A tabu search algorithm for the power system islanding problem

Abstract

The power system islanding problem aims to divide the power system into several different islands after serious disturbances occur. The objective of this problem is to minimize the total generation-load imbalance of all islands while placing the coherent generators in the same island and maintaining the connectivity of each island. Two main challenges of solving this problem are the large scale of the power system and the requirement of a short computation time. In this study, we propose a tailored tabu search algorithm to solve this problem, which employs a novel initial solution generation procedure and a neighborhood operation based on the movement of the boundary nodes. The numerical experiments on 15 test instances show that this algorithm can deal with the power systems with up to 3120 nodes within 0.7 s. Then, the comparisons with some existing islanding methods based on the IEEE 39-bus system and the IEEE 118-bus system prove the validity and accuracy of our method. Finally, time-domain simulations based on three power systems demonstrate the importance of our tabu search algorithm in minimizing the impacts of the disastrous disturbances. The computational results imply that our tabu search algorithm is very effective and efficient and satisfies the requirements of islanding power systems of various sizes.