Robust tuning of fixed-structure controller for disk drives using statistical model and multi-objective genetic algorithms

This paper proposes a non-gradient based method for the parameter optimization of fixed-structure controllers in hard disk drives (HDDs). Besides satisfying multiple frequency-domain constraints, the primary target of HDD servo design aims at the minimization of position error signal (PES) for a large population of drives. This is made possible by adopting a new statistical disturbance model inside the optimization loop to evaluate the time-domain performance of candidate controllers. The convexity of multi-dimension searching space is lost because the controller structure is fixed. This non-convex multiobjective optimization problem (MOP) is solved by multiobjective genetic algorithms (MOGA), which are genetic algorithms (GA) combined with the concept of Pareto optimality. Multiple optimal solutions with trade-offs are provided to support decision making. A design example of tuning a track following controller is used to demonstrate the effectiveness of the proposed method.