基于神经网络的并网光伏电站自适应距离保护  

Adaptive Distance Protection of Grid-Connected Photovoltaic Power Plant Based on Neural Network

作　　者：薛风华 徐微微 王洪寅 庞吉年 祁言嘉 王宝华[2] XUE Feng-hua;XU Wei-wei;WANG Hong-yin;PANG Ji-nian;QI Yan-jia;WANG Bao-hua(Suqian Power Supply Branch of State Grid Jiangsu Electric Power Co.,Ltd,Suqian 223800,China;School of Automation,Nanjing University of Science and Technology,Nanjing 210094,China)

机构地区：[1]国网江苏省电力有限公司宿迁供电分公司,江苏宿迁223800 [2]南京理工大学自动化学院,江苏南京210094

出　　处：《电工电气》2020年第4期25-29,共5页Electrotechnics Electric

摘　　要：光伏电站并入配电网后,原先配电网的距离保护动作特性发生变化,其分支系数随着光伏电站的出力变化而变化,给含有分布式电源的配电网保护整定计算带来困难。针对某实际光伏电站结构,分析了光伏电站并网后作为分布式电源对配电网距离保护的影响,利用BP神经网络建立光伏电站的等效阻抗模型,通过BP神经网络建立其输出功率与等效阻抗的关系,并以此得到自适应距离保护的整定方法,实例验证了该方法的正确性与实用性。The distance protection action characteristics of the original distribution network will change after the photovoltaic power station is integrated into the distribution network,and its branching coefficient will change with the output of the photovoltaic power station,which makes it difficult to calculate the distribution protection of distributed power grids.Aiming at the structure of certain actual photovoltaic power station,this paper analyzed the impacts of distributed generation on the distribution network distance protection,used BP neural network to establish the equivalent impedance model in photovoltaic power station and the relationship between the output power and the equivalent impedanceby BP neural network,and obtained the tuning method of adaptive distance protection on this account.The correctness and practicability of this method are verified by an example.

关 键 词：光伏发电 配电网 分支系数 距离保护 神经网络 

分 类 号：TM615[电气工程—电力系统及自动化]