基于功率波动特征和SSA-GRU联合优化的短期光伏功率预测

马艺玮; 马维兴

doi:10.19912/j.0254-0096.tynxb.2025-0512

PDF(1539 KB)

太阳能学报 ›› 2026, Vol. 47 ›› Issue (8) : 161-168. DOI: 10.19912/j.0254-0096.tynxb.2025-0512

基于功率波动特征和SSA-GRU联合优化的短期光伏功率预测

马艺玮¹, 马维兴^1,2

作者信息 +

SHORT-TERM PV POWER PREDICTION BASED ON POWER FLUCTUATION FEATURES AND SSA-GRU JOINT OPTIMIZATION

Ma Yiwei¹, Ma Weixing^1,2

Author information +

文章历史 +

摘要

光伏发电功率的随机波动特性是造成预测精度低的一项重要原因,为此提出一种结合光伏功率波动特征和SSA-GRU模型联合优化的短期光伏功率预测方法。首先依次建立4种不同的光伏功率波动特征模型和基于FCM的相似功率波动模式聚类算法,以获得具有良好分布特性的各种输入子数据集;然后根据光伏功率波动特性构建融合奇异谱分析法（SSA）和门控循环单元网络（GRU）的SSA-GRU预测模型,并提出改进的浣熊优化算法（ICOA）对其联合优化,发挥二者的协同优势以提高光伏功率预测性能。为验证所提方法的优越性,选用宁夏某实际光伏电站数据进行比较实验,结果充分表明所提方法是科学有效的,比其他模型具有更好的预测性能。

Abstract

The random fluctuation characteristics of PV power are an important reason for low prediction accuracy. Therefore, this paper proposes a short-term PV power prediction method that combines PV power fluctuation features and joint optimization of SSA-GRU model. Firstly, four different PV power fluctuation feature models and an FCM-based clustering algorithm for similar power fluctuation patterns are built to obtain various input sub datasets with good distribution characteristics. Secondly, based on PV power fluctuation features, an SSA-GRU prediction model is constructed by integrating singular spectrum analysis （SSA） and gated recurrent unit （GRU）, and an improved coati optimization algorithm （ICOA） is given to jointly optimize the model to fully utilize their synergistic advantages for improving PV power prediction performance. To verify the superiority of the proposed method, a comprehensive comparative experiment is conducted using data from a real PV power station in Ningxia. The results fully demonstrate that the proposed method is scientifically effective and has better predictive performance than comparative models.

导出引用

马艺玮, 马维兴. 基于功率波动特征和SSA-GRU联合优化的短期光伏功率预测[J]. 太阳能学报. 2026, 47(8): 161-168 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0512

Ma Yiwei, Ma Weixing. SHORT-TERM PV POWER PREDICTION BASED ON POWER FLUCTUATION FEATURES AND SSA-GRU JOINT OPTIMIZATION[J]. Acta Energiae Solaris Sinica. 2026, 47(8): 161-168 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0512

中图分类号： TM615 TP183

参考文献

[1] Bae K Y, Jang H S, Sung D K.Hourly solar irradiance prediction based on support vector machine and its error analysis[J]. IEEE Transactions on Power Systems, 2017, 32(2): 935-945.
[2] 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.
Wang Z H, Wang C, Cheng L, et al.Photovoltaic power combined prediction based on ensemble empirical mode decomposition and deep learning[J]. High Voltage Engineering, 2022, 48(10): 4133-4142.
[3] 刘洁, 林舜江, 梁炜焜, 等. 基于高阶马尔可夫链和高斯混合模型的光伏出力短期概率预测[J]. 电网技术, 2023, 47(1): 266-274.
Liu J, Lin S J, Liang W K, et al.Short-term probabilistic forecast for power output of photovoltaic station based on high order Markov chain and Gaussian mixture model[J]. Power System Technology, 2023, 47(1): 266-274.
[4] Liu J, Shi Q, Han R L, et al.A hybrid GA-PSO-CNN model for ultra-short-term wind power forecasting[J]. Energies, 2021, 14(20): 6500.
[5] Tuerxun W, Xu C, Guo H Y, et al.A wind power forecasting model using LSTM optimized by the modified bald eagle search algorithm[J]. Energies, 2022, 15(6): 2031.
[6] 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.
Chen Q M, Liao H F, Sun Y K, et al.Photovoltaic power prediction model based on GWO-GRU[J]. Acta Energiae Solaris Sinica, 2024, 45(7): 438-444.
[7] 周满国, 黄艳国, 段锦锋. 基于GRU-RF模型的太阳辐照度短时预测[J]. 太阳能学报, 2022, 43(7): 166-173.
Zhou M G, Huang Y G, Duan J F.Short term prediction of soral irradiance based on GRU-RF model[J]. Acta Energiae Solaris Sinica, 2022, 43(7): 166-173.
[8] 李宇豪, 杨建卫, 李佳瑞, 等. 基于CNN-GRU与特征增强的超短期光伏功率预测[J]. 计算机仿真, 2024, 41(10): 83-88.
Li Y H, Yang J W, Li J R, et al.Ultra-short-term power prediction of PV based on CNN-GRU with feature augmentation[J]. Computer Simulation, 2024, 41(10): 83-88.
[9] Wang K J, Qi X X, Liu H D.Photovoltaic power forecasting based LSTM-convolutional network[J]. Energy, 2019, 189: 116225.
[10] 戴长春, 郑晓亮, 杨铖, 等. 基于改进DBO-BiLSTM-GRU的光伏发电功率预测方法研究[J]. 自动化应用, 2024, 65(19): 89-93.
Dai C C, Zheng X L, Yang C, et al.Research on photovoltaic power generation prediction method based on improved DBO-BiLSTM-GRU[J]. Automation Application, 2024, 65(19): 89-93.
[11] 黄雨薇, 彭道刚, 姚峻, 等. 基于SSA和K均值的TD-BP神经网络超短期光伏功率预测[J]. 太阳能学报, 2021, 42(4): 229-238.
Huang Y W, Peng D G, Yao J, et al.Ultra-short-term photovoltaic power forecast of TD-BP neural network based on SSA and K-means[J]. Acta Energiae Solaris Sinica, 2021, 42(4): 229-238.
[12] 杨国华, 张鸿皓, 郑豪丰, 等. 基于相似日聚类和IHGWO-WNN-AdaBoost模型的短期光伏功率预测[J]. 高电压技术, 2021, 47(4): 1185-1194.
Yang G H, Zhang H H, Zheng H F, et al.Short-term photovoltaic power forecasting based on similar weather clustering and IHGWO-WNN-AdaBoost modal[J]. High Voltage Engineering, 2021, 47(4): 1185-1194.
[13] 何铖, 余成波, 龙铖, 等. 基于VMD-TCN-LSTM模型的短期光伏功率预测[J]. 计算机仿真, 2024, 41(7): 113-117, 139.
He C, Yu C B, Long C, et al.Short-term photovoltaic power prediction based on VMD-TCN-LSTM model[J]. Computer Simulation, 2024, 41(7): 113-117, 139.
[14] 张蕊, 李安燚, 刘世岩, 等. 基于波动特征提取下云层分型的短期光伏发电功率预测方法[J]. 太阳能学报, 2024, 45(11): 330-342.
Zhang R, Li A Y, Liu S Y, et al.A short-term photovoltaic power generation power prediction method based on cloud classification using wave feature extraction[J]. Acta Energiae Solaris Sinica, 2024, 45(11): 330-342.
[15] 樊围国, 陈珂翰. 基于浣熊算法优化双向长短期记忆网络的碳价预测[J]. 电力科学与工程, 2023, 39(7): 34-41.
Fan W G, Chen K H.Carbon price prediction of bidirectional long short term memory network optimized by coati optimization algorithm[J]. Electric Power Science and Engineering, 2023, 39(7): 34-41.
[16] Goh H H, Luo Q W, Zhang D D, et al.Hybrid SDS and WPT-IBBO-DNM based model for ultra-short term photovoltaic prediction[J]. CSEE Journal of Power and Energy Systems, 2023, 9(1): 66-76.