Assembled Methods for the Prediction of the Incident GHI over the City of Puno in Tilted Solar Panels

Loayza-Pizarro F.G.; Nuñez-Medrano Y. · Proceedings of the LACCEI international Multi-conference for Engineering, Education and Technology (2022) · DOI: 10.18687/laccei2022.1.1.529

Resumen

Resumen-This paper focuses on the comparison of Ensemble Machine Learning (ML) methods to forecast the incident GHI on a solar panel tilted at 30° (sexagesimal degrees). In addition, the performance was compared with commonly used simple ML regression methods. A pre-calculation of the GHI (variable used as a target by the models) was performed for 30° tilting surfaces. The optimized Bagging Ensemble and Extra Trees methods obtained better results using the evaluation metrics MSE and R2. In general, the Ensemble models showed satisfactory results and can be used for global solar radiation estimation at locations where only sunshine hours data are available.

Instituciones: Universidad Nacional de Ingeniería