On the efficiency of an alternative codification for multiobjective service restoration in electric distribution networks

Abstract

This paper proposes an alternative codification to solve the service restoration in electric power distribution networks using a SPEA2 multiobjective evolutionary algorithm, assuming the minimization of both the load not supplied and the number of switching operations involved in the restoration plan. Constrains as the line, power source and voltage drop limits in order to avoid the activation of protective devices are all included in the proposed algorithm. Experimental results have shown the convenience on considering these new representations in the sense of feasibility maintenance and also in the sense of better approximation to the Pareto set.