A Convex Optimization Approach to Robust Controller Design for Active Magnetic Bearing Suspension Systems

摘要

Active magnetic bearings (AMBs) which suspend high speed rotor systems present a challenge in control design. A controller with robustness to uncertainty and capable of adjusting itself according to the rotor speed is essential. Based on a prototype for a flywheel energy storage system constructed at University of Virginia, a convex approach to design a robust linear parameter varying (LPV) controller is proposed. As a result, the robust LPV controller can maintain its stabilizing capability and robust performance for the whole speed range. One impediment for implementing the LPV controller is the computational costliness. Based on the convex approach, an effective methodology for implementing the LPV controller on-line is proposed. Simulation and experimental results are in agreement with the theoretical derivation.