A. Vargas–Martínez*¹, L. E. Garza–Castañón²

^1,2 Tecnológico de Monterrey, Campus Monterrey Av. E. Garza Sada # 2501, 64849, Monterrey, México. *E–mail: A00777924@itesm.mx, legarza@itesm.mx

ABSTRACT

We present the integration of artificial intelligence, robust, nonlinear and model reference adaptive control (MRAC) methods for fault–tolerant control (FTC). We combine MRAC schemes with classical PID controllers, artificial neural networks (ANNs), genetic algorithms (GAs), H∞ controls and sliding mode controls. Six different schemas are proposed: the first one is an MRAC with an artificial neural network and a PID controller whose parameters were tuned by a GA using Pattern Search Optimization. The second scheme is an MRAC controller with an H∞ control (H∞). The third scheme is an MRAC controller with a sliding mode controller (SMC). The fourth scheme is an MRAC controller with an ANN. The fifth scheme is an MRAC controller with a PID controller optimized by a GA. Finally, the last scheme is an MRAC classical control system. The objective of this research is to generate more powerful FTC methods and compare the performance of above schemes under different fault conditions in sensors and actuators. An industrial heat exchanger process was the test bed for these approaches. Simulation results showed that the use of Pattern Search Optimization and ANNs improved the performance of the FTC scheme because it makes the control system more robust against sensor and actuator faults.

Keywords: Fault–tolerant control, MRAC, ANN, PID, GA, H∞, SMC.

Se presenta la integración de métodos de Inteligencia Artificial, Control Robusto, Control No lineal y Control Adaptable por Modelo de Referencia (MRAC) en el Control Tolerante a Fallas (FTC). Se combinan diferentes esquemas de MRAC con controladores PID clásicos, redes neuronales, algoritmos genéticos, controladores H∞ y controladores por modo deslizante. Se proponen seis diferentes esquemas: el primer esquema es un MRAC con una red neuronal y un PID cuyos parámetros fueron optimizados con un algoritmo genético utilizando Optimización por Búsqueda de Patrones. El segundo esquema es un controlador MRAC con un controlador H∞. El tercer esquema es un controlador MRAC con un controlador por modo deslizante. El cuarto esquema es un controlador MRAC con una red neuronal. El quinto esquema es un controlador MRAC con un controlador PID optimizado por un algoritmo genético. Finalmente, el último esquema es un controlador clásico MRAC. El objetivo de esta investigación es generar métodos de FTC más poderosos y comparar su desempeño bajo diferentes condiciones de fallas tanto en sensores como en actuadores. Se utilizó un intercambiador de calor industrial para realizar dichos experimentos. Los resultados obtenidos de la simulación demostraron que el uso de Optimización por Búsqueda de Patrones con redes neuronales mejoró el desempeño del esquema FTC, ya que el sistema de control se volvió más robusto ante fallas en sensores y en actuadores.

DESCARGAR ARTÍCULO EN FORMATO PDF

References

[1] Fradkov A., Andrievsky B., & Peaucelle, D., Adaptive Control Design and Experiments for LAAS "Helicopter" Benchmark, European Journal of Control, Vol. 14, No. 4, 2008, pp. 329–339.         [ Links ]

[2] Stengel R., Intelligent failure–tolerant control, IEEE Control System Magazine, Vol. 11, No. 4, June, 1991, pp. 14–23.         [ Links ]

[3] Patton R., Lopez–Toribio C., & Uppal F., Artificial intelligence approaches to fault diagnosis, IEE Colloquium on Update on Developments in Intelligent Control, 1998, pp. 3/1–312, London, United Kingdom, October.         [ Links ]

[4] Schroder P., Chipperfield A., Fleming P., & Grum, N., Fault Tolerant Control of Active Magnetic Bearings, IEEE International Symposium on Industrial Electronics, 1998, pp. 573–578, Pretoria, South Africa, July.         [ Links ]

[5] Yu D., Chang T., & Yu D. W., Adaptive Neural Model–Based Fault Tolerant Control for Multi–Variable Processes, Engineering Applications of Artificial Intelligence, Vol. 18, No. 4, June, 2005, pp. 393–411.         [ Links ]

[6] Nieto J., Garza–Castañon L., Rabhi A., El Hajjaji A., & Morales–Menendez R., Vehicle Fault Detection and Diagnosis combining AANN and ANFIS, Seventh IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, 2009, pp. 1079–1084, Barcelona, Spain, June.         [ Links ]

[7] Polycarpou M., & Helmicki A., Automated fault detection and accommodation: A learning systems approach, IEEE Transactions on Systems, Vol. 25, No. 11, November, 1995, pp. 1447–1458.         [ Links ]

[8] Pashilkar A., Sundararajan N., & Saratchandran P., A Fault–tolerant Neural Aided Controller for Aircraft Auto–landing, Aerospace Science and Technology, Vol. 10, No. 1, January, 2006, pp. 49–61.         [ Links ].

[9] Perhinschi M., Napolitano M., Campa G., Fravolini M., & Seanor B., Integration of Sensor and Actuator Failure Detection, dentification, and Accommodation Schemes within Fault Tolerant Control Laws, Control and Intelligent Systems, Vol. 35, No. 4, 2007, pp. 309–318.         [ Links ]

[10] Patan K., & Korbicz J., Fault detection and accommodation by means of neural networks. Application to the boiler unit. Seventh IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, 2009, pp. 119–124, Barcelona, Spain, June.         [ Links ]

[11] Sugawara E., Fukushi M., & Horiguchi S., Fault Tolerant Multi–layer Neural Networks with GA Training, Eighteenth IEEE International Symposium on Defect and Fault Tolerance in VLSI systems, 2003, pp. 328–335, November.         [ Links ]

[12] Liang B., & Duan G., Robust H∞ fault–tolerant control for uncertain descriptor systems by dynamical compensators, Journal of Control Theory and Applications, Vol. 2, No. 3, August, 2004, pp. 288–292.         [ Links ]

[13] Yen G., & Ho L., Fault Tolerant Control: An Intelligent Sliding Mode Control Strategy, Proceeding of the American Control Conference, 2000, pp. 4204–4208, Chicago, Illinois, United States, June.         [ Links ]

[14] Nagrath J., Control Systems Engineering, 3rd Ed., Anshan Ltd, 2006, pp. 715.         [ Links ]

[15] Whitaker H., Yamron J., & Kezer A., Design of Model Reference Adaptive Control Systems for Aircraft, Report R–164, M. I. T. Press, 1958.         [ Links ]

[16] Nguyen H., Nadipuren P., Walker C., & Walker E., A First Course in Fuzzy and Neural Control, Chapman & Hall/CRC Press Company, 2002, pp. 165.         [ Links ]

[17] Ruan, D., Intelligent Hybrid Systems: Fuzzy Logic, Neural Networks, and Genetic Algorithms, Kluwer Academic Publishers, 1997, pp. 9.         [ Links ]

[18] Priddy K., and Keller P., Artificial neural networks, SPIE Press, 2005.         [ Links ]

[19] Goldberg D., Genetic algorithms in search, optimization, and machine learning, Addison–Wesley, 1989.         [ Links ]

[20] Khalil, H., Nonlinear Systems, 3rd Ed., Prentice Hall, 2002, pp. 552.         [ Links ]

[21] Zames G., Feedback and optimal sensitivity: model reference transformations, multiplicative seminorms, and approximate inverse, IEEE Transactions on Automatic Control, Vol. 26, No. 2, April, 1981, pp. 301–320.         [ Links ]

[22] Skogestad S., & Postlethwaite I., Multivariable Feedback Control. Analysis and Design, Wiley Ed., 2005, pp. 376.         [ Links ]

[23] McFarlane D., & Glover K., Robust Controller Design Using Normalized Coprime Factor Plant Descriptions, Springer Verlag, 1989.         [ Links ]

[24] Jia L., & Jingping J., The Model Reference Adaptive Control Based on the Genetic Algorithm, International Conference on Neural Networks,1997,pp. 783–787, Houston, Texas, United States, June.         [ Links ]

[25] Ahmed M., Neural Net based MRAC for a Class of Nonlinear Plants, Neural Networks, Vol. 13, No. 1, January, 2000, pp. 111–124.         [ Links ]

[26] Zhu O., Kim B., House B., & Kim, K. J., An Adaptive Controller for Wolsong NGS Bulk Liquid Zone Control of RRS, Nuclear Science Symposium, 1999, pp. 16991703, Seattle, Washington, United States, October.         [ Links ]

[27] Hongjie H., & Bo Z., A New MRAC Method based on Neural Network for High–Precision Servo System, IEEE Vehicle Power and Propulsion Conference, 2008, pp. 15, Harbin, China, September.         [ Links ]

[28] Lian K., Chiu C., & Liu P., Semi–Decentralized Adaptive Fuzzy Control for Cooperative Multirobot Systems with H∞ Motion/Internal Force Tracking Performance, IEEE Transaction on System, Man, and Cybernetics– Part B: Cybernetics, Vol. 32, No. 3, June, 2002, pp. 269–280.         [ Links ]

[29] Yu W., H∞ Tracking–based adaptive fuzzy–neural control for MIMO uncertain robotic systems with time delays, Fuzzy Sets and Systems, Vol. 146, 2004, pp. 375–401.         [ Links ]

[30] Miyasato Y., Model Reference Adaptive H∞ for Distributed Parameter Systems of Hyperbolic Type by Finite Dimensional Controllers – construction with unbounded observation operator, Forty–sixth IEEE Conference on Decision and Control, 2007, pp. 13381343, New Orleans, Louisiana, United States, December.         [ Links ]

[31] Vilela J., Costa R., & Hsu L., Cooperative Actuators for Fault Tolerant Model–Reference Sliding Mode Control, IEEE International Symposium on Industrial Electronics, 2003, pp. 690–695, June.         [ Links ]

[32] Haijun G., Tianping Z., & Qikun S., Decentralized model reference adaptive sliding mode control based on fuzzy model, Journal of Systems Engineering and Electronics, Vol. 17, No. 1, March, 2006, pp. 182–186.         [ Links ]