基于CEEMDAN-ARIMA-BiLSTM-SelfAttention与误差修正的风电功率预测模型

刘吉成; 鲍红焉; 刘子毅; 朱琳

doi:10.19805/j.cnki.jcspe.2026.250186

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于CEEMDAN-ARIMA-BiLSTM-SelfAttention与误差修正的风电功率预测模型

新能源 | 更新时间：2026-10-10

- 基于CEEMDAN-ARIMA-BiLSTM-SelfAttention与误差修正的风电功率预测模型
- A Wind Power Prediction Model Based on CEEMDAN-ARIMA-BiLSTM- SelfAttention with Error Correction Strategy
- 动力工程学报 2026年46卷第5期页码：97-106
- 作者机构：
- 作者简介：
  
  刘吉成(1963—),男,宁夏中卫人,教授,博士,研究方向为清洁能源价值链、新能源功率预测,E-mail:ljc29@vip.163.com。
- 基金信息：
  
  国家自然科学基金资助项目(71771085)
- DOI：10.19805/j.cnki.jcspe.2026.250186
  中图分类号： TK83
- 纸质出版：2026
- 稿件说明：
移动端阅览
刘吉成, 鲍红焉, 刘子毅, 朱琳. 基于CEEMDAN-ARIMA-BiLSTM-SelfAttention与误差修正的风电功率预测模型[J]. 动力工程学报, 2026, 46(5): 97-106.
刘吉成, 鲍红焉, 刘子毅, 朱琳. 基于CEEMDAN-ARIMA-BiLSTM-SelfAttention与误差修正的风电功率预测模型[J]. 动力工程学报, 2026, 46(5): 97-106. DOI： 10.19805/j.cnki.jcspe.2026.250186.

DOI： 10.19805/j.cnki.jcspe.2026.250186

摘要

为有效应对风电功率随机性给预测带来的困境

提出了一种计及误差修正的新型混合模型。该方法利用自适应噪声完全集成经验模态分解(CEEMDAN)对风电功率原始数据及其预测误差进行分解

通过t检验以衡量分解后所得各子序列的特性。应用自注意力机制(SelfAttention)增强的双向长短期记忆网络(BiLSTM)预测具有高频特性的序列

采用自回归积分滑动平均(ARIMA)模型处理具有低频特性的序列。最后

将初始风电功率预测结果与相应的误差预测值相加

得到基于误差修正的风功率预测值。选取不同风场中的功率样本

并将其与12个模型进行对比

以验证所提模型CEEMDAN-ARIMA -BiLSTM-SelfAttention的性能。结果表明:所提模型在WF1秋季场景下较单一模型在均方根误差(RMSE)、平均绝对误差(MAE)、平均绝对百分比误差(MAPE)以及决定系数(R2)指标上分别至少提升了70.19%、68.67%、25.41%和7.59%

因此误差修正与分解策略可有效提升风电功率预测的性能;与基准模型相比

所提模型在多场景下均为最可靠的预测方法。

Abstract

A new hybrid model incorporating error correction was proposed to effectively address the challenges posed by the randomness of wind power in forecasting. The raw wind power data and prediction errors were first decomposed through complete ensemble empirical mode decomposition with adaptive noice (CEEMDAN)

t-test was used to measure the characteristics of each subsequence obtained after decomposition. A bidirectional long short-term memory network (BiLSTM) enhanced by SelfAttention mechanism was applied to predict sequences with high-frequency characteristics

and the autoregressive integrated moving average (ARIMA) model was used to process sequences with low-frequency characteristics. Finally

the initial wind power prediction results were added to the corresponding error prediction values to obtain the wind power prediction values based on error correction. Power samples from different wind fields were selected and compared with other 12 models to validate the performance of the proposed model CEEMDAN-ARIMA-BiLSTM-SelfAttention. Results show that the proposed model significantly improves the performance of wind power forecasting by at least 70.19%

68.67%

25.41% and 7.59% in terms of root mean square error (RMSE)

mean absolute error (MAE)

mean absolute percentage error (MAPE) and coefficient of determination (R2) respectively under the WF1 autumn scenario compared to the single model. Therefore

the error correction and decomposition strategy effectively enhances forecasting accuracy. Compared to the baseline model

the proposed model demonstrates the most reliable forecasting method in multiple scenarios.

关键词

Keywords

references

GAO Wei, HAN Mengyao, CHEN Lijuan, et al. Life cycle carbon emission accounting of a typical coastal wind power generation project in Hebei Province, China[J]. Energy Conversion and Management, 2025, 324: 119243.

MOU Dunguo, WANG Zining. A systematic analysis of integrating variable wind power into Fujian power grid[J]. Energy Policy, 2022, 170: 113237.

付炳喆, 王玮, 任国瑞, 等. 考虑增强特征选择的深度卷积-时序网络短期风功率预测[J]. 动力工程学报, 2024, 44(10): 1565-1573.FU Bingzhe, WANG Wei, REN Guorui, et al. Short-term wind power prediction with deep convolutional-temporal networks considering enhanced feature selection[J]. Journal of Chinese Society of Power Engineering, 2024, 44(10): 1565-1573.

YANG Mao, HAN Chao, ZHANG Wei, et al. A short-term power prediction method based on numerical weather prediction correction and the fusion of adaptive spatiotemporal graph feature information for wind farm cluster[J]. Expert Systems with Applications, 2025, 274: 126979.

WANG Da, YANG Mao, ZHANG Wei, et al. Short-term power prediction method of wind farm cluster based on deep spatiotemporal correlation mining[J]. Applied Energy, 2025, 380: 125102.

YANG Mao, LI Xiangyu, FAN Fulin, et al. Two-stage day-ahead multi-step prediction of wind power considering time-series information interaction[J]. Energy, 2024, 312: 133580.

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.LIU Jixiong, ZOU Rui, XU Siwei. Power generation forecasting of wind-generated electricity based on ARIMA[J]. Modern Information Technology, 2025, 9(4): 157-161, 166.

ZHANG Wanqing,[KG(-2x] LIN Zi, LIU Xiaolei. Short-term offshore wind power forecasting-a hybrid model based on discrete wavelet transform (DWT), seasonal autoregressive integrated moving average (SARIMA), and deep-learning-based long short-term memory (LSTM)[J]. Renewable Energy, 2022, 185: 611-628.

王维高, 魏云冰, 滕旭东. 基于VMD-SSA-LSSVM的短期风电预测[J]. 太阳能学报, 2023, 44(3): 204-211.WANG Weigao, WEI Yunbing, TENG Xudong. Short-term wind power forecasting based on VMD-SSA-LSSVM[J]. Acta Energiae Solaris Sinica, 2023, 44(3): 204-211.

JIANG Jiading, WANG Feng, TANG Rui, et al. TS_XGB: ultra-short-term wind power forecasting method based on fusion of time-spatial data and XGBoost algorithm[J]. Procedia Computer Science, 2022, 199: 1103-1111.

张昊立, 张菁, 倪建辉, 等. 引入注意力机制的LSTM-FCN海上风电功率预测[J]. 太阳能学报, 2024, 45(6): 444-450.ZHANG Haoli, ZHANG Jing, NI Jianhui, et al. LSTM-FCN offshore wind power forecasting with introduction of attention mechanism[J]. Acta Energiae Solaris Sinica, 2024, 45(6): 444-450.

骆钊, 吴谕侯, 朱家祥, 等. 基于多尺度时间序列块自编码Transformer神经网络模型的风电超短期功率预测[J]. 电网技术, 2023, 47(9): 3527-3536.LUO Zhao, WU Yuhou, ZHU Jiaxiang, et al. Wind power forecasting based on multi-scale time series block auto-encoder transformer neural network model[J]. Power System Technology, 2023, 47(9): 3527-3536.

周婕, 张春美, 郭红戈. 基于EMD-PF-GRNN的短期风电功率预测研究[J]. 太原科技大学学报, 2023, 44(1): 8-14.ZHOU Jie, ZHANG Chunmei, GUO Hongge. Research on short-term wind power forecasting based on EMD-PF-GRNN[J]. Journal of Taiyuan University of Science and Technology, 2023, 44(1): 8-14.

SAREEN K, PANIGRAHI B K, SHIKHOLA T, et al. An imputation and decomposition algorithms based integrated approach with bidirectional LSTM neural network for wind speed prediction[J]. Energy, 2023, 278: 127799.

YAN Yi, WANG Xuerui, REN Fei, et al. Wind speed prediction using a hybrid model of EEMD and LSTM considering seasonal features[J]. Energy Reports, 2022, 8: 8965-8980.

周丹, 袁至, 李骥, 等. 考虑平抑未来时刻风电波动的混合储能系统超前模糊控制策略[J]. 发电技术, 2024, 45(3): 412-422.ZHOU Dan, YUAN Zhi, LI Ji, et al. An advanced fuzzy control strategy for hybrid energy storage systems considering smoothing of wind power fluctuations at future moments[J]. Power Generation Technology, 2024, 45(3): 412-422.

DING Yunfei, CHEN Zijun, ZHANG Hongwei, et al. A short-term wind power prediction model based on CEEMD and WOA-KELM[J]. Renewable Energy, 2022, 189: 188-198.

赵敏, 王孟军, 刁海岸, 等. 基于CEEMD-CNN-GRU的风电功率预测[J]. 四川轻化工大学学报(自然科学版), 2024, 37(4): 68-74.ZHAO Min, WANG Mengjun, DIAO Haian, et al. Wind power prediction based on CEEMD-CNN-GRU[J]. Journal of Sichuan University of Science & Engineering (Natural Science Edition), 2024, 37(4): 68-74.

王渝红, 史云翔, 周旭, 等. 基于时间模式注意力机制的BiLSTM多风电机组超短期功率预测[J]. 高电压技术, 2022, 48(5): 1884-1892.WANG Yuhong, SHI Yunxiang, ZHOU Xu, et al. Ultra-short-term power prediction for BiLSTM multi wind turbines based on temporal pattern attention[J]. High Voltage Engineering, 2022, 48(5): 1884-1892.

XIONG Jinlin, PENG Tian, TAO Zihan, et al. A dual-scale deep learning model based on ELM-BiLSTM and improved reptile search algorithm for wind power prediction[J]. Energy, 2023, 266: 126419.

XU Zhiqi, XUE Tong, CHEN Xinyu, et al. Wind power correction model designed by the quantitative assessment for the impacts of forecasted wind speed error[J]. Advances in Climate Change Research, 2025, 16(1): 73-81.

YANG Mao, GUO Yunfeng, HUANG Tao, et al. Wind farm cluster power prediction based on graph deviation attention network with learnable graph structure and dynamic error correction during load peak and valley periods[J]. Energy, 2024, 312: 133645.

徐睿麟, 郑建勇, 梅飞, 等. 基于谱聚类和多元变分模态分解的风电机组功率预测[J]. 电网技术, 2024, 48(5): 2043-2053.XU Ruilin, ZHENG Jianyong, MEI Fei, et al. Power prediction of wind turbine units based on spectral clustering and multivariate variational mode decomposition[J]. Power System Technology, 2024, 48(5): 2043-2053.

韩博, 李长青, 刘卫亮, 等. 基于二次分解和BiGRU的超短期光伏发电功率预测[J]. 动力工程学报, 2025, 45(1): 62-69, 79.HAN Bo, LI Changqing, LIU Weiliang, et al. Ultra-short-term photovoltaic power forecasting based on secondary decomposition and BiGRU[J]. Journal of Chinese Society of Power Engineering, 2025, 45(1): 62-69, 79.

CHEN Yongbao, XU Junjie. Solar and wind power data from the Chinese state grid renewable energy generation forecasting competition[J]. Scientific Data, 2022, 9(1): 577.

齐绍洲, 赵鑫, 谭秀杰. 基于EEMD模型的中国碳市场价格形成机制研究[J]. 武汉大学学报(哲学社会科学版), 2015, 68(4): 56-65.QI Shaozhou, ZHAO Xin, TAN Xiujie. A study on the formation mechanism of Chinese carbon market price based on EEMD model[J]. Wuhan University Journal (Philosophy & Social Science), 2015, 68(4): 56-65.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

计入工业园区电热负荷需求的源网荷储优化调度方法研究

基于WPHM和非支配性遗传算法的燃气轮机透平叶片多目标优化维护决策

基于BKA-Transformer-BiGRU的垃圾焚烧炉SNCR脱硝系统动态建模

基于物理模型与机器学习求解叶栅流场的降阶模型

芦竹与典型煤协同燃烧过程中灰成分与结渣特性研究