Analytical and neuro-fuzzy modeling for fault detection and identification for nonlinear systems: application to robot manipulator

• Autorzy: Lazeregue M. S. , Salhi H. , Tadjine M.
• Języki publikacji: EN

Abstrakty

EN

This work deals with modeling and fault detection and identification for robot manipulator. We have used for a dynamical system a hybrid approach. The model is decomposed into two parts: first, a certain part modeled using classical analytical theory and it is preferable to be linear. Second, an uncertain part representing the nonlinearities neglected in the first part, which is modeled using neuro-fuzzy modeling. Both analytical redundancy and neuro-fuzzy modeling are used to improve robustness. The analytical redundancy is used to generate residuals for the fault detection and location procedure. The neuro-fuzzy modeling is used to model modeling errors and faults, which allows performing the robustness and the sensitivity. Thanks to neuro-fuzzy modeling the errors of modeling are compensated and the faults are well identified as it is shown through the results of simulation.

Słowa kluczowe

EN

dynamical system; robot manipulator; neuro-fuzzy modeling; identification; error compensation; fault detection and identification

• Wydawca: Polish Academy of Sciences, Committee of Automatic Control and Robotics
• Czasopismo: Archives of Control Sciences
• Rocznik: 2009
• Tom: Vol. 19, no. 3
• Strony: 325--350
• Opis fizyczny: Bibliogr. 34 poz., rys.

Twórcy

autor

Lazeregue M. S.

autor

Salhi H.

autor

Tadjine M.

• Laboratoire Systemes Dept. Electonique, Universite Saad Dahlab, Soumaa, Blida, Algerie

Bibliografia

• [1] A. ISIDORI: Nonlinear control systems; An introduction. (2nd Ed.) Springer Ver-lag, 1989.
• [2] G. BORNARD: Modelisation physiąue systemes non lineaires. 1 Edite par Fossard AJ et Normand-Cyrot, Masson, 1993, (in French).
• [3] G. BORNARD, F. CELLE-COUENNE and G. GILLES: Observabilite et observa-teurs systemes non lineaires. 1 Edie par Fossard A.J et Normand-Cyrot, Masson, 1993, (in French).
• [4] B. BOUZOUIA: Commande dynamique des robots manipulateurs: Identification des parametres et etude de strategies adaptatives. PhD thesis, l’UPB de Toulouse, 1989, (in French).
• [5] C. DE WIT BROGLIATO, DE LUCA SICILIANO and W. KHALIL: Control ofrigid manipulators. Theorie de la Commande des Robots, Ecoled’ete d’Automatique de Grenoble, 1992.
• [6] D. BRUNET, D. JAUME, M. LABARRERE, A. RAULTand M. VERGE: Detection et diagnostic de Pannes. Approches par modelisation. Edition Hermes, 1990, (in French).
• [7] E. Y. CHOW and A. S. WILSKY: Analytical redundancy and the design of robust failure detection systems. IEEE Trans, on Automatic Control, AC-29(7), (1984).
• [8] Li Chunshien and KUO-HSIANG CHENG: Recurrent neuro-fuzzy hybridlearning approach to accurate system modeling. Fuzzy Sets and Systems, 158 (2007), 174-212.
• [9] P. Coiffet: La robotique, principes et applications. Edition Hermes, 1986, (in French).
• [10] D. J. KAZAKOS, S.A. MANESIS and T.G. PEMENIDES: Nonlinear observer for fermentation process and bioreactor. ECC France, 1993, 280-283.
• [11] E. DOMBRE and W. KHALIL: Modelisation et commande des robots. Editios Hermes, 1988, (in French).
• [12] B. DUBUISSON: Diagnostic et reconnaissance des formes. Edition Hermes, 1990, (in Frrench).
• [13] F. J.UPPAL and R. J. PATTON: Fault diagnosis of an electro-pneumaticValve actua-tor using a neural network with fuzzy capabilities. ESANN’2002, Bruges, Belgium, (2002), 501-506.
• [14] P. M. FRANK and J. WUNNENBERG: Roubust fault diagnosis using unknown input observer Schemes. Fault diagnosis in dynamie systems; theory and applications. Edite par Ron Patton, Paul Frank and Robert Clark, Prentice Hall, 1989.
• [15] P. M. FRANK: Advances in observer-based fault diagnosis. Tooldiag 93, 3 Toulouse, France, (1993).
• [16] G. C. MOUZOURIS and J.M. MENDEL: Dynamie non-singleton fuzzy logie systems for nonlinear modeling. IEEE Trans, on Fuzzy Systems, 5(2), (1997), 199-207.
• [17] J. GERTLER and M. KUNWER: Optimal residual decoupling for fault diagnosis. Tooldiag 93, Toulouse, France, (1993).
• [18] G. I. SAINZ PALMERO, J.JUEZ SANTAMARIA, E.J. MOYA DE LA TORRE and J.R. PERAN GONZALEZ: Fault detection and fuzzy rule extraction in AC motors by a neuro-fuzzy ART-based system. Engineering Applications ofArtificial intelli-gence, 18 (2005).
• [19] H. K. KHALIL: Robustness issues in output feedback control of feedback linearis-able systems. ECC France, (1993), 58-60.
• [20] R. ISERMANN: Integration of fault detection and diagnosis methods. IFAC Sympo-sium Safeprocess’'94, Helsinki Finland, 2 (1994), 597-612.
• [21] J. B. WALLER, J. HU and K. HIRASAWA: Nonlinear model predictive control uti-lizing a neuro-fuzzy Predictor. IEEE'2003, (2003), 3459-3464.
• [22] B. KOPPEN and P.M. FRANK: Application of observer-based fault detection schemes to inverted pendulum. Tooldiag 93, Toulouse, France, 3 (1993), 979-986.
• [23] S. LABIOD: Contribution ala commande adaptative floue des systemes Non Lineaires. PhD thesis. FENP (el Harach) Algerie, 2004, (in French).
• [24] M. S. LAZEREGUE, F. NOUREDDINE and D. NOYES: Detection d'erreurs dans un systeme dynamiąue. Application robotiąue. 4 e Colloque Maghrebin sur les Mod-eles Numeriques de 1’Ingenieur, U.S.T.H.B, Aller, 1 (1993), 298-303, (in French).
• [25] K. NAND, S.P. SINGH and A.S. RAGHUVANSHI: Adaptive intelligent hydro speed identification with water and random load disturbances. Engineeńng Application ofartificial intelligence. (2007), 1-14.
• [26] B. NARENDRA, A.G. KOTHARI and D.P. KOTHARI: ANFIS based HDVC control and fault identification. HAITJ. of Science and Engineeńng B, 2 (2005), 673-689.
• [27] F. NOUREDDINE, M.S. LAZEREGUE and D. NOYES: A fault detection-location of a robot using scheme of observers. IFAC Symposium Safeprocess’94, Helsinki, Finland 1(1994), 341-346.
• [28] P. CONTI: Contribution a la commande dynamique adaptative des robots manipu-lateurs. PhD thesis, 1’UPB de Toulouse, France, 1987, (in French).
• [29] R. BABUSKA H. VERBRUUGEN: Neuro-fuzzy methods for nonlinear system identification. Annual Reviews in Control, 27 (2003), 73-85.
• [30] G. I. SAINZ PALMERO, J.JUEZ SANTAMARIA, E.J. MOYA DE LA TORRE and J.R. PERAN GONZALEZ: Fault detection and fuzzy rule extraction in AC motors by a neuro-fuzzy ART-based system. Engineeńng Applications ofArtificial intelligence, 18 (2005).
• [31] M. STAROŚWIECKI: La problematique et les approches de la surveillance des sys-temes technologiąues. Journees’etude S3. Edition LAIL Cite scientifique, (1994), (in French).
• [32] S. G. TZAFESTAS: System fault diagnosis using the Knowledge-based methodol-ogy. In R. PATTON, P. FRANK and R. CLARK (Eds.) Fault diagnosis in dynamie systems; Theory and applications. Prentice Hall, 1989.
• [33] Y. DIAO and K.M. PASSINO: Intelligent faul-tolerant control using adaptive and learning mehods. Control Engineeńng Practice, 10 (2002), 801-817.
• [34] Y. DIAO and K.M. PASSINO: Fault diagnosis for turbine engine. Control Engineeńng Practice, 12 (2004), 1151-1165.