1 bond graph model based for diagnosis belkacem ould bouamama professeur : ecole polytechnique de...

1

BOND GRAPH MODEL BASED FOR DIAGNOSIS

BOND GRAPH MODEL BASED FOR DIAGNOSIS

Belkacem OULD BOUAMAMAProfesseur : Ecole Polytechnique de Lille (poltech-lille.fr)

Recherche : Laboratoire d'Automatique, Génie Informatique

et Signal (LAGIS - UMR CNRS 8021)

Coordonnées :[email protected]

Tel: (33) (0) 3 28 76 73 87 , mobile : (33) (0) 6 60 12 30 20

mailto:[email protected]

2Belkacem Ould Bouamama, Damadics Vacation Scholl, Liille, 9-11 april.

PLANPLAN

INTRODUCTION

RESIDUAL GENERATION USING BOND GRAPH

RESIDUAL AND MODEL BUILDER USING

SYMBOLS SOFTWARE

ONLINE INDUSTRIAL APPLICATION

CONCLUSIONS AND DISCUSSION


PART 1PART 1

INTRODUCTION


RESIDUAL AND MODEL BUILDER USING SYMBOLS SOFTWARE




Given presentation

1.1. FDI Algorithms1.1. FDI Algorithms

FDI Algorithms FDI Algorithms

Non-Model Based Techniques Model Based

Partie 2Partie 2

ObserversIdentification Information Redundancy

analytical redundancy

Hardware redundancy


1.2. Model-based FDI1.2. Model-based FDI

S E N SO R SS E N SO R S

Process actual operation

RESIDUALGENERATORRESIDUAL

GENERATOR

MODEL OF THE

NORMAL OPERATION ALARM GENERATION

0

IsolationIsolation IdentificationIdentification

ALARM INTERPRETAION

DetectionDetection

6

R

1.3. Hardware and analytical redundancy

1.3. Hardware and analytical redundancy

SS1 1 or Sor S22

0P

.P*Q 111

dt

dCR

SS22

Hardware redundancyHardware redundancy

Detection IsolationSensors

0S

.S*Fr 1111

dt

dCR

S3 S2 S1

F2

F1

0P*Q2 R 0S*Fr 122 R

Analytical redundancyAnalytical redundancy

?

LeakageLeakageSS11FF11 ValveValve FF22

r1

r2

11 11

00

11

1100

11

11

00

11

Monitorability analysis


1.4. Issues1.4. Issues

MODELLINGModelling step is most important in FDI design

obtaining the model is a difficult task

The constraints are not deduced in a systematic way

It is not trivial in the real systems to write the model under a "beautiful" form x=f(x,u,θ).

RESIDUAL GENERATIONEliminate the unknowns : analytic redundancy approach

– Existing methodology : parity space for linear, elimination theory (constraints under polynomial forms)

Variables to be considered : all quantities constrained by the system components (process, actuators, sensors, algorithms

How to generate directly from the process ARRs and models : Bond

graph tool well suited because of its causal and structural properties.


1.5. Bond graph and diagnostic1.5. Bond graph and diagnosticBOND GRAPH FOR MODELLING (1961)

Control (1990) Diagnostic

Qualitative Approach (1993)

Quantitative approach (1995)

Open loop Linear model Sensor and actuator faults

Monoenergy Bond Graph Multienergy bond graph

Closed loopSensor, process and actuator faultsImplementation


PART 2PART 2

INTRODUCTION






2.1. Bi-partite graph representation and bond graph interpretation

2.1. Bi-partite graph representation and bond graph interpretation

),,( ZSS

State equations)(

)),(),((

xcy

tutxFx

yuxZ

Structural description

Z

m

a

c

b

s

K

u

MSfMSe

DfDe

SfSe

X

fe

fe

fe

nene

,

.

.

,

,

22

11

Bi-partite graph

How to define and Z ?


A) Constraints (1/2) A) Constraints (1/2)

njSGYTFjjs R ,10= nj is the number of junctions

Structural equations S : Informations about the structure

Measurement equations m

nsmDfDem R ,= ns is the number of sensors

Behavioral equations b : Informations about the behavior

NcsNcmNe

b

RSICRb

*

= Ncs: is the number of simple components

Ncm : Nbre of multiport components

Ne : nbre of energies, (Ne=2),


A) ConstraintS (2/2) A) ConstraintS (2/2)

Control algorithm equations c nc

cncccc R ,...21= nc is the number of controllers

anncnsnenjR

Controllled sources equations a

naanaMSejMSfMSfMSfa R ,......=

21

na is the number of controlled sources


B) VariablesB) Variables

Z=X∪KZ=X∪K USED VARIABLES:

• Known variables K

nsncnaRKuDfDeSeSfMSfMSeK= ,

na is the number of actuators

nc is the number of controllers

ns is the number of sensors

nensncnaRZ .2

• Unknown variables X

)(,...)(,)(, 2211 tf(t)etf(t)etf(t)eX(t)= nene ne is the nbre of elementsNcsNcmNeX(t) *2**2


2.2. Hydraulic academic example 2.2. Hydraulic academic example

R

f3

f2 e2

f1

MSf

Ps=0

FIPI

LI


2.2.1. Bond graph model 2.2.1. Bond graph model

0MSf

C

1

R

Env

iron

nem

ent

Env

iron

nem

ent

5

Se1

2

3

4

Z

0 1

C RDe:P Df:FDe:P Df:F

J0

J1

C

R

mP

mF

e2

f2

e4

f4

X

s

b

mK

MSfSeDeDf

SeMSf

De:P Df:F


2.2.2. ARRs generation matching and incidence matrix

2.2.2. ARRs generation matching and incidence matrix

f2 e2 f4 e4 MSf Se De Df

J0 1 0 1 0 1 0 0 0

J1 0 1 0 1 0 1 0 0

C 1 1 0 0 0 0 0 0

R 0 0 1 1 0 0 0 0

mP 0 1 0 0 0 0 1 0

mF 0 0 1 0 0 0 0 1

Causal matching w.r.t all unknown variables but not w.r.t all the constraints

Z=XK

X K

1

1

1

1

RRA1

RRA2


2.2.3. Oriented graph associated with a matching

2.2.3. Oriented graph associated with a matching

mP

e2 f2De C

MSf

Df mF

f4

J0RRA1

MSf

mPDe

Se

Df mF

f4

e2

J1RRA2

Se

R

e4


2.2.4. ARRS generation : Bi-partite graph and BG approach

2.2.4. ARRS generation : Bi-partite graph and BG approach

1) Unknown variables elimination order in the oriented graph

2) Initial step for ARR generation : difficult to fix

1) Covering causal path in the BG which is a particular matching according to the affected causality

2) From Energy conservation law from “0” or “1” junction

The goal is to study all the causal paths relating the considered junction to the sources and the sensors

Bond graphBond graphBi-partite graphBi-partite graph


2.2.5. ARRS generation using BG approach

(1/2) 2.2.5. ARRS generation using BG approach

(1/2)

Env

iron

nem

ent

Env

iron

nem

ent

0 151

2

3

4

0 1

C RDe:P Df:F

SeMSf

A) Bond graph model in integral causalityA) Bond graph model in integral causality


2.2.5. ARRS generation using BG approach (2/2)

2.2.5. ARRS generation using BG approach (2/2)

0*231 Defff:ΦJ0 ?231 ,, fffX

?1f

?2f

MSff 11- MSf

dtDedf c /)(2 2-C-2-De

?3f Dff 33-Df

)/)(( dtDedDfMSf c:ARR1 )(DfD R-Se-e:ARR2

0*543 Dfeee:ΦJ1 ?543 ,, eeeX

0 1

Env

iron

nem

ent

Env

iron

nem

ent

51

2

3

4

0 1

C R

SeMSf

De*:P Df*:F

?3e Dee 33- De

?5e See 55-Se

?4e )(4 Dfe R4-R-4-Df

Bond graph model in derivative causality sources

are dualised


2.3. Thermofluid process2.3. Thermofluid process

PI

LC

u1

TCu2

FI


2.3.1. Generalized causal path2.3.1. Generalized causal path

mSfh :

TSe:

1

Rc

m

H

0h

0t

De:T

De:P

C

It is a causal path that can follow power links or informational links, or both.It is a causal path that can follow power links or informational links, or both.

TDeSfh :

PDeSfh :


2.3.2. Bond graph model2.3.2. Bond graph model

QSft:

TQ

T

H0t

Env

iron

nem

ent

C

m

R

11P

mSfh : 0h

Rc T

H

TSe:

CpMTF

:

QSft:

TQQSft

:TQ

T

H0t

Env

iron

nem

ent

C

m

R

11P

mSfh : 0h

Rc T

H

TSe:

CpMTF

:

T

HT

HT

H0t0t

Env

iron

nem

ent

C

m

R

11P

mSfh : 0h

Env

iron

nem

ent

C

m

R

11P

mSfh : 0h

C

m

R

11P

mSfh : 0h

C

m

R

11P

mSfh : 0h

R

11P

mSfh :mSfh : 0h

Rc T

HRcRc T

H

TSe:

CpMTF

:

TSe:

CpMTF

:

TSe:

CpMTF

:

TSe:

CpMTF

:

De:P

0

TC

De:T

u2De*:u2 0

Dfg

TF:

LC

u1

0 De:L0De*:u1


2.3.3. Monitoring of controlled system and material redundancy

2.3.3. Monitoring of controlled system and material redundancy

gTF:

LC

u1

0 De:L0De*:u1 1 2 3 4

De*:P

mSf :

Level sensor

Pressure sensor

Fictive sensor : control signal

Mat

eria

l re

du

nd

ancy

Th

erm

oflu

id

Pro

cess

Ifor

mat

ion

sys

tem

C

1 0

PgDeLDe ::: RRA3

)(1 LCFu :RRA4

25

SOFTWARE IMPLEMENTATIONSOFTWARE IMPLEMENTATION

2.4. RRAs generation algorithm

Bond Graph Junction structure Bond Graph Junction structure

P&IDiagram

Behavioral equations Measurement equations Control algoritm equations

For each junction

Structural equations

Substitutions

ARR

Is ARR independent ? Add to ARRs set

YesNo

Output=F(inputs)

ARR


PART 3PART 3

INTRODUCTION






3.1. How it works ?3.1. How it works ?

ONLINE

OF LINE

GUIP

I

D

Arc

hit

ectu

ral m

odel

Data Process GUI

Behavioral modelResiduals

Monitorability


3.2. Main activity3.2. Main activity

SOFTWARE

Formal residuals

Formal dynamic model


Technical specifications

P&IDiagram

XML format


3.3. HOW TO BUILD ARCHITECTURAL MODEL ?3.3. HOW TO BUILD ARCHITECTURAL MODEL ?

PID

PID

Generic data base

Select process plant item

Interconnect process plant item

Check architectural consistency


3.4. Used Software3.4. Used Software

SYMBOLS SYstem Modeling by BOndgraph Language and Simulation.

It is a modeling, simulation and control systems software for a variety of scientific and engineering applications.

Created by : HighTech Consultants, Indian Institute of Technology from 1980s.

http://www.symbols2000.com


3.5. Functionalities used from symbols

3.5. Functionalities used from symbols

EXISTINGUser interfacesEquations manipulation routines (Symbolic algebra in linear case) Submodels facilities (creating capsules)Solvers

CREATEDGeneric items data baseResidual generation programLinking with symbols format


3.6. Toolbox architecture3.6. Toolbox architecture

Internal activities

Data BAseData BAse

FDIPADFDIPAD

Outputs

Matlab S-function



Structural matrix

Structural matrix

Online monitoring

Sensors

Code en C

Offline Simulation

Create New Submodel capsule

Bond graph

FDIPADFDIPAD

Behavioral model in normal or faulty mode

Behavioral model in normal or faulty mode

RRAsRRAs


3.7. Graphical User Interface (1/2)3.7. Graphical User Interface (1/2)

Architectural model

Behavioral model

Data base


3.7. Graphical User Interface (2/2)3.7. Graphical User Interface (2/2)

Fault signature

Residuals


3.8. Demonstration using academic examples

3.8. Demonstration using academic examples

R

f3

f2 e2

f1

MSf

P s= 0

FIPI

LI

R

f3

f2 e2

f1

MSf

P s= 0

FIFIFIFIFIPIPIPI

LILI

PI

LC

u1

TCu2

FIPIPIPI

LC

u1

LCLCLCLC

u1

TCu2 TCu2 TCTCu2

FIFIFI


PART 4PART 4

INTRODUCTION






4.1. Application to a steam generator supervision

4.1. Application to a steam generator supervision


4.2. Panorama CCOM Interface4.2. Panorama CCOM Interface

Panorama Supervision

Controls FaultsResiduals

FCTINTPP Archives Mistral Alarms

Data Server

CCOM C++ Interface

CCOM Server

CCOM Client

DD

E

Process Data, Residues & Faults

Commands & Reconfigurations12

3

4


4.3. Integration in the supervision system

4.3. Integration in the supervision system

Process Panorama

CCOMG2 CCOMTB 3.7

Java CCOMTB 3.2

Java CCOMTB 5.2

G2 CCOMData Manager

FctIntppC++ CCOM

TB 7.2C++ CCOM

Data

TBS

Faults

Reconfiguration

Data

Faults

FaultsAll

Process Provider

XML TB 5.1

Mo

de

l

Model

Temporal Band sequences

Reconfiguration


PART 5PART 5

INTRODUCTION






CONCLUSIONSCONCLUSIONS

In supervision tasks, human operators do not consider the running process in terms of its mathematical behavior, but of its P&Ids or functions

The interest of the presented approach :consists in the use of only one representation (bond graph modelling) for ARRs and dynamics models generation in symbolic format.the industrial designer can easily (because of integration of the functional tool as interface with the human operator) build the thermofluid dynamic model and ARRsPropose to the user a sensor placement to satisfy a given technical specificationTo add a new component in the data base in a generic way


References References

B. Ould Bouamama, (2002). Bond Graph Appraoch as analysis tool in thermofluid model library conception. Journal of Franklin Institute. 28 pages (accepté pour publication en 2003).

B. Ould Bouamama, M. Staroswiecki, K. G. Dauphin-Tanguy and A.K. Samantary (2002). Model builder using Functional and bond graph tools for FDI design submitted to CEP

B. Ould Bouamama. Mémoire à Diriger les Recherches N°H360: "Modélisation et Supervision des Systèmes en Génie des Procédés -- Approche Bond Graphs" , Université des Sciences et Technologies de Lille, Soutenue le 20 Décembre 2002.

Ould Bouamama B, Samanatary A.K, G. Dauphin-Tanguy, M. Staroswiecki (2002). Causality Inversion Approach in Derivation of Analytical Redundancy Relations for Fault Detection and Isolation. International Conference on Bond Graph Modelling and Simulation, ICBGM'2003, Orlando, Florida, vol. 35, n°.2,pp. 104-109, 19-23 january 2003.

Busson, F. « Modélisation et surveillance par bond graph des processus thermofluides. », Thèse de doctorat de l’Université des Sciences et Technologies de Lille, Décembre 2002, France.


ENDEND


PIDiagramPIDiagram

FIR

10PR11

PIR16

TR17

PC2

PR14

PR15

V3

Opérateur

PR13

PR12

ZC1

V2

V11 LIR

9LIR

8

LG1

TR5

PC1

PIR

7

TR6

Q4

Résistance thermique

LC1

V10

FIR

3

P2

P1

V9

Réservoir

TIR2

LIR

1

LG3

V8

Condensate

V4

V6

LG2

LC2

Aero-refrigerator

TIR26

Environment FIR

23

FIR

24

TIR27

TIR21

Cooling water

P3

P4

TIR22

TC5

PR27

TIR20

LIR19

LIR18

V5

TIR25

V1

Détente de la vapeur

Condenseur

Circuit d’alimentation

Chaidière

Circuit de refroidissement

FIR

10

FIR

10

FIR

10PR11PR11

PIR16

PIR16

TR17TR17TR17

PC2

PC2

PR14PR14PR14

PR15PR15

V3

Opérateur

PR13PR13PR13

PR12PR12PR12

ZC1

ZC1

V2

V11 LIR

9LIR

8

LIR

8

LG1

LG1

TR5

TR5

TR5

PC1

PC1

PIR

7

PIR

7

TR6

TR6

Q4Q4

Résistance thermique

LC1

LC1

V10

FIR

3

P2

P1

V9

Réservoir

TIR2

LIR

1

LG3

FIR

3

FIR

3

FIR

3

P2

P1

V9

Réservoir

TIR2

LIR

1

LG3

P2

P1

V9

Réservoir

TIR2

LIR

1

LG3

P1P1P1

V9

Réservoir

TIR2

LIR

1

LG3

V9

Réservoir

TIR2

TIR2

LIR

1

LIR

1

LG3

LG3

V8

Condensate

V4

V6

LG2

LC2

Aero-refrigerator

TIR26

TIR26

Environment FIR

23

FIR

23

FIR

24

FIR

24

FIR

24

TIR27

TIR27

TIR21

TIR21

Cooling water

P3

P4

TIR22

TIR22

TC5

TC5

PR27PR27

TIR20

TIR20

LIR19

LIR19

LIR18

LIR18

V5

TIR25

TIR25

TIR25

V1

Détente de la vapeur

Condenseur

Circuit d’alimentation

Chaidière

Circuit de refroidissement


Dynamic modelDynamic model


ResidualsResiduals


Technical speficiationsTechnical speficiations


Monitorability analysisMonitorability analysis

Fault indicatorsC

ompo

nent

s

Messages : Which component is Monitorable Isolable

1 bond graph model based for diagnosis belkacem ould bouamama professeur : ecole polytechnique de...

Documents

bond graph model

bond graph residual

damadics vacation scholl

diagnostic bond graph

bond graph tool

model builder

discussion slide

analytical redundancy