Optimal Tuning for Fractional-Order Controllers: An Integer-Order Approximating Filter Approach

Abstract

This paper offers a systematic framework for the design of suboptimal integer-order controllers based on fractional-order structures. The proposed approach is built upon the integer-order approximations that are traditionally used to implement fractional-order controllers after they are designed. Accordingly, the fractional-order structures are exploited to derive a suitable parameterization for a fixed-structure integer-order controller. The parameters, which describe the structure of the fixed-order integer controller, are the same as those used to describe the original fractional structure. Optimal tuning is then performed in the parameter space of the fractional structure and the resultant set of optimum parameters will determine the optimal integer-order controller. The proposed approach outperforms traditional implementations of optimal fractional controllers and presents an alternative that attempts to capture merits of both fractional-order and integer-order structures.