Research On Strategies For Service Restoration Of Distribution Networks With DGs Based On Islanding Operation

To reduce environmental pollution and improve the reliability of power supply, it is encouraged that the construction of smart distribution network need the massive introduction and making full use of DGs. Different from that network reconfiguration is usually used to recover the power supply of non-fault loss area in traditional distribution system, islanding operation can be used as a new approach for service restoration of distribution networks with DGs. After a fault occurs, DGs could ensure the power supply of the priority loads when the area is isolated from the main grid, and some harm to system security caused by DGs could be avoid ed by reasonable planning and effective control.In this thesis, researches on strategies for service restoration of dist ribution networks with DGs based on islanding operation are carried out to deal with different fault situations of distribution networks. When serious external fault of distribution system is detected, in order to avoid the blackout, the distribution system should be divided into several islands according reasonable splitting points by islanding to ensure the continual power supply of priority loads. In the situation of blackout, the black-start of distribution network is used to recover the power system by starting one DG with black-start ability to mobilize other DGs in the distribution system.Firstly, a heuristic islanding algorithm based on Distance-Weight is proposed. Distance-Weight which represents the cost to recover the power supply of one load is introduced to evaluate the distance between the objective load and the current island network. The initial island is composed of all the DGs and some branches between them, and then it will be enlarged by merging the nearest load among all the considered loads. The feasible islanding scheme will be carried out soon after the island network has been satisfied with all the constraints.Then, considering the limitations of heuristic algorithm which could not find out the best solution commonly in solving complex network, an islanding algorithm for distribution networks with DGs based on adaptive genetic annealing algorithm is proposed in the third chapter. Take account of the priority and con trollability of loads, the islanding problem model is constructed by considering all the safe constraints and taking the maximum priority load as the main objective. Each branch of distribution network is coded as 0-1, and breadth-first search is adopted to decode an islanding scheme. After the identification of all the areas, the fitness of a solution will be calculated while the infeasible solution should be punished. The best islanding scheme will be acquired by the search of adaptive genetic annealing algorithm.Considering the black-start ability of DGs, the fuzzy analytic hierarchy process is used to evaluate the performance of each DG. Three criteria including starting ability, energy supply and operation status are adopted to analyze the performance of each DG. The ranking of DG performance can be acquired by computing the weight coefficient and the value of each index to support the optimization of DGs starting sequence during the black-start process.Finally, the black-start strategy to recover the power supply of part area in distribution network after blackout is proposed. The black-start area is designed based on the islanding scheme. The adaptive genetic annealing algorithm is used to optimize the starting sequence of DGs and recovery routes. The loads in the island will be recovered gradually after the main network with all DGs is constructed.Simulation results of PG&E 69-node distribution system demonstrate that the distribution system could be recovered under different fault situations by the proposed algorithms, which proves the effectiveness of the proposed algorithms in this thesis.