基于矩阵广义逆的电缆线芯温度动态计算方法

Dynamic Calculation of Cable Core Temperature Based on Generalized Inverse of Matrix

为更好地对电缆线芯温度进行间接测量,提出一种以电缆运行电流和表皮温度为输入的线芯温度动态计算方法。首先建立了电缆传热的简化热路模型,并在误差敏感性分析的基础上引入了线芯电阻随温度变化的二阶修正;然后对模型表征的微分方程进行离散化,得到仅有4个模型参数的计算公式;最后以电缆实验(或运行)数据为样本构造学习矩阵,并通过矩阵的广义逆计算模型参数,代入公式完成整个动态计算方法的构建。分别以室内实验电缆、在线110 k V高压电缆为对象进行了实验与分析,结果表明,相较于传统方法,该方法能够更加简单、准确地计算电缆线芯温度,有助于实现电力电缆工况的实时监测。

To implement indirect measurement of cable core temperature better, a dynamic calculation method, which takes cable’s operating current and surface temperature as inputs, to compute core temperature is proposed. Firstly, traditional thermal circuit model of cable heat transfer is highly simplified, and on the basis of error sensitivity analysis a two-order correction of cable core resistance varying with temperature is led in; secondly, the differential equation characterized by the simplified model is discretized to obtain the calculation formula with only 4 model parameters; finally, taking the experimental or operating data of cable as samples a learning matrix is constructed, and then the four parameters are estimated via generalized inverse of the learning matrix and the estimated values for model parameters are substituted into the calculation formula, thus the establishment of the whole dynamic calculation method is completed. Taking the cable for indoor experiment and the online 110kV high voltage cable as objects respectively, the experiments and analysis are performed and the results show that using the proposed method the cable core temperature can be calculated more simply and accurately, and it contributes to implement the real-time monitoring of operating conditions of the power cable.