摘要
针对典型遗传算法处理拓扑动态变化配电网故障区段定位及隔离问题时存在的若干问题,提出了一种基于改进遗传算法的拓扑动态变化配电网故障区段定位及隔离模型。引入专用适应度函数,采集训练数据预测误差并映射为个体适应度值,建立最优适应度值与最优个体的逻辑对应关系,采用BP神经网络捕获最优个体并对网络初始权值和阈值进行赋值,借助定位及隔离函数输出目标结果。对模型进行效能仿真验证与工程应用分析,可以较好实现拓扑动态变化配电网故障区段精确定位与自动隔离,具有故障信息感知全面、拓扑动态变化适应性强、自主决策性高等优势。
In view of some questions of typical genetic algorithm in dealing with the problem of fault section location and isolation in distribution network with topology dynamic change,a fault section location and isolation model based on improved genetic algorithm is proposed.The special fitness function is introduced to collect the prediction error of training data and map it to individual fitness value.The logical correspondence between the optimal fitness value and the optimal individual is established.The BP neural network captures the optimal individual and assigns the initial weight and threshold value of the network.With the help of the positioning and isolation function,the objective result is output.Through the effectiveness simulation verification and engineering application analysis of the model,it can achieve the accurate location and automatic isolation of the fault section of the distribution network with dynamic topology change,and has the advantages of comprehensive fault information perception,strong adaptability to dynamic topology change,and high autonomy in decision⁃making.
作者
陈颢
陆玉军
刘恢
李澄
葛永高
王宁
CHEN Hao;LU Yujun;LIU Hui;LI Cheng;GE Yonggao;WANG ning(Jiangsu Frontier Electric Power Technology Co.,Ltd.,Nanjing 211100,China;Jiangsu Runhe Zhirong Technology Co.,Ltd.,Nanjing 210012,China)
出处
《电子设计工程》
2020年第21期114-119,共6页
Electronic Design Engineering
基金
江苏方天电力技术有限公司研发基金项目(YF201901)。
关键词
低压配电网
故障区段定位
故障隔离
改进遗传算法
BP神经网络算法
low voltage distribution network
fault location
fault isolation
improved genetic algorithm
BP neural network algorithm
