Alocação ótima de capacitores em sistemas elétricos de distribuição desequilibrados através de análise de sensibilidade com limitação do espaço de busca

Abstract

The problem of optimal allocation of capacitors in unbalanced electrical distribution systems is not so simple to solve, and this is due, in large part, to its highly combinatorial, non-linear and non-convex nature, which still includes restrictions that must represent technical and economic aspects from the distribution systems. Therefore, it is of great interest to formulate more elaborate optimization tools that contribute to such an activity. This work proposes an algorithm for optimizing unbalanced electrical distribution systems through the optimal allocation of capacitor banks along the feeders for a given planning horizon. The method is divided into three steps, the first step is intended to calculate the fixed reactive compensation demand necessary for the proper operation of the systems, the second step is intended for a sensitivity analysis to define the most sensitive nodes to the function objective and then, finally, the third step is aimed at the optimal allocation of capacitor banks in the distribution systems. The proposed method uses a three-phase power flow and a genetic algorithm to solve the optimal power flow problem and considers as optimization variables the capacity of the banks in each possible allocation location. The objective of the proposed methodology is to maximize the savings in operating costs, in relation to the reduction of technical energy losses, in a planning horizon, with a minimum investment cost, mitigating the voltage unbalance and respecting the voltage limits on the nodes of the systems. A contribution of this work is to deal with the redefinition of the search space in order to improve the efficiency of the GA, thus increasing the quality of the results and potentially reducing the computational effort in large systems.