Neural Network Fuzzy Adaptive Control Of Permanent Magnet Synchronous Motor

With the advantages of simple structure, high reliability and easy maintenance, the permanent magnet synchronous motor(PMSM), including the surface permanent magnet synchronous motor(SPMSM) and the interior permanent magnet synchronous motor(IPMSM), has become the main power equipment in the modern AC motor control system. In order to improve the dynamic and stable performance of the system, the neural network control, fuzzy control and adaptive control are adopted to solve the control problems of the traditional control methods in this paper.First of all, the purpose and significance of this research are introduced in this part. Then, the research about the PMSM servo system at the domestic and foreign are introduced. Mover, the research about the four quadrant PMSM speed servo system at the domestic and foreign are described.Secondly, the mathematical models of PMSM are introduced in detail. The mathematical models of four quadrant PMSM speed servo systems are described in this part, including the topology of four quadrant voltage type converter, the mathematical model of grid-side converter and the mathematical model of motor-side converter.Thirdly, the PMSM position servo system is the main research in the third part. Aiming at the unsatisfactory control PMSM position servo system which is caused by fixed parameters of proportional integral derivative(PID) controller, a position controller combined with fuzzy PI control and Radial basis function(RBF) neural network PID control based on the smooth switching is proposed. The simulation results verify that the PMSM position servo system adopted fuzzy RBF neural network controller has good dynamic performance and high precise positioning under the occurrence of external disturbance. Further, the system has better anti-interference to load disturbance variation. To overcome the low steady precision of fuzzy control, a new position controller combining the improved single neuron PID control and fuzzy PI control based on the trapezoidal membership function of fuzzy switching is designed. The simulation results show that the PMSM position servo system combining fuzzy and improved single neuron control based on fuzzy switching has high steady precision, strong anti-interference ability under the occurrence of external disturbance. Aiming at the parameters of the shakeless fuzzy controller, the single neuron control is introduced to adjust the output scaling factor online. Then, in order to verify the performance of the proposed algorithm controller, the IPMSM position servo system using the improved shakeless fuzzy controller is simulated. The simulations show that the proposed algorithm controller has faster position response and better anti-interference ability.Fourthly,the PMSM speed sensorless position servo system is the main research in the fourth part. The system controllers are designed based on RBF neural network and backstepping. The speed observer is designed according to the model reference adaptive system(MRAS). The simulation model is built and the simulation results are analyzed. The results show that the method has good control performance.Fifthly,the PMSM speed servo system is the main research in the fifth part. Aiming at the control problems due to the traditional control strategies adopted in the four quadrant PMSM speed drive system, the voltage loop controller in the grid-side is designed based on the single direct MRAS control strategy, and the speed controllers are designed according to the backstepping. Then, the high control performance is achieved with the designed controllers in the four quadrant PMSM speed drive system.At last, the research content of this paper is summarized.