Adaptive fault detection for a class of nonlinear systems based on output estimator design
IFAC Proceedings Volumes (IFAC-PapersOnline)
1 PART 1
Adaptive system and control, Fault detection and diagnosis, Nonlinear adaptive control
This paper considers output estimator based fault detection problem for a class of nonlinear systems with unknown system parameters. Because observer design for such systems is extremely difficult if not impossible, output estimator design is used for the purpose of fault detection. In order to achieve output estimator design using adaptive approaches, a multi-input multi-output (MIMO) nonlinear system is first decomposed into a group of multi-input single-output (MISO) nonlinear systems. For each MISO nonlinear system, an output equation is derived through filtering the output, the inputs, and those nonlinear functions of the outputs, which depends linearly on all unknown system parameters. Based on the output equation and using adaptive approaches, an adaptive output estimator is designed for the corresponding output. By defining residuals using the adaptive output estimation errors resulting from the output estimators, a fault detection scheme is proposed. The efficacy of the proposed fault detection scheme is tested on a single-link flexible robot manipulator model thorough computer simulations. Copyright © 2007 International Federation of Automatic Control All Rights Reserved.
Chen, Wei Tian and Saif, Mehrdad. (2008). Adaptive fault detection for a class of nonlinear systems based on output estimator design. IFAC Proceedings Volumes (IFAC-PapersOnline), 17 (1 PART 1).