Signal-based H∞ optimal control for AMB system based on genetic algorithm

This paper addresses a signal-based H_∞ robust centralized control for multi-input multi-output active magnetic bearing systems. An often problematic and time-consuming weight selection is solved with a genetic algorithm. The properties of the obtained H_∞ controller are compared with a reference linear-quadratic and a loop transfer recovery controller. In order to ensure a credible comparison of the closed-loop systems, the same genetic algorithm is used when tuning the weights for both control designs. The robust stability of a multi-input multi-output system with multiple perturbations is analyzed. A good agreement between the analytical and the experimental results is verified.