Research On The Key Techniques Of High Performance Torque Control For Permanent Magnet Synchronous Motors

Permanent magnet synchronous motors (PMSM) have been widely used in the industrial field due to its simple construction, high torque/inertia ratio, high efficiency and low loss and the research of permanent magnet synchronous motors system with high performance deserves vast potential and important meaning. The high performance torque control strategies of PMSM were studied, then the methods in PMSM AC speed regulating system of reduce flux and torque ripples, keep switching frequency constant in inverters, realize sensorless control have been analyzed in details and verified through simulation and experiments. The main contents are as follows:Firstly, the stator flux plane is divided into six sectors in traditional direct torque control (DTC), the suitable voltage vector is selected according to the position of stator flux and the requirement of flux and toque. However, there are only six basic voltage vectors with fixed amplitude to be selected. Therefore, the control of flux and torque is not very exactly, the effect to flux is asymmetric, and results in large ripples. Moreover, flux and torque ripples will even larger and the inverter has inconstant switching frequency due to flux and torque digital hysteresis controllers.The stator voltage vectors to be selected are limited lead to large flux and torque ripples, so a DTC method based on twelve voltage vectors was proposed. Effect of stator voltage vectors on flux have been analyzed in detail. Only six basic voltage vectors could not meet the requirements of the system, and wrong voltage vectors maybe select in some cases. Voltage vectors increased method was proposed. Twelve voltage vectors, included six new resultant vectors and six original vectors, could be selected. In this method, the voltage vector selection is more detailed. And the effect of new twelve voltage vectors to stator flux is relatively flat which is good for accuracy control in bang-bang control system. Simulation results have shown that this method can control flux and torque more accurately and can decrease flux and torque ripples effectively.A DTC method based on eighteen sectors is proposed so as to solve the problem of asymmetrical effect of voltage vectors on flux in traditional DTC. Flux variable quantity cared about sampling period, voltage vector amplitude in a sampling period and the included angle between voltage vector and flux. Flux variable quantity asymmetry in the sectors borderline. So the six original sectors are subdivided into eighteen sectors to figure it out. Simulation results showed that this method has reduced flux and torque ripples, and improved the flux trajectories.Space voltage vector modulation technology is introduced to the traditional DTC for the sake of keeping inverter switching frequency constant and reducing flux and torque ripples. Inverter has constant switching frequency due to space voltage vector module instead of hysteresis controllers and voltage vector switching table. Suitable voltage vectors will be calculated on the basis of flux and torque error. Simulation results showed that this method has reduced flux and torque ripples significantly.Secondly, stator flux observer based on Extended Kalman filter (EKF) is presented in order to overcome the disadvantages of error accumulation and saturation in traditional DTC. The EKF observer is modeled with the input state of PMSM current and voltage and output of speed and rotor position on the basis of two-phase static coordinate. The observer combined with PMSM current model to estimate stator flux and motor speed in real time. The proposed system has less flux and torque ripples and stronger robustness because EKF algorithm can deal with system interference and measurement noise very well. Simulation and experimental results have shown that the flux and torque ripples were decreased effectively, the flux and speed can be estimated accurately and sensorless control system is achieved.Thirdly, stator flux estimator based on particle filter (PF) algrithm is designed. Particle filter algorithm is an effective tool for non-linear and non-gaussian estimation problems, and has become focus. The application of particle filter technique to the stator estimation is explored. The particle filter algorithm is analyzed in detail. The PMSM stator flux estimation model based on particle filter is designed. Simulation results have verified this method.Fourthly, speed feedback control system without sensor on the basis of reduced order model reference adaptive system (MRAS) is designed. Model reference adaptive technique and popov hyperstability theory is given and analyzed. The model reference adaptive system is modeled with the reference model of voltage equation and adjustable model of PMSM reduced order state equation. The adjustable model approach to the reference model gradually, and speed and rotor position information will be get at last. Simulation and experimental results have shown that this method has high estimation accuracy, strong anti-interference ability, and fine system performance.