Pilot-bus Selection and Network Partitioning Optimization Method Considering Randomness and Relevance of Source and Load Power Change

• 1、State Key Laboratory of Power Transmission Equipment & System Security and New Technology(Chongqing University), Shapingba District, Chongqing 400044, China

Abstract：To ensure efficient and reliable operation and maintenance of an AVC system, a network partitioning plan is frequently required to keep unchanged during a long period for engineering application and the corresponding network partitioning plan shall adapt to changes of the source & load power during a long period. Therefore, randomness and relevance of the source & load power changes is considered and the coordination optimization method for pilot-bus selection and network partitioning is proposed in this paper. Besides coupling between nodes inside and outside partitions, this model shall also consider the network partitioning balance requirement when the relevance of the source & load power randomly changes. Randomness and relevance of the load and wind power are expressed by using the normal distribution, Weibull distribution and relevance coefficient and the satisfaction degree of the reactive balance is expressed by using the chance constraint. Based on the Latin hypercube sampling of the source & load power relevance, the scene compression technology is proposed to consider the maximal reactive requirement of the partition and simulate typical scenes of the source-load power. Next a immune genetic algorithm based on target relative dominance is used to solve the optimization model. The simulation results of the IEEE39 system are used to demonstrate effectiveness of the proposed method.

Keywords： network partitioning pilot-bus randomness immune genetic algorithm