Machine Learning-Based Prediction and Benefit Analysis of Photovoltaic Power Generation

Ziyi Qing; Jiangshan Li; Man Wang; Ruihuan Wang; Zixun Qing

doi:10.54097/02qh0m42

Authors

Ziyi Qing
Jiangshan Li
Man Wang
Ruihuan Wang
Zixun Qing

DOI:

https://doi.org/10.54097/02qh0m42

Keywords:

Domestic Photovoltaic Panels, Machine Learning, Prediction, Economic Benefits.

Abstract

The increasing global demand for energy and the environmental impact of fossil fuels have propelled the advancement of solar photovoltaic (PV) technology. This paper emphasizes the importance of accurately predicting the efficiency and cost-effectiveness of solar PV systems for informed market and investment decisions. We adopted a machine learning approach due to the inadequacy of traditional models, utilizing historical data from household PV panels. We evaluated three models: Gradient Boosted Regression Tree (GBRT), Bi-Directional Gated Recurrent Unit (BiGRU), and Random Forest (RF). The results indicated that the BiGRU model excels in prediction accuracy, while the RF model demonstrates superior stability and robustness to outliers. Furthermore, the study assesses the economic and environmental returns of household PV systems, concluding that their installation offers significant environmental and financial benefits under a fully grid-connected operation.

Downloads

Download data is not yet available.

References

Yang Xiaozhan, editor-in-chief; Feng Wenlin, and Ran Xiuzhi, associate editors. Introduction to New Energy and Sustainable Development [M]. Chongqing: Chongqing University Press,2019

Wang Huan, Ma Bing, Jia Lingxiao, et al. Role of key minerals in the clean energy transition under the carbon neutrality target, supply and demand analyses and their suggestions[J]. China Geology,2021,48(06):1720-1733.

LAI Bo,CHEN Minyu,ZHONG Haiwang,et al. Overview of long-term planning for new power systems with a high proportion of renewable energy[J]. Chinese Journal of Electrical Engineering,2023,43(02):555-581.

MA Jingwen, LI Shuqing, XIA Mengyao. A research review on the application of machine learning in GDP forecasting[J]. Science and Technology Intelligence Research,2022,4(03):73-94.

ZHANG Chi,GUO ¥,LI Ming. A review of the development and application of artificial neural network models[J]. Computer Engineering and Applications,2021,57(11):57-69.

Park S, Jung S, Lee J, et al. A Short-Term Forecasting of Wind Power Outputs Based on Gradient Boosting Regression Tree Algorithms[J]. Energies, 2023, 16(3): 1132.

Fu Lianlian, Wu Jian. Pig price prediction based on gradient boosting regression model[J]. Computer Simulation,2020,37(01):347-350.

ZHANG Changfan, LIU Jiafeng, HE Jing, et al. Bearing fault diagnosis based on improved convolutional bidirectional gated recurrent network[J]. Journal of Electronic Measurement and Instrumentation,2021,35(11):61-67.

REN Liqiang, JIA Shuyi, WANG Haipeng, et al. A review of deep learning based time series classification research[J/OL]. Journal of Electronics and Information,1-23[2024-04-28].

Zhu Xingyu. Technical and economic analysis of rooftop photovoltaic power generation system for universities based on dual-carbon target [D]. Qilu University of Technology,2023.

Gao Ruiming. Research on the operation mode of the new energy enterprise of CGN under the background of dual-carbon target [D]. Xi'an University of Technology, 2023.

Yang Yuxin. Analysis of environmental benefits of campus rooftop photovoltaic power generation under the background of "dual carbon"[J]. Automation Application,2023,64(01):41-42+70.