Dynamic neural network-based estimator for fault diagnosis in reaction wheel actuator of satellite attitude control system

This paper presents an approach to simultaneous fault detection and isolation in the reaction wheel actuator of the satellite attitude control system. A model-based adaptive nonlinear parameter estimation technique is used based on a highly accurate reaction wheel dynamical model while each parameter is an indication of a specific type of fault in the system. The estimation is based on the nonlinear finite-memory filtering strategy that is solved for optimal estimation functions. To make the optimization feasible for on-line application, the optimal estimation functions are approximated by MLP neural networks thus reducing the functional optimization problem to a nonlinear programming problem, namely, the optimization of the neural weights. The well-known standard back-propagation algorithm and backpropagation through-time algorithm were employed inside the neural adaptation algorithms to obtain the required gradients. Simulation results show the effectiveness of the methodology for the proposed application.

[1]  Thomas Parisini,et al.  Neural approximators for nonlinear finite-memory state estimation , 1997 .

[2]  T. Brotherton,et al.  Automated rule extraction for engine vibration analysis , 1999, 1999 IEEE Aerospace Conference. Proceedings (Cat. No.99TH8403).

[3]  Rolf Isermann,et al.  Fault diagnosis of machines via parameter estimation and knowledge processing - Tutorial paper , 1991, Autom..

[4]  John L White.  ACTIVE CONTROL STUDIES OF STRUCTURALLY TRANSMITTED ENGINE VIBRATION ON COMMERCIAL AIRPLANES , 2001 .

[5]  Angelo Alessandri.  Fault diagnosis for nonlinear systems using a bank of neural estimators , 2003, Comput. Ind..

[6]  Asoke K. Nandi,et al.  Real-time classification of rotating shaft loading conditions using artificial neural networks , 1997, IEEE Trans. Neural Networks.

[7]  J. L. White,et al.  EXPERIENCE IN ROTOR BALANCING OF LARGE COMMERCIAL JET ENGINES , 2002 .

[8]  J. L. White,et al.  A study of engine excited non-linear vibro-acoustics in commercial airplanes , 1997 .

[9]  Amrane Houacine.  Regularized fast recursive least squares algorithms for finite memory filtering , 1992, IEEE Trans. Signal Process..

[10]  John S. Mitchell,et al.  An introduction to machinery analysis and monitoring , 1981 .

[11]  M. Sanguineti,et al.  Approximating Networks and Extended Ritz Method for the Solution of Functional Optimization Problems , 2002 .

[12]  A. Jazwinski.  Limited memory optimal filtering , 1968 .

[13]  Donald C. Wunsch,et al.  Vibration analysis via neural network inverse models to determine aircraft engine unbalance condition , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..

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

[15]  Thomas G. Habetler,et al.  An unsupervised, on-line system for induction motor fault detection using stator current monitoring , 1994, Proceedings of 1994 IEEE Industry Applications Society Annual Meeting.

[16]  T. I. Liu,et al.  Intelligent monitoring of ball bearing conditions , 1992 .

[17]  John T. Renwick,et al.  Vibration Analysis---A Proven Technique as a Predictive Maintenance Tool , 1985, IEEE Transactions on Industry Applications.

[18]  R. Froelich,et al.  Reaction wheel attitude control for space vehicles , 1959 .

[19]  Alexander Medvedev.  State estimation and fault detection by a bank of continuous finite-memory filters , 1998 .

[20]  R. J. Patton,et al.  Artificial intelligence approaches to fault diagnosis , 1998 .