Short-term Wind and Solar Power Forecasting via Integrated Frequency-Time Feature Analysis and Multi-scale Sliding Windows

CHEN Wenying; WU Mengda; LIU Shuai; LIU Weiliang; ZHANG Qiliang; LIU Changliang; WANG Xin; XU Jiahao

doi:10.19805/j.cnki.jcspe.2026.250162

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Short-term Wind and Solar Power Forecasting via Integrated Frequency-Time Feature Analysis and Multi-scale Sliding Windows

更新时间：2026-06-11

- Short-term Wind and Solar Power Forecasting via Integrated Frequency-Time Feature Analysis and Multi-scale Sliding Windows
- Chinese Journal of Power Engineering Vol. 46, Issue 6, Pages: 60-69(2026)
- 作者机构：
  
  1. 华北电力大学控制与计算机工程学院,河北,保定,071000
  2. 华北电力大学保定市综合能源系统状态检测与优化调控重点实验室,河北,保定,071000
  3. 国家能源集团新能源技术研究院有限公司,北京,102209
- 作者简介：
- 基金信息：
- DOI：10.19805/j.cnki.jcspe.2026.250162
  CLC： TM614
- Published：2026
- 稿件说明：
移动端阅览
陈文颖, 吴梦达, 刘帅, 刘卫亮, 张启亮, 刘长良, 王昕, 徐家豪. 频-时域特征解析协同多尺度滑动窗口的短期风光功率预测[J]. 动力工程学报, 2026, 46(6): 60-69.
陈文颖, 吴梦达, 刘帅, 刘卫亮, 张启亮, 刘长良, 王昕, 徐家豪. 频-时域特征解析协同多尺度滑动窗口的短期风光功率预测[J]. 动力工程学报, 2026, 46(6): 60-69. DOI： 10.19805/j.cnki.jcspe.2026.250162.

DOI： 10.19805/j.cnki.jcspe.2026.250162

摘要

针对风光功率预测中频域特征挖掘不足及动态特征捕捉难题

提出了一种融合频-时域特征解析与多尺度滑动窗口特征融合的卷积-双向长短期记忆网络(CNN-BiLSTM)预测算法。首先

数据经孤立森林算法去噪后

采用皮尔逊相关系数与梯度提升回归(GBRT)的加权集成特征评估方法筛选出高相关输入特征;随后

经多尺度滑动窗口提取高相关特征时序统计量以捕获数据动态变化趋势;最后

构建基于快速傅里叶变换(FFT)频域解析与自注意力(SA)特征聚焦的CNN-BiLSTM算法

通过FFT实现频域特征提取

同时利用SA机制动态分配时序特征权重

进而实现频-时域协同表征学习

提升功率预测精度。北方某风电场与光伏电站的算例结果表明:所提算法具备较高的预测精度

在风功率预测中

平均绝对误差(MAE)和均方根误差(RMSE)分别平均降低了35.43%和31.47%

决定系数(R2)平均提高了8.04%;在光功率预测中

MAE和RMSE分别平均降低31.48%和19.39%

R2平均提高了2.19%。

Abstract

To address insufficient frequency-domain feature extraction and dynamic feature capture challenges in renewable power forecasting

a convolutional neural network-bidirectional long short-term memory network (CNN-BiLSTM) algorithm integrating frequency-time domain analysis with multi-scale sliding window feature fusion was proposed. The methodology involved three phases: 1) raw data was denoised using an isolation forest algorithm

and high-relevance input features were selected through a weighted ensemble evaluation combining Pearson correlation coefficients and gradient boosting regression trees (GBRT); 2) temporal statistics were extracted via multi-scale sliding windows to capture dynamic evolution patterns; 3)the CNN-BiLSTM was augmented with fast Fourier transform (FFT)-based frequency-domain feature extraction and self-attention (SA) mechanisms for adaptive temporal weight allocation

enabling synergistic frequency-time representation learning and improved power prediction accuracy. Case studies at wind and solar farms in northern China demonstrate significant accuracy improvements. For wind power

the proposed model achieves average reductions of 35.43% and 31.47% in mean absolute error (MAE) and root mean square error (RMSE)

respectively

and an increase of 8.04% in coefficient of determination (R2). For solar power

average reductions in MAE and RMSE are 31.48% and 19.39%

respectively

and R2 improves by 2.19%.

关键词

Keywords

references

中华人民共和国生态环境部. 中国应对气候变化的政策与行动2024年度报告[R]. 北京: 中华人民共和国生态环境部, 2024.

潘玲颖, 陈锦奇, 柴博涵. "双碳"目标下新能源发电产业集聚对区域绿色发展的影响研究[J]. 动力工程学报, 2022, 42(11): 1051-1060.PAN Lingying, CHEN Jinqi, CHAI Bohan. Research on the impact of new energy power generation industry agglomeration on regional green economic efficiency[J]. Journal of Chinese Society of Power Engineering, 2022, 42(11): 1051-1060.

陈海鹏, 李赫, 阚天洋, 等. 考虑风电时序特性的深度小波-时序卷积网络超短期风功率预测[J]. 电网技术, 2023, 47(4): 1653-1662.CHEN Haipeng, LI He, KAN Tianyang, et al. DWT-DTCNA ultra-short-term wind power prediction considering wind power timing characteristics[J]. Power System Technology, 2023, 47(4): 1653-1662.

周鑫, 李燕, 曾永辉, 等. 基于SARIMAX-SVR的光伏发电功率预测[J]. 电力系统及其自动化学报, 2024, 36(5): 1-8. ZHOU Xin, LI Yan, ZENG Yonghui, et al. Forecasting of photovoltaic power generation based on SARIMAX-SVR[J]. Proceedings of the CSU-EPSA, 2024, 36(5): 1-8.

HILL D C, MCMILLAN D, BELL K R W, et al. Application of auto-regressive models to U.K. wind speed data for power system impact studies[J]. IEEE Transactions on Sustainable Energy, 2012, 3(1): 134-141.

董坤, 冉鹏, 刘旭, 等. 基于权参数优化的并行深度学习光伏功率预测[J]. 动力工程学报, 2024, 44(1): 91-98.DONG Kun, RAN Peng, LIU Xu, et al. Photovoltaic power prediction by weight parameter optimization-based parallel deep learning[J]. Journal of Chinese Society of Power Engineering, 2024, 44(1): 91-98.

XIN Peiqi, WANG Hongtao, WANG Haobo. Short-term wind power forecasting based on VMD-QPSO-LSTM[C] //Proceedings of 2024 IEEE 4th International Conference on Power, Electronics and Computer Applications. Shenyang, China: IEEE, 2024: 474-478.

RUBASINGHE O, ZHANG Xinan, CHAU T K, et al. A novel sequence to sequence data modelling based CNN-LSTM algorithm for three years ahead monthly peak load forecasting[J]. IEEE Transactions on Power Systems, 2024, 39(1): 1932-1947.

QUAN Hao, ZHANG Wei, ZHANG Wenjie, et al. An interval prediction approach of wind power based on skip-GRU and block-bootstrap techniques[J]. IEEE Transactions on Industry Applications, 2023, 59(4): 4710-4719.

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.

李士哲, 王霄慧, 刘帅. 考虑多变量相关性改进的风电场Transformer中长期预测模型[J]. 智慧电力, 2024, 52(4): 62-68, 107.LI Shizhe, WANG Xiaohui, LIU Shuai. Improved Transformer medium and long term prediction model of wind farm considering multivariate correlation[J]. Smart Power, 2024, 52(4): 62-68, 107.

QIU Jiyang, CHEN Yuanyuan, ZHANG Yanli, et al. Short-term prediction of wind power by TCN-SA model based on dual feature extraction[C] //Proceedings of 20244th Power System and Green Energy Conference. Shanghai, China: IEEE, 2024: 164-168.

魏乐, 戴泽, 陈远野, 等. 考虑多对一时空特征的短期风功率组合预测模型[J]. 动力工程学报, 2024, 44(12): 1869-1877.WEI Le, DAI Ze, CHEN Yuanye, et al. Combined prediction model of short-term wind power by considering many-to-one spatial-temporal features[J]. Journal of Chinese Society of Power Engineering, 2024, 44(12): 1869-1877.

李运龙, 金怀平, 范守元, 等. 在线选择性集成即时学习风电功率自适应预测[J]. 太阳能学报, 2024, 45(10): 487-496.LI Yunlong, JIN Huaiping, FAN Shouyuan, et al. Adaptive wind power forecasting based on online selective ensemble just-in-time learning[J]. Acta Energiae Solaris Sinica, 2024, 45(10): 487-496.

OU Zhongxi, ZHAO Xiaoyan, HUANG Zhixin, et al. Enhanced XGBoost model for accurate wind power forecast considering multiple schedule windows[C] //Proceedings of 2024 IEEE International Conference on Energy Internet (ICEI). Zhuhai, China: IEEE, 2024: 565-569.

YU Weilong, LI Shuaibing, ZHANG Hao, et al. Ultra-short-term wind-power forecasting based on an optimized CNN-BILSTM-attention model[J]. iEnergy, 2024, 3(4): 268-282.

NIU Tong, WANG Jianzhou, DU Pei, et al. WPFSAD: wind power forecasting system integrating dual-stage attention and deep learning[J]. IEEE Transactions on Industrial Informatics, 2023, 19(11): 11252-11264.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Ultra-short-term Photovoltaic Power Forecasting Based on Secondary Decomposition and BiGRU

Acoustic Tomography Temperature Measurement Based on Grid Non-uniform Discretization

Two-stage Short-term Wind Power Prediction with Multi-factor Joint Correction Considering Spatio-temporal Features

Pelletizing and Performance of Non-calcined Fe-based Oxygen Carriers Using Circulating Fluidized Bed Fly Ash as an Inert Support

Research on Hybrid Energy Storage Primary Frequency Control Strategy and Capacity Configuration Based on Fuzzy Adaptive Control

Related Author

HAN Bo

LI Changqing

LIU Weiliang

LIU Shuai

LIU Changliang

XU Jiahao

WANG Xin

Related Institution

Baoding Key Laboratory of State Detection and Optimization Control of Integrated Energy System

New Energy Technology Research Institute, CHN Energy

AI问答

⁰