Dynamic database approach for fault tolerant control using dual heuristic programming

In this paper a hierarchical architecture that combines a high degree of reconfigurability and long-term memory is presented as a fault tolerant control algorithm for complex nonlinear systems. Dual heuristic programming (DHP) is used for adapting to faults as they happen for the first time in an effort to prevent the build up of a general failure and also as tuning device after switching to a known scenario. A dynamical database, initialized with as much information of the plant as available, oversees the DHP controller. The decisions of which environments to record, when to intervene and where to switch are autonomously made based on the specifically designed quality indexes. Results of the application of the complete algorithm to a proof-of-the-concept numerical example help to illustrate the fine interrelations between each of its subsystems.

[1]  George G. Lendaris,et al.  Training strategies for critic and action neural networks in dual heuristic programming method , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[2]  Kumpati S. Narendra,et al.  Adaptation and learning using multiple models, switching, and tuning , 1995 .

[3]  N. E. Wu,et al.  Concepts and methods in fault-tolerant control , 2001, Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).

[4]  Edgar N. Sánchez,et al.  Adaptive recurrent neural control for nonlinear system tracking , 2000, IEEE Trans. Syst. Man Cybern. Part B.

[5]  George G. Lendaris,et al.  More on training strategies for critic and action neural networks in dual heuristic programming method , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[6]  Yixin Diao,et al.  Fault tolerant system design using adaptive estimation and control , 2000 .

[7]  George G. Lendaris,et al.  Globally convergent approximate dynamic programming applied to an autolander , 2001, Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).

[8]  Paul J. Werbos,et al.  Backpropagation Through Time: What It Does and How to Do It , 1990, Proc. IEEE.

[9]  Kumpati S. Narendra,et al.  Adaptive control using multiple models , 1997, IEEE Trans. Autom. Control..

[10]  S. J. Qin,et al.  Detection and identification of faulty sensors with maximized sensitivity , 1999, Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251).

[11]  Mogens Blanke,et al.  FAULT MONITORING AND RE-CONFIGURABLE CONTROL FOR A SHIP PROPULSION PLANT , 1998 .

[12]  Roberto A. Santiago,et al.  Adaptive critic designs: A case study for neurocontrol , 1995, Neural Networks.