Estimation of Sensor Bias Fault in Adaptive Control of Power System

Based on the simulation results, steady-state tracking faults are improved. Control of indeterminate systems, despite the actuator and sensor bias, has been and remains a major challenge. Sensor fault can cause process fault. Among the cases where sensor bias is common, air velocity measurements and gyroscope rates can be mentioned. Although considerable research efforts have previously focused on adapting the fault, the bias correction of the sensor appears to be relatively limited. However, the cause of several crashes was the sensor fault (due to radio altimeter fault, angle of attack sensor fault, airspeed speed sensor fault). Also, finding a way to fix the sensor bias problem is of the utmost importance. The direct model reference adaptive control (MRAC) method is used to control uncertain systems using controllers that are adapted to achieve a performance close to a reference model. However, these controllers maintain system stability and provide close tracking of the reference model response. In this paper, we intend to address the problem of unknown fault bias matching by adjusting the direct reference model adaptive control for state-feedback for state-tracking (SFST). Also, to obtain an asymptotic stable bias fault estimator, we use the Kalman filter to estimate the bias sensor fault. Based on the simulation results, steady-state tracking faults are improved.

Keywords:

Model Reference Adaptive Control , Sensor Bias , Kalman Filter , fault estimation