基于DC-CNN-PE-SSA-Informer的电缆缆芯温度预测研究  

Cable Core Temperature Prediction Based on DC-CNN-PE-SSA-Informer

作　　者：鲍克勤 赵欣妍 刘擘 王仕博[3] 郝海斌 BAO Keqin;ZHAO Xinyan;LIU Bo;WANG Shibo;HAO Haibin(Faculty of Automation Engineering,Shanghai University of Electric Power,Shanghai 200090,China;China Ruilin Engineering Technology Co.,Ltd.,Nanchang 330038,China;Faculty of Metallurgical and Energy Engineering,Kunming University of Science and Technology,Kunming 650093,China)

机构地区：[1]上海电力大学自动化工程学院,上海200090 [2]中国瑞林工程技术有限公司,江西南昌330038 [3]昆明理工大学冶金与能源工程学院,云南昆明650093

出　　处：《昆明理工大学学报(自然科学版)》2025年第2期116-125,共10页Journal of Kunming University of Science and Technology(Natural Science)

基　　金：上海市自然基金面上项目(20ZR1421300);上海市电站自动化技术重点实验室项目(13D0Z2273800)。

摘　　要：针对电缆缆芯温度不易直接测量,且预测精确度不足的问题,本文提出了DC-CNN-PE-SSA-Informer混合预测模型,该模型利用扩展因果卷积网络(DC-CNN)增强对时间序列数据局部特征的捕捉能力,并将提取的特征传递至Informer模块以捕获长期依赖关系,通过引入相对位置编码(PE)加强Informer模型对时间序列中相对位置信息的捕捉能力,最后由麻雀搜索算法(SSA)进行参数优化。通过对电缆温度场进行有限元分析,求解出不同条件下的缆芯温度作为仿真实验的样本数据。仿真结果表明,DC-CNN-PE-SSA-Informer模型相比常见的预测模型在电缆缆芯温度预测方面具有更高的预测精度,为电力调度的运行方式提供了依据。In response to the challenges of directly measuring and accurately predicting cable core temperatures,this paper proposes a hybrid predictive model named DC-CNN-PE-SSA-Informer.This model employs a Dilated Causal Convolutional Neural Network(DC-CNN)to enhance the capture of local features in time series data and conveys these features to the Informer module to grasp long-term dependencies.By integrating Relative Positional Encoding(PE),the model strengthens the Informer’s ability to discern relative positional information within the time series.Optimization of parameters is performed using the Sparrow Search Algorithm(SSA).Finite element analysis of the cable temperature field was conducted to solve for core temperatures under various conditions,serving as sample data for simulation experiments.The simulation results indicate that the DC-CNN-PE-SSA-Informer model achieves higher predictive accuracy in forecasting cable core temperatures compared to common predictive models,and provides a basis for the operational methods of power dispatch.

关 键 词：电力电缆 温度预测 扩展因果卷积网络(DC-CNN) INFORMER 麻雀搜索算法(SSA) 位置编码(PE) 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TM75[自动化与计算机技术—控制科学与工程]