Islanding Method for Power Restoration of Distribution Network Based on Flexible Adaptive Genetic Algorithm

Abstract

Nowadays, when the distribution network is disconnected from the main network due to faults, it can restore the power supply to the recoverable loads around by distributed generations (DGs). In this paper, the distribution network is modeled as a tree and 0-1 array is used to represent the on-off between nodes. A flexible adaptive genetic algorithm is used to determine the islanded power restoration area. In the selection process, the improved pointer random sampling is used to replace the traditional roulette wheel sampling, and the flexibility in mutation is considered in order to improve individual fitness. Load recovery effect, network loss, and switching action cost are taken as objective functions to island areas optimally under the constraints of voltage, line capacity, network connection, and islanding capacity. The simulation results show that the proposed method can quickly form a recovery scheme and is feasible in power systems.