Control Optimization of Linear Maglev Synchronous Motor Based on RBF Neural Network
The exciting linear motor of CNC machine tool feed control system is selected as the research object to construct the mathematical model according to the operation mechanism of the system.Error function and RBF neural network are set to realize the approximation control,the operating stability of the system is verified by the adaptive law,and the simulation analysis is carried out.Under no-load starting,RBF control reaches the fastest response rate and enters the designated suspension height after 0.17 s.Compared with PID control and SMC control,the adjustment efficiency increases by 42.2%and 24.1%respectively.Under sudden loading,the suspension air gap height under RBF control decreases by 5.0×10-5 m,and recovers to the original state after 0.065 s.Compared with the previous PID and SMC control,the dynamic landing is significantly reduced,and the recovery time is evidently shortened as well.Under end disturbance,RBF control forms a basically stable response,which helps to obtain a more stable air gap height,so that the control system can significantly resist the end effect.The test shows that the anti-interference performance of the system can be remarkablely improved by using the control strategy.
numerical control machine toolmotormagnetic levitation systemRBF neural network
万方数据
维普
