A Friction Disturbance Compensation Method for Electromechanical Actuator Based on Fractional Order Adaptive Neural Network

Friction torque disturbance affects the tracking performance of electromechanical actuator servo system, bringing position and speed tracking errors, and even may leading to instability of the servo system. Aiming at the problem of poor tracking performance of electromechanical actuator servo system under friction torque disturbance, a FOANN friction compensation algorithm is proposed to estimate and compensate the friction torque. Firstly, base on LuGre friction model, a electromechanical actuator model is established, and the unmeasured state variable in the LuGre model is estimated by radial basis function neural network. Secondly, a FOANN controller is designed, and the stability of corresponding closed-loop system is proved by Lyapunov stability theory. Finally, through simulation and experimental platform, the dynamic performance of FOANN is compared with those of traditional PD and MRAC. The simulation and experimental results show that, with the proposed FOANN friction torque compensation algorithm, the tracking errors of both position and velocity of electromechanical actuator servo system are greatly reduced. FOANN algorithm can effectively estimate and compensate friction torque, reduce the impact of friction disturbance and enhance the dynamic performance of the servo system.