Automated Switch Placement in Power Distribution Systems: Analysis of Binary PSO Parameters

Abstract

Currently, electric utilities have improved their operational performance using automated switches (ASWs). These devices make it possible to transfer loads remotely, restoring energy supply to the greatest number of consumers. However, utilities do not have precise criteria to define the number of ASWs and their respective locations. This paper proposes a binary particle swarm optimization (BPSO) to explore ASW allocation alternatives seeking the best solution that minimizes the system Mean Interruption Frequency Index (SAIFI), System Mean Interruption Duration Index (SAIDI) and Unsupplied Energy (ENS). A probabilistic methodology based on Monte Carlo Simulation is proposed to estimate SAIFI, SAIDI and ENS. An analysis of the BPSO parameters, such as inertia weight (w) and acceleration coefficients (c1, c2) is performed to select values suitable for application in electrical networks. The results show that the parameters c1 and c2 between 3.0 and 3.3 and w = 1 improve the behavior of the BPSO.