作者:\n\t苏畅，施刚，张建文，刘海春，周剑桥\n

Authors:\n\tSU Chang,SHI Gang,ZHANG Jianwen,LIU Haichun,ZHOU Jianqiao\n
摘要:\n\t基于电力电子器件实现的柔性互联装置可独立、连续调节有功和无功功率，具有提供无功支撑、调节潮流分布、平衡馈线负载、降低网络损耗等多种功能，已成为配电网发展的核心装置。为使其在配电网运行中充分发挥优势，对其实时运行优化问题开展了研究，提出了一种粒子群优化-人工神经网络两阶段方法：一阶段采用具有全局寻优能力的粒子群优化算法求解若干典型工况，构建训练集为人工神经网络提供数据支持；二阶段采用人工神经网络，可离线进行网络训练并在接收到待求解工况时快速在线求解。以一9节点配电网络为算例，验证了柔性互联装置的调节效果，以及所提方法的准确性与快速性。\n

Abstract:\n\tFlexible interconnection device (FID), based on power electronic technology,allows independent and continuous regulation of active and reactive power. With the functions of reactive power support, power flow regulation, load balancing and network loss reduction, it has become the core device of the development of distribution network.To give full play to its advantages in the operation of distribution network, the real-time operation optimization problem is studied, and a PSO-ANN two-step method based on particle swarm optimization (PSO) and artificial neural network (ANN) is proposed: in step 1, PSO is used to solve some typical working conditions, and construct the training set to provide data support for ANN; in step 2, ANN is adopted, which can be trained offline and solve the working condition received quickly online.A case study on a 9node distribution network verifies the regulation capability of FID and the accuracy and rapidity of the proposed method.\n


PDF全文下载地址:

可免费Download/下载PDF全文