基于TCN-NCP-CFC神经网络的超短期风电功率预测

赵飞; 吴文标; 汪燕毅

doi:10.19912/j.0254-0096.tynxb.2024-2249

PDF(2705 KB)

太阳能学报 ›› 2026, Vol. 47 ›› Issue (4) : 376-386. DOI: 10.19912/j.0254-0096.tynxb.2024-2249

基于TCN-NCP-CFC神经网络的超短期风电功率预测

赵飞¹, 吴文标¹, 汪燕毅²

作者信息 +

ULTRA-SHORT-TERM WIND POWER PREDICTION BASED ON TCN-NCP-CFC NEURAL NETWORK

Zhao Fei¹, Wu Wenbiao¹, Wang Yanyi²

Author information +

文章历史 +

摘要

针对风力发电功率的不确定性和波动性，提出一种结合时间卷积网络（TCN）、闭式连续时间网络（CFC）和神经回路策略（NCP）的超短期风电功率预测方法。首先，使用TCN对原始数据进行初步学习，提取时间序列中的关键信息；随后，将其输入NCP-CFC网络中，利用NCP独特的分层类脑递归连接结构及CFC高效的求解机制和防梯度消失特性进行预测；最后，通过全连接层调整输出范围和维度，获得最终预测结果。通过模型结构的消融实验以及与RNN类模型的对比实验验证各模块的必要性。通过两组算例分析（内蒙古某风电场的算例：MSE=25.70 MW2、RMSE=5.07 MW、MAE=3.92 MW、SMAPE=32.51%、R²=0.92；开源算例：MSE=27.38 MW2、RMSE=5.23 MW、MAE=3.71 MW、SMAPE=38.52%、R²=0.84），验证了所提模型在超短期风电功率预测中的有效性。

Abstract

To address the uncertainties and fluctuations associated with wind power generation, an ultra-short-term wind power forecasting method integrating temporal convolutional networks （TCN）, closed-form continuous-time networks （CFC）, and neural circuit policies （NCP） is proposed. Initially, TCN is utilized to conduct preliminary learning from raw data, extracting crucial information from the time series. Subsequently, the processed data is fed into the NCP-CFC network, which leverages the unique hierarchical brain-like recursive connections of NCP and the efficient solving mechanism and anti-gradient vanishing properties of CFC for forecasting. Finally, a fully connected layer adjusts the output range and dimensions to produce the final forecast. The necessity of each module is validated through ablation studies and comparative experiments with RNN-based models. Two case studies are conducted to demonstrate the effectiveness of the proposed model in ultra-short-term wind power forecasting： one involving a wind farm in Inner Mongolia （MSE=25.70 MW², RMSE=5.07 MW, MAE=3.92 MW, SMAPE=32.51%, R²=0.92） and another using open-source data （MSE=27.38 MW², RMSE=5.23 MW, MAE=3.71 MW, SMAPE=38.52%, R²=0.84）.

导出引用

赵飞, 吴文标, 汪燕毅. 基于TCN-NCP-CFC神经网络的超短期风电功率预测[J]. 太阳能学报. 2026, 47(4): 376-386 https://doi.org/10.19912/j.0254-0096.tynxb.2024-2249

Zhao Fei, Wu Wenbiao, Wang Yanyi. ULTRA-SHORT-TERM WIND POWER PREDICTION BASED ON TCN-NCP-CFC NEURAL NETWORK[J]. Acta Energiae Solaris Sinica. 2026, 47(4): 376-386 https://doi.org/10.19912/j.0254-0096.tynxb.2024-2249

中图分类号： TM614

参考文献

[1] 国家能源局. 国家能源局发布2024年1-8月份全国电力工业统计数据[J]. 电力勘测设计, 2024(9): 39.
The National Energy Administration. The National Energy Administration released the statistical data of the national electric power industry from January to August2024[J]. Electric power survey & design, 2024(9): 39.
[2] 乔颖, 鲁宗相, 闵勇. 提高风电功率预测精度的方法[J]. 电网技术, 2017, 41(10): 3261-3269.
QIAO Y, LU Z X, MIN Y.Research & application of raising wind power prediction accuracy[J]. Power system technology, 2017, 41(10): 3261-3269.
[3] HANIFI S, LIU X L, LIN Z, et al.A critical review of wind power forecasting methods: past, present and future[J]. Energies, 2020, 13(15): 3764.
[4] 兰昆, 吴战波, 赵泽妮, 等. 基于多因子的太阳辐照度预测方法研究进展[J]. 太阳能学报, 2024, 45(5): 593-601.
LAN K, WU Z B, ZHAO Z N, et al.Advance in prediction methods for solar irradiance based on multiple-factors[J]. Acta energiae solaris sinica, 2024, 45(5): 593-601.
[5] 贾凌云, 云斯宁, 赵泽妮, 等. 神经网络短期光伏发电预测的应用研究进展[J]. 太阳能学报, 2022, 43(12): 88-97.
JIA L Y, YUN S N, ZHAO Z N, et al.Recent progress of short-term forecasting of photovoltaic generation based on artificial neural networks[J]. Acta energiae solaris sinica, 2022, 43(12): 88-97.
[6] 赵泽妮, 云斯宁, 贾凌云, 等. 基于统计模型的短期风能预测方法研究进展[J]. 太阳能学报, 2022, 43(11): 224-234.
ZHAO Z N, YUN S N, JIA L Y, et al.Recent progress in short-term forecasting of wind energy based on statistical models[J]. Acta energiae solaris sinica, 2022, 43(11): 224-234.
[7] 向玲, 刘佳宁, 苏浩, 等. 基于CEEMDAN二次分解和LSTM的风速多步预测研究[J]. 太阳能学报, 2022, 43(8): 334-339.
XIANG L, LIU J N, SU H, et al.Research on multi-step wind speed forecast based on CEEMDAN secondary decomposition and LSTM[J]. Acta energiae solaris sinica, 2022, 43(8): 334-339.
[8] 傅质馨, 王宝池, 刘皓明, 等. 基于风速-功率组合分解重构的海上风电功率预测[J]. 太阳能学报, 2024, 45(11): 418-426.
FU Z X, WANG B C, LIU H M, et al.Offshore wind power forecasting based on wind speed-power combination decompositon[J]. Acta energiae solaris sinica, 2024, 45(11): 418-426.
[9] 时培明, 郭轩宇, 杜清灿, 等. 基于TCN-BiLSTM-Attention-ESN的光伏功率预测[J]. 太阳能学报, 2024, 45(9): 304-316.
SHI P M, GUO X Y, DU Q C, et al.Photovoltaic power prediction based on TCN-BiLSTM-Attention-ESN[J]. Acta energiae solaris sinica, 2024, 45(9): 304-316.
[10] 符杨, 任子旭, 魏书荣, 等. 基于改进LSTM-TCN模型的海上风电超短期功率预测[J]. 中国电机工程学报, 2022, 42(12): 4292-4302.
FU Y, REN Z X, WEI S R, et al.Ultra-short-term power prediction of offshore wind power based on improved LSTM-TCN model[J]. Proceedings of the CSEE, 2022, 42(12): 4292-4302.
[11] 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.
QIAO K L, DONG C, CHE J F, et al.Short-term prediction method of wind power clusters based on graph convolution neural network under spitio-temporal characteristics[J]. Acta energiae solaris sinica, 2024, 45(5): 95-103.
[12] 钟吴君, 李培强, 涂春鸣. 基于尾流关联的动态超图风电功率超短期预测方法[J]. 中国电机工程学报, 2025, 45(12): 4693-4706.
ZHONG W J, LI P Q, TU C M.Dynamic hypergraph wind power ultra short term prediction method based on wake correlation[J]. Proceedings of the CSEE, 2025, 45(12): 4693-4706.
[13] BAI S J, KOLTER J Z, KOLTUN V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling[EB/OL].2018: arXiv preprint arXiv:1803.01271, http://arxiv.org/abs/1803.0127/
[14] HASANI R, LECHNER M, AMINI A, et al.Liquid time-constant networks[J]. Proceedings of the AAAI conference on artificial intelligence, 2021, 35(9): 7657-7666.
[15] HASANI R, LECHNER M, AMINI A, et al.Closed-form continuous-time neural networks[J]. Nature machine intelligence, 2022, 4(11): 992-1003.
[16] LECHNER M, HASANI R, AMINI A, et al.Neural circuit policies enabling auditable autonomy[J]. Nature machine intelligence, 2020, 2(10): 642-652.
[17] 李林艳, 毕贵红, 孔凡文, 等. 计及相似日和时间相关性的深度学习短期电力负荷预测[J]. 电力科学与工程, 2025, 41(1): 41-52.
LI L Y, BI G H, KONG F W, et al.Short-term power load forecasting of deep learning based on similar daily clustering and time correlation[J]. Electric power science and engineering, 2025, 41(1): 41-52.
[18] BARTLETT P L, MONTANARI A, RAKHLIN A.Deep learning: a statistical viewpoint[J]. Acta numerica, 2021, 30: 87-201.
[19] 姚琦, 柳玉, 白恺, 等. 风电场功率预测水平的多指标综合评价方法研究[J]. 太阳能学报, 2019, 40(2): 333-340.
YAO Q, LIU Y, BAI K, et al.Study of multi-index comprehensive evaluation method for wind farm power prediction level[J]. Acta energiae solaris sinica, 2019, 40(2): 333-340.
[20] DING Y.Data science for wind energy[M]. Boca Raton, FL: Chapman and hall/CRC, 2019.
[21] GU A, DAO T. Mamba: linear-time sequence modeling with selective state spaces[EB/OL].2023: arXiv: 2312.00752. https://arxiv.org/abs/2312.00752
[22] ZHAO Z N, YUN S N, JIA L Y, et al.Hybrid VMD-CNN-GRU-based model for short-term forecasting of wind power considering spatio-temporal features[J]. Engineering applications of artificial intelligence, 2023, 121: 105982.
[23] JIA L Y, YUN S N, ZHAO Z N, et al.Improving short-term forecasting of solar power generation by using an EEMD-BiGRU model: a comparative study based on seven standalone models and six hybrid models[J]. International journal of green energy, 2024, 21(14): 3135-3158.
[24] MENG Y, YUN S N, ZHAO Z N, et al.Short-term electricity load forecasting based on a novel data preprocessing system and data reconstruction strategy[J]. Journal of building engineering, 2023, 77: 107432.
[25] GUO J X, YUN S N, MENG Y, et al.Prediction of heating and cooling loads based on light gradient boosting machine algorithms[J]. Building and environment, 2023, 236: 110252.