基于电力大数据平台的三层数据过滤机制研究及应用  

Research on three-layer data filter based on big data platform in digital planning of urban grid

作　　者：郑毅[1] 李元楷 李强 王婧 李婧 李温静 杨镜非[1] ZHENG Yi;LI Yuankai;LI Qiang;WANG Jing;LI Jing;LI Wenjing;YANG Jingfei(Derpartment of Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;School of Economics Fudan University,Shanghai 200433,China;State Grid Information&Telecommunication Group Co.,Ltd.,Beijing 102211,China)

机构地区：[1]上海交通大学电气工程系,上海200240 [2]复旦大学经济学院,上海200433 [3]国网信息通信产业集团,北京102211

出　　处：《供用电》2022年第3期32-39,共8页Distribution & Utilization

基　　金：国家自然科学基金项目“基于互逆映射对偶性的电力网络方程多解和可解性问题研究”(51777121)。

摘　　要：大数据技术是城市电网数字化规划中的重要工作内容,具有重要的理论和实践价值。但是这些数据体积大、类型多、更新速度快,为防止海量数据带来的分析干扰,提出了基于电力大数据平台的三层数据过滤机制,对过滤数据构建分段式小样本回归模型。该模型在上海某风力发电厂风机发电功率预测中进行应用,算例结果表明三层数据过滤机制有效地解决了庞大数据量的无序、可用性较低等问题,对整体数据具有良好的表征性,提升了预测精度,促进了电网信息价值密度的高质量提升。The big data technology is an important work content in the digital planning of urban grids and has significant theoretical and practical value.However,this data is large,diverse,and fast in updating.To prevent the analysis interference caused by massive data,this paper proposes a three-layer data filtering mechanism based on the power big data platform to build a segmented small sample regression model for the filtered data.The model is applied to predicting wind turbine power generation power of a wind power plant in Shanghai.The example results show that the three-layer data filtering mechanism effectively solves the problems of disorder and low availability of enormous data volume.This mechanism also has the excellent characterization of the overall data,improves the prediction accuracy,and promotes the high-quality improvement of the power grid information's value density.

关 键 词：数据窄窗口 数据过滤 关联度分析 电力大数据平台 价值密度 

分 类 号：TM72[电气工程—电力系统及自动化]