摘要
为挖掘电网大数据的实用价值,探索电网故障的时空分布特征及其风险影响因素,本文采用时空热点趋势分析相关分析和地理加权回归分析等多种方法,对广州市2016年至2018年的电网故障数据进行研究。研究结果表明:广州电网故障具有峰值时段在6月至8月等时间分布规律;空间分布具有电网故障高密度区域逐年向中心城区集中和电网故障分布范围逐年扩大等特征,且时空热点趋势表明中心城区仍将是电网故障热点区域。在风险分析方面,强风强雨天气与电网故障数量总体呈现出较强的正向相关关系,高温与电网故障数量的相关性表现为强度逐年降减弱的正向相关;各类兴趣点对电网故障风险影响的差异水平不同,其中电力通信设施、政府机关和科教文化设施类兴趣点影响的区位差异较大。
In order to explore the practical value of power grid big data,the spatiotemporal distribution characteristics of power grid faults and risk factors,this paper adopts various methods such as spatial and temporal hot spot trend analysis,correlation analysis and geographical weighted regression analysis,etc.,the fault data of Guangzhou power grid from 2016 to 2018 are studied.The results show that Guangzhou power grid fault has the distribution rule of peak period from June to August;the spatial distribution is characterized by the centralization of high-density grid fault areas to the central area and the expansion of grid fault distribution scope year by year.Moreover,the spatiotemporal hotspot trend indicates that the central area will remain the power grid fault hotspot.In terms of risk analysis,there is a strong positive correlation between strong wind and rain weather and the number of power grid faults on the whole,while the correlation between high temperature and the number of power grid faults is a positive correlation of decreasing intensity year by year.The influence of various points of interest on the power grid fault risk are different,among which the influence of power communication facilities,government agencies and scientific,educational and cultural facilities is greatly different.
作者
裴求根
杨舒涵
卢宾宾
PEI Qiugen;YANG Shuhan;LU Binbin(Guangdong Power Grid Co.,Ltd.,Guangzhou 510060 Guangdong,China;School of Remote Sensing and Engineering,Wuhan University,Wuhan 430072 Hubei,China)
出处
《电力大数据》
2020年第11期1-8,共8页
Power Systems and Big Data
基金
国家自然科学基金项目(41871287,42071368)。
关键词
电网故障分布
电网故障风险
时空分布
回归分析
兴趣点
power grid fault distribution
power grid fault risk
spatiotemporal distribution
regression analysis
points of interest
