基于聚合模态分解和TCN-BiGRU的光伏功率预测模型

李梦阳; 陈柳; 史蒙; 赵玉娇

doi:doi:10.19912/j.0254-0096.tynxb.2024-1799

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 >

基于聚合模态分解和TCN-BiGRU的光伏功率预测模型

更新时间：2026-03-06

- 基于聚合模态分解和TCN-BiGRU的光伏功率预测模型
- 基于聚合模态分解和TCN-BiGRU的光伏功率预测模型
- 太阳能学报 2026年47卷第2期页码：91-99
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：doi:10.19912/j.0254-0096.tynxb.2024-1799
  中图分类号： TM615
- 网络首发：2026-03-05，
  
  纸质出版：2026-03-05
- 稿件说明：
移动端阅览
李梦阳, 陈柳, 史蒙, 等. 基于聚合模态分解和TCN-BiGRU的光伏功率预测模型[J]. 太阳能学报, 2026,47(2):91-99.

李梦阳, 陈柳, 史蒙, et al. 基于聚合模态分解和TCN-BiGRU的光伏功率预测模型[J]. 2026, 47(2): 91-99.
李梦阳, 陈柳, 史蒙, 等. 基于聚合模态分解和TCN-BiGRU的光伏功率预测模型[J]. 太阳能学报, 2026,47(2):91-99. DOI： doi:10.19912/j.0254-0096.tynxb.2024-1799.

李梦阳, 陈柳, 史蒙, et al. 基于聚合模态分解和TCN-BiGRU的光伏功率预测模型[J]. 2026, 47(2): 91-99. DOI： doi:10.19912/j.0254-0096.tynxb.2024-1799.

摘要

针对光伏发电功率随机性强、波动性高导致预测精度低的问题

提出一种基于聚合模态分解（AMD）、时间卷积网络（TCN）和双向门控循环单元（BiGRU）的光伏功率组合预测模型。该模型使用完全自适应噪声集合经验模态分解（CEEMDAN）对原始光伏序列进行处理

得到多个频率不同的子序列

通过样本熵（SE）对子序列进行区分

保留含信号的低频、中频分量。将CEEMDAN分解得到的高频分量用逐次变分模态分解（SVMD）进行二次分解

降低序列不平稳度。最后

构建不同分量的TCN-BiGRU网络模型

得到各分量的预测值进行加和后输出最终预测结果。通过算例分析进行实验表明

对比其他模型

所提出的组合预测模型在光伏发电功率预测中具有较高的预测精度和稳定性。

Abstract

To address the issue of low prediction accuracy caused by the strong randomness and high volatility of photovoltaic power generation

a hybrid prediction model based on aggregated mode decomposition（AMD）

temporal convolutional network （TCN） and bidirectional gated recurrent unit （BiGRU） is proposed. In view of the randomness and high volatility of photovoltaic power generation

a complete ensemble empirical mode decomposition with adaptive noise （CEEMDAN） is used to decompose the original photovoltaic sequence data for the first time

and a series of sub-sequences with different frequencies are obtained. Sample entropy （SE） is used to segment the molecular sequence

retaining the low-frequency and medium-frequency components of the signal. The high-frequency components obtained by CEEMDAN is decomposed secondarily by successive variational mode decomposition （SVMD） to reduce the sequence instability. Finally

the TCN-BiGRU model is used to predict each component

and final photovoltaic power prediction result is obtained by superimposing the prediction of each component. Experiments based on the analysis of arithmetic instances demonstrate that the forecasting accuracy and stability of the proposed hybrid prediction model outperform other approaches in photovoltaic power forecasting.

关键词

Keywords

references

VAN DER MEER D W, SHEPERO M, SVENSSON A, et al. Probabilistic forecasting of electricity consumption, photovoltaic power generation and net demand of an individual building using Gaussian Processes[J]. Applied energy, 2018, 213: 195-207.

HAO D N, QI L F, TAIRAB A M, et al.Solar energy harvesting technologies for PV self-powered applications: a comprehensive review[J]. Renewable energy, 2022, 188: 678-697.

HAN S, QIAO Y H, YAN J, et al.Mid-to-long term wind and photovoltaic power generation prediction based on copula function and long short term memory network[J]. Applied energy, 2019, 239: 181-191.

SAINT-DRENAN Y M, GOOD G H, BRAUN M, et al. Analysis of the uncertainty in the estimates of regional PV power generation evaluated with the upscaling method[J]. Solar energy, 2016, 135: 536-550.

SHENG H M, XIAO J, CHENG Y H, et al.Short-term solar power forecasting based on weighted Gaussian process regression[J]. IEEE transactions on industrial electronics, 2018, 65(1): 300-308.

YANG D Z, JIRUTITIJAROEN P, WALSH W M.Hourly solar irradiance time series forecasting using cloud cover index[J]. Solar energy, 2012, 86(12): 3531-3543.

YANG X Y, WANG S C, PENG Y, et al.Short-term photovoltaic power prediction with similar-day integrated by BP-AdaBoost based on the Grey-Markov model[J]. Electric power systems research, 2023, 215: 108966.

高寒旭, 袁祖晴, 张淑婷, 等. 基于LSTM模型的短期光伏功率预测[J]. 太阳能学报, 2024, 45(6): 376-381.

陈庆明, 廖鸿飞, 孙颖楷, 等. 基于GWO-GRU的光伏发电功率预测[J]. 太阳能学报, 2024, 45(7): 438-444.

JAIHUNI M, BASAK J K, KHAN F, et al.A novel recurrent neural network approach in forecasting short term solar irradiance[J]. ISA transactions, 2022, 121: 63-74.

WANG K J, QI X X, LIU H D.A comparison of day-ahead photovoltaic power forecasting models based on deep learning neural network[J]. Applied energy, 2019, 251: 113315.

王开艳, 杜浩东, 贾嵘, 等. 基于相似日聚类和QR-CNN-BiLSTM模型的光伏功率短期区间概率预测[J]. 高电压技术, 2022, 48(11): 4372-4388.

张雲钦, 程起泽, 蒋文杰, 等. 基于EMD-PCA-LSTM的光伏功率预测模型[J]. 太阳能学报, 2021, 42(9): 62-69.

王振浩, 王翀, 成龙, 等. 基于集合经验模态分解和深度学习的光伏功率组合预测[J]. 高电压技术, 2022, 48(10): 4133-4142.

ZHANG J L, LIU Z Y, CHEN T.Interval prediction of ultra-short-term photovoltaic power based on a hybrid model[J]. Electric power systems research, 2023, 216: 109035.

DU P D, ZHANG G, LI P L, et al.The photovoltaic output prediction based on variational mode decomposition and maximum relevance minimum redundancy[J]. Applied sciences, 2019, 9(17): 3593.

RESHEF D N, RESHEF Y A, FINUCANE H K, et al.Detecting novel associations in large data sets[J]. Science, 2011, 334(6062): 1518-1524.

YANG Y, LI S S, LI C, et al.Research on ultrasonic signal processing algorithm based on CEEMDAN joint wavelet packet thresholding[J]. Measurement, 2022, 201: 111751.

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.

NAZARI M, SAKHAEI S M.Successive variational mode decomposition[J]. Signal processing, 2020, 174: 107610.

WANG Y Y, CHEN J, CHEN X Q, et al.Short-term load forecasting for industrial customers based on TCN-LightGBM[J]. IEEE transactions on power systems, 2021, 36(3): 1984-1997.

CHO K, VAN MERRIENBOER B, BAHDANAU D, et al.On the properties of neural machine translation: encoder-decoder approaches[C]//Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation. Doha, Qatar, 2014: 103-111.

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

基于多元气象信息和改进组合神经网络的分布式光伏短期功率预测模型

基于黑翅鸢算法和VMD的短期风电功率预测

基于改进VMD和CSABO-TCN-BiGRU的中短期光伏功率预测

基于PIA数据修复和聚类的MVMD-CNN-BiLSTM-ATT短期光伏功率预测

基于注意力时间卷积神经网络的光伏功率概率预测