浙江电力2025，Vol.44Issue(9)：58-69,12.DOI:10.19585/j.zjdl.202509006
基于TimeVAE和迁移学习的综合能源系统负荷预测方法
A load forecasting method for IESs using TimeVAE and transfer learning
摘要
Abstract
In this paper,an integrated learning approach based on time variational autoencoder(TimeVAE)and transfer learning(TL)is proposed to address the accuracy degradation in integrated energy systems(IESs)load forecasting caused by insufficient historical data in newly built systems.The TimeVAE-based variational autoen-coder generates load data to enhance the diversity of the target-domain dataset,while the source-domain knowledge with abundant historical data is leveraged to optimize the target-domain load forecasting model through a frozen train-ing strategy.Based on IES data from multiple campuses of Arizona State University,the effectiveness of the pro-posed method is validated.Experimental results demonstrate that the method significantly improves load forecasting accuracy under few-shot conditions,providing critical references for achieving efficient and reliable IES load fore-casting.关键词
TimeVAE/迁移学习/小样本负荷预测/综合能源系统/数据增强/冻结训练策略/多能负荷预测Key words
TimeVAE/transfer learning/few-shot load forecasting/IES/data augmentation/frozen training strat-egy/multi-energy load forecasting引用本文复制引用
陈哲,周金辉,靳东辉,陈积光,马恒瑞,张嘉鑫,朱苏洵..基于TimeVAE和迁移学习的综合能源系统负荷预测方法[J].浙江电力,2025,44(9):58-69,12.基金项目
国家自然科学基金(61933005) （61933005）
国网浙江省电力有限公司科技项目(2006CB200303) （2006CB200303）
