基于数据中台的智能化电力监控数据应用模型

Intelligent power monitoring data application model based on data center

为了精准采集低开销智能化电力监控数据,有效应用电力监控数据,评估智能化电网运行状态,设计基于数据中台的智能化电力监控数据应用模型,该模型中数据技术层利用基于压缩传感的数据采集方法,采集低开销智能化电力监控数据,按照结构化与非结构化数据形式存储采集的数据;统一数据层利用贴源层按照存储数据类型,建立数据库,分类存储各类型数据;通过共享层清洗整合各类型数据,获取各类型数据的业务明细数据;采用分析层分析业务明细数据,为数据分析应用提供所需数据;通过统一数据服务引擎为不同数据分析应用提供数据服务支撑;数据分析应用中利用K-means算法与多元线性回归法预测智能化电网线损,通过博弈论集对云方法,评估与预测智能化电网运行状态。实验结果证明:该模型可精准采集低开销电力监控数据,在不同数据稀疏度与压缩长度时,数据采集的累积相对估计误差均较低;可有效聚类线损相关特征属性数据,精准预测电网线损,有效评估电网运行状态。

In order to accurately collect low-cost intelligent power monitoring data, effectively apply power monitoring data and evaluate the operation status of intelligent power grid, an intelligent power monitoring data application model based on data center is designed. In the model, the data technology layer uses the data acquisition method based on compressed sensing to collect low-cost intelligent power monitoring data, and stores the collected data in the form of structured and unstructured data;The unified data layer uses the paste source layer to establish a database according to the stored data type and store various types of data by classification;Clean and integrate various types of data through the sharing layer to obtain the business details of various types of data;The analysis layer is used to analyze the business detail data to provide the required data for the data analysis application;Provide data service support for different data analysis applications through unified data service engine;In the application of data analysis, K-means algorithm and multiple linear regression method are used to predict the line loss of intelligent power grid, and the operation state of intelligent power grid is evaluated and predicted by game theory set pair cloud method. The experimental results show that the model can accurately collect low overhead power monitoring data, and the cumulative relative estimation error of data acquisition is low under different data sparsity and compression length;It can effectively cluster the characteristic attribute data related to line loss, accurately predict the line loss of power grid and effectively evaluate the operation state of power grid.