Machine‐Learning Accelerating the Development of Perovskite Photovoltaics

Abstract

Perovskite solar cells (PSC) are a potential candidate for next‐generation photovoltaic technology. Despite the significant advancements in the field of PSCs, the ongoing development of stable and efficient metal halide perovskite materials, along with their successful integration into photovoltaic applications, remains challenges. These challenges originate from the diverse range of device structures and perovskite compositions, requiring meticulous consideration and optimization. Traditional trial‐and‐error methods are characterized by their sluggishness and labor‐intensive nature. Recently, the emergence of extensive datasets and advancements in computer hardware have facilitated the utilization of machine learning (ML) across multiple domains, including in various fields for material discovery and experimental optimization. Herein, the fundamental procedure of ML is briefly introduced, and latest progress of ML in the materials development and solar cell fabrication is comprehensively reviewed. The utilization of ML in PSCs at all stages of design can be categorized into four main areas: screening perovskite material, fabrication process optimization, device structure optimization, and understanding mechanism. The challenges and outlooks on the future development of ML are finally discussed. It is highly expected that this review can offer valuable guidance for the design and development of highly efficient and stable PSCs.