An Enhanced Active Disturbance Rejection Control of BPMSM Based on Neural Network Parameters Dynamic Adjustment Method
Dr. Xin Wang
School of Electrical and Information Engineering
Progress In Electromagnetics Research C, Vol. 131, 159-169, 2023
An enhanced linear active disturbance rejection control (E-LADRC) method with dynamically adjust is proposed to improve the observer gain and observation effect in the convenient linear active disturbance rejection control (C-LADRC), reduce the sensitivity of the observer to interference, and find the appropriate observer gain coefficient. Firstly, the mathematical model of bearingless permanent magnet synchronous motor (BPMSM) and the C-LADRC algorithm are described and analyzed. Secondly, the E-LADRC algorithm is designed to overcome the shortcomings of the C-LADRC. Thirdly, the back propagation neural network (BPNN) algorithm with strong self-learning and adaptive ability is used to dynamically adjust the parameters of the E-LADRC, so as to improve the performance of the control system. Finally, the whole control system is analyzed, and the effectiveness of the proposed algorithm is verified on the experimental platform. The experimental results show that the proposed control algorithm can effectively reduce the jitter amplitude of speed and displacement.