In this paper a novel approach for the fault detection and diagnosis of the actuators in nonlinear systems is presented. The faults are assumed to be any unexpected changes of the parameters in the actuators, which include either abrupt changes or slowly drifting changes. Firstly a known nonlinear system is considered, where an adaptive diagnostic model incorporating the estimate of fault is constructed. The diagnostic algorithm is then developed which produces the estimate of the fault so that the error between the system output and that of the model is minimised. Unknown nonlinear systems are then studied using a feedforward neural network trained to estimate the system under healthy conditions. Taking the trained neural network as the neuro model of the system, similar detection and diagnostic algorithms to that of known systems are obtained which are still capable of producing a good estimate for the fault. The method is demonstrated using a simulated nonlinear system and good results are obtained.
Chen, S., and Billings, S.A.1994. 'Neural networks for nonlinear system modelling and identification '. In: C. J. Harris (Ed), Advances in Intelligent Control, Taylor & Francis, London, 85-112.
2.
Chow, E.Y., Lou, X.C., Verghese, G.C., and Willsky, A.S.1986. 'Redundancy relations and robust failure detection'. In: Basseville, M. and Benveniste, A. (Eds) Detection of Abrupt Changes in Signals and Dynamic Systems, LNCIS, 77, Springer Berlin, 275-294.
3.
Clark, R.N.1978. 'A simplified instrument failure detection scheme', IEEE Trans Aero Elec Syst, 14, 558-563.
4.
Clark, R.N.1989. 'State estimation schemes for instrument fault detection' . In: Patton, R. J., Frank, P., and Clark, R. N. (Eds), Fault diagnosis in Dynamic Systems: Theory and Application, Prentice Hall International, Series in Syst Contr Engg.
5.
Frank, P.M.1990. 'Fault diagnosis in dynamic systems using analytical knowledge-based redundancy: a survey and some new results', Automatica, 26, 459-474.
6.
Frank, P.M.1994. 'Enhancement of robustness in observer-based fault detection' , Int J Contrel, 59, 955-981.
7.
Freyermuth, B., and Isermann, R.1991. 'Model incipient fault diagnosis of industrial robots via parameter estimation and feature classification', Proc of 1991 European Control Conference, Grenoble, France.
8.
Girosi, F., and Poggio, T.1990. 'Networks and the best approximation property', Biol Cybern, 63, 169-176.
9.
Isermann, R.1984. 'Process fault diagnosis with Parameter estimation methods' , Proc IFAC Digital Computer Application to Process Control, Vienna, 51-60.
10.
Leontaritis, I.J., and Billings, S.A.1987. 'Model selection and validation methods for non-linear systems', Int J Control, 45, 311-341.
11.
Linkens, D.A., and Mahfouf, M.1994. 'Supervisory generalised predictive control and fault detection for multivariable anaesthesia', IEE-ProcControl Theory and Applications, 141, 70-82.
12.
Mironovski, L.A.1979. 'Functional diagnosis of linear dynamic systems', Automn Remote Control, 40, 120-128.
Montague, G.A., Wills, M.J., Morris, A.J., and Tham, M.T.1991. 'Artificial neural network based multivariable predictive control', Proc of ANN'91, Bournemouth , UK.
15.
Ono, T., Kumamaru, T., Maeda, A., Sagara, S., and Kumamaru, K.1987. 'Influence matrix approach to fault diagnosis of parameters in dynamic systems', IEEE Trans Indus Electron, 34, 285-291.
16.
Park, J., and Rizzoni, G.1994. 'An eigenstructure assignment algorithm for the design of fault detection filters', IEEE Trans on Automat Control , 39, 1521-1524.
17.
Patton, R.J., and Chen, J.1991. 'A robust parity space approach to fault diagnosis based on optimal eigenstructure assignment', Proc Int Conf on Control '91, 2, 1056-1061.
18.
Patton, R.J., Chen, J., and Siew, T.M.1994. 'Fault diagnosis in nonlinear systems via neural neturorks' , Proc Int Conf on Control '94, 2, 1346-1351.
19.
Patton, R.J., Frank, P., and Clark, R.N.1989. Fault Diagnosis in Dynamic Systems: Theory and Application , Prentice Hall International, Series in Syst Contr Engg.
20.
Rumelhart, D.E., Hinton, G.E., and Williams, R.J.1986. 'Learning internal representations by error propagation' . In: Rumelhart, D. E., and McClelland, J. L. (Eds), Parallel Distributed Processing, 1, chap 8, Cambridge, MA: MIT Press.
21.
Wang, H., and Daley, S.1993. 'An approach to fault detection using nonlinear modelling and estimation', Proc 12th IFAC World Congress , Sydney, 8, 291-294.
22.
Wang, H., Brown, M., and Harris, C.J.1994. 'Fault detection for a class of non-linear systems via associative memory networks', ProcIMechE, 208, 101-108.
23.
Wang, H., Liu, P.G., Harris, C.J., and Brown, M.1995. Advanced Adaptive Control, to be published by Elsevier Science, Oxford.
24.
Widrow, B., and Lehr, M.A.1990. '30 years of adaptive neural networks: perceptron, madaline, and back-propagation', Proc IEEE, 78, 1415-1441.
25.
Willis, M.J., Massimo, D.C., Montague, G.A., Tham, M.T., and Morris, A.J.1991. 'Artificial networks in process engineering', IEE.ProcControl Theory and Applications, 138, 256-266.
26.
Zhang, J., and Roberts, P.D.1992. 'On-line process fault diagnosis using neural network techniques', Trans InstMeasurement and Control, 14, 179-188.
