A Study On Sliding Mode Observer And Control Of Permanent Magnet Synchronous Motor

In traditional control system of permanent magnet synchronous motor (PMSM), it need takes the speed and position signals of the motor as feed-back signals, and the signals usually can be obtained by mechanical sensors on the motor, while the sensors will cause the problems of installation, volume change, interference effect and reliability decrease,etc. In addition,PMSM application is expected to have high performance in dynamic and static response. Now the study on new type of high performance control algorithms that based on the two basic control strategies, vector control and direct torque control, improves the control performance and quality, which of the PMSM system is extremely urgent to settle.For these reasons, this dissertation focuses on the usage of sliding mode variable structure control theory. The main results are listed in the following aspects.First of all, the basic working principle of PMSM is briefly introduced, and the mathematical models of PMSM which in different coordinate systems are analyzed. The basic system structure and main function module of vector control are also analyzed, the work provides a research foundation for the new sliding mode of the PMSM, whose variable structure speed position observer algorithm and the sliding mode variable structure control algorithm.Secondly, aiming at the problems of PM SM sliding mode observer on chattering, angular deviation and speed calculation disturbance, two kinds of new-type sliding mode observer have been proposed. One is extended Kalman filter sliding mode observer. Through combining Sigmoid function, extended Kalman filter and phase-locked loop technology, the system chattering has been weakened, and the compensation requirement and angular deviation of low pass filter have been eliminated. It is verified that Phase-locked loop used in extracting speed signal has successfully weakened disturbance caused by speed calculation.The second one is extended state adaptive sliding mode observer, whose principle is expanding the observation state from the original single current signal to current and back emf, and combining with model reference adaptive control theory to extract angle signal from counter emf. At the same time, the speed signal is extracted by adaptive rate algorithm, which has weakened the system chattering and omitted low pass filter, and has also solved the problems of angular deviation and speed disturbance.Moreover, nonsingular terminal sliding mode control is an algorithm developed on the basis of the terminal sliding mode control, which has solved the singularity problem of the latter one, however, the convergence rate of the system state is not suitable compared to terminal sliding mode control, also the system chattering problem exists. In order to improve the convergence rate of the system state, the pseudoindex reaching law has been used. And the extended observer has been brought in to observe system disturbance to weaken system chattering. Based on such theoretical analysis, the extended state observer nonsingular terminal sliding mode control algorithm has been designed using in permanent magnet synchronous motor speed loop control. The simulation result has verified the validity and correctness of the proposed algorithm.Finally, combined with multi-rate sampled control theory, the sliding mode predictive control is researched, then it is proposed that the input multiple sampling rate sliding mode predictive control algorithm of PMSM. This algorithm can solve problem that the dynamic response of the speed loop and current loop mismatch in the permanent magnet synchronous motor vector control system.