Research Of Control Algorithm Of Magnetic Levitation Ball Control System

Nowadays the whole field of science and technology is a big step forward. The magnetic levitation technology is also followed the pace of the development science and technology. Magnetic levitation technology is the key in the application area such as magnetic bearing, maglev train, high-speed magnetic levitation motor. The magnetic levitation technology as the core of maglev technology has also been more attention.The magnetic levitation system is a nonlinear, uncertainty and unstable system. The development of magnetic levitation controller is more and more difficult because of the factors. The magnetic levitation ball system,which is strict to real-time control, is used as the object to implement and verify all kinds of control algorithm. The maglev ball control system is taken as the object in the research, aiming to find more high-precision control algorithm and more targeted performance evaluation.First of all, it is important to understand the character, composition and working principle of the magnetic levitation ball system.The system of nonlinear mathematical modeling is established using mechanics, electromagnetism and other knowledge. And the ability, observability and stability of the system is analyzed with the tool called MATLAB.The fuzzy PID controller is used to solve the limitations of the PID controller, when the controller is used in the the magnetic levitation ball system. The fuzzy PID control algorithm has better performance because its the combination of the fuzzy control algorithm and PID algorithm. There are three kinds of improvement means to solve the problems of the conventional fuzzy PID control. The variable universe adaptive fuzzy PID controller is one of the improvement mean,which optimizes the range of the fuzzy variables. The simulation result shows that the variable universe adaptive fuzzy PID controller has better stability, higher control precision. And the result is very important theoretical basis for subsequent real-time control test.In the magnetic levitation ball control system, there are two problems of the existing system that the degree of satisfactory optimization is low and the performance evaluation index of the control system lacks pertinence for different input signals. Based on this, a multi-class satisfactory optimization control algorithm is proposed. In this method, the PID controller based on genetic algorithm is used. We can get global optimal solution through this controller. In addition, Kp, Ki, and Kd are selected as the variable to be optimal of the system. These parameters are evaluated in class through the preset performance evaluation index and satisfactory function in the input of step, sine, square, sawtooth and random signals. Finally optimal results are obtained through this system.Finally, the two algorithms is verified in the real-time control platform of magnetic levitation ball control system.The test is aiming to verify the effectiveness of the controller. The result based on the variable universe adaptive fuzzy PID controller shows that the method is more effective than the conventional control algorithm. And the result shows the quantitative performance evaluation index. The result based on the multi-class satisfactory optimization controller shows that the genetic algorithm PID control is more effective at no mutation class tracking signal than mutation class tracking signal. And it gives the the different evaluation performance indicators for corresponding signal and the satisfaction of each index. Through the intuitive display of the corresponding index data, the evaluation of the control performance of the magnetic levitation ball control system becomes more direct and systematic.