基于PCA和XGBoost的D-PMU实时扰动预测研究
    点此下载全文
引用本文:袁智勇1,熊 瑶2,葛宁超2,秦 拯2,于 力1,徐 全1,林跃欢1.基于PCA和XGBoost的D-PMU实时扰动预测研究[J].计算技术与自动化,2022,(4):140-143
摘要点击次数: 200
全文下载次数: 0
作者单位
袁智勇1,熊 瑶2,葛宁超2,秦 拯2,于 力1,徐 全1,林跃欢1 (1.南方电网科学研究院广东 广州 5100802.湖南大学湖南 长沙 410082) 
中文摘要:随着具有高频数据流特性的D-PMU设备的广泛应用和普及,配电网系统中的量测数据量爆炸式增长,对大数据处理技术的要求越来越高。为了配电网能够可靠稳定运行,准确并及时对D-PMU存在的扰动进行在线预测分类显得非常重要。因此,提出了基于Spark的计算平台,采用PCA算法对D-PMU时间序列特征进行提取,结合XGBoost算法对D-PMU主要的扰动特征进行预测分类。实验结果表明提出的方法提高了D-PMU扰动分类的准确性,并且算法的计算速度也有显著的提升,确保了数据处理的实时性。
中文关键词:D-PMU流数据  Spark  扰动预测  XGBoost  PCA
 
Research on Real Time Disturbance Prediction of D-PMU Based on PCA and XGBoost
Abstract:With the wide application and popularization of D-PMU equipment with high frequency data flow characteristics, the amount of measurement data in the distribution network system grows explosively, and the demand for big data processing technology is getting higher and higher. In order to ensure the reliable and stable operation of distribution network, it is very important to accurately and timely predict and classify the disturbance existing in D-PMU online. Therefore,a computing platform based on spark is proposed, PCA algorithm is used to extract the features of D-PMU time series, and XGBoost algorithm is used to predict and classify the main disturbance features of D-PMU. The experimental results show that the proposed method improves the accuracy of D-PMU disturbance classification, and the computing speed of the algorithm is significantly improved to ensure the real-time data processing.
keywords:D-PMU flow data  Spark  disturbance prediction  XGBoost  PCA
查看全文   查看/发表评论   下载pdf阅读器