改进DARMmcr模型的输电线路安全运行数据比较分析  

Comparative analysis of transmission line safety operation data based on improved DARMmcr model

作　　者：王慧 孙国法 付龙海[1] WANG Hui;SUN Guofa;FU Longhai(Yantai Vocational College,Yantai 264000,Shandong,China;Qingdao University of Technology Qingdao 266000,Shandong,China)

机构地区：[1]烟台职业学院,山东烟台264000 [2]青岛理工大学,山东青岛266000

出　　处：《粘接》2023年第5期183-187,共5页Adhesion

基　　金：山东省高等学校青创人才引育计划团队项目(项目编号:鲁教人字[2019]号-23);烟台职业学院2022年度校本课题(项目编号:YZXB01)。

摘　　要：针对现有故障预测模型无法兼顾不同类型数据的特性,预测精度和效率较低等问题,提出了一种将计及罕见变量的动态关联规则挖掘模型和模糊推理系统相结合用于输电线路故障预测。使用外部环境数据(连续和离散特征)用作输入数据,连续特征由集成模型处理,离散特征由计及罕见变量的动态关联规则挖掘模型处理。通过算例对预测方法的性能进行比较分析,验证了该方法的优越性。结果表明,与传统的预测方法相比,该预测模型考虑了不同类型输入数据的特征,可以进一步提高预测效果,有一定的参考价值。Aiming at the problems that the existing fault prediction models can not take into account the characteris⁃tics of different types of data and low prediction accuracy and efficiency,a dynamic association rule mining model considering rare variables and fuzzy reasoning system were proposed for transmission line fault prediction.External environment data(continuous and discrete features)were used as input data,with the continuous features pro⁃cessed by the integrated model,and the discrete features processed by the dynamic association rule mining model considering rare variables.The performance of the prediction method was compared and analyzed through an exam⁃ple to verify the superiority of the method.The results showed that compared with the traditional prediction meth⁃ods,the prediction model considered the characteristics of different types of input data,which could further im⁃prove the prediction effect and has a certain reference value.

关 键 词：输电线路 故障预测 环境数据 动态关联 推理系统 

分 类 号：TM711.2[电气工程—电力系统及自动化]