Advancements in Bifacial Photovoltaics: A Review of Machine Learning Techniques for Enhanced Performance

Abstract

Bifacial photovoltaics have gained a lot of popularity in recent years given their ability to utilize scattered and reflected solar radiation from both sides of the panel. Although the price of a Bifacial module is generally higher than conventional mono-photovoltaic panel, it compensates for the higher energy generation per unit area. Even though the potential of Bifacial photovoltaics market is promising, their applications are still limited compared to mono-photovoltaics. However, researchers have been experimenting with Bifacial photovoltaics to exploit their capabilities in different applications and working scenarios, especially with Artificial Intelligence (AI) models. This study will focus on reviewing different Machine Learning (ML) algorithms that have been exploited and modified in order to be used with Bifacial system applications in the last three years of literature. Moreover, most popular ML algorithms are presented and discussed with respect to different Bifacial system parameters. Finally, a conclusion of future prospects and the potential of ML in bifacial photovoltaic industry and applications is presented.