Research On Key Technology Of High-Performance Speed Control Of AC Servo System

With the widely application of permanent magnet synchronous motors (PMSM) in these fields of CNC machine tools, industrial robots, aerospace and other fields, people have higher requirement for speed control accuracy of PMSM servo system. However, as the limitations of domestic servo control technology, high-end servo systems are still monopolized by foreign servo manufacturers. Under this background, this dissertation proposed a state estimator based on extended Kalman filter for high-precision speed detection. With the technologies of parameters identification and self-tuning based on the model reference adaptive system, a high-performance speed control system was designed to achieve high-performance speed control of PMSM servo system.Servo drive system usually uses closed-loop control structure, which includes the current loop, speed loop and position loop, to obtain the good dynamic following and load-disturbance resistance performance. The speed loop is the basis of the servo system. It is used to achieve precise positioning as the inner loop of position loop and provide current reference for current loop. So the speed control is an extremely important aspect of the AC servo control system, and its control performance is an important component of the overall performance of the servo system. The speed control of AC servo system usually uses closed-loop control structure. The accuracy of speed feedback plays an important role for the control performance of servo system. The thesis, combining with the research on the parameters self-tuning by the other members of the team, focuses on studying the high-precision speed detection technology and designing a high-performance speed control system of AC servo system.Firstly, the study analyzed the mathematical models of PMSM and the rotor oriented vector control strategy. Based on the analysis of the mathematical models of PMSM, three-loop control model was established. Then, an expanded M/T speed detection method was designed based on analyzing some commonly-used speed detection method to eliminate the error from non-complete pulse period of the conventional M/T speed detection method. Besides, in order to make up the speed detection failure with the expanded M/T speed detection at a lower speed, the state estimator based on Kalman filter, which is suitable for a wide speed and high noise environments was built. Because the state estimator based on Kalman filter can’t do state estimation with the nonlinear model, a high-precision speed detection algorithm based on extended Kalman filter was applied in this dissertation. Experimental results show that the algorithm has a higher speed detection accuracy and faster dynamic response capabilities, and it has great significance for improving the speed detection accuracy and the control performance of servo system.With the high-precision speed detection algorithm based on the extended Kalman filter and the technologies of parameter self-tuning, a safe, reliable and stable AC servo driver was designed. In order to verify the control accuracy and dynamic response of the proposed algorithm and the overall performance of the servo drive system, simulation analysis and experiments were carried out and the simulation and experimental results show the high performance speed control method is effective and practical.