Active Disturbance Rejection Controller Based on Neural Network in the Permanent Magnetic Synchronous Motor Servo System

Abstract:

Permanent magnetic synchronous motor (PMSM) is a strong coupling and non-linear system. In the speed-loop of PMSM servo system controlled by active disturbance rejection controller (ADRC), it is difficult to ensure the estimation precision of disturbance by extended state observer (ESO) in the case of big disturbance. In the paper, the ADRC based on the artificial neural network (ANN) is applied to the speed-loop of PMSM servo system. The result of simulation shows that this algorithm has better anti-load-disturbance performance.

• Periodical: Applied Mechanics and Materials (Volumes 63-64)
• Pages: 874-877
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.63-64.874
• Online since: June 2011
• Authors: Song Wang, Xiao Na Wu
• Keywords: Active Disturbance Rejection Controller, Neural Network (NN), Non-Linear Function, Permanent Magnetic Synchronous Motor

References

[1] Han Jingqing. Active disturbance rejection control technique - the technique for estimating and compensating the uncertainties, [M]. Beijing: Defense industry press. (2008).

[2] Huang Yi, Han Jingqing. Analysis and design of nonlinear continuous second-order extended state observer, [J]. Chinese Science Bulletin, 2000, 45(13) , pp.1373-1379(in Chinese).

DOI: 10.1007/bf02909682

[3] Zhang Rong, Han Jingqing. Active disturbance rejection controller based on neural network, [J]. Journal of System Simulation，2000，12(2) , pp.149-151(in Chinese).

[4] Wang Tao, Yang Xiaoping, Yu Xiangyang, et al. ADRC of hydraulic turbine governing system based on neural network, [J]. Journal of Hydroelectric Engineering, 2006, 25(3), pp.125-129(in Chinese).