工程设计学报2025，Vol.32Issue(3)：281-295,15.DOI:10.3785/j.issn.1006-754X.2025.04.179
基于数字孪生的变压器热点温度预测预警技术研究
Research on transformer hotspot temperature prediction and warning technology based on digital twin
摘要
Abstract
The hotspot temperature of transformers has a direct impact on the reliability and stability of the power grid system.In response to the problems of complex traditional transformer management mode and high cost,low computational efficiency and high computational error in the transformer hotspot temperature prediction methods,a transformer hotspot temperature prediction and warning technology based on digital twin is proposed.Firstly,a six-dimensional digital twin model of the transformer was built to achieve functions such as system data sharing,multi-source fusion and virtual-real interaction.Then,a digital twin system driven by perception-interaction that could support artificial intelligence and machine learning algorithms was constructed.The chaotic adaptive particle swarm optimization(CAPSO)algorithm was adopted to optimize the weights and thresholds of the BP(back propagation)neural network,which accelerated the convergence speed of the original network.Meanwhile,a transformer hotspot temperature prediction model based on CAPSO-BP was established.Finally,the on-site monitoring data of transformers were used for simulation on the virtual engine platform,and the development and application of various functions of the transformer hotspot temperature prediction and warning system were implemented.Concurrently,the feasibility and effectiveness of the prediction model were verified.The research results provide new ideas and theoretical basis for the transformation of the digital twin transformer system from digitalization to intelligence.关键词
变压器/数字孪生/人工智能/机器学习/混沌自适应粒子群优化/反向传播神经网络/温度预测Key words
transformer/digital twin/artificial intelligence/machine learning/chaotic adaptive particle swarm optimization/back propagation neural network/temperature prediction分类
信息技术与安全科学引用本文复制引用
李佰霖,马云帆,陈昱锐,罗远林,褚凡武,付文龙..基于数字孪生的变压器热点温度预测预警技术研究[J].工程设计学报,2025,32(3):281-295,15.基金项目
国家自然科学基金资助项目(51741907) （51741907）
梯级水电站运行与控制湖北省重点实验室开放基金资助项目(2021KJX04) （2021KJX04）
