Detecting faults in a hydraulic system using neural network and observer approaches

• Autorzy: Shields D.N. , Du S. , Gaura E.
• Języki publikacji: EN

Abstrakty

EN

A robust fault detection observer (RFDO), which includes a robust residual, is designed for a general nonlinear system with polynomial-type nonlinearities. Two main theoretical results are given for the existence, stability and the detectability of the RFDO. The approach is applied to a three-tank hydraulic system where two fault scenarios are considered. Three different degree observers, with corresponding residuals, are assessed. This observer approach is then compared to the performance of a neural network approach under the same fault conditions.

Słowa kluczowe

EN

robust fault detection observer; nonlinear systems; polynomial-type nonlinearities; stability; detectability; three-tank hydraulic system; neural network; continuous-time system

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Systems Science
• Rocznik: 2002
• Tom: Vol. 28, nr 1
• Strony: 85--106
• Opis fizyczny: Bibliogr. 27 poz.

Twórcy

autor

Shields D.N.

• Maths-MIS, Voventry University, Priory St, Voventry, CV15FB, U.K.

autor

Du S.

• Maths-MIS, Voventry University, Priory St, Voventry, CV15FB, U.K.

autor

Gaura E.

• Maths-MIS, Voventry University, Priory St, Voventry, CV15FB, U.K.

Bibliografia

• [1] Astrom K., Albertos P„ Blankę M., Isidori A., Schaufeberger W., Sanz R., Control of Compto Springer-Yerlag, New York 2001.
• [2] Ashton S. a., Methods for fault detection and isolation with application to hydraulic pipelines, Ph.D. thesis, Coventry University, 1999.
• [3] Ashton S. A. Shields D. N., Fault detection observers for a class of noniinear systems, [in;] H. Nijmeijer and T. I. Fossen (eds.), New Directions In Noniinear Observer Design, Springer-Yerlag, London, 1999.
• [4] Benkhedda H., Patton R., Information Fuzzion in Fault Diagnosis Based on B-spline Networks, IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, Safeprocess97. Kingston Upon Hull, Yol. 2, 1997,681-686.
• [5] De Persis C., Isidori A., A geometrie approach to noniinear fault detection and isolation, Proceedings of IFAC Safeprocess’2000,2000,209-214.
• [6] Edelmajer a., Boker J., Szigeti F., Keyicsky L., Robust detection fdter design in the presence of time-varying system perturbations, Automatica, Yol. 33,1997,471-475.
• [7] Edwards C., Spurgeon S. K., Patton R., Sliding modę observers for fault detection and isolation, Automatica, Yol. 36,2000, 541-553.
• [8] Frank P. M., On-line Fault Detection in Uncertain Noniinear Systems Using Diagnostic Observers: A Survey, International Journal of Systems Science, Yol. 25,1994, 2129-2154.
• [9] Hac a.. Design of disturbance decoupled observer for bilinear systems, ASME, J. Dynamie Syst. Measure. Control, Yol. 114, No. 12,1992, 556-562.
• [10] Han z., Frank P. M., Physical parameter estimation based FDI with neural networks, Proceedings of IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, Safeproc- ess’97. Kingston Upon Hull, Yol. 1,1997, 294-299.
• [11] Hammouri H., Kinnaert M., Yaagoubi E., Observer-based approach to fault detection and isolation for noniinear systems, IEEE Trans. Automat. Contr., Yol. 44,1998, 1879-1884.
• [12] Hammouri H., Kinnaert M., Yaagoubi E., Fault detection and isolation for state affine systems, European Journal of Control, Yol. 4,1999, 2-16.
• [13] Kinnaert M., Peng Y., Hammouri H., The fundamental problem of residual generation for bilinear systems up to injection. Proc. IFAC Conf. ECC’95, Romę, Italy, 1995, 3777-3782.
• [14] Kinnaert M., Robust fault detection based on observers for bilinear systems, Automatica, Yol. 35, 1999,1829-1842.
• [15] Krener a. J., ISODORI A., Linearization by output injection and noniinear obserrers, Systems and Control Letters, Yol. 3, 1983,47-52.
• [16] Massoumnia M., Yerghese G., Willsky A., Fault detection and identification, IEEE Trans, on Automat. Control., Yol. 46,1998, 853-865.
• [17] Numeijer H., Yan der Schaft A., Noniinear Dynamical Control Systems, Springer-Yerlag, New York 1990.
• [18] Patton R., Frank P., Clark R., Fault Diagnosis in Dynamical Systems, Theory and Applications, Prentice Hall, 1989.
• [19] Seliger ,R., Frank P. M., Robust componeni fault detection and isolation in nonlinear dynamie Systems using nonlinear unknown input obsen/ers, Preprints of Safeprocess’91, Baden-Baden, FRG, Sept. 10-13,1991, Vol. 1,313-318.
• [20] Shields D. N., Ashton S. A., Daley S., Design of observers for detecting faults in hydraulic sub- sea pipełines, Control Engineering Practice, VoI. 9, 2001, 297-311.
• [21] Staroswieki M., Comtet-Varoa G., Analytic redundancy relations for fault detection and isolation in algebraic dynamie Systems, Automatica, Vol. 37,2000, 687-699.
• [22] XiA X. H., Gao w. B., Nonlinear obsen/er design by obseryer error linearization, SIAM J. of Control and Optimization, Vol. 27,1989,199-216.
• [23] Yu D., Shields D. N., Bilinear fault detection observer and its application to a hydraulic system, International Journal of Control, Vol. 64,1996,1023-1047.
• [24] Yu D., Shields D. N., Daley S., A hybrid fault diagnosis approach using neural networks, Neural Computing and Applications, Vol. 4,1996,21-26.
• [25] Yu D., Shields D. N., A bilinear fault detection observer, Automatica, Vol. 32, 1996,1597-1602.
• [26] Yu D., Shields D. N., Extension of parity-space method to fault diagnosis of bilinear systems, Int. J. of Systems Science, Vol. 33,2001, 953-962.
• [27] Zhirabok A. N., Fault diagnosis in nonlinear systems with uncertainties. Proc. of IFAC Symp., Safeprocess’97, Hull University, Vol. 1,1994, 528-533.