Volume 2,Issue 3
数据驱动的风电功率在线预测方法
随着风电装机容量的快速增长和新能源大规模并网运行，对风电功率预测的实时性与准确性提出了更高要求，传统离线建模方法难以适应运行环境的动态变化，为此本文提出一种基于深度学习的风电功率在线预测方法。首先，利用历史数据对预测模型进行批量训练，构建初始模型；然后引入Wasserstein 距离度量新数据与历史数据之间的分布差异，并据此检测数据分布是否发生变化；当检测到显著变化时，动态更新深度学习模型参数，从而实现基于滚动时间窗口的在线预测。最后，通过对两个风电场的实测数据进行仿真模拟，实验结果表明，所提出的方法在多种预测模型上表现出更高的预测精度与稳定性，验证了本文所提方法的有效性。
[1] 史彭珍, 魏霞, 张春梅, 等. 基于 VMD-BOA-LSSVM-AdaBoost 的短期风电功率预测[J]. 太阳能学报, 2024, 45(1): 226-233.
[2] 夏卫平, 邓艾东, 薛原, 等. 基于 VMD 和多时间尺度分类预测的风电单机短期功率预测研究[J]. 太阳能学报, 2025, 46(12): 554-563.
[3] 赵洪山, 杨铎, 刘欣雨, 等. 基于动态深度学习的风电功率在线预测方法[J]. 太阳能学报, 2025, 46(9): 171-180.
[4] 李翠萍, 张冰, 李军徽, 等. NWP 辅助复合神经网络预测误差修正的风储系统日前上报策略[J]. 太阳能学报, 2024, 45(10): 86-96.
[5] 杨明, 李梦林, 王勃, 等. 面向新型电力系统的数值天气预报技术及应用综述[J]. 电力系统自动化, 2025, 49(17): 1-20.
[6] 陈勇. 考虑气象相似性与数值天气预报修正的海上风功率预测研究[J]. 电力设备管理, 2025(12): 150-152.
[7] 刘吉雄, 邹瑞, 许思为. 基于 ARIMA 的风电发电量预测[J]. 现代信息科技, 2025, 9(4): 157-161, 166.
[8] 高志远, 郝泽, 张超, 等. 电力市场中新能源发电概率预测的实施研究和应用展望[J/OL]. 电网技术, 2026: 1-13[2026-02-03].
[9] 苗长新, 王霞, 李昊, 等. 基于数值天气预报风速误差修正的风电功率日前预测[J]. 电网技术, 2022, 46(9): 3455-3464.
[10] 林铮, 刘可真, 沈赋, 等. 考虑海上风电多机组时空特性的超短期功率预测模型[J]. 电力系统自动化, 2022, 46(23): 59-66.
[11] MEI Y, CHE J, SUN Q, et al. An overview of short-term wind power forecasting: Multi-scale decomposition and multi-model deep learning fusion[J]. Energy Strategy Reviews, 2026, 63: 102021.
[12] 王继恒, 胡阳, 宋子秋, 等. 基于多模态混合深度学习的大型风电机组入流风场预测[J/OL]. 系统仿真学报, 2026: 1-14[2026-02-03].
[13] 范竞敏, 贺广林, 王新刚, 等. 基于 VMD-TCN-GRU-AM 的超短期风电预测[J]. 太阳能学报, 2025, 46(6): 538-547.
[14] 乔宽龙, 董存, 车建峰, 等. 时空特性下基于图卷积神经网络的风电集群功率短期预测方法[J]. 太阳能学报, 2024, 45(5): 95-103.
[15] 徐武, 范鑫豪, 沈智方, 等. 多尺度特征提取的 Transformer 短期风电功率预测[J]. 太阳能学报, 2025, 46(2): 640-648.
[16] ZENG A, CHEN M, ZHANG L, et al. Are transformers effective for time series forecasting?[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2023, 37(9): 11121-11128.
[17] ZHANG T, ZHANG Y, CAO W, et al. Less is more: fast multivariate time series forecasting with light sampling-oriented MLP structures[EB/OL]. arXiv:2207.01186, 2022[2026-02-03].
[18] WANG R, WU J, CHENG X, et al. Adaptive expert fusion model for online wind power prediction[J]. Neural Networks, 2025, 184: 107022.
[19] SHEN J, QU Y, ZHANG W, et al. Wasserstein distance guided representation learning for domain adaptation[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2018, 32(1).
[20] XIANG Q, ZI L, CONG X, et al. Concept drift adaptation methods under the deep learning framework: a literature review[J]. Applied Sciences, 2023, 13(11): 6515.
[21] CHEN Y, XU J. Solar and wind power data from the Chinese state grid renewable energy generation forecasting competition[J]. Scientific Data, 2022, 9(1): 577.
[22] HONG T, PINSON P, FAN S, et al. Probabilistic energy forecasting: global energy forecasting competition 2014 and beyond[J]. International Journal of Forecasting, 2016, 32(3): 896-913.
[23] DEY R, SALEM F M. Gate-variants of gated recurrent unit (GRU) neural networks[C]//2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS). 2017: 1597-1600.
[24] GREFF K, SRIVASTAVA R K, KOUTNÍK J, et al. LSTM: a search space odyssey[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, 28(10): 2222-2232.
[25] HEWAGE P, BEHERA A, TROVATI M, et al. Temporal convolutional neural (TCN) network for an effective weather forecasting using time-series data from the local weather station[J]. Soft Computing, 2020, 24(21): 16453-16482.
[26] LIU Y, HU T, ZHANG H, et al. iTransformer: inverted transformers are effective for time series forecasting[EB/OL]. arXiv:2310.06625, 2023[2026-02-03].
[27] PIAO X, CHEN Z, MURAYAMA T, et al. Fredformer: frequency debiased transformer for time series forecasting[C]//Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2024: 2400-2410.
[28] ZHANG Y, YAN J. Crossformer: transformer utilizing cross-dimension dependency for multivariate time series forecasting[C]//The Eleventh International Conference on Learning Representations. 2023.