Adaptive sliding mode–based diagnosis of actuator faults for LPV systems

Abstract

In this paper, an adaptive sliding mode observer is developed for actuator fault diagnosis of linear parameter-varying systems. The main advantage of the proposed approach is its ability to cope with time-varying distribution matrix in linear parameter-varying systems. Furthermore, the proposed adaptive observer is characterized by its output robustness against parameter uncertainties and disturbances without any a priori knowledge about their bounds. The efficiency of the proposed fault diagnosis approach is validated using simulation studies.

DC-Link Voltage and Catenary Current Sensors Fault Reconstruction for Railway Traction Drives

“…Despite different methods presented in the literature for SMO design, as linear matrix inequality (LMI) [ 42 , 43 ], or adaptive SMO [ 44 ], which can avoid the overestimation of discontinuous gain and reduce chattering in the estimated variables, this article proposes a simple solution to design the observer, for railway application requirements. Low computational requirements and easiness to tune, in the case of input filter parameter variation and maximum fault magnitude to be detected, are the main factors for the observer design selection.…”

Adaptive parallel fractional sliding mode control

“…In Reference 17, an adaptive sliding mode control for a class fuzzy system is proposed. And, an adaptive sliding mode scheme for faults diagnosis is proposed in Reference 18.…”