摘要
当前配电网单相接地故障检测模型知识表达能力较差,其产生过程只依据正常配电网样本数据,无需单相接地故障数据,未充分利用先验知识,导致检测精度低。提出新型配电网单相接地故障自动检测模型。提取配电网单相接地故障特征,设计配电网单相接地故障自动化检测模型,将故障特征向量作为输入向量,引入输入向量与权值向量元素之间的匹配程度,将其作为第二层节点的输入,将第二层的输出作为第三层的输入,对模型进行训练,输出的结果即为配电网单相接地故障检测结果。实验结果表明,所提模型具有很高的检测精度。
The current distribution network of single-phase grounding fault detection model knowledge expres-sion ability is poorer, the production process on the basis of normal distribution network sample data, only need sin-gle phase ground fault data, did not make full use of prior knowledge, lead to low accuracy. New power single-phase earth fault automatic detection model is put forward. Power single-phase earth fault feature extracting, the de-sign of distribution network automated testing model of single-phase earth fault fault feature vector as input vector, the introduction of input vector and weight vector elements, the matching degree between them as the second node of the input and the output of the second layer as the third layer of input, was carried out on the model training. The output is the result of the distribution network of single-phase grounding fault detection results. The experimen-tal results show that the proposed model is of high precision.
出处
《科学技术与工程》
北大核心
2017年第4期218-222,共5页
Science Technology and Engineering
关键词
配电网
单相接地故障
自动化
检测
power distribution network single phase ground fault automation detection
