Optimization of a phase stabilization controller for an ultra-precision positioning stage using a structured genetic algorithm

Abstract

A practical controller optimization method is proposed to improve control performance in response speed and positioning accuracy of an ultra-precision positioning stage. The structural resonance characteristics as well as the time delay effect severely limit the control bandwidth, and the direct drive feature enabled by aerostatic bearing and voice coil motors makes the stage very sensitive to various disturbances. Therefore, a phase stabilization controller is employed owing to its superior performance in suppressing resonances compared with a traditional notch-based controller. By developing candidates of controller structure based on phase stabilization design criteria, the controller design involves the determination of both structure and parameters, which can achieve better performance than merely tuning parameters of a controller with predefined structure. Consequently, the design of structure and parameters of the controller are formulated into an optimization problem, and the objective function considers both control bandwidth and disturbance rejection to obtain balanced improvements of both response speed and positioning accuracy. Then a structured genetic algorithm is applied to optimize the structure and parameters of the controller simultaneously. Experiments are carried out for comparison of the proposed strategy with the traditional notch-based proportional–integral–derivative controller and the manually tuned phase stabilization controller. Comparative experimental results validate that the optimized controller achieves the smallest positioning error and shortest settling time, which demonstrates the effectiveness of the proposed method in practical application.