Speed Loop Modeling And Pi Parameter Tuning Of Servo Systems Based On DSA

PMSM servo system has been used widely in CNC machine tools, robotics, laser processing, radar and other fields for its excellent performance. Modeling of the PMSM servo system laid the theoretical basis of parameter tuning of the servo system. Since AC servo system is a multi-variable, strong coupling nonlinear system, theoretical modeling process often overlooked the influence of certain parameters or simplified system architecture, leading to that the established model was not precise enough, thus not suitable for direct application in practical engineering. On the other hand, whether PID controller parameters settings reasonable or not directly affect the static and dynamic performances of the entire servo system. While the widespread used traditional PID parameter tuning methods have poor adaptability with tedious tuning process. Therefore, to solve the above two problems thus improving the performance servo control system, the main studies in this paper are as follows:Firstly, the mathematical model of PMSM servo system and its control strategies were presented. After obtaining the speed loop and current loop mathematical model of the servo drive through theoretical model, used Sisotool to design the speed loop PID controller and its parameter tuning was proposed.Secondly, due to difficultly in establishing accurate mathematical model of Servo system with theoretical modeling, proposed using Agilent35670A dynamic signal analyzer to obtain precise servo drive transfer function. The principle measuring precise speed loop servo drive transfer function with the dynamic signal analyzer were discussed in detail. Finally, a dynamic characteristics test platform for PMSM servo system was built for controlled object model identification experiments and PID parameter tuning experiments. Experimental results illustrated that:(1) Agilent35670A dynamic signal analyzer capable of measuring the actual model of the speed loop of the servo system, and the measured model was more accurate than the theoretical model, and more suitable for practical application.(2) Better speed controller PID parameters could be obtained base on the measured controlled object model while using Sisotool controller design and tuning methods.