Remaining useful life prediction of reaction wheel motor in satellites

Abstract

For advanced attitude control of the satellite, reaction wheels are used, which are actuated by motor. Fault detection and prediction of Remaining Useful Life (RUL) of the motor is of great importance. This study aims to demonstrate health management of reaction wheel motor in satellites by estimating the RUL of the damping coefficient. Multi-scale Extended Kalman Filter (EKF) is employed for this purpose using the data of input current and output velocity measured during the life test of the motor. The motor dynamic behavior is modeled by the ordinary differential equations. The damping coefficient is taken as the health indicator, which is obtained by solving the micro EKF problem. To set threshold for the damping coefficient, degradation behavior of characteristic curve that evaluates the motor performance is used to correlate with the degradation of the damping coefficient. Then, the RUL is predicted by solving the macro EKF problem.