摘要
温度作为评估电力电缆是否能安全、稳定运行的重要参数,但电缆受到安装环境等因素的影响,导致其电缆缆芯温度难以直接测得。基于此,提出依托有限元开展电缆缆芯温度预测的方法。下文在阐述有限元及BP神经网络相关概念基础上,利用有限元结合电缆热参数、边界条件等相关数据,创建电缆温度场有限元模型,并求出各环境及负荷参数下的缆芯温度,获得相应的样本数据。随之,借助样本数据训练神经网络构建相应的温度预测模型。实验结果证实,文中所提出的缆芯预测方法能准确预测其温度,能够对于电网实现主动预测性管理提供重要的参考。
Temperature is an important parameter to evaluate whether the power cable can operate safely and stably,but the cable is affected by the installation environment and other factors,resulting in the difficulty to directly measure the temperature of the cable core.Based on this,the method of cable core temperature prediction based on finite element is proposed.On the basis of explaining the relevant concept of finite element and BP neural network,the cable temperature field finite element model uses the relevant data,and find the cable core temperature under each environment and load parameters to obtain the corresponding sample data.Subsequently,the neural network is trained to build the corresponding temperature prediction model.The experimental results confirm that the proposed cable core prediction method can accurately predict the temperature and provide an important reference for realizing the active predictive management.
作者
谭静
吴叶弘
田鹏
TAN Jing;WU Ye-Hong;TIAN Peng(Direc t Curre nt Company,Hubei Elec trie Power Co.,L td,State Grid,Yi chang 443000;Shanghai Roye Electrical Co.,Ltd.,Shanghai 201802)
出处
《环境技术》
2023年第1期126-131,共6页
Environmental Technology
关键词
有限元
缆芯
电力电缆
温度预测
BP神经网络
finite element
cable core
power cable
temperature forecast
BP neural network
