Research On Disturbance Rejection Based Control Methods For Permanent Magnet Synchronous Motorspeed-Regulation System

With the development of power electronics technology, micro-electronic technology, micro-processor technology and modern motor control theory, permanent magnet synchronous motor(PMSM) has been widely applied in AC servo drive system, especially in high precision control field, such as robotics, aerospace, machine tools and industrial automation. In the practical application, there are a variety of disturbances caused by different factors. These disturbances are mainly in two forms:periodic disturbances, slowly varying non-periodic disturbances. These two types of disturbances affect the PMSM system’s static and dynamic performances. In order to improve system’s performance, the research on suppressing these two types of disturbances is carried out in this paper.Firstly, this paper introduces the development of the AC speed regulation system, the mathematical model of PMSM, control strategies and control methods of PMSM, and research status of disturbance rejection for PMSM speed-regulation system. Secondly, a detailed analysis of the mechanism of torque ripples which make PMSM’s steady-state velocity ripples occur. And the equivalent disturbances of these mechanisms are shown in the PMSM vector control system. Thirdly, according to the roles of the equivalent disturbances, composite controller of speed loop and current loop are designed to eliminate steady-state ripples. The simulation and experimental results show that suppressing the steady-state ripples in both speed loop and current loop, which depend on the mechanism of PMSM’s torque ripples, has a better performance than those methods of suppressing the steady-state ripples that only in speed loop. Finally, in order to suppress the periodic disturbances and slowly varying non-periodic disturbances at the same time, a new observer based on internal model principle(IMP) and extended state observer(ESO) is designed, which can estimate these two types of disturbances respectively. And a composite controller based on the estimated disturbance values is also designed. The simulation and experimental results show that the observer can quickly estimate each periodic disturbances and slowly varying non-periodic disturbances respectively. And the composite controller can also effectively suppress these two types of disturbances at the same time, the performance of dynamic and steady state also become better.