Dissipativity-based fault estimation for switched non-linear systems with process and sensor faults

摘要

This study investigates the fault estimation problem for switched non-linear systems, which have process fault and sensor fault. The Takagi-Sugeno fuzzy models are applied to construct the non-linear dynamics in switched non-linear systems. A novel fault estimation observer, which has an adjustable dimension, is designed to estimate the system state, process fault, and sensor fault, simultaneously. In the observer design process, only the input and output are used, and the differential of the output is not needed. Under an arbitrary switching signal, sufficient conditions can be achieved to ensure that the estimation error dynamic is strict (Q, S, R)-gamma-dissipative, and the poles of the error system located in a given region. Finally, the simulation examples are provided to verify the proposed method.