基于特征选择的养殖行业异常用电行为检测  

Detection of Abnormal Electricity Consumption Behavior in the Aquaculture Industry Based on Feature Selection

作　　者：孙耀 易校石 Sun Yao;Yi Xiaoshi(Yili Normal University,Yining,China)

机构地区：[1]伊犁师范大学,新疆伊宁

出　　处：《科学技术创新》2024年第24期13-16,共4页Scientific and Technological Innovation

基　　金：大学生创新创业训练计划项目支持(S2023107640036)。

摘　　要：随着智能电网和物联网技术的发展,异常用电行为检测变得尤为重要。本文提出了一种基于TOPSIS特征选择的异常用电行为检测方法。利用养殖行业用户用电数据计算各特征与理想点和负理想点的相对接近度,对特征进行排序并选取最具代表性的特征。实验结果表明,该方法能够有效提升异常检测模型的性能,具有较高的准确性和可靠性,为养殖行业的用电安全管理提供了有力支持。With the development of smart grid and Internet of Things technology,abnormal electricity consumption behavior detection has become particularly important.This paper proposes an abnormal electricity consumption behavior detection method based on TOPSIS feature selection.By using electricity consumption data from users in the aquaculture industry to calculate the relative closeness between each feature and the ideal point and negative ideal point,the features are sorted and the most representative features are selected.The experimental results show that this method can effectively improve the performance of the anomaly detection model,with high accuracy and reliability,providing strong support for electricity safety management in the aquaculture industry.

关 键 词：TOPSIS 特征选择 异常用电行为检测 养殖行业 

分 类 号：TM73[电气工程—电力系统及自动化] TP311.1[自动化与计算机技术—计算机软件与理论]