Research on thrust fluctuation compensation for permanent magnet linear motor based on extreme learning machine
Yanwei Huang · Fuzhou daxue xuebao. Ziran kexue ban · 2012
In view of non-linear character of the permanent magnet linear motor thrust ripple,a PID+ELM feed-forward compensation of permanent magnet linear motor position control system is designed in this paper.The sample for training ELM network is acquired through the analysis of the ripple characteristics and its law.The trained network is used as a feed-forward compensation link to compensate for the thrust fluctuations in the control system.The results of simulation demonstrate that the proposed control system is practicable and effective.Although the tracking accuracy of the proposed control scheme is slightly lower than that of the SVM and RLS methods,it takes much shorter training time and satisfies strict real-time requirements.