Research On Robust Control For Permanent Magnetic Synchronous Motor Systems Via Convex Optimization Algorithm

With high power density, high efficiency, small volume, low power consumption, high reliability and simple structure, the permanent magnetic synchronous motors (PMSM) have received extensive attention and are widely applied into national defense, aviation, industrial production and our daily life. However, due to the multivariable, nonlinear and strong coupling characteristics of PMSM, some disadvantageous factors including current coupling, parameter variations and external disturbance still exist in PMSM, which will seriously influence the control performance of PMSM. As a result, the high performance control requirement for PMSM is difficult to be realized. In order to meet the demands in engineering applications, a lot of breakthrough in the field of PMSM control needs to be made to ensure the rapid development of PMSM. The main research work and contributions of this thesis are listed as follows:(1) The research background and theoretical foundation are provided. Some related control strategies and theoretical basis for PMSM control system are also introduced in this thesis. Among these strategies, AC control technology, sliding mode control, robust control, PI/PID control, backstepping control, disturbance observer control and optimization algorithm are highlighted and their latest research status and development prospects in motor control system are also discussed.(2) The basic mathematical model and the position control problem for PMSM system are explored. In order to achieve an approximate decoupling motor model, the corresponding position control model of PMSM is established based on the assumption that is id=0.On this basis, a generalized PI position controller based on convex optimization algorithm is designed such that the closed-loop PMSM systems are asymptotically stable and the good position tracking performance can also be satisfied. To achieve the satisfactory robustness, the L1 performance optimization index is considered in the optimization algorithm. Moreover, in order to overcome the problems of sensor installation costs and space defects, this thesis proposes a reduced-order Luenberger observer and then sensorless position regulation for PMSM can be achieved.(3) The speed control problem of PMSM system with uncertainty term is studied in this chapter. By combining linear matrix inequality (LMI) optimization algorithm with L1 performance optimization index, a generalized PI feedback controller is designed for the PMSM model with parameter perturbation and external disturbance. As a result, the multi-objective control requirements including stability, speed tracking performance and robustness can be satisfied simultaneously. Moreover, by using power electronic toolbox, a speed control simulation platform of PMSM is built based on space vector pulse width modulation (SVPWM) algorithm, which further verifies the effectiveness of the control algorithm.(4) The speed tracking control and anti-disturbance problem of PMSM model are discussed based on disturbance observer design method in this thesis. In order to solve the mismatch problem between the control input and the load disturbance, this chapter constructs a cascade speed control model with multi-source disturbances by a series of variable conversions. The disturbances in cascaded model are formulated into two types, where the first type is described by an exogenous system, and the other type is supposed to satisfy H2-norm bounded condition. Thus, a hierarchical composite anti-disturbance control strategy is introduced for the cascade speed control model. By combining the nonlinear disturbance observer with L1 optimization index, the first type of disturbances can be estimated and rejected and the second type of disturbances can also be attenuated simultaneously. Moreover, based on improved convex optimization algorithm, the feedback control input consisting of speed controller and disturbance observation is designed such that multi-objective control requirements including stability, robustness, dynamical tracking and anti-disturbance performance for cascaded speed regulation system can be achieved.(5) Based on Matlab software, linear matrix inequality (LMI) toolbox, Simulink real-time simulation platform and power electronic toolbox (SimPowerSystem), corresponding simulation examples in each chapter are given to demonstrate the effectiveness of the proposed control algorithms.