Service restoration of distribution systems using enhanced NDE with LCO

When applying modern intelligent optimization algorithms to solve the service restoration problem, algorithms can often generate many infeasible solutions (associated with non-radial configurations). Consequently, such algorithms need to analyze the topological structure for every proposed solution, which is time-consuming. To avoid these disadvantages, this paper introduces the Node-Depth Encoding (NDE) technique into the Non-Dominated Sorting Genetic Algorithm-II (NSGA-II). This enhanced algorithmic procedure substitutes restructuring operations based on the NDE technique for the traditional crossover and mutation operations, and hence it can ensure a radial configuration, and determines the topological structure of the network within a reduced calculation time. This paper also introduces a Load Curtailment Operation (LCO) into the restructuring operations which can make effective load curtailment recommendations. The simulation results show the validity of LCO and demonstrate the improved performance of the enhanced algorithmic procedure in convergence and diversity when compared with the previous NSGA-II approach.