Predicting Solar Power Generation from Direction and Tilt Using Machine Learning XGBoost Regression. Issue 1 (1st June 2022)

• Record Type: Journal Article
• Authors: Kim, Yongjun; Byun, Yungcheol
• Is Part Of: Journal of physics. Volume 2261:Issue 1(2022)
• Publication Date: 2022-06-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2261/1/012003
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Electricity production using photovoltaic panels is widely used all over the world. Electricity generation using solar power is helping a lot in reducing the Earth's carbon production. However, many photovoltaic panels are often installed improperly to produce maximum efficiency. This not only reduces the amount of electricity produced, but also wastes resources required to make solar panels. To solve this problem, the electricity production data using the longitude and latitude of the actual solar panel installation location in 5 regions in Korea, and the direction and tilt angle of the installation site are used to predict the electricity production by using the machine learning XGBoost regression model. A study was conducted to determine the direction and tilt angle of the installation site to generate maximum electricity production in longitude and latitude. Through this, the longitude and latitude of the solar panel installation location for maximum electricity production were analysed, and in the future, more local data will be used to get out of the limited area, and specific research efficiency will be brought.