计及罕见变量的配电电缆线路故障风险预警方法  

Distribution cable line fault risk warning method with rare variables

作　　者：董章 李思尧 陈雅旎 刘惠 何明 DONG Zhang;LI Siyao;CHEN Yani;LIU Hui;HE Ming(Luohu Power Supply Bureau,Shenzhen Power Supply Bureau Co.,Ltd,Shenzhen,Guangdong 518000)

机构地区：[1]深圳供电局有限公司罗湖供电局,广东深圳518000

出　　处：《电气技术》2023年第5期35-40,共6页Electrical Engineering

摘　　要：本文将关联规则挖掘(ARM)算法应用于配电电缆线路的故障风险预警中。为将预测效果做进一步提升,提出一种计及罕见变量的关联规则挖掘方法。首先,针对故障在不同季节时段的分布,对传统ARM算法的重要度诊断标准阈值设定进行改进,使罕见季节时段中的风险元素也能够被挖掘;其次,根据罕见元素在不同环境特征下的分布,对传统形式的重要度诊断标准计算方法进行改进,以有效应对来自罕见环境元素的影响;最后,通过实例仿真将所提出的方法与传统ARM算法进行预测效果对比,仿真结果表明所提方法具有更好的准确性和可行性。In this paper,the association rule mining(ARM)algorithm is applied to the fault risk warning of distribution cable lines.To further improve the prediction effect,an association rule mining method that takes into account rare variables is proposed.Firstly,the importance threshold of the traditional ARM algorithm is improved for the distribution of faults in different seasonal periods,so that the risk elements in rare seasonal periods can also be mined.Secondly,according to the distribution of rare elements under different environmental characteristics,the traditional form of important diagnostic criteria calculation method is improved,so that it can effectively respond to the influence of rare environmental elements.Finally,the proposed method is compared with the traditional ARM algorithm by example simulation,and the simulation results show that the proposed method has better accuracy and feasibility.

关 键 词：电缆线路故障 关联规则挖掘(ARM) 罕见时段 罕见环境元素 故障风险预警 

分 类 号：TM75[电气工程—电力系统及自动化]