Integrated feasible direction method and genetic algorithm for optimal planning of harmonic filters with uncertainty conditions

This paper presents an integrated approach of feasible direction method and genetic algorithm (FDM + GA) to optimize the planning of large-scale passive harmonic filters (PHF). The objective is to minimize the total demand distortion of harmonic currents, total harmonic distortion of load bus voltages, filters loss and the cost of total installation of LC tuned filters. Reactive power compensation and constraints of individual harmonics have been considered in the design process, and the constraints of harmonics with orders lower than the filter tuned-points have been set stricter to avoid amplifying non-characteristic harmonics as well. In order to determine a set of weights of objective function representing the relative importance of each term, the simplest and most efficient form of triangular membership functions have been employed. The designed FDM + GA heuristic is applied to the harmonic problems in a chemical plant with three 6-pulse rectifiers. Three design schemes are compared to demonstrate the performance of the FDM + GA heuristic. Finally, expectations and standard deviations of objective values are used to present the effects of filter parameter detuning, loading uncertainty, and changing of system impedance.