A decomposition heuristic algorithm for dynamic reconfiguration after contingency situations in distribution systems considering island operations

Abstract

This paper presents a decision-making model for the dynamic reconfiguration of distribution networks after contingency situations based on formulations of the self-healing problem with islanding. This paper develops a three-stage algorithm. The first stage applies a greedy reconfiguration heuristic, which calculates the criteria values for each viable network configuration in each hour. To address the variety of criteria, the analytic hierarchy process method is applied to define the weights. The second stage addresses the problem statically to reduce the number of configurations and, therefore, the combinations of step 3. In the third stage, the problem is approached dynamically, generating the optimal sequence of topologies for the analyzed period, while considering the time-varying operational conditions of the system. The proposed heuristic significantly reduces the complexity of the problem while minimizing the following criteria: energy not supplied, losses, number of interrupted consumers and switch operations. The physical constraints of the power grid considered in the proposed model are the line limits, protective equipment settings, voltage limits, distributed generators (DG) limits and network radiality. The proposed model is analyzed in different scenarios in the IEEE 123-bus system, modified with the insertion of DGs. The proposed method is compared with a conventional benchmark model.