Synthesis of fixed-structure robust controllers using the distributed particle swarm optimizer with cyclic-network topology

This paper discusses a new meta-heuristic approach with high reliability to the synthesis problem of fixed structure robust controllers satisfying multiple control specifications. For this purpose, first, the particle swarm optimizer (PSO) with cyclic-network topology is developed. Such a neighborhood topology can ensure a good trade-off between exploration and exploitation ability of the swarm, which results in a significant reduction of the probability of premature convergence to local optima. Second, the proposed distributed PSO algorithm is incorporated with the simple constraint handling method [8] using a virtual objective function to handle multiple control specifications. Then, it is shown how to find optimal parameters of a fixed-structure controller guaranteeing the given specifications based on the developed PSO technique using cyclic-network topology. Third, a typical numerical example to demonstrate its effectiveness is given, which clearly shows that the proposed distributed PSO scheme gives a novel and powerful impetus to the fixed structure robust controller synthesis.