Power Prediction of Photovoltaic Clusters Based on Spatio-Temporal Feature Extraction and Cross-Modal Fusion

WANG Jian; LIU Huiyuan; ZHANG Zhanxi; SHEN Fu; WANG Kaizheng; CAI Zilong

doi:10.12204/j.issn.1000-7229.2025.11.011

PDF(1381 KB)

Electric Power Construction ›› 2025, Vol. 46 ›› Issue (11) : 121-129. DOI: 10.12204/j.issn.1000-7229.2025.11.011

Renewable Energy and Energy Storage

Power Prediction of Photovoltaic Clusters Based on Spatio-Temporal Feature Extraction and Cross-Modal Fusion

Author information +

History +

Abstract

[Objective] Photovoltaic （PV） power forecasting is a critical component of grid-connected PV dispatch and optimization. However， existing forecasting methods inadequately capture the spatial correlations between power plants， particularly in scenarios with multiple plants exhibiting strong correlations， in which the prediction accuracy remains suboptimal. [Methods] To this end， this study proposes a PV power prediction model， the spatiotemporal multimodal fusion network （ST-MoFNet）， that combines spatiotemporal feature extraction and a cross-modal fusion attention module. The model extracts spatiotemporal features from the spatial and temporal dimensions using a graph convolutional network （GCN） and a temporal convolutional network （TCN-Informer）， respectively， and efficiently fuses multimodal information to capture the complex spatiotemporal dependencies among power plants using a cross-modal fusion attention module. [Results] The experimental results showed that the ST-MoFNet exhibited superior prediction performance compared with the other models， achieving the best results in 1-， 3-， and 5-step predictions. The average R squared （R2） value was 0.896， with a substantial improvement in accuracy of 6-16%. [Conclusions] The combined ST-MoFNet model effectively solved the shortcomings of traditional prediction methods in cluster prediction through its advantages in spatiotemporal feature extraction and information fusion and significantly improved the accuracy and reliability of PV power prediction.

Key words

photovoltaic power generation / graph convolution / cluster prediction / Informer

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

WANG Jian , LIU Huiyuan , ZHANG Zhanxi , et al . Power Prediction of Photovoltaic Clusters Based on Spatio-Temporal Feature Extraction and Cross-Modal Fusion[J]. Electric Power Construction. 2025, 46(11): 121-129 https://doi.org/10.12204/j.issn.1000-7229.2025.11.011

References

List( Publishing order | Descend order by publishing year | Descend order by cited within ) Chart analysis

[1]	国家发展改革委,国家能源局. 国家发展改革委国家能源局关于印发《“十四五”现代能源体系规划》的通知[EB/OL].(2022-01-29)[2024-02-27]. https://www.gov.cn/zhengce/zhengceku/2022-03/23/content_5680759.htm. https://www.gov.cn/zhengce/zhengceku/2022-03/23/content_5680759.htm Cited in this article [2]

[2]	LIU L Y, ZHAO Y, CHANG D L, et al. Prediction of short-term PV power output and uncertainty analysis[J]. Applied Energy, 2018, 228: 700-711. Cited in this article [1]

[3]	SHIVASHANKAR S, MEKHILEF S, MOKHLIS H, et al. Mitigating methods of power fluctuation of photovoltaic (PV) sources:a review[J]. Renewable and Sustainable Energy Reviews, 2016, 59: 1170-1184. Cited in this article [1]

[4]	ANTONANZAS J, OSORIO N, ESCOBAR R, et al. Review of photovoltaic power forecasting[J]. Solar Energy, 2016, 136: 78-111.

[5]

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.

https://doi.org/10.19912/j.0254-0096.tynxb.2021-0501

Abstract

准确的太阳能发电功率短期预测是保证电力调度和大规模光伏并网的关键。该文对近年来光伏发电功率短期预测研究进展进行综述,并对影响光伏发电功率的各种气象因素进行相关性分析。针对用于光伏发电短期功率预测的人工神经网络模型和深度学习模型进行总结和评述。太阳辐照度是影响预测模型精度的主要气象参数。在光伏发电功率短期预测中,神经网络及其组合模型均表现出较好的预测精度,但组合模型整体上优于单一预测模型。

JIA

Lingyun

, YUN

Sining

, ZHAO

Zeni

, et al. Recent progress of short-term forecasting of photovoltaic generation based on artificial neural networks[J]. Acta Energiae Solaris Sinica, 2022, 43(12): 88-97.

https://doi.org/10.19912/j.0254-0096.tynxb.2021-0501

Abstract

Accurate short-term forecasting of photovoltaic generation is crucial to ensure power dispatching and large-scale photovoltaic grid connection. This review paper presents an extensive review on recent progress in the short-term forecasting of solar power generation. The correlation analysis of various meteorological factors affecting on solar power generation is implemented. The artificial neural network models and deep learning models for solar power forecasting are summarized and reviewed. The solar irradiance is the main meteorological parameter affecting the accuracy of forecasting models. In the short-term forecasting of solar power generation, both neural network models and hybrid models demonstrate a satisfactory prediction accuracy, whereas the hybrid models perform better than the single forecasting models in the prediction accuracy.

[6]	刘甚臻, 马超. 基于小波变换和混合深度学习的短期光伏功率预测[J]. 可再生能源, 2023, 41(6): 744-749. LIU Shenzhen, MA Chao. Short-term photovoltaic power prediction based on wavelet transform and hybrid deep learning[J]. Renewable Energy Resources, 2023, 41(6): 744-749. Cited in this article [1]

[7]	赵宏程, 张志鹏. 大规模间歇性新能源接入对西藏电网的影响研究[J]. 电力设备管理, 2020(12): 124-125, 128. ZHAO Hongcheng, ZHANG Zhipeng. Research on the impact of large-scale intermittent new energy integration on Tibet's power grid[J]. Electric Power Equipment Management, 2020(12): 124-125, 128. Cited in this article [1]

[8]	田壁源, 李永新, 文玉玲, 等. 光伏功率预测技术与方法[J]. 电工技术, 2019(15): 25-27. TIAN Biyuan, LI Yongxin, WEN Yuling, et al. Technology and method of PV power prediction[J]. Electric Engineering, 2019(15): 25-27.

[9]	BEHERA M K, NAYAK N. A comparative study on short-term PV power forecasting using decomposition based optimized extreme learning machine algorithm[J]. Engineering Science and Technology, an International Journal, 2020, 23(1): 156-167.

[10]	张晓珂, 张辉, 戴小然, 等. 基于层次聚类和BILSTM的光伏短期功率预测模型[J]. 智慧电力, 2024, 52(9): 41-48. ZHANG Xiaoke, ZHANG Hui, DAI Xiaoran, et al. Photovoltaic short-term power forecasting model based on hierarchical clustering & BILSTM[J]. Smart Power, 2024, 52(9): 41-48.

[11]

江卓翰, 周胜瑜, 何禹清, 等. 基于K-I-ELM多模型集成的分布式光伏出力短期预测方法[J]. 电力科学与技术学报, 2024, 39(4): 146-152.

JIANG

Zhuohan

, ZHOU

Shengyu

, HE

Yuqing

, et al. Short-term prediction method of distributed PV output power based on K-I-ELM multi-model integration[J]. Journal of Electric Power Science and Technology, 2024, 39(4): 146-152.

[12]	邹港, 赵斌, 罗强, 等. 基于PCA-VMD-MVO-SVM的短期光伏输出功率预测方法[J]. 电力科学与技术学报, 2024, 39(5): 163-171. ZOU Gang, ZHAO Bin, LUO Qiang, et al. Prediction method of short-term PV output power based on PCA-VMD-MVO-SVM[J]. Journal of Electric Power Science and Technology, 2024, 39(5): 163-171.

[13]	李庆生, 张裕, 龙家焕, 等. 基于IMFO-LSTM模型的光伏功率预测研究[J]. 电力科学与技术学报, 2024, 39(3): 199-206. LI Qingsheng, ZHANG Yu, LONG Jiahuan, et al. Photovoltaic power prediction based on IMFO-LSTM model[J]. Journal of Electric Power Science and Technology, 2024, 39(3): 199-206.

[14]	周强, 张晓忠, 陈久益, 等. 基于遗传算法小波神经网络的光伏电站发电量预测方法[J]. 智慧电力, 2024, 52(4): 78-84. ZHOU Qiang, ZHANG Xiaozhong, CHEN Jiuyi, et al. Power generation forecasting methods of photovoltaic power plants based on genetic wavelet neural network method[J]. Smart Power, 2024, 52(4): 78-84.

[15]	邱书琦, 蹇照民, 方立雄, 等. 基于变分模态分解和集成学习的光伏发电预测[J]. 智慧电力, 2024, 52(3): 32-38. QIU Shuqi, JIAN Zhaomin, FANG Lixiong, et al. Photovoltaic power generation forecasting based on variational modal decomposition and ensemble learning[J]. Smart Power, 2024, 52(3): 32-38. Cited in this article [1]

[16]

叶洪吉, 卫东, 郭倩, 等. 基于设计信息的分布式光伏电站预测发电量计算方法[J]. 太阳能学报, 2021, 42(4): 253-259.

Hongji

, WEI

Dong

, GUO

Qian

, et al. Calculation method of power generation forecast of distributed pv power station based on design information[J]. Acta Energiae Solaris Sinica, 2021, 42(4): 253-259.

Cited in this article [1]

[17]	李余琪, 张刚林, 甘敏. 基于函数系数自回归模型的风速时间序列预测[J]. 数学的实践与认识, 2017, 47(8): 161-166. LI Yuqi, ZHANG Ganglin, GAN Min. Wind speed prediction based on functional coefficient autoregressive models[J]. Mathematics in Practice and Theory, 2017, 47(8): 161-166. Cited in this article [1]

[18]	陈文进, 陈菁伟, 钱建国, 等. 气象特征频繁变化区域的光伏功率预测方法[J]. 浙江电力, 2023, 42(3): 37-46. CHEN Wenjin, CHEN Jingwei, QIAN Jianguo, et al. A photovoltaic power prediction method for regions with frequent changes of meteorological characteristics[J]. Zhejiang Electric Power, 2023, 42(3): 37-46.

[19]	SHARADGA H, HAJIMIRZA S, BALOG R S. Time series forecasting of solar power generation for large-scale photovoltaic plants[J]. Renewable Energy, 2020, 150: 797-807.

[20]	PIERI E, KYPRIANOU A, PHINIKARIDES A, et al. Forecasting degradation rates of different photovoltaic systems using robust principal component analysis and ARIMA[J]. IET Renewable Power Generation, 2017, 11(10): 1245-1252. Cited in this article [1]

[21]	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. SHI Peiming, GUO Xuanyu, DU Qingcan, et al. Photovoltaic power prediction based on TCN-BiLSTM-Attention-ESN[J]. Acta Energiae Solaris Sinica, 2024, 45(9): 304-316. Cited in this article [1]

[22]	王舒雨, 李豪, 马刚, 等. 基于TCN和DLinear的光伏发电功率多步预测模型[J]. 电力建设, 2025, 46(4): 173-184. WANG Shuyu, LI Hao, MA Gang, et al. Multistep prediction model for photovoltaic power generation based on time convolution and DLinear[J]. Electric Power Construction, 2025, 46(4): 173-184.

[23]

张家安, 王琨玥, 陈建, 等. 基于空间相关性的分布式光伏出力预测[J]. 电力建设, 2020, 41(3): 47-53.

https://doi.org/10.3969/j.issn.1000－7229.2020.03.006

Abstract

随着分布式光伏在配电网的渗透率不断上升，其出力波动将成为调度运行中不可忽略的一项不确定因素。基于同一地区光伏出力变化的相关性，提出一种基于空间相关性的分布式光伏出力预测方法。先对同一地区集中式、分布式光伏出力历史数据做无遮归一化，以无遮系数表征光伏出力不确定性；再由K-means聚类方法对天气情况分类，建立基于Copula函数的各类天气工况下光伏出力的相关性模型；最后根据集中式光伏出力信息实现分布式光伏出力预测。以我国北部某城市光伏电站数据为算例，验证了该方法的有效性。

ZHANG

Jiaan

, WANG

Kunyue

, CHEN

Jian

, et al. Research on prediction of distributed photovoltaic output considering spatial relevance[J]. Electric Power Construction, 2020, 41(3): 47-53.

https://doi.org/10.3969/j.issn.1000－7229.2020.03.006

Cited in this article [1] Abstract

With the increasing proportion of distributed photovoltaic （DPV） power in distribution network， the fluctuation of its power output will become a non-negligible uncertain factor in power grid dispatch and operation. On the basis of the correlation of photovoltaic power generation in one region， a prediction method for distributed photovoltaic output is proposed on the basis of spatial correlation. Firstly， the historical data of centralized and distributed photovoltaic output in the same region are normalized to uncovered coefficient which represents the randomness of photovoltaic output. Then， the weather conditions are classified by K-means clustering. According to Copula theory， the correlation model of photovoltaic output under various weather conditions is established. Finally， the distributed photovoltaic output is predicted according to the information of centralized photovoltaic output. The validity of the proposed method is verified by using an example of a photovoltaic power station in a city of northern China.

[24]

赵如意, 王晓辉, 郑碧煌, 等. 基于特征优化和混合改进灰狼算法优化BiLSTM网络的短期光伏功率预测[J]. 电网技术. 2025, 49(1): 209-222.

ZHAO

Ruyi

, WANG

Xiaohui

, ZHENG

Bihuang

, et al. Short-term photovoltaic power prediction based on feature optimization and hybrid improved grey wolf algorithm-optimized BiLSTM network[J]. Power System Technology. 2025, 49(1): 209-222.

Cited in this article [1]

[25]	文贤馗, 何明君, 张俊玮, 等. 基于K均值聚类的光伏集群发电功率超短期预测研究[J]. 电力系统保护与控制, 2025, 53(12): 165-172. WEN Xiankui, HE Mingjun, ZHANG Junwei, et al. Research on ultra-short-term power forecasting of photovoltaic clusters based on K-means clustering[J]. Power System Protection and Control, 2025, 53(12): 165-172.

[26]	朱涛, 李俊伟, 朱元富, 等. 基于LSTM和误差修正的光伏发电短期功率预测[J]. 哈尔滨理工大学学报, 2025, 30(2): 122-130. ZHU Tao, LI Junwei, ZHU Yuanfu, et al. Short-term power prediction of photovoltaic power generation based on LSTM and error correction[J]. Journal of Harbin University of Science and Technology, 2025, 30(2): 122-130.

[27]	王玲芝, 李晨阳, 刘婧, 等. 基于GRO-SSA-LSTM的短期光伏发电功率预测[J]. 太阳能学报, 2025, 46(2): 401-409. WANG Lingzhi, LI Chenyang, LIU Jing, et al. Short term photovoltaic power prediction based on gro-ssa-lstm[J]. Acta Energiae Solaris Sinica, 2025, 46(2): 401-409.

[28]

陈船宇, 熊国江, 方厚康, 等. 基于MODWT-CEEMDAN-LSTM的短期光伏功率区间预测模型[J]. 太阳能学报, 2025, 46(2): 416-424.

CHEN

Chuanyu

, XIONG

Guojiang

, FANG

Houkang

, et al. Short-term photovoltaic power interval prediction model based on modwt-ceemdan-lstm[J]. Acta Energiae Solaris Sinica, 2025, 46(2): 416-424.

Cited in this article [1]

[29]	GONG M J, YAN C C, XU W, et al. Short-term wind power forecasting model based on temporal convolutional network and Informer[J]. Energy, 2023, 283: 129171. Cited in this article [1]

[30]

赵雪锋, 张宇宁, 詹巍, 等. 基于TCN-Informer的多源融合虚拟最优气象源光伏功率预测[J]. 水力发电, 2025, 51(3): 97-104, 124.

ZHAO

Xuefeng

, ZHANG

Yuning

, ZHAN

Wei

, et al. Photovoltaic power prediction based on TCN-informer and multi-source fusion of virtual optimal weather sources[J]. Water Power, 2025, 51(3): 97-104, 124.

Cited in this article [1]

[31]	WANG Y H, ZHANG C, FU Y Y, et al. Hybrid solar radiation forecasting model with temporal convolutional network using data decomposition and improved artificial ecosystem-based optimization algorithm[J]. Energy, 2023, 280: 128171. Cited in this article [1]

[32]	冯秀坤, 刘国通, 李占峰, 等. 基于CNN-BiLSTM-Attention的光伏功率预测模型[J]. 电工技术, 2025(4): 50-52, 57. FENG Xiukun, LIU Guotong, LI Zhanfeng, et al. Forecasting of photovoltaic power generation model based on CNN-BiLSTM-attention[J]. Electric Engineering, 2025(4): 50-52, 57. Cited in this article [1]

[33]	VENKATESWARAN D, CHO Y. Efficient solar power generation forecasting for greenhouses: a hybrid deep learning approach[J]. Alexandria Engineering Journal, 2024, 91: 222-236. Cited in this article [1]

[34]	李豪, 马刚, 李天宇, 等. 基于时空相关性的短期光伏出力预测混合模型[J]. 电力系统及其自动化学报, 2024, 36(5): 121-129. LI Hao, MA Gang, LI Tianyu, et al. Hybrid model for short-term photovoltaic output prediction based on spatiotemporal correlation[J]. Proceedings of the CSU-EPSA, 2024, 36(5): 121-129. Cited in this article [1]

[35]	张亮, 周立洋, 徐晓春, 等. 一种基于GCN的光伏短期出力预测方法研究[J]. 太阳能学报, 2024, 45(8): 289-294. ZHANG Liang, ZHOU Liyang, XU Xiaochun, et al. Research on short-term PV output prediction method based on gcn[J]. Acta Energiae Solaris Sinica, 2024, 45(8): 289-294. Cited in this article [1]

[36]

刘运超, 杨宁, 崔承刚, 等. 基于时空信息的区域内光伏电站功率预测[J]. 科学技术与工程, 2023, 23(34): 14596-14602.

LIU

Yunchao

, YANG

Ning

, CUI

Chenggang

, et al. Power prediction of regional photovoltaic plants based on spatio-temporal information[J]. Science Technology and Engineering, 2023, 23(34): 14596-14602.

Cited in this article [1]

[37]	YU Y J, HU G P. Short-term solar irradiance prediction based on spatiotemporal graph convolutional recurrent neural network[J]. Journal of Renewable and Sustainable Energy, 2022, 14(5): 053702. Cited in this article [1]

[38]	ZHUANG W, LI Z H, WANG Y, et al. GCN-informer: a novel framework for mid-term photovoltaic power forecasting[J]. Applied Sciences, 2024, 14(5): 2181. Cited in this article [1]