基于分区选择的主动配电网分布式最优潮流分析

Distributed Optimal Power Flow Analysis of Active Distribution Networks Based on Intelligent Partition Strategy

基于交替方向乘子法(ADMM)在分布式形式下解决主动配电系统最优潮流问题(OPF),针对分布式算法性能受到配电系统区域划分影响的问题,提出了一种基于量测数据驱动的电网分区方法,以加速优化算法的收敛速度。与传统的ADMM算法依赖于全局信息不同,本文引入了一致性方法来协调区域交界的平衡问题,从而实现最优潮流问题的完全分布式求解。此外,本文采用LinDistFlow(Linearized Distribution Flow)交流近似模型来应对配电网最优潮流问题的非凸性挑战。通过在不同规模的IEEE配电网案例上进行测试,验证了所提方法的有效性,且其在优化算法的迭代次数、计算时间和误差精度等性能上均优于其他分区方法。

With the increasing penetration of distributed energy in power systems,the optimal power flow(OPF)analysis-based distribution network energy management system faces challenges such as high data dimensionality,long solution time,and large computational load.In response to these challenges,this paper aims to use the alternating direction method of multipliers(ADMM)to solve the OPF problem of active distribution systems in a fully distributed fashion.Considering the impact of distribution system partitioning on the performance of distributed algorithms,a data-driven network partitioning method based on measurement data is proposed to accelerate the convergence rate of the optimization algorithm.Moreover,different from traditional ADMM algorithms that rely more or less on certain global information,this paper integrates a consensus-based method to coordinate the balance problem at the interface between regions,thereby achieving a fully distributed solution to the OPF problem.In addition,this paper employs the LinDistFlow AC power flow approximation model to resolve the non-convexity challenges of the OPF problem.Finally,the effectiveness of the proposed method is verified through tests on IEEE distribution network cases of different sizes,and its performance in terms of error accuracy and computation time of the optimization algorithm is superior in comparison to other partitioning methods.