Sensor fault detection based on fuzzy singularly perturbed model

Abstract

Fuzzy singularly perturbed modelling and sensor fault detection for nonlinear singularly perturbed systems (NSPSs) are investigated. A continuous-time fuzzy singularly perturbed model with superimposed faults on state variables and external disturbances are established to describe NSPSs with sensor faults. Based on the model, a novelty $H_{\mathrm{\infty }}$ fuzzy fault detector (FFD) is designed, and the sufficient conditions for $H_{\mathrm{\infty }}$ performance are derived. The above approaches are applied to a CE150 helicopter with abrupt and intermittent faults on sensors. The innovations of this paper are as follows: (i) The nonlinear and slow–fast time-scales are described by a united continuous-time fuzzy singularly perturbed model, and based-it a FFD is designed, which can avoid the incomplete decoupling problem caused by decomposition methods; (ii) The sensor faults are expressed by the superimposed faults on state variables when the continuous-time fuzzy singularly perturbed model is constructed, which is less conservative than existing methods; (iii) In the FFD design, the first-order derivative model of the estimate function for sensor faults is established, and it is combined with a fuzzy observer, which can improve the detection accuracy of the FFD; (iv) $H_{\mathrm{\infty }}$ control technology is introduced into the FFD design process to suppress external disturbances.

Keywords:

• Fuzzy singularly perturbed model
• sensor fault detector
• nonlinear singularly perturbed systems
• LMIs
• $H_{\mathrm{\infty }}$

Acknowledgments

Thank Area Editor and anonymous reviewers for their constructive comments and suggestions.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the articles (Chen, 2014; Jun, 2007; Qiao & Yang, 2018).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Chunxiao He

Chunxiao He received his B.S. degree from China University of Geosciences, Beijing, P.R. China, in 2004. and M.S. degree from University of Science and Technology Beijing, Beijing, P.R. China, in 2009. Now he is a Ph.D. candidate in University of Science and Technology Beijing. His main research interest is fault-tolerant control for complex systems?

Jinxiang Chen

Jinxiang Chen received the Ph.D. degree from the Department of Automation, University of Science and Technology Beijing, Beijing, China in 2009. She was a postdoctoral fellow in Tsinghua University, China in 2009–2011. Now she is a professor in China Iron & Steel Research Institute Group, Beijing, China. Her research interests include fuzzy control for nonlinear multiple time-scales systems, big data analysis and prediction, shape and gauge integrated control for hot ultra-thin strip mills.

Xisheng Li

Xisheng Li received his B.S. , M.S. and Ph.D. degrees from University of Science and Technology Beijing, Beijing, P.R. China, in 1991,1994 and 2000, respectively. Now he is a professor in University of Science and Technology Beijing. His main research interests include sensor signal processing, weak signal detection and fault-tolerant control for complex systems.