PMSM Auto Disturbance Rejection Control System Based On Fuzzy Control

The permanent magnet synchronous motor is widely used for its advantages of small size, light weight, simple structure. However, the load disturbance has become a major obstacle to its further development. Therefore, this paper studies the auto disturbance rejection control technology, and proposes a new kind of auto disturbance rejection control system based on fuzzy control for permanent magnet synchronous motor(PMSM).First, the model of PMSM speed control system is established. T technology are analyzed. And the auto disturbance rejection controller model is established based on the principle and composition of the auto disturbance rejection control. Due to the auto disturbance rejection controller parameters being difficult for the actual operation and setting, the fuzzy control is introduced. Combined with the characteristics of two kinds of control methods, the system model, based on fuzzy control of auto disturbance rejection, is established.The other part is the identification of moment of inertia. First of all, the integral identification method, combined with the specific speed signal, is put forward. The moment of inertia identification expression, which contains no other independent variables, is extracted from the motor equations of motion. For extending the applying range of the algorithm and not affecting the normal operation, an identification algorithm based on discrete Landau recursive is put forward. By selecting the appropriate adaptive gain, the moment of inertia of the system is accurately identified.Next part is the load torque observation. Two kinds of load torque observers are introduced, which is the direct observation and the load state observer. Then we use MATLAB to simulate this two methods and as result the latter one is chosen.At last, we use MATLAB to simulate the proposed algorithm and establish the permanent magnet synchronous motor speed control system physical platform with DSP as the MPU, on which we test the identification of moment of inertia, the load torque observation and the system model.