Distribution System Fault Recovery with Controllable Distributed Generations

With the increasing installed capacity of distributed generators (DGs) in distribution systems, its impact on distribution systems is becoming more significant. The conventional methods for distribution systems fault recovery have to be upgraded due to more DGs integration. In this paper, a novel restoration strategy with a two-stage process is proposed. In the first stage, islands are established with power supplied from DGs. In the second stage, the islands formed in first stage will be modeled as dispatchable loads. These dispatchable loads and other traditional loads that are not within the islands will be restored together. This problem described as a multi-objective problem with the minimal system loss, number of switch operation and potential loss of loads. Fast elitist non-dominated sorting partheno-genetic algorithm is employed to solve this multi-objective problem. IEEE 33-bus system is used as an example. The results show that the models and algorithms proposed in this paper is efficient.