Multicriteria analysis for optimal reconfiguration of a distribution network in case of failures

This work presents a methodology for the optimal reconfiguration of a distribution network in the presence of a failure, through a multicriteria optimization algorithm. For this purpose, the optimal network reconfiguration alternative is verified in the IEEE 33 bus test system and 123 bus test system. The variables analyzed within the multicriteria decision matrix include the total interruption time per nominal kVA installed (TITK), mean frequency of interruption per nominal kVA installed (MFIK), reset time for reconfiguration, energy not supplied, total losses in the lines of the system and operation and maintenance costs. The result enables selecting the best scenario based on analyzing every decision criterion; the multicriteria decision algorithm is developed in the Matlab environment. Subsequently, the winning reconfiguration alternatives are validated through simulations in Cymdist for different failure scenarios. In the analysis of results, metrics are presented that enable us to observe a significant improvement in the typical problems that occur in an electric system.