基于改进CNN-BiLSTM-Attention的两阶段短期风电功率预测
1. 国电和风风电开发有限公司, 辽宁沈阳 110000
2. 哈尔滨工业大学, 机电工程学院, 黑龙江哈尔滨 150001
| 摘 要： | 为了提升短期风电功率预测的准确性，本文研究了一种基于逐次变分模态分解（SVMD）方法、贝叶斯优化（BO）算法和CNN-BiLSTM-Attention模型结合的两阶段短期风电功率预测方法。该方法具体原理为：在第一阶段，利用SVMD方法对经过数据预处理后的风电功率数据进行分解，将得到的模态分量作为CNN-BiLSTM-Attention短期功率预测模型的输入，然后引入BO算法对预测模型的学习率、隐含层节点和正则化参数进行调优，利用训练好的模型进行预测，得到初始风电功率预测值；在第二阶段，采用预测模型对误差序列进行误差补偿预测，得到初始误差功率预测数据，将初始预测功率和误差预测功率求和得到最终的风电功率预测结果。结果表明：相比于CNN-BiLSTM-Attention模型，改进CNN-BiLSTM-Attention模型两阶段预测模型的MAE值、MAPE值和RMSE值分别下降了85.7%、75.2%和77.3%，表明了本文研究的改进CNN-BiLSTM-Attention模型的两阶段短期风电功率预测方法的有效性。 |
| 关 键 词： | 短期风电功率预测; 逐次变分模态分解; 贝叶斯优化算法; 误差补偿; 预测模型 |
| DOI： | 10.57237/j.jsts.2025.02.002 |
View:
Downloads:
收稿日期: 2025-10-18
接受日期: 2025-12-08
在线出版日期: 2025-12-24
Two-Stage Short-Term Wind Power Prediction based on Improved CNN-BiLSTM-Attention
1. Guodian and Wind Power Development Co., Ltd, Shenyang 110000, China
2. School of Mechanical and Electrical Engineering, Harbin Institute of Technology, Harbin 150001, China
| Abstract: | To enhance the accuracy of short-term wind power prediction, this paper proposes a novel two-stage forecasting framework that integrates Sequential Variational Mode Decomposition (SVMD), Bayesian Optimization (BO), and a CNN-BiLSTM-Attention model. In the first stage, the preprocessed wind power historical data is decomposed into several modal components via SVMD. These components serve as inputs to the CNN-BiLSTM-Attention model, whose hyperparameters—including the learning rate, number of hidden units, and regularization coefficient—are automatically tuned using the BO algorithm. The output of this stage is the initial power prediction. In the second stage, the prediction error sequence from the first stage is analyzed and similarly processed (decomposed and modeled) to generate an error compensation term. The final prediction is obtained by summing the initial power prediction and the predicted error compensation. Results show that compared with the CNN-BiLSTM-Attention model, the MAE, MAPE and RMSE values of the improved CNN-BiLSTM-Attention two-stage prediction model decreased by 85.7%, 75.2% and 77.3%, respectively, demonstrating the effectiveness of the two-stage short-term wind power prediction method of the improved CNN-BiLSTM-Attention model studied in this paper. |
| Keywords: | Short-term Wind Power Forecasting; Successive Variational Mode Decomposition; Bayesian Optimization Algorithm; Error Compensation; forecastIng Model |
参考文献
| [1] | 曾娜梅, 霍志红, 许昌, 等. 基于改进HHT的分钟级超短期风速预测 [J]. 动力工程学报, 2021, 41(04): 309-315+329. |
| [2] | 朱菊萍, 魏霞, 谢丽蓉, 等. 基于VMD和改进BiLSTM的短期风电功率预测 [J]. 太阳能学报, 2024, 45(06): 422-428. |
| [3] | 刘金朋, 邓嘉明, 高鹏宇,等. 基于VMD-IMPA-SVM的超短期风电功率预测 [J]. 智慧电力, 2024, 52(07): 24-31+79. |
| [4] | 罗潇远, 刘杰, 杨斌,等. 基于改进鱼鹰优化算法与VMD-LSTM的超短期风电功率预测 [J]. 太阳能学报, 2025, 46(03): 652-660. |
| [5] | 李文鑫. 基于深度神经网络的超短期风电功率预测研究 [D]. 保定: 华北电力大学控制与计算机学院, 2022. |
| [6] | 王继东, 于俊源, 孔祥玉. 基于双重分解和双向长短时记忆网络的中长期负荷预测模型 [J]. 电网技术, 2024, 48(08): 3418-3426. |
| [7] | 王霞. 基于数据驱动的风电特征提取和功率预测方法研究 [D]. 北京: 中国矿业大学电气工程学院, 2023. |
| [8] | 谷学静, 陈洪磊, 孙泽贤, 等. 基于VMD-RL-LSTM的短期风功率预测 [J]. 计算机仿真, 2023, 40(04): 89-93+309. |
| [9] | Junqiang W; Xuejie, W; Tianming Y, et al. Ultra-short-term forecasting of wind power based on multi-task learning and LSTM [J]. International Journal of Electrical Power & Energy Systems. 2023, 149, 109073. |
| [10] | Blazakis K, Katsigiannis Y, Stavrakakis G. One-day-ahead solar irradiation and windspeed forecasting with advanced deep learning techniques [J]. Energies, 2022; 15: 4361. |
| [11] | 唐清苇, 向月, 代佳琨, 等. 基于CNN–LSTM的风电场发电功率迁移预测方法 [J]. 工程科学与技术, 2024, 56(02): 91-99. |
| [12] | Zhang Y M, Wang H. Multi-head attention-based probabilistic CNN-BiLSTM for day-ahead wind speed forecasting [J]. Energy, 2023, 278, 127865. |
| [13] | 康田雨, 覃智君. 基于超参数优化和双重注意力机制的超短期风电功率预测 [J]. 南方电网技术, 2022, 16(05): 44-53. |
| [14] | Wang Z J, Wang L, Revanesh M, et al. Short-Term Wind Speed and Power Forecasting for Smart City Power Grid With a Hybrid Machine Learning Framework [J]. IEEE Internet of Things Journal, 2023, 10(21): 18754-18765. |
| [15] | Yan Y, Wang X R, Ren F, et al. Wind speed prediction using a hybrid model of EEMD and LSTM considering seasonal features [J]. Energy Reports, 2022, 8: 8965-8980. |
| [16] | Zhang Y J, Zhang L, Sun D, et al. Short-Term Wind Power Forecasting Based on VMD and a Hybrid SSA-TCN-BiGRU Network [J]. Applied Sciences, 2023, 13(17): 988. |
| [17] | 马志侠, 张林鍹, 巴音塔娜, 等. 基于自适应二次分解与CNN-BiLSTM的超短期风电功率预测 [J]. 太阳能学报, 2024, 45(06): 429-435. |
| [18] | 何柳青. 基于深度学习的短期风电功率区间预测 [D]. 西安:西安理工大学水利水电学院, 2023. |
| [19] | Reziwan K, Wumaier T. Yan F, et al. Hospital Outpatient Volume Prediction Model Based on Gated Recurrent Unit Optimized by the Modified Cheetah Optimizer [J]. IEEE Access, 2023, 11: 139993-140006. |
| [20] | 雷柯松, 吐松江•卡日, 伊力哈木•亚尔买买提, 等. 基于WGAN-GP和CNN-LSTM-Attention的短期光伏功率预测 [J].电力系统保护与控制, 2023, 51(09): 108-118. |
| [21] | 常青松, 杨昭, 杨熠辉, 等. 基于相似日聚类的超短期光伏功率组合预测模型 [J]. 热力发电, 2023, 52(11): 123-131. |
| [22] | 刘益豪,石宇强. 基于注意力机制的CNN-BiLSTM复杂设备剩余寿命预测 [J]. 机械设计, 2024, 41(S1): 58-63. |
| [23] | 荆志宇, 李培强, 林文婷. 结合贝叶斯优化及通道注意力的双端优化时序式风功率预测模型 [J]. 电力系统及其自动化学报, 2024, 36(08): 39-47+59. |
