Fault estimation based on observer for fractional-order Lu system with bifurcation problem

Abstract

Control systems have been susceptible to faults in recent years due to the complexities of their equipment and long-term function. These faults can reduce system performance, impose losses and casualties, and/or reduce final product quality. In order to detect such defects in a control system, fault estimation (FE) has been adopted in several studies. In this paper Estimating the sensor-based fault of a fractional-order Lu system with disturbances such as noise was investigated. This fault estimation problem is implemented using a bifurcation problem in the fractional-order Lu system. This study tries to synthesize the perception of bifurcation problem and fault estimation. Noise disturbance in nonlinear systems makes fault diagnosis a complex problem that should be managed over Takagi-Sugeno (TS) approximated fuzzy models. This approximation method can predict the faults and states of system. To model the observer, the TS fuzzy model was incorporated with a proportional plus integral. The results proved that the observer has a desirable performance to measure states and fault of sensor. Based on the estimations, the actual value of the parameters gives us the deviation value, which is equal to the low amount of deviation here, due to the bifurcation that happens in the system.