, BIOMATH  1, USC 
0.5, LEMAIRE 0.4, DOMECQ UK 0.9
Deliverable (end of 2001) Report on
characteristics of sensors Budget Measuring
quality Maintenance
Specifications existing/future sensors (hard/soft)
3
Workpackage 2
Workpackage 2.2 Data validation

• Objectives (7.5 MM)
• Assessment of quality of data collected at
  full-scale and pilot plants
• Comparison with off-line data
• Confrontation with known interrelationships
  (e.g. mass balances)
• Critical review of state-of-the-art of data
  validation/consistency checking
• Selection of appropriate data filtering
  techniques

Partners BIOMATH  7.5 MM

• Deliverable (March 2003)
• Procedure to validate, filter and check sensor
  data consistency
• Implementation in new sensors data treatment

4
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors
Objective (28.7 MM) Develop new hardware
sensors based on robust probes (pH, gas flow, ...)
Partners APPLITEK  15.7 BIOMATH  13
5
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors
Relevant parameters in anaerobic treatment
Analysis method

• IN LIQUID PHASE
• alkalinity
• volatile fatty acids (VFA)
• ammonia
• IN GAS PHASE
• CH4
• CO2
• H2S ?

AUTOMATIC TITRATION
SPECIFIC SENSORS
6
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors

• Titrimetric sensor
• Study lab titrimetric sensor Biomath (PhD Van
  Vooren) (limitations, capabilities)
• Development new titrimetric sensor
• Hardware building new analyser
• Process Control Software automation new
  analyser
• Implement titrimetric methodology
• Testing titrimetric sensor
• Compare VFA measurements with GC
• Alkalinity ammonia with standards

7
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors

• Titrimetric sensor
• consecutive acid/base additions with
  pH-measurement
• wet part sample preparation
• down-titration to pH 3 (VFA alkalinity)
• stripping CO2
• up-titration to pH 11 (NH4)
• data processing part data analysis

development adaption algorithm
8
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors
Operating mode
9
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors
Evaluation titration curve see separate
presentation
10
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors

• Previously obtained results
• Effluents
• Surface waters
• Algal mixed liquor
• Manure
• Comparison Lab / Titrimeter

11
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors

• Gas sensor
• Study available specific gas sensors
• robustness
• capabilities
• interferences (e.g. H2S)
• Development new gas sensor
• Hardware combination specific gas sensor
  flow sensor
• Process Control software data output
• Testing gas sensor reference gas measurements

12
Workpackage 2
Workpackage 2.3 Adaptive aspecific sensors

• Deliverable (27 M)
• Prototype adaptive sensor based on robust probes

13
Workpackage 2
Workpackage 2.4 Software sensors

• Objectives (23.9 MM)
• Design and evaluate robust software sensors
  for estimation of process variables
• Provide a confidence interval for the estimates
• Establish the basis of the sensor network
• Integrate fault residues in sensors

Partners INRIA 11.3, SPES 5.8, UDG  6.8
14
Workpackage 2
Workpackage 2.4 Software sensors
Software sensor design Models for various
working mode Model parameter uncertainties Measure
d quantities
Set of software sensors with confidence interval
backup software sensor based on fuzzy logic
15
Workpackage 2
Workpackage 2.4 Software sensors
Sensor network and fault residues example
Software sensor 1
Software sensor 1
Model 1
Model 2
estimates of the VFA
pH
Qgas
Alkalinity
Qch4
Real sensors
Real sensors
Same predictions ?
16
Workpackage 2
Workpackage 2.4 Software sensors
Validation
- Extensive simulation - Off line data

• Deliverable (27 M)
• Software sensors for process prediction with
  fault residues

17
Workpackage 2
Workpackage 2.5 Sensors validation/location

• Objectives (24.5 MM)
• Validation of the new sensors in pilot-plant
• Determination of the optimum location of all
  sensors in WWTP

Partners APPLITEK  2.5, BIOMATH  4, USC  4,
CESAME  10, INRIA 2, UDG 2

• Deliverable (33 M)
• Final report on validation results and optimum
  location

18
Workpackage 2
Workpackage 2.6 Sensors integration

• Objectives (8.5 MM)
• Integration of new sensors in TELEMAC
  supervision system

Partners APPLITEK  4, SPES  2.5, LEMAIRE  2

• Deliverable (36 M)
• Final report on full-scale performance of
  sensors in TELEMAC supervision system

19
Workpackage 2
Action plan next 6 months

• Sensor specifications
• evaluation titrimetric sensor Biomath
• evaluation commercially available gas sensors
• new design titrimetric and gas sensor
• literature review for data filtering

20
Workpackage 2
Open questions global coherence checking

• How many prototypes ? (WP 2.5 WP 2.6)
• Optimum location ? (WP 2.1 WP 2.5)
• Titrimetric and gas sensor in 1 analyser cabinet
  or separate ?
• Development other sensors ?
• Measurements in influent versus in digester ?
• Sensors integration (WP 2.6)
• simple I/O
• Remote control communication
  adjustment configuration analyser via network ?

Upgraded software
21
Workpackage 2
Human and material resources

• Biomath (WP2 25.5 MM WP1 5.5 MM WP3 4.5 MM)
• Ir. Usama Zaher project leader modelling
  (100 )
• Prof. Peter Vanrolleghem general supervision
  (10 )
• Ir. Dirk De Pauw parameter est. exp. design
  (25 )
• Lic. Paul Willems technical assistance
  titrimeter (10 )
• Portable PC consumables

22
Workpackage 2
Human and material resources

• AppliTek (24 MM)
• Dr. Kristof De Neve project leader (50 )
• Dr. X development new analyser (80 )
• Ing. Annick Beyaert validation tests (60 )
• William Delva technical assistance software
  (70 )
• Dr. Ir. Henk Vanhooren sensor specifications
  (20 )
• Lab GC, 2 bioreactors consumables (sensors,
  electrodes)

23
Workpackage 2
Human and material resources

• INRIA (13.3 MM)
• Drs. Victor Alcaraz development and validation
  of software sensor (100 )
• Dr. Olivier Bernard software sensor (10 )
• Engineer technical assistance (10 )

24
Workpackage 2
Human and material resources

• CESAME (10 MM)
• Prof. Dr. Denis Dochain (100 ?)

25
Workpackage 2
Human and material resources

• USC (4.5 MM)
• Prof. Dr. Juan M. Lema (100 ?)
• Dr. Enrique Roca (10 )
• On-line COD, PLC, computers and flow meters

26
Workpackage 2
Human and material resources

• SPES (8.3 MM)
• Ing. Paolo Ratini (100 )
• PC, 16bit emulator 8bit emulator

27
Workpackage 2
Human and material resources

• Lemaire (2.4 MM)
• Pierre Lemaire (100 )

28
Workpackage 2
Human and material resources

• UDG (8.8 MM)
• Dr. V. Gonzalez-Alvarez coördinator (50 )
• Dr. C. Pelayo Ortiz (40 )
• M.Sc. V. Zuniga Partida (30 )
• Dr. O. Gonzalez-Reynoso (50 )
• Dr. R. Femat-Flores (30 )
• Dr. B. Castillo-Toledo (20 )

29
Workpackage 2
Human and material resources

• UDG (8.8 MM)
• M. Sc. student L. Galdino Garcia (100 )
• M. Sc. student A. Gonzalez (100 )
• Drs. H.O. Mendez Acosta (100 )
• Drs. A. Carrasco (100 )
• Drs. A. Benitez (100 )
• M. Sc. student J.P. Garcia Sandoval (100 )
• M. Sc. student R. Martinez (100 )
• 3 computers

30
Workpackage 2
Human and material resources

• DOMECQ UK (0.9 MM)
• Jean-Francois Lavigne (100 )

31
Workpackage 2
Summary of deliverables

• T3 Sensor Specifications
• T6 Study lab titrimetric sensor
• T12 1st version software sensor design
  adaptive sensor
• T18 filtering sensor data consistency softw
  are sensor (confidence interval) construction
  adaptive sensor (hardware/software) 1st version
  backup fuzzy observer

32
Workpackage 2
Summary of deliverables

• T20 software state forecasting
• T24 fault residues sensor network
• T27 final version software sensor final
  version adaptive sensor