A short-term power prediction method based on the transformation of multi-source spatiotemporal feature for photovoltaic cluster.
Saved in:
| Title: | A short-term power prediction method based on the transformation of multi-source spatiotemporal feature for photovoltaic cluster. |
|---|---|
| Authors: | Zhou, Chaohong1 (AUTHOR) xiaomeidedian2020@163.com, Zhang, Fan1 (AUTHOR), Gu, Shenhui1 (AUTHOR), Zhao, Zexi1 (AUTHOR) |
| Source: | International Journal of Green Energy. 2025, Vol. 22 Issue 12, p2663-2679. 17p. |
| Subject Terms: | *Solar cells, Electric power production forecasting, Random forest algorithms, Hilbert-Huang transform, Spatiotemporal processes, Artificial neural networks, Hierarchical clustering (Cluster analysis), Machine learning |
| Abstract: | A short-term power prediction method for photovoltaic cluster based on transformation of multi-source spatiotemporal feature is proposed to overcome the problem of insufficient mining spatiotemporal feature by traditional short-term power prediction methods for photovoltaic cluster. Firstly, the random forest algorithm is used to analyze the importance of every feature of numerical weather prediction, and a topology graph is generated based on the geographical coordinates of the photovoltaic power station to guide the most important feature of numerical weather prediction to divide the photovoltaic cluster into several sub-clusters by improving deep attention embedded graph clustering. Then, the photovoltaic power and numerical weather prediction are decomposed by Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise, and the decomposed components are reconstructed based on permutation entropy. Finally, the short-term power prediction result for photovoltaic cluster is obtained by time transformation network. Through simulation verification, the experimental results show that the root mean square error and the mean absolute error of the proposed method, respectively, reduces 0.0165 and 0.0170 in average, and the accuracy rate improves 1.63% compared with the other method. It can make greater contributions to large-scale photovoltaic grid connection and regional power supply. [ABSTRACT FROM AUTHOR] |
| Copyright of International Journal of Green Energy is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | GreenFILE |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | A short-term power prediction method for photovoltaic cluster based on transformation of multi-source spatiotemporal feature is proposed to overcome the problem of insufficient mining spatiotemporal feature by traditional short-term power prediction methods for photovoltaic cluster. Firstly, the random forest algorithm is used to analyze the importance of every feature of numerical weather prediction, and a topology graph is generated based on the geographical coordinates of the photovoltaic power station to guide the most important feature of numerical weather prediction to divide the photovoltaic cluster into several sub-clusters by improving deep attention embedded graph clustering. Then, the photovoltaic power and numerical weather prediction are decomposed by Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise, and the decomposed components are reconstructed based on permutation entropy. Finally, the short-term power prediction result for photovoltaic cluster is obtained by time transformation network. Through simulation verification, the experimental results show that the root mean square error and the mean absolute error of the proposed method, respectively, reduces 0.0165 and 0.0170 in average, and the accuracy rate improves 1.63% compared with the other method. It can make greater contributions to large-scale photovoltaic grid connection and regional power supply. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 15435075 |
| DOI: | 10.1080/15435075.2025.2469143 |
