I. Introduction
The aim of capacitors in radial distribution systems is to reduce power losses, voltage profile improvement, system capacity release, reactive power compensation, power factor correction and reliability enhancement. To achieve these advantages, it is necessary an optimal capacitors placement, that determines locations and sizes of capacitors to be installed in the system [1]. Optimal capacitors placement is a combinatorial problem where the search space grows exponentially with the size of the system [2]. Mathematically, the capacitor placement is a nonlinear problem, whose solution is non-convex and may present different optimal local solutions [3]. This problem has been investigated since the 60’s, but in the 90’s several approaches have been developed through heuristic and metaheuristic techniques: Simulated Annealing [4], Genetic Algorithms [5], Tabu Search [6], Ant Colony Algorithm [3] and Particle Swarm Optimization (PSO). PSO is a metaheuristic technique that has been highlighted due to its efficiency to solve combinatorial problems [3]. PSO has also been found to be robust in solving problems featuring non-linearity, non-differentiability and high dimensionality [7].