Volume 09,Issue 03

Predicting Power Output of Solar Panels Using Machine Learning Algorithms

Authors

Lutfu S. Sua, Figen Balo

Abstract
Prediction of the energy output of solar energy systems is a field of research drawing significant attention due to its increasing share in the energy market among other factors. By integrating many techniques and simulating complicated dynamics, machine learning algorithms have emerged as an option to conventional approaches to handle these problems and offer resolutions that enhance the efficiency of photovoltaic systems. Research on machine learning and photovoltaic systems, in particular, has advanced rapidly in the last five years thanks to the involvement of deep learning in photovoltaic systems. Today, more potent models are being used to analyze structured data, including multidimensional time series, pictures, and videos. This necessitates the review of novel approaches that tackle issues in photovoltaic systems utilizing cutting-edge machine learning modellings. Machine learning methods are proving to be very efficient in various energy-related applications including consumption prediction, intrusion detection, and output prediction. This research aims to compare a number of machine learning algorithms in predicting the power output of solar panels using 13 different parameters. The paper contributes to the field by building a comprehensive model of output prediction providing an accurate comparison of several methods.

Keyword: Machine learning, Solar panel, Output prediction, Renewable energy, Solar energy.

PDF [ 363.47 Kb ] | Endnote File | XML