Robust unknown input observer design for state estimation and fault detection using linear parameter varying model

13th European Workshop on Advanced Control and Diagnosis

Li, Shanzhi^1,3 ; Wang, Haoping^1 ; Aitouche, Abdel^2 ; Tian, Yang^1 ; Christov, Nicolai^3

Automation School, Nanjing University of Technology and Science, Nanjing

210094, China^1

CRIStAL UMR CNRS 9189, HEI-Lille, Lille

59014, France^2

CRIStAL UMR CNRS 9189, University of Lille 1, Lille

59650, France^3

关键词: Actuator fault;    Linear matrix equality;    Linear Matrix Inequalities (LMIs);    Linear parameter varying models;    Original state;    Physical systems;    Unknown input observer;    Wind turbine systems;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/783/1/012001/pdf
DOI  :  10.1088/1742-6596/783/1/012001

来源: IOP

【 摘 要 】

This paper proposes a robust unknown input observer for state estimation and fault detection using linear parameter varying model. Since the disturbance and actuator fault is mixed together in the physical system, it is difficult to isolate the fault from the disturbance. Using the state transforation, the estimation of the original state becomes to associate with the transform state. By solving the linear matrix inequalities (LMIs)and linear matrix equalities (LMEs), the parameters of the UIO can be obtained. The convergence of the UIO is also analysed by the Layapunov theory. Finally, a wind turbine system with disturbance and actuator fault is tested for the proposed method. From the simulations, it demonstrates the effectiveness and performances of the proposed method.