基于模糊测度融合算法的配电网接地故障判断研究

Research on ground fault diagnosis of distribution network based on fuzzy measure fusion algorithm

配电网接地故障类型复杂多样,误判会严重威胁配电网的运行安全,论文基于模糊测度融合算法提出了一种新的配电网接地故障判断方法。对历史数据进行模糊聚类分析,获得故障类和非故障类聚类中心。采用两种样本相似度测度来分析待测样本集与历史样本之间的相似度,形成模糊测度融合矩阵,通过二级评判指标体系获得二级故障测度综合评判矩阵和一级故障测度综合评判矩阵,实现对配电网接地故障的判断。将提出的配电网接地故障判断模型应用于10kV配电网实物试验中,结果表明该方法能够有效地对配电网故障进行判断,具有良好的适应性和较强的鲁棒性。

The types of distribution network grounding faults are complex and diverse,and misjudgment will seriously threaten the operation safety of distribution network.Based on fuzzy measure fusion algorithm,a new distribution network grounding fault judgment method is proposed in this paper.Fuzzy clustering analysis is carried out on the historical data to obtain the clustering centers of fault class and non-fault class.Two sample similarity measures are used to analyze the similarity between the sample set to be tested and the historical samples to form a fuzzy measure fusion matrix.Through the two-level evaluation index system,the two-level fault measure comprehensive evaluation matrix and the one-level fault measure comprehensive evaluation matrix are obtained to judge the grounding fault of distribution network.The proposed distribution network grounding fault judgment model is applied to the physical test of 10kV distribution network.The results show that this method can effectively judge the distribution network fault,and has good adaptability and strong robustness.