A parallel tabu search based approach to optimal network reconfigurations for service restoration in distribution systems

This paper proposes a new method for optimizing the network reconfigurations for service restoration in distribution systems. The service restoration may be expressed in a complicated combinatorial optimization problem. In this paper, meta-heuristics are considered to solve the problem efficiently. The proposed method makes use of parallel tabu search (PTS) that considers a couple of strategies in tabu search (TS). One strategy is to decompose the neighborhood into subneighborhoods to reduce computational effort. The other is to introduce several tabu lengths into TS to make solution candidates more diverse. The proposed method is applied to the 69-node distribution systems. A comparison is made between the conventional and the proposed methods in terms of computational effort and solution accuracy.