含光伏发电和时变负荷的主动配电网孤岛划分  

ISLANDING DIVISION OF ACTIVE DISTRIBUTION NETWORK WITH PHOTOVOLTAIC GENERATION AND TIME-VARYING LOAD

作　　者：田晶京[1] 李安乐 高锋阳[1,2] 赵峰 耿芳[1] Tian Jingjing;Li Anle;Gao Fengyang;Zhao Feng;Geng Fang(School of Automation and Electrical,Lanzhou Jiaotong University,Lanzhou 730070,China;Gansu Jiaoda Engineering Testing Technology Co.,Ltd.,Lanzhou 730070,China)

机构地区：[1]兰州交通大学自动化与电气工程学院,兰州730070 [2]甘肃交达工程检测科技有限公司,兰州730070

出　　处：《太阳能学报》2021年第9期125-131,共7页Acta Energiae Solaris Sinica

基　　金：甘肃省重点研发计划(18YF1FA058);兰州市人才(2017-RC-95)-电网状态检测和故障诊断关键技术。

摘　　要：针对含有光伏发电和负荷功率动态变化的主动配电网孤岛划分问题,首先建立光储系统和负荷的时变模型,其次采用基于模拟退火遗传算法的模糊C-均值聚类算法根据时变负荷的内在相似性进行聚类,完成故障恢复时段划分,并在此时段内以各节点聚类中心值表示各节点负荷状态的静态孤岛划分问题。在此基础上建立孤岛数学模型并确定光储系统孤岛供电可行域,最后以IEEE33节点系统为算例,采用改进的二进制粒子群算法对其模型求解,得到各故障恢复时段内的孤岛划分方案。Aiming at islanding division of active distribution network with dynamic changes of photovoltaic generation and load power.Firstly,the time-varying model of optical storage system and load is established.Secondly,the fuzzy C-means clustering algorithm based on simulated annealing genetic algorithm is used to cluster according to the internal similarity of time-varying loads to complete fault recovery period division.And the problem of static islanding division of load status of each node is represented by the cluster center value of each node.On this basis,the islanding mathematical model is established and the feasible area of islanding power supply for optical storage system is determined.Finally,taking the IEEE 33 node system as an example,the improved binary particle swarm optimization algorithm is used to solve the model,and the islanding division scheme in each fault recovery period is obtained.

关 键 词：光伏发电 配电网 孤岛 粒子群算法 模糊聚类 

分 类 号：TM734[电气工程—电力系统及自动化]