摘要
针对目前变压器状态准确性和时效性不高的问题,提出了一种结合概率图模型和改进Apriori算法的关联规则挖掘方法,用于挖掘变压器状态参数的关联规则,以清晰表示数据之间的关联程度,减小计算量,提高关联规则挖掘效率。利用该方法分析四川某地500kV变压器的实测参数,挖掘状态参数之间的关联规则,并用挖掘出来的关联规则修正单个状态参数预测结果。结果表明,将关联规则应用于状态参数预测可提高预测精度,时效性和可行性较优,为变压器状态评估提供了一种新方法。
Aiming at the problem that the accuracy and timeliness of transformer state are not high at present,an association rule mining method based probability graph model and improved Apriori algorithm was proposed to mine the association rules of transformer state parameters,which can clearly express the degree of correlation between data,reduce the amount of calculation and improve the efficiency of association rule mining.The proposed method was used to analyze the measured parameters of the 500 kV transformer in Sichuan Province.The association rules between the state parameters were excavated and then the mining association rules were adopted to correct the single parameter prediction values.The results indicate that the prediction accuracy is obviously improved,and the method show a better effectiveness and feasibility,which provides a new method for transformer state assessment.
作者
周步祥
袁岳
张致强
黄河
ZHOU Bu-xiang;YUAN Yue;ZHANG Zhi-qiang;HUANG He(School of Electrical Engineering and Information,Sichuan University,Chengdu 610065,China)
出处
《水电能源科学》
北大核心
2019年第3期164-167,163,共5页
Water Resources and Power
关键词
变压器
关联规则
概率图模型
状态预测
transformer
association rules
probabilistic graph model
state prediction
