Control Design For Magnetic Levitation Systems

In recent years, because of the merits such as non-contact, no mechanical friction, high-precision positioning, low power consumption, no need of lubrication and longevity etc, magnetic levitation technology is becoming an attractive topic. For the performance of the controller will directly determine the performance of the magnetic levitation system including system stability, stiffness and capability of anti-jamming etc. it is a key technology and important task to construct a high performance feedback controller to regulate the position of the levitated object. Magnetic levitation system is usually uncertain, open-loop unstable and nonlinear, due to the nonlinearity of the magnetic force. All of these add the difficulty of designing the maglev controller. So investigation of controller design approach and exploration of new magnetic levitation controller is of great theoretical and practical significance.This thesis considers magnetic levitation systems with one degree of freedom. This system has a simple structure, and is easy to be modeled, and performance analysis is relatively easy which can greatly shorten the research cycle. Also, for multi-degree of freedom magnetic levitation system, we can use the research results obtained by researching one degree of freedom magnetic levitation system. The contributions of this thesis are as follows:1. We gave a detailed analysis of system structure and working principle, then established its mathematical model. The linear processing is made to the system model by Taylor series expansion and feedback linearization.2. Based on linear model of magnetic levitation systems, we design PID controller and state feedback controller for the systems. It is observed from the simulation result that both of PID controller and state feedback controller having good performance.3. We cast the controller design as the output regulation problem, and a nonlinear maglev controller is designed by model transformation and backstepping approach. Theoretical analysis and system simulation results show that it has more superior performance than the linear controller such that all signals of system are bounded and the air-gap of the levitated object asymptotically approaches to a given value. For the high-frequency and low-frequency interference, it has a satisfactory inhibitory effect. Keywords:magnetic levitation system, feedback linearization, backstepping approach, model transformation, output regulation...