Perovskite solar cells empowered by machine learning

Perovskite solar cells (PSCs) have attracted considerable interest due to their excellent optoelectronic properties. However, while single-junction PSCs have achieved remarkable efficiencies, factors such as a limited range of developed perovskite materials and immature fabrication processes have constrained their commercialization. Achieving the development of perovskite materials and the preparation of high-performance devices at low cost is a key challenge for the commercialization of PSCs. To address this challenge, machine learning (ML) has been widely applied in the field of PSCs. This paper briefly introduces the basic workflow of ML, providing a foundational understanding for further research on its applications in the PSCs domain. Subsequently, the paper systematically reviews the relevant applications of ML in the PSCs field. Finally, it summarizes the key factors that need to be considered for ML-empowered PSCs and highlights the future directions that should be continuously monitored for development.