Modeling and robust control of a magnetic bearing system

Active magnetic bearing technology offers a number of advantages over conventional bearings including high speed, low power losses, and low maintenance costs. These benefits can, however, only be achieved by advanced signal processing and control. In this paper, a systematic approach is used to model and control a magnetic bearing system. An analytic model of the magnetic bearing system is derived whose parameters are then optimized to correlate with the measurement data. The control system is then designed based upon the model and the associated modeling error. A multi-objective control design method that accounts for a set of robust stability and performance requirements is used to design a controller, which is then implemented using a digital signal processor board. The design is verified to improve the performance of the existing system.