The current electricity metering device is composed of power transformers,energy meters,and secondary connections.Due to the different attributes of data from multiple devices,and the intermediate links that can easily lead to errors,power theft,and other factors affecting the results of automatic data integration.Propose an automatic data integration method for electric energy mete-ring devices based on SOM network.Firstly,the training process of the SOM network is utilized to continuously change the connection weights of neurons,calculate and complete the classification and processing of data from electric energy metering devices;Then,ana-lyze the constraints of automatic integration,and under the constraints of data attribute parameters(i.e."T value")and device ter-minal parameters(i.e."M value"),decompose the data integration task into multiple task subsets to obtain the attributes of the e-lectricity metering device data,ensuring the consistency and reliability of the classified data;Finally,according to the constraints,the automatic integration of terminal data is achieved through the autonomous learning ability of the SOM network.The experimental results show that the method proposed in this paper has a low packet loss rate and error,and can efficiently complete the task of auto-matic data integration for power metering devices.
关键词
自组织映射神经网络/电能计量装置/数据自动集成/自主学习训练/任务终端矩阵
Key words
self-organizing mapping neural network/electric energy metering device/automatic data integration/autonomous learning training/task terminal matrix
维普
万方数据