基于关联规则修正的电网不良运行数据辨识方法  

Identification Method of Power Grid Bad Operation Data Based on Association Rule Correction

作　　者：温炜 汪江 WEN Wei;WANG Jiang(Training Center of State Grid Ningxia Electric Power Co.,Ltd.,Yinchuan 750001 China)

机构地区：[1]国网宁夏电力有限公司培训中心,宁夏银川750001

出　　处：《自动化技术与应用》2024年第12期30-33,38,共5页Techniques of Automation and Applications

基　　金：国网宁夏电力有限公司培训中心教育培训开发项目(9129JP21N014)。

摘　　要：研究基于关联规则修正的电网不良运行数据辨识方法,保障电网的稳定运行与平稳持续供电。经模糊C均值聚类算法由电网历史运行数据中挖掘负荷数据,频域分解后获得四种负荷分量,运用温度因子线性回归模型,结合各分量对应时间的温度数据预测未来各分量负荷值,相加后得到总负荷基础预测值,运用温度变化幅度与电网负荷变化幅度的关联规则修正此基础预测值,运用修正后预测值实现电网实时运行不良数据的辨识。结果显示,该方法可依据关联规则修正所得负荷预测值,辨识出电网实时运行中的不良负荷数据,避免电网运行故障、保障其平稳供电。The identification method of power grid bad operation data based on association rule modification is studied to ensure the stable operation of power grid and steady and continuous power supply.The fuzzy C-means clustering algorithm is used to mine the load data from the historical operation data of the power grid,and four load components are obtained after frequency domain decomposition.The temperature factor linear regression model is used to predict the future load values of each component by combining the temperature data of the corresponding time of each component,and the basic predicted value of the total load is obtained after adding.Using temperature change and the change in power load correction based prediction,association rules using the revised forecast grid real-time data to identify defective operation,and the results show that this method can be fixed income load forecast based on association rules,identify the bad load data real-time power network,to avoid electrical fault,guarantee the stable power supply.

关 键 词：关联规则修正 电网 不良运行数据 模糊C均值 频域分解 线性回归 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TM737[自动化与计算机技术—控制科学与工程]