Implementation of controller optimization using S-function based metaheuristic algorithms

Abstract

This paper presents general technique for implementing metaheuristic algorithms using S-Functions and its application in controller parameter optimization. The proposed method provides a suitable platform for parameter tuning in MATLAB/Simulink and PLECS software platforms. The proposed technique is suitable for exchanging optimization parameters when the objective function is not expressed in closed mathematical form. The S-Function based technique takes advantage of the software framework by embedding function calls of the optimization algorithm in callback methods which are executed at predefined time intervals. This leads to great improvement in algorithm efficiency in terms of simulation speed and ease of implementation. Validity of the proposed method is verified by implementing the particle swarm optimization algorithm (PSO) as an S-Function. The S-Function based PSO algorithm is applied to tune PI controllers for converter control in a doubly fed induction generator (DFIG) wind generation system.