论文信息 - Diagnostic Decision Using Recurrent Neural Networks

Diagnostic Decision Using Recurrent Neural Networks

Abstract This contribution presents the results of an electrical motor on-line diagnosis. The implementation of a dynamic diagnostic decision is interesting in the case of closed-loop systems for which the residual responses may be transient. A neural network classifier is developed to diagnose the sensor and the actuator faults from some residuals. Simulation results are presented to show the detection and isolation ability. The classifier has been designed to cope with transient behaviours of the residuals generated by abrupt faults. The robustness with respect to unknown input (torque disturbance) and the sensitivity with respect to drift faults are experimented.

Sylviane Gentil | Christophe Combastel | Monica Alexandru

[1] Paul M. Frank,et al. Fault diagnosis in dynamic systems using analytical and knowledge-based redundancy: A survey and some new results , 1990, Autom..

[2] Ah Chung Tsoi,et al. Locally recurrent globally feedforward networks: a critical review of architectures , 1994, IEEE Trans. Neural Networks.

[3] Rolf Isermann,et al. Trends in the Application of Model Based Fault Detection and Diagnosis of Technical Processes , 1996 .

[4] Heikki N. Koivo,et al. Fault Diagnosis of Dynamic Systems Using Neural Networks , 1993 .

[5] Paul M. Frank,et al. Analytical and Qualitative Model-based Fault Diagnosis - A Survey and Some New Results , 1996, Eur. J. Control.

[6] Kumpati S. Narendra,et al. Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[7] Christophe Combastel,et al. A symbolic reasoning approach to fault detection and isolation applied to electrical machines , 1998, Proceedings of the 1998 IEEE International Conference on Control Applications (Cat. No.98CH36104).