广东电力2025，Vol.38Issue(3)：104-112,9.DOI:10.3969/j.issn.1007-290X.2025.03.011
基于FP-Growth数据挖掘的直流串联故障电弧特征提取及检测方法
Feature Extraction and Detection Method of DC Series Fault Arc Based on FP-Growth Data Mining
摘要
Abstract
In order to solve the problem that traditional time-frequency analysis methods are difficult to effectively extract the fault arc characteristics under various electrode materials,leading to the failure of fault arc detection devices and causing fires,this paper proposes a DC fault arc detection algorithm based on frequent pattern growth(FP-Growth)data mining,which enables effective extraction of features under multiple electrode materials.Based on the improved particle swarm optimization(IPSO),the accuracy rate,detection time and memory size of the detection algorithm are collaboratively optimized to achieve the selection of the optimal parameters of the fault arc detection algorithm under the condition of limited hardware resources.Test results show that this method can quickly and effectively detect DC series fault arcs under different electrode materials,providing references for the hardware implementation of fault arc feature extraction and detection methods based on material differences.关键词
故障电弧/频繁模式增长/协同优化/改进粒子群/硬件实现/机器学习Key words
fault arc/frequent pattern growth(FP-Growth)/collaborative optimization/improved particle swarm optimization(IPSO)/hardware implementation/machine learning分类
信息技术与安全科学引用本文复制引用
杨晓华,费正源,代盛国,刘家欣,陈思磊,李兴文..基于FP-Growth数据挖掘的直流串联故障电弧特征提取及检测方法[J].广东电力,2025,38(3):104-112,9.基金项目
云南电网有限责任公司科技项目(YNKJXM20222147) （YNKJXM20222147）
