基于GBDT算法的变电器重过载精准预测研究  

Research on Accurate Prediction of Transformer Heavy Overload Based on GBDT Algorithm

作　　者：李淑玲[1] LI Shuling(Xi'an Eurasia University,Xi'an 710065,China)

机构地区：[1]西安欧亚学院,陕西西安710065

出　　处：《现代信息科技》2023年第7期144-146,共3页Modern Information Technology

基　　金：2021年陕西省计算机教育学会项目(21-18)。

摘　　要：为了实现变电器稳定安全的运行,解决设备预警的时效性差、精准度低、故障后抢修成本高等问题,文章基于配电设备历史运行数据和机器学习等相关知识,采用梯度提升树(GBDT)算法,对设备重过载情况进行预测。研究结果表明,建立设备运行状态智能感知模型和设备重过载预测模型,能够精准预测设备重过载现象,优化设备检修维护策略,降低设备检修维护成本,可实现保障电网经济运行的目的。In order to realize the stable and safe operation of the transformer and solve the problems of poor timeliness,low accuracy and high cost of emergency repair after failure of the equipment early warning,this paper uses the gradient lifting tree(GBDT)algorithm to predict the equipment overload based on the historical operation data of the distribution equipment and machine learning and other relevant knowledge.The research results show that the establishment of intelligent perception model of equipment operation status and equipment heavy overload prediction model can accurately predict the phenomenon of equipment heavy overload,optimize equipment maintenance strategies,reduce equipment maintenance costs,and can achieve the purpose of ensuring the economic operation of the power grid.

关 键 词：变电器 重过载 GBDT算法 

分 类 号：TP39[自动化与计算机技术—计算机应用技术] TM715[自动化与计算机技术—计算机科学与技术]