Fault detection for sector-bounded non-linear systems with servo inputs and sensor stuck faults

Abstract

This paper addresses the fault detection (FD) problem for discrete-time sector-bounded non-linear systems with finite-frequency servo inputs. The non-linear systems are firstly modelled as multi-models. The main contributions are that a novel FD filter which combines the finite-frequency $H_{\infty }$ and $H_{-}$ indices is designed. Different from the existing methods, the proposed FD scheme guarantees that the generated residuals are robust against the servo inputs in fault-free case and sensitive to them in faulty cases. Thus, the small sensor stuck faults including outage faults can be detected. For this class of systems, the existing finite-frequency FD filter design methods are invalid. A new lemma is developed to characterise the system performances in finite-frequency domain. In addition, sufficient conditions for the existence of such a FD filter are derived by introducing slack variable and linear matrix inequalities techniques. Finally, an example is presented to demonstrate the method and its effectiveness.

Keywords:

• Fault detection
• servo inputs
• sensor stuck faults
• sector-bounded nonlinearity
• finite-frequency domain

Notes

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported in part by the Funds of the National Natural Science Foundation of China [grant number 61621004], [grant number 61420106016]; and the Research Fund of State Key Laboratory of Synthetical Automation for Process Industries [grant number 2013ZCX01].