基于改进XGBoost模型的极端场景配电网电能质量多种扰动源检测方法  

Multiple Disturbance Sources Detection Method for Power Quality in Extreme Scenario Distribution Network Based on Improved XGBoost Model

作　　者：吴斌 赵哲 郝晓辉 王荟敬 李景涛 WU Bin;ZHAO Zhe;HAO Xiaohui;WANG Huijing;LI Jingtao(Shijiazhuang Power Supply Branch of State Grid Hebei Electric Power Co.,Ltd.,Shijiazhuang 050000,China;School of Electrical Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,China)

机构地区：[1]国网河北省电力有限公司石家庄供电分公司,河北石家庄050000 [2]河北科技大学电气工程学院,河北石家庄050018

出　　处：《微型电脑应用》2024年第12期170-173,182,共5页Microcomputer Applications

摘　　要：针对三相电压波形有可能出现畸变或偏移的问题,为了提高扰动源的检测精度,精准控制导致三相电压波形畸变或偏移的扰动源,基于改进XGBoost模型提出一种极端场景配电网电能质量多种扰动源检测方法。在配电网环境下,通过对配电网运行信号的分析,检测配电网的电能质量。针对存在扰动现象的配电网信号,利用改进XGBoost模型提取存在扰动现象的配电网信号,对其进行特征匹配并确定类型,输出检测结果。实验表明:该方法能够将配电网在极端场景的电压偏差和频率偏差控制在预设值以下,具有良好的检测性能和应用性能。In order to improve the detection accuracy of disturbance sources and accurately control the distortion or offset of three-phase voltage waveforms,a multiple disturbance source detection method for extreme scenario distribution network power quality is proposed based on an improved XGBoost model to address the problem of distortion or offset in three-phase voltage waveforms.In the distribution network environment,the power quality of the distribution network is detected by the analysis of the distribution network operation signal.Aimed at the distribution network signals with disturbance phenomenon,the improved XGBoost model is used to extract the distribution network signals with disturbance phenomenon,match the feature and determine the type,and output the detection results.The experiment shows that the methed can control the voltage deviation and frequency deviation of the distribution network in extreme scenarios below the preset values,and has good detection performance and application performance.

关 键 词：改进XGBoost模型 极端场景 配电网电能质量 扰动源检测 

分 类 号：TM73[电气工程—电力系统及自动化]