摘要
为降低电网数据挖掘难度,确保电网高效运行,构建基于网络拓扑的电网资源挖掘推荐模型。分析电网系统基本组成结构,明确电网资源来源,通过自下向上的形式构建资源库,确保数据有效储存;将网络拓扑描述为有向图形式,利用二维邻接矩阵表述相邻节点之间的连接状态,为资源推荐提供有效路径;经过归一化与k—均值聚类处理,统一资源格式,实现相同类型资源聚类;利用爬虫技术设置挖掘权限匹配阈值,结合抓取规则完成聚类谱智能排序,实现在爬虫协议控制下电网资源能够被准确挖掘推荐。仿真实验表明,该方法可提高数据预处理效果,减少延迟推荐现象,同时推荐结果的平均绝对误差较低。
In order to reduce the difficulty of power grid data mining and ensure the efficient operation of power grid,a power grid resource mining recommendation model based on network topology is constructed.Analyze the basic composition and structure of the power grid system,clarify the source of power grid resources,and build a resource database in the form of bottom-up to ensure the effective storage of data.The network topology is described as a directed graph,and the connection state between adjacent nodes is expressed by two-dimensional adjacency matrix,which provides an effective path for resource recommendation.After normalization and k-means clustering,the resource format is unified to realize the clustering of the same type of resources.The crawler technology is used to set the mining permission matching threshold,combined with the capture rules to complete the intelligent sorting of the clustering spectrum,so that the power grid resources can be accurately mined and recommended under the control of the crawler protocol.Simulation results show that this method can improve the effect of data preprocessing,reduce the phenomenon of delayed recommendation,and the average absolute error of recommendation results is low.
作者
赵越
程伟华
赵申
吴小虎
Zhao Yue;Cheng Weihua;Zhao Shen;Wu Xiaohu(State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210000,China;Jiangsu Electric Power Information Technology Co.,Ltd.,Nanjing 210000,China)
出处
《能源与环保》
2022年第5期174-179,186,共7页
CHINA ENERGY AND ENVIRONMENTAL PROTECTION
关键词
网络拓扑
电网资源
数据挖掘推荐
聚类算法
网络爬虫
network topology
power grid resources
data mining recommendation
clustering algorithm
web worm
