A New Representative Power Station Selection Method in Distributed Photovoltaic Cluster Power Forecasting
DOI:
https://doi.org/10.13052/dgaej2156-3306.405611Keywords:
Photovoltaic power forecasting, distributed photovoltaic clusters, vector error correction model, spatial temporal graph convolutional networkAbstract
With the rapid development of distributed photovoltaic power generation systems, photovoltaic cluster power forecasting plays a vital role in the stable operation of power grids and optimal energy dispatching. Due to the large number of power stations in the PV cluster, wide distribution and high data dimension, it is not only computationally complex to directly predict all power stations, but also may reduce the forecasting accuracy due to data redundancy, so it is particularly important to reasonably select representative power stations in the cluster forecasting. In order to improve the accuracy of cluster forecasting, this paper proposes a new representative power station selection method based on vector error correction model (VECM), and uses spatiotemporal graph convolutional network to predict the power of distributed photovoltaic clusters. Firstly, a VECM is constructed, and the correlation between each power station and the cluster is analyzed by combining the results of variance decomposition method, and the power stations related to the cluster are selected as the representative power stations. Then, the heron optimization algorithm is introduced to calculate the optimal weight distribution of each representative power station. Finally, the spatiotemporal graph convolutional network is constructed by using the historical data of the representative power stations to realize the feature extraction of complex spatiotemporal data between photovoltaic power stations, and the power forecasting data of each representative power station is output, and the cluster forecasting power is obtained through the optimal weight calculation. An example is carried out in a distributed photovoltaic cluster in a province to prove the effectiveness of the proposed method in cluster forecasting.
Downloads
References
TIAN Zheng-qing, ZHANG Yong, LIU Xiang, et al. Effects of Photovoltaic Power Station Construction on Terrestrial Environment: Retrospect and Prospect [J]. Environmental Science, 2024, 45(01):239–247.
Niu, X., and Luo, X. (2023). Policies and Economic Efficiency for Distributed Photovoltaic and Energy Storage Industry. Distributed Generation & Alternative Energy Journal, 38(04), 1197–1222.
Zheng Wanting, Xiao Hao, Pei Wei. Probabilistic power forecasting for regional distributed photovoltaic systems using NGBoost and enhanced weight optimization[J]. Distribution & Utilization, 2024, 41(07):19–28.
Zhou Yi, Xiao Xianyong, Zhao Qinghua, et al. Photovoltaic power forecasting based on combined data cleaning and improved attention mechanism[J]. Distribution & Utilization, 2024, 41(10):31–37+49.
Li Ye, Influence of Accessed Distributed Photovoltaic Power Generation on Power Quality of Distribution Network[J]. Popular Utilization of Electricity, 2024, 39(03): 35–36.
He Yihui, Guan Lin, Wang Tong, et al. Quantitative Evaluation and Analysis of the Planning and Construction of the New Power System in China Southern Power Grid[J]. Southern Power System Technology, 2024, 18(10): 40–53.
Wang Wenjing, Wang Sicheng. Status and Prospect of Chinese Distributed Photovoltaic Power Generation System[J]. Bulletin of Chinese Academy of Sciences, 2016, 31(02): 165–172.
Xu, Z., Xiang, K., Wang, B., and Li, X. (2024). Study on PV Power Prediction Based on VMD-IGWO-LSTM. Distributed Generation & Alternative Energy Journal, 39(03), 507–530.
Li, H., Li, Z., Wang, C., Xia, L., Tan, H., and Li, K. (2024). Based on Deep Learning Model and Flink Streaming Computing Short Term Photovoltaic Power Generation Prediction for Suburban Distribution Network. Distributed Generation & Alternative Energy Journal, 39(04), 789–806.
Zheng Xiaoyu, Ji Yu, Zhang Ying, et al. Coordinated Optimal Control of Distributed Photovoltaic Cluster Based on Model Predictive Control [J]. Power System and Clean Energy, 2019, 35(07):66–74.
Boussif O, Boukachab G, Assouline D, et al. Improving day-ahead Solar Irradiance Time Series Forecasting by Leveraging Spatio-Temporal Context[J]. Advances in Neural Information Processing Systems, 2024, 36.
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(05):121–129.
Han Xiao, Wang Tao, Wei Xiaoguang, et al. Ultrashort-term photovoltaic output forecasting considering spatiotemporal correlation between arrays[J]. Power System Protection and Control, 2024, 52(14):82–94.
Yang X, Zhao Z, Peng Y, et al. Research on distributed photovoltaic power prediction based on spatiotemporal information ensemble method[J]. Journal of Renewable and Sustainable Energy, 2023, 15(3).
Wang Youjia, Lu Zongxiang, Qiao Ying, et al. Short-Term Regional Wind Power Statistical Upscaling Forecasting Based on Feature Clusterin [J]. Power System Technolo, 2017, 41(05):1383–1389.
Chen Ying, Chen Fugui, Chen Xu, et al. The regional photovoltaic power forecasting method based on statistical up-scaling approach[J]. Renewable Energy Resources, 2012, 30(11):20–23.
Song Yang, Chen Chen. Study on the relationship between water resources utilization and economic growth based on VECM model[J]. Yellow River, 2022, 44(S2):52–54
Guo Yuan, Xia Xiangyang, Yue Jiahui, et al. Battery Cluster Inconsistency Detection Method and Intelligent O&M Scheme Based on Vector Error Correction Model[J]. Electric Power, 2024, 57(06):9–17+44.
Guo Yaoyang, Zhang Li, Wei Yusi, et al. Research on Electricity Load Forecasting Technique Considering Inter-industry Coupling and Correlation Characteristics[J]. Electric Power Information and Communication Technology, 2025, 23(02):1–10.
Fu Y, Liu D, Chen J, et al. Secretary bird optimization algorithm: a new metaheuristic for solving global optimization problems[J]. Artificial Intelligence Review, 2024, 57(5):1–102.
Xiang Yubo. Impact Factor of Chinese Carbon Emission Allowance Price, Based on Cointegration Test and VECM [D]. Shanghai: Shanghai University Of Finance and Economics, 2022.
Tian Wenfuhui. The dynamic relationship between energy con-sumption and economic growth in Yunnan: An empirical study based on the vector error correction model[J]. Science Technology and Industry, 2022, 22(1):295–302 (in Chinese).
