An optimal multirate control design with robustness specification for sampled-data HDD servo systems

Abstract

One major contributor of the track mis-registration in a hard disk drive (HDD) servo system is the position error signal (PES) 3/spl sigma/, or 3 times the standard deviation of the PES. The paper presents an optimal robust multirate control design to minimize 3/spl sigma//sub PES/ in HDD servo systems with stochastic distribution of plant parameters via genetic algorithm and random neighborhood search considering robust stability restrictions. The expected H/sub /spl infin// norm of the sampled-data system was set as the robust performance index and the H/sub /spl infin// norm of weighted complementary sensitivity function of the plant input was set as the robust stability index for the multiplicative perturbation. The genetic algorithm with sharing scheme and tabu list which could improve diversity of solutions and global search ability was employed for a coarse optimization of the controller parameters and random neighborhood search did a fine-tuning of the controller. The numerical results show that, the proposed method could improve the robust performance of the sampled-data HDD servo system considerably while keeping the robust stability.