基于改进K-means算法的电力大数据系统研发

Research and development of power big data system based on improved K-means algorithm

随着智能电网和大数据技术的发展,对电力大数据进行有效分析,为电力系统的运行提供相应的指导意见越来越重要。提出了基于改进K-means算法的电力大数据系统研发,根据电力大数据分析的实际需求情况,设计了电力大数据的功能模块,主要分为前端和后端两个模块,前端为人机交互界面,后端用于实现数据采集、存储、分析,以及与业务的接口功能。对传统的K-means算法进行了改进,实验对比发现,改进的K-means算法聚类更准确,误分率降低到1%以下。根据电力大数据的需求分析和功能模块设计,开发了一套电力大数据分析平台,平台在福建电力公司实际运行结果验证了此系统的实用性,采用改进的K-means算法对电力设备监测数据进行聚类分析,准确检测出各个设备当前的受污染状况,为电力工作人员实现对设备的管理规划提供了参考意见。

With the development of smart power grid and big data technology,it is more and more important to provide the corresponding guidance for the operation of power system.In this paper,the research and development of power big data system based on improved k-means algorithm is proposed.According to the actual demand of power data analysis,the function module of power big data is designed,mainly divided into the front and back end two modules.The front-end is the human-computer interaction interface,and the back-end is used to implement data acquisition,storage,analysis,and interface functions of the business.The traditional k-means algorithm was improved,and the experimental comparison found that the improved k-means algorithm clustering was more accurate and the error rate dropped below 1%.According to the demand analysis and function module design of power big data,a set of power data analysis platform is developed.The practical results of the platform in Fujian electric power company verified the practicability of the system.With the improved K-means algorithm for electric power equipment monitoring data clustering analysis,accurate detection of infected status of the current of each device,for electric power staff to realize the management of equipment planning provides guidance.