*3.1. Methodology*

With an intention to provide the most relevant and comprehensive review, a proper selection criterion is needed; therefore, different bibliographic databases were searched: Web of Science, Scopus, Google Researcher and Arxiv. With the aim of finding relevant works, a selection was performed searching keywords related with AI and PV systems. The articles with fewer than 8 citations were excluded since they were not considered relevant enough to the state of the art. As a exception to this rule, the articles published in 2021 were selected even if they did not have enough citations. After removing duplicates and nonrelated papers, 250 articles were obtained. The articles tackled different problems found in PV systems, but most of them were focused on four different problems due to their importance:


In this section, these different problems are addressed by explaining the contributions of each paper in order to provide a global vision of the state of the art of each problem.
