基于同步型ADMM的含海上风电场电力系统分布鲁棒无功优化OA北大核心CSTPCD

The output of large-scale offshore wind power is highly random and volatile,which increases the difficulty of reactive power and voltage control of the power system.In this regard,the distance between the true probability and the reference probability distribution of wind speed in offshore wind farms was first measured based on Kullback-Leibler divergence,thereby constructing a probability distribution ambiguity set of wind speed.Then,taking into account the operating constraints of the transmission network and offshore wind farms,a distributed robust reactive power optimization model for the power system containing offshore wind farms was established.In order to maintain the information privacy of the transmission grid and offshore wind farms during the optimization calculation process,the transmission grid and offshore wind farm areas were spatially decoupled based on the synchronous alternating direction multiplier method(ADMM),thereby decomposing the original centralized optimization model into corresponding regions.The sub-model was solved iteratively in a distributed manner.Among them,the sub-models of each area were multi-layer models,which were solved iteratively by embedding column and constraint generation algorithms in distributed algorithms.Finally,the correctness and effectiveness of the proposed model were verified in a modified IEEE 39 bus system containing two offshore wind farms.The results indicate that the proposed model decision results can effectively reduce the system network loss and voltage deviation while considering the uncertainties of random variables.