基于Stacking集成学习的电力系统故障诊断  

Power System Fault Diagnosis Based on Stacking Ensemble Learning

作　　者：徐福聪 李泽科 范海威 许力[2] 张章学 XU Fucong;LI Zeke;FAN Haiwei;XU Li;ZHANG Zhangxue(State Grid Fujian Electric Power Co.,Ltd.,Fuzhou 350003,China;Fujian Provincial Key Lab of Network Security and Cryptology,Fujian Normal University,Fuzhou 350117,China;Fujian Strait Information Technology Co.,Ltd.,Fuzhou 350003,China)

机构地区：[1]国网福建省电力有限公司,福建福州350003 [2]福建师范大学福建省网络安全与密码技术重点实验室,福建福州350117 [3]福建省海峡信息技术有限公司,福建福州350003

出　　处：《福建师范大学学报（自然科学版）》2024年第6期40-46,共7页Journal of Fujian Normal University：Natural Science Edition

基　　金：国家自然科学基金资助项目(62471139);中央引导地方科技发展专项(2021L3032)。

摘　　要：针对电力系统故障分类复杂度高、特征选择难度大、用单一模型进行预测诊断的准确率低的问题,提出一种基于Stacking集成学习的电力系统故障诊断的方法。通过对电路中的三相电流和电压进行分析并提取故障特征,采用Stacking集成学习方法对不同类型模型进行融合,并且通过硬投票法对电力系统中的故障进行检测与分类,以解决传统单一模型在电路系统故障类型检测中准确率低的问题。模型分类结果表明,基于多分类器融合的方法能够有效地检测和分类输电线路上的故障,所提模型的准确率可以达到97.8%,可以有效地分类故障类型。In order to address the challenges of high complexity in power system fault classification,the difficulty in feature selection,and the low accuracy in predictive diagnosis using single models,this paper proposes a method for power system fault diagnosis based on Stacking ensemble learning.By analyzing the three-phase current and voltage in circuits and extracting fault features,this method applies Stacking ensemble learning to fuse different types of models,and a hard voting method is used to detect and classify faults within the power system to overcome the accuracy limitations of traditional single-model approaches in circuit system fault detection.The model classification results show that the multi-classifier fusion approach effectively detects and classifies faults in transmission lines,achieving an accuracy of up to 97.8%in fault type classification.

关 键 词：故障分类 Stacking融合 硬投票 电力系统 

分 类 号：TM73[电气工程—电力系统及自动化]