基于FOX-VMD联合小波阈值去噪的短期光伏功率预测研究

郭文凯; 王果; 闵永智; 苏鹏飞; 刘昕玥

doi:doi:10.19912/j.0254-0096.tynxb.2024-2044

English Version Reader LoginJournal Login

您当前的位置：

首页 >

文章列表页 >

基于FOX-VMD联合小波阈值去噪的短期光伏功率预测研究

更新时间：2025-12-02

- 基于FOX-VMD联合小波阈值去噪的短期光伏功率预测研究
- Vol. 46, Issue 6, Pages: 260-270(2025)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：doi:10.19912/j.0254-0096.tynxb.2024-2044
  CLC： TM615
- Published Online：07 July 2025，
  
  Published：2025
- 稿件说明：
移动端阅览
郭文凯, 王果, 闵永智, et al. 基于FOX-VMD联合小波阈值去噪的短期光伏功率预测研究[J]. 2025, 46(6): 260-270.
DOI：

郭文凯, 王果, 闵永智, et al. 基于FOX-VMD联合小波阈值去噪的短期光伏功率预测研究[J]. 2025, 46(6): 260-270. DOI： doi:10.19912/j.0254-0096.tynxb.2024-2044.

摘要

针对光伏功率数据中的噪声干扰、蕴含信息难以提取以及预测模型误差较大等问题

提出一种优化数据处理以及预测误差修正的多阶段短期光伏功率预测模型。首先

采用组合赋权法计算气象特征相关性

利用赤狐优化算法（FOX）优化变分模态分解方法（VMD）参数

结合最优小波阈值方法（WT）进行数据预处理;其次

对每个固有模态函数（IMF）分量构建双向长短期记忆网络（BiLSTM）模型

叠加重构得到初步预测结果;最后

建立误差修正模型

修正初步预测结果

获得最终预测值。算例分析表明

实验数据测试集的均方根误差（RMSE）、平均绝对误差（MAE）和平均绝对百分比误差（MAPE）值分别为5.21 kW、3.01 kW和0.01%

相比原始BiLSTM模型降低81.01%、82.80%和88.89%

证明所提模型可有效提取信息

减少噪声干扰

降低预测误差。

Abstract

Aiming the problems of noise interference

difficulty in extracting the information contained in PV power data and the error of prediction models

a short-term PV power prediction method based on optimization data processing and prediction error correction was proposed. Firstly

the correlation coefficients of meteorological features affecting PV power were calculated using the combination assignment method. Secondly

the parameters of the variational mode decomposition （VMD） were optimized using the FOX optimization algorithm

and the optimal parameters of the wavelet thresholding method （WT） were determined through experiments

completed data decomposition and denoising. Subsequently

a bidirectional long short-term memory （BiLSTM） model was constructed for each intrinsic mode function （IMF） component

and the initial prediction results were obtained by superposition reconstruction. Finally

an error prediction model based on the BiLSTM neural network was established

to obtain the final PV short-term power prediction results by using error compensation method. Example analyses demonstrated that the root mean square error （RMSE）

mean absolute error （MAE） and mean absolute percentage error （MAPE） values of the test set of the proposed method in this paper are 5.21 kW

3.01 kW and 0.01%

respectively

which are 81.01%

82.80%

and 88.89% lower than the original BiLSTM model

which means the method proposed in this paper can effectively extract the information

reduce the noise interference within the data

and mitigate the intrinsic error of the BiLSTM prediction model.

关键词

Keywords

references

董存, 王铮, 白捷予, 等. 光伏发电功率超短期预测方法综述[J]. 高电压技术, 2023, 49(7): 2938-2951.

周建国, 周路明, 王剑宇, 等. 基于优化分解与误差修正的光伏发电功率预测[J]. 太阳能学报, 2024, 45(7): 502-509.

汪鸿, 朱正甲, 陈建华, 等, 基于人工智能技术与物理方法结合的新能源功率预测研究[J]. 高电压技术, 2023, 49(S1): 111-117.

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.

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.

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.

王涛, 李薇, 许野, 等. 计及相似日的VMD-FE-LSTM光伏出力组合预测模型研究[J]. 太阳能学报, 2024, 45(5): 490-499.

王福忠, 王帅峰, 张丽. 基于VMD-LSTM与误差补偿的光伏发电超短期功率预测[J]. 太阳能学报, 2022, 43(8): 96-103.

黄泽, 毕贵红, 谢旭, 等. 基于MBI-PBI-ResNet的超短期光伏功率预测[J]. 电力系统保护与控制, 2024, 52(2): 165-176.

毕贵红, 张梓睿, 赵四洪, 等. 基于多模式分解和多分支输入的光伏功率超短期预测[J]. 高电压技术, 2024, 50(9): 3837-3849.

龙小慧, 秦际赟, 张青雷, 等. 基于相似日聚类及模态分解的短期光伏发电功率组合预测研究[J]. 电网技术, 2024, 48(7): 2948-2957.

王谦, 高海波, 左文. 基于概率稀疏自注意力的船舶短期电力负荷预测[J]. 大连海事大学学报, 2024, 50(1): 134-142.

杨汪洋, 魏云冰, 罗程浩. 基于CVMD-TCN-BiLSTM的短期电力负荷预测[J]. 电气工程学报, 2024, 19(2): 163-172.

马莹莹, 靳雪振. 基于EEMD和小波阈值的短时交通流预测研究[J]. 重庆交通大学学报(自然科学版), 2022, 41(6): 22-29.

王东风, 刘婧, 黄宇, 等. 结合太阳辐射量计算与CNN-LSTM组合的光伏功率预测方法研究[J]. 太阳能学报, 2024, 45(2): 443-450.

王晓霞, 俞敏, 冀明, 等. 基于气候相似性与SSA-CNN-LSTM的光伏功率组合预测[J]. 太阳能学报, 2023, 44(6): 275-283.

张青山, 王丽婕, 郝颖, 等. 基于卫星云图和晴空模型的分布式光伏电站太阳辐照度超短期预测[J]. 高电压技术, 2022, 48(8): 3271-3281.

GE Y D, ZHANG C, WANG Y W, et al.A novel hybrid model based on multiple influencing factors and temporal convolutional network coupling ReOSELM for wind power prediction[J]. Energy conversion and management, 2024, 313: 118632.

MOHAMMED H, RASHID T.FOX: a FOX-inspired optimization algorithm[J]. Applied intelligence, 2023, 53(1): 1030-1050.

DONG H C, LIU S, LIU D L, et al.Enhanced infrasound denoising for debris flow analysis: integrating empirical mode decomposition with an improved wavelet threshold algorithm[J]. Measurement, 2024, 235: 114961.

蒋田勇, 喻晨宇, 黄可, 等. 基于AO算法优化VMD参数联合小波阈值的桥梁信号去噪方法[J]. 中国公路学报, 2023, 36(7): 158-168.

LEI W, WANG G, WAN B Q, et al.High voltage shunt reactor acoustic signal denoising based on the combination of VMD parameters optimized by coati optimization algorithm and wavelet threshold[J]. Measurement, 2024, 224: 113854.

姜建国, 杨效岩, 毕洪波. 基于VMD-FE-CNN-BiLSTM的短期光伏发电功率预测[J]. 太阳能学报, 2024, 45(7):462-473.

尚海昆, 许俊彦, 李宇才, 等. CEEMDAN在变压器振动信号提取中的应用[J]. 控制理论与应用, 2022, 39(3): 459-468.

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.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

基于GSS的RIME-CNN-LSTM-Attention光伏出力组合预测模型

SHORT-TERM PHOTOVOLTAIC POWER FORECASTING BASED ON ATTENTION-GRU MODEL

SHORT-TERM PHOTOVOLTAIC POWER FORECASTING BASED ON FLUCTUATION CHARACTERISTICS MINING

SHORT-TERM PV POWER FORECASTING BASED ON IMPROVED VMD AND SNS-ATTENTION-GRU

SHORT TERM PHOTOVOLTAIC POWER PREDICTION BASED ON SIMILAR DAY CLUSTERING AND PCC-VMD-SSA-KELM MODEL

Related Author

黄毅

马一力

苗安康

袁越

Liu Guohai

Sun Wenqing

Wu Zhenfei

Chen Zhaoling

Related Institution

Jiangsu Zhen An Power Equipment Co., Ltd.

School of Electrical and Information Engineering, Jiangsu University

College of Automation, Beijing Information Science and Technology University

College of Information and Electrical Engineering, China Agricultural University

School of Electrical Engineering, Xinjiang University

AI问答

⁰