An Efficient Genetic Algorithm for Loss Minimum Distribution System Reconfiguration

Abstract

This article describes a chromosome coding method of genetic algorithm (GA) for minimizing losses of a radial distribution system when it is subjected to network reconfiguration. By simulating the survival of the fittest among the strings, the optimum string is searched by randomized information exchange between strings by performing crossover and mutation. A vector-based load flow method for distribution systems using sparsity technique, which requires less memory and less number of iterations to converge is used, as this problem requires faster solution. Reciprocal of active power losses is taken as fitness function. The proposed algorithm improves the performance of GA and reduces the number of bits and iterations for distribution system reconfiguration. Results have shown that proposed algorithm has more advantages over previous algorithms when large power distribution systems are considered.