Experimental studies on H∞ control for magnetic levitation system with friction compensation

ECP Model 730 magnetic levitation system is a typical nonlinear experiment system. Due to uncertainties and nonlinearities which existed in the system, it is difficult to achieve high performances. In order to improve the performance, a new improved H∞ control scheme with friction compensation is proposed. The magnetic levitation system is modeled and the parameters are identified with experiments. A normal H∞ controller is designed for the linearization system with nonlinear compensation, but there still exists steady state error in actual experiment. To the problem, friction identification experiment is performed and Stribeck curve of LuGre model is fitted by genetic algorithm. Friction nonlinear effects are compensated mostly with the model. Furthermore, regarding friction uncertainties as system's input disturbance, a new H∞ controller which has the ability to reject input disturbance is designed. The effectiveness of this new control scheme is verified through experiments.