基于数据负荷序列聚类的配电网电力营销实时信息系统测试

The Construction of Distribution Network Power Marketing Real-Time Information System Based on Data Load Sequence Clustering

伴随现代信息技术的快速发展,在电力营销工作中构建信息系统对优化电网正常工作、推动营销活动顺利实施具有十分关键的影响。构建一种基于数据负荷序列聚类的配电网电力营销实时信息系统,当数据接口模块通过现场设备采集电力数据之后,传输至数据操作模块进行数据统计、存储并传输至基本操作模块,设备管理人员可以在基本操作模块刷新配电网电力数据;然后基于该数据通过预警模块使用基于数据挖掘多层次细节分解的负荷序列聚类分析方法,预测电力信息是否存在异常,若存在异常情况,预警模块将及时通知设备管理人员实施修复。经验证,在谐波、电压闪变两种不同的异常电力环境下,所构建系统预测精度较高、时延较短,可实现高精度、高效率的配电网电力营销实时信息管理。

With the rapid development of modern information technology,the construction of information system in power marketing has very important impacts on optimizing the normal work of power grid and promoting the smooth implementation of marketing activities.In this paper,a realtime information system of distribution network power marketing based on data load sequence clustering is constructed.After the power data is collected through the field equipment it is transmitted to the data operation module to implement data statistics,storage and transmission to the basic operation module.The equipment manager can refresh the power data on the basic operation module.Basis on the refreshed data,the load sequence clustering analysis method based on the multilevel detail decomposition of data mining is used to predict whether there is any abnormality in the power information.In case of an abnormality,the early warning module will inform the equipment manager in real time to implement the repair.It has been proved that the system can achieve high accuracy and high efficiency of the real-time information management of distribution network power marketing.