Search

Several authors have treated the optimization of passive filters in electric distribution systems. Optimization methods like: sequential quadratic programming (SQP), simulated annealing (SA), differential evolution (DE), artificial neural networks (ANN), particle swarm optimization (PSO), genetic algorithm (GA), etc., have been employed for optimizing certain configurations of passive filters. These optimization methods have been employed to solve several formulations of the problem of the project of filters. These formulations can be classified in: formulations of one or several objectives. The objective of the present work is to show the evolution of the formulation of this problem in the lasts years respect to the objective functions and constraints used. This analysis shows how the formulations employed have been upgraded from single-objective to multi-objective formulations to achieve a better representation of this complex problem.

Different methods of optimization have been used for the selection and localization of capacitor banks in distribution circuits. However, it is necessary to consider the distortion taken place by the harmonics of the no-linear loads and the possible appearance of resonances between the inductive elements of the system and the existent capacitor banks. On the other hand, it has not been integrated in a single methodology the use of capacitor banks and of passive harmonic filters to compensate the reactive power in harmonic polluted distribution systems. The present work intends as objective to develop this methodology. The developed program is based on the multiobjective optimization algorithm NSGA-II that it is proven with success in several examples.