PCA和CNN的电力计量装置断流故障自动监测方法  

Automatic monitoring method for power metering device interruption fault based on PCA and CNN

作　　者：杨晓波 韩硕辰 刘岩 刘志凯 YANG Xiao-bo;HAN Shuo-chen;LIU Yan;LIU Zhi-kai(State Grid JiBei Marketing Service Center,Beijing 100045,China;State Grid JiBei Electric Power Co.,Ltd.,Beijing 100052,China;Beijing China-Power Information Technology Co.,Ltd.,Beijing 100020,China)

机构地区：[1]国网冀北电力有限公司营销服务中心,北京100045 [2]国网冀北电力有限公司,北京100052 [3]北京中电普华信息技术有限公司,北京100020

出　　处：《信息技术》2025年第2期110-115,共6页Information Technology

摘　　要：针对电力计量装置在测量机械设备电能消耗的过程中易出现断流故障的问题,文中提出PCA和CNN的电力计量装置断流故障自动监测方法。填充处理电力计量装置运行数据缺失值,获取完整电力计量装置运行数据,分析数据的相关性,进行数据集分段。利用PCA提取故障数据特征,结合CNN建立故障监测模型,采用差分进化方法优化监测模型参数,将故障特征输入模型中,获取模型训练最佳输出值,实现电力计量装置断流故障精准监测。实验结果表明,该方法可缩减装置故障监测时的收敛速度,减少迭代次数和误报数量,能有效提高故障监测效果和性能。To solve the problem that power metering devices are prone to cause current interuption in the process of measuring the power consumption of mechanical equipment,an automatic monitoring method of power metering devices’current interuption fault based on PCA and CNN is proposed.The missing values of the operation data of the power metering device is filled in to obtain complete operation data of the power metering device for analyzing the correlation of the data and segmenting the dataset.Nextly,PCA is used to extract fault data features,and a fault monitoring model is established by using CNN.The monitoring model parameters are optimized using differential evolution method,and fault features are input into the model to obtain the optimal output value of model training,and achieve accurate monitoring of power metering device interruption faults.The experiment results show that this method can reduce the convergence speed during device fault monitoring,reduce the number of iterations and false alarms,and effectively improve the effectiveness and performance of fault monitoring.

关 键 词：PCA CNN 电力计量装置 断流故障 自动监测 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]