Automatic restoration of active distribution networks based on tabu search specialized algorithm

Abstract

This paper presents a new technical approach based on mixed integer nonlinear mathematical formulation for the automatic restoration problem of distribution networks with distributed generation (DG), using node-depth encoding (NDE) and tabu search (TS) metaheuristic. The mathematical model considers the possibility of DG isolated operation and two principal objectives of restoration problem, that are to minimize number of consumers without energy supply outside the affected area, and to minimize the number of required switching. This methodology provides a switching logic sequence, which guaranties to attend operational aspects of distribution grids avoiding creating loops or generating short-circuits throughout the manually or automatic switching process. Finally, results of the proposed methodology for a 135 bus-system are presented in single-faulted and multiple-faulted scenarios, showing the effectiveness of the algorithm.