The Research Of The Thrust Ripple Suppression Of Pmlsm

With the wide application of linear servo system in modern industry,especially in some high control performance occasions,such as robots,CNC machine tools,the advantages of linear motors have been paid more and more attention by scholars at home and abroad.But,the thrust ripple caused by the end effect,the cogging effect,the ripple force and the friction force.It will affect the positioning accuracy and anti-jamming performance of the linear servo system.For reducing permanent magnet linear synchronous motor thrust ripple,improving the performance of the system,a PID neural network(Proportional-Integral-Derivative Neural Network PIDNN)controller is designed in this paper.At the same time the bacterial foraging optimization algorithm to optimize the PIDNN(BFO-PIDNN)controller is also presented.The simulation experiment results verify the effectiveness of PIDNN and BFO-PIDNN controller.Firstly,the structure and working principle of permanent magnet linear synchronous motor(PMSM)and the principle of vector control are introduced in this paper.The thrust fluctuation,especially the thrust ripple caused by end effect,ripple force,cogging force and friction force is analyzed,and the corresponding mathematical description is adapted to describe them.On this basis,the complete mathematical model of permanent magnet linear synchronous motor with the thrust fluctuation is achieved.Secondly,the network structure and the PIDNN algorithm is introduced.The PIDNN controller is designed for the PMLSM speed tracking system.PIDNN network contains proportion neuron,integral neuron and differential neuron.The network has good static performance and dynamic performance,and it can deal with dynamic information effectively.The PIDNN controller is compared with the traditional PID controller and BP controller.The simulation results show that the PIDNN controller has better performance.On this basis,the bacterial foraging optimization algorithm which has strong global search ability is proposed to training PIDNN.The BFO-PIDNN controller is designed to improve the ability of deal with complex system.At the same time,The BFO-PIDNN controller is compared with the GA-BP controller and the PIDNN controller.The results show that theBFO-PIDNN controller is more effective in suppressing the thrust ripple of permanent magnet linear synchronous motor(PMSM).Finally,the experiment platform based on NI CompactRIO is built,and the actual control effects of PIDNN and BFO-PIDNN are tested.From the experimental results,it can be seen that the two control algorithms of PIDNN and BFO-PIDNN can suppress the thrust fluctuation effectively,and also has good control effect.