一种基于数字孪生技术的水电企业数据采集传输方法  

A data collection and transmission method for hydropower enterprises based on digital twin technology

作　　者：王刚 WANG Gang(China Yangtze Power Co.,Ltd.,Yichang 443002,China)

机构地区：[1]中国长江电力股份有限公司,湖北宜昌443002

出　　处：《电子设计工程》2024年第23期71-75,共5页Electronic Design Engineering

基　　金：长江电力大数据平台建设项目(1619010001)。

摘　　要：输电线路热容量的限制对水电企业的安全可靠性起着至关重要的作用。为了准确采集、传输并估计输电线路的温度数据,现有的基于物理的标准是根据多个传感器测量的环境和线路条件来实现,计算成本高、估计精度低,基于此,提出基于数字孪生技术的数据采集与传输方法。该方法通过物理传感器数据与实际导体温度数据之间的输入输出关系来进行机器学习,作为基于物理标准的数字等效。通过对实际数据的实验评估,将所提出的方法与IEEE738标准进行比较,结果表明,该方法可明显降低预测温度的均方根误差,数字孪生体提供了更准确和可靠的估计,可以作为传统方法的补充或潜在替代方法。The limitation of thermal capacity of transmission lines plays a crucial role in the safety and reliability of hydropower enterprises.In order to accurately collect,transmit,and estimate temperature data of transmission lines,existing physical based standards are implemented based on the environment and line conditions measured by multiple sensors,with high computational costs and low estimation accuracy.Based on this,a data collection and transmission method based on digital twin technology is proposed.This method uses the input-output relationship between physical sensor data and actual conductor temperature data for machine learning,as a digital equivalent based on physical standards.Through experimental evaluation of actual data,the proposed method was compared with the IEEE 738 standard.The results showed that this method can significantly reduce root mean square error of the predicted temperature,and digital twins provide more accurate and reliable estimates,which can serve as a supplement or potential alternative to traditional methods.

关 键 词：动态热线额定值 数据驱动 数字孪生技术 机器学习 

分 类 号：TN912.3[电子电信—通信与信息系统]