IEEE Access (Jan 2019)
Mixed-Integer Distributed Ant Colony Multi-Objective Optimization of Single-Tuned Passive Harmonic Filter Parameters
Abstract
The extended ant colony known as the Mixed Integer Distributed Ant Colony Optimization (MIDACO) is presented in this paper as a new application of solving multi-objective single-tuned passive filters design problems. This paper presents a new non-dominated solution for the optimization of four independent objective functions which are maximized power factor, minimized total harmonic voltage distortion, minimized total demand distortion, and minimized investment cost of the filter. The global solution is achieved by maintaining the quality factor of the filter in a specified range, avoiding the harmonic resonance and maintaining the capacitor's capability limits within the standard limits. The attained parameters of the filter are used to weigh the performance of the system, and the robustness of the proposed algorithm is verified by comparing the results with three different highly competitive evolutionary techniques. Also, the proposed algorithm attains the Pareto front of the problem and tolerates the selection of its parameters to the most effective solution. The numerical results specify the comprehensive passive filter design through possible multi-objective approaches, and the improvements of multi-objective over single-objective optimization are also presented in this paper.
Keywords
