Neural Network Fault Detection System for Dynamic Processes

• Autorzy: Korbicz J. , Patan K. , Obuchowicz A.
• Języki publikacji: EN

Abstrakty

EN

The neural model-based Fault Detection and Isolation (FDI) system for dynamic non-linear processes is considered. The emphasis is placed upon the use of Artificial Neural Networks (ANN's) for residual generation. The proposed network is constructed with the Dynamic Neuron Model (DNM) which contains local memory. Similar to server based schemes, a network is applied to build the nominal and fault models of the investigated system. The output residuals between the process and the models bank are use to detect and identify faults in the system. The modelling efficiency based on the multilayer feedforward Network of Dynamic Neurons (NDN) is compared with the Elman and recurrent network with outside feedbacks. Finally, the NDN and the cascade NDN architectures are applied to build Neural-Residual Generators (NRG) of the two tank system.

Słowa kluczowe

EN

fault detection system; dynamic neural networks; observer based scheme; cascade network architectures optimization; two-tank system

• Wydawca: Polska Akademia Nauk, Wydział IV Nauk Technicznych
• Czasopismo: Bulletin of the Polish Academy of Sciences. Technical Sciences
• Rocznik: 2001
• Tom: vol. 49, nr 2
• Strony: 301--321
• Opis fizyczny: bibliogr. 30 poz., rys., tab.

Twórcy

autor

Korbicz J.

• Institute of Control and Computation Engineering, Technical University of Zielona Góra, ul. Podgórna 50, 65-246 Zielona Góra, Poland (Politechnika Zielonogórska, Wydział Robotyki i Technik Programowania)

autor

Patan K.

• Institute of Control and Computation Engineering, Technical University of Zielona Góra, ul. Podgórna 50, 65-246 Zielona Góra, Poland (Politechnika Zielonogórska, Wydział Robotyki i Technik Programowania)

autor

Obuchowicz A.

• Institute of Control and Computation Engineering, Technical University of Zielona Góra, ul. Podgórna 50, 65-246 Zielona Góra, Poland (Politechnika Zielonogórska, Wydział Robotyki i Technik Programowania)

Bibliografia

• [1] A. Alessandri, T. Parisini, Direet model-based fault diagnosis using neural filters, Prep. IFAC Symp. Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS'97, Hull, DK, 1 (1997) 353-358.
• [2] M. Ayoubi, Fault diagnosis with dynamie neural structure and application to a turbocharger, Proc. IFAC Symp. Fault Detection, Supervision and Safety for Technical Process SAFEPROCESS'94, Espoo, Finland, 2 (1994) 618-623.
• [3] J. Chen, R. Patton, Robust model-based fault diagnosis, Kluwer Academic Publishers, Boston 1998.
• [4] W. Cholewa, J. Kaźmierczak, Teehnical diagnostics of engine, (in Polish), Features Processing of Signals, Silesian University Press, Gliwice 1992.
• [5] G. C Ybenko, Approximation by superpositions of a sigmoidal function, Math. Contr. Signals and Syst., 2 (1989) 303-314.
• [6] A. Doering, M. Galicki, H. Witte, Structure optimization of neural networks with A*-algorithm, IEEE Trans. Neural Networks, 8, 6, (1997) 1434-1445.
• [7] S. E. Fahlman, C. D. Lebiere, The eascade-correlation learning architecture, Advances in NIPS2, ed.: D. Toutetzky, (1990) 524-532.
• [8] A. Ficola, L. C. Michele, F. Magnino, An approach to fault diagnosis for nonlinear dynamic systems using neural networks, Prep. IFAC Symp. Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS'97, Hull, DK, 1(1997) 365-370.
• [9] R. Galar, Handicapped individua in evolutionary processes, Biol. Cybern., 51 (1985) 1-9.
• [10] D. E. Goldberg, Genetic algorithms in search, optimization, and machine learning, Addison-Weley Publishing Company, Inc., Reading, Massachusetts.
• [11] P. M. Frank, Fuzzy supervision. Application of fuzzy logic to process supervision and fault diagnosis, Proc. Int. Workshop Fuzzy Technologies in Automation and Intelligent Systems. Fuzzy Duisburg'94, Duisburg, Germany, (April 7-8 1994) 36-59.
• [12] P. M. Frank, B. Koppen-Seliger, New developments using AI in fault diagnosis, Engng Applic. Artif. Intell., 10, l, (1997) 3-14
• [13] R.. Isermann, Supervision, fault detection and diagnosis methods. An introduction, Control Eng. Practice, 5, 5, (1997) 639-652.
• [14] R. Isermann, P. Balle, Trends in application of model based fault detection and diagnosis of technical processes, Control Eng. Practice, 5, 5, (1997) 709-719.
• [15] H. N. Koivo, Artificial neural networks in fault diagnosis and control, Control Eng. Practice, 2, 7, (1994) 89-101.
• [16] J. Korbicz, Neural networks and their application in fault detection and diagnosis, Prep. IFAC Symp. Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS'97, Hull, UK, 1(1997) 377-382.
• [17] J. Korbicz, Cz. Cempel (Eds.), Analytical and knowledge-based redundancy in fault detection and diagnosis, Special issue of Applied Mathematics and Computer Science, 3, 3, (1993).
• [18] J. Korbicz, A. Obuchowicz, K. Patan, Network of dynamic neurons in fault detection systems, Proc. IEEE Int. Conf. System, Man and Cybernetics, San Diego, USA, (October 11-14, 1998), 1862-1867.
• [19] J. Korbicz, A. Obuchowicz, D. Uciński, Artificial neural networks. Foundations and applications (in Polish), Akademicka Oficyna Wydawnicza, Warszawa 1994, PLJ.
• [20] J. M. Kościelny, Fault isolation in industrial process by dynamie table of state methods, Automatica, 31, 5, (1995) 747-753.
• [21] I. J. Leontaritis, S. A. Bilings, Input-output parametric models for non-lincar systems, lut. J. of Control, 41 (1985) 303-344.
• [22] A. Ligęza, P. Fuster Parra, And/or/not causel graphs - A model for diagnostic reasoning, Appl. Math. and Comp. Sci., 7, 1, (1997) 185-203.
• [23] T. Marcu, Pattern recognition techniques using fuzzly labeled data for process fault detection, Appl. Math. and Comp. Sci., 6, 4, (1996) 815-840.
• [24] K. S. Narendra, K. Parthasarathy, Identifieation and control of dynamical system using neural networks, IEEE Trans. Neural Networks, 1(1990) 12-18.
• [25] A. Obuchowicz, J. Korbicz, Evolutionary search with soft selection and forced direction of mutation, Proc. 7th Int. Symp. Intelligent Information Systems, Malbork, Poland, (15-19 June 1998), 300-309.
• [26] S. Osowski, Neural networks. Algorithmic approach (in Polish), WNT, Warszawa 1996.
• [27] K. Patan, J. Korbicz, Dynamie neural network with filters of different orders, Proc. 4th Int. Symp. Methods and Models in Automation and Robotics, Międzyzdroje, Poland, 2 (1997) 745-750.
• [28] K. Patan, A. Obuchowicz, J. Korbicz, Cascade network of dynamic neurons in fault detection systems, Proc. European Control Conference ECC'99, Karlsruhe, Germany, (August 31-Sept. 3, 1999) (accepted).
• [29] R. J. Patton, P. M. Frank, R. N. Clark (Eds.), Fault diagnosis in dynamic systems. Theory and Application, Prentice Hall, New York 1989.
• [30] R. J. Patton, J. Chen, T. M. Siew, Fault diagnosis in nonlinear dynamic systems via neural networks, Proc. IEE Int. Conf. CONTROL'94, Coventry, UK, (March 21-24, 1994) 1346-1351.