Experiment And Research In Low Speed Stability Of Permanent AC Servo System

Permanent magnet AC servo system is widely used in CNC (Computer Numerical Control) machine tools, industrial robots, astronomy turn-table, elevator, aeronautics&astronautics, and other modern control systems. As a driving devise, the permanent magnet servo motor plays a key role in the AC serve system. The problem of low speed quiet running property in AC system has been the central task in the servo system research for a long time. Low speed jitter is the main unsteady phenomenon of the system when it is running in low speed or super low speed. It is the phenomenon that speed jumps by pulse model when the serve system is running in a low speed which is lower than a critical value. One of the most possible reasons of this is phenomenon is that there is a variational friction torque disturbing in the servo system.Aiming to give a clarify view of the low speed steady problem in servo system, in this thesis, a number of experiments had been done on an experiment platform based on PMAC (programmable multiplied Axis controller)&IPC(Industrial Personal Computer) controlled system. By researching the phenomena, the article found out the relationship between the motor output torque, speed and the steady characters of the servo system. According gathering experiments data, the paper evaluated a minimum stable speed of the experiment platform. By turning the commend speed from high to low and gathering the data, we plotted a relational curve of the speed and torque, and gave a proof that it is the relationship between the friction torque and speed. And then the paper analyzed the reasons that led the servo system to be unsteady when running in low speed.After giving an introduction of the friction characters and presenting static and dynamic friction models, the paper chose stribeck friction model and coulomb+viscous+static friction model and combined them with the advantageous fuzzy PID control strategy to make the simulation. First, this article described how to make a model based on friction model and how to realize it in matlab simulink tool box. Second, the paper gave four simulations, separately simulated on traditional PID control strategy based on the two friction models and Fuzzy PID control strategy based on the same two friction models. There are there conclusions of the simulation results. First, fuzzy PID controller played a better performance than the traditional PID controller in the controlling of the system to be steady. Second, to compare with the coulomb+viscous+static friction model, the stribeck friction model can give a more precise description of the friction existing in the servo system. Third, fuzzy PID control strategy is more feasible using in the servo system disturbing by friction torque than the traditional PID control strategy. Because the traditional PID controller could not eliminate the distortion phenomenon occurred when the speed varying across zero arousing by friction torque, while the fuzzy PID controller could.