Permanent Magnet Synchronous Motor Control System Based On RBF Neural Network

With the continuous improvement of industrial automation,permanent magnet synchronous motor(PMSM)occupies an increasing proportion in the industrial control industry due to its advantages of light weight,high efficiency,reliable and stable operation.At present,most of the industrial control field is still using Proportional-Integral-Differential(PID)control,PID control structure is simple,high reliability of operation,so the majority of automation control experts and scholars favor.But in the face of complex industrial environment,and in the case of large load fluctuations and high requirements for motor speed accuracy,traditional PID control will not be able to meet the requirements of high precision control.In order to obtain better control performance of motor control system,the paper based on the vector control,combined with Radial Basis Function(RBF)neural network to control permanent magnet synchronous motor,and uses the self-learning ability of neural network to adjust PID parameters in real time with the actual running state of the motor.In this paper,the mathematical model of permanent magnet synchronous motor is firstly introduced.Then,according to the structure and learning algorithm of RBF neural network,the parameters of RBF neural network are modified by gradient descent method.Finally,after the training and learning of the RBF neural network,the Jacobian information of the controlled object is obtained,and the PID parameters are further adjusted to meet the high performance control requirements for the permanent magnet synchronous motor system.According to the above theory,this paper built the simulation model of each module in MATLAB/SIMULINK,and carried out the simulation test on the permanent magnet synchronous motor under the conditions of no-load start,sudden load and acceleration and deceleration.At the same time,TMS320F28335 as the main control chip,the paper uses C language in the CCS6.0 software development environment to write the algorithm,and builds the relevant hardware and software experimental platform.The algorithm of RBF neural network PID controller is verified by experiment,simulation and experimental results prove the feasibility and effectiveness of RBF neural network in the motor control system.