面向电力传动设备运作状态监测的数据融合方法  

ata fusion method for operation status monitoring of electric drive equipment

作　　者：柴耀军 王亮 魏刚 李茂盛 CHAI Yaojun;WANG Liang;WEI Gang;LI Maosheng(Zhunneng Group Gangue Power Generation Company,National Energy Group,Ordos 010300,China)

机构地区：[1]国家能源集团准能集团矸石发电公司,内蒙古鄂尔多斯010300

出　　处：《电子设计工程》2024年第18期101-104,109,共5页Electronic Design Engineering

摘　　要：电力传动设备通过监测平台判断其运行状态,而通过数据融合能够使平台监测信息处理结果更加精确,提高电网的稳定性。为此,设计面向电力传动设备运作状态监测的数据融合方法。利用CMSE准则改进EMD降噪技术,降噪处理传感器信号数据。通过多传感器时空配准将各传感器数据变换至同一个参考框架中。利用烟花算法改进BP神经网络,构建FWA-BP神经网络模型,实现电力传动设备运作状态监测数据融合。测试结果表明,设计方法设备监测数据融合的平均均方根误差低于1.0,平均绝对误差均值在1.5以下,设备运作状态监测的数据融合误差较小。Electric drive equipment can judge its operation status through the monitoring platform,and data fusion can make the platform monitoring information processing results more accurate and improve the stability of the power grid.Therefore,a data fusion method for monitoring the operation status of electric drive equipment is designed.The CMSE criterion is used to improve the EMD noise reduction technology and process the sensor signal data.The data of each sensor is transformed into the same reference frame through multi-sensor spatiotemporal registration.Using the fireworks algorithm to improve the BP neural network,build the FWA-BP neural network model,and realize the data fusion of the operation status monitoring of electric drive equipment.The test results show that the average root-mean-square error of equipment monitoring data fusion in the design method is less than 1.0,the average absolute error is less than 1.5,and the data fusion error of equipment operation status monitoring is small.

关 键 词：电力传动设备 CMSE EMD降噪技术 运作状态监测 烟花算法 数据融合 

分 类 号：TN929.5[电子电信—通信与信息系统]