Performance Assessment of FACsSnI ₃ Perovskite Solar Cells via Deep Learning

Abstract

The development of lead-free perovskite solar cells (PSCs) presents a promising path for sustainable energy and engineering applications such as embedded and aerospace systems. However, understanding and mitigating photovoltaic mechanisms while optimizing device performance remain significant challenges. In this study, we employ numerical simulation techniques combined with deep learning (DL) techniques to gain insights into both the design optimization and performance analysis of high-efficiency lead-free FACsSnI <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$_{3}$</tex> solar cells. By using SCAPS-1D numerical simulations, we analyze key-performance parameters such as open-circuit voltage, short-circuit current, fill factor, and power conversion efficiency (PCE) under varying conditions influenced by photogeneration and interfacial loss processes. Our approach based on the investigation of the impact of the mole fraction variations of the absorber layer on the device performance. This approach can predict critical performance pathways and identify the most effective structural configurations for enhancing the efficiency. The results demonstrate that the proposed investigation not only improves the accuracy of performance predictions but also accelerates the design process, reducing development time and cost. This study provides a new framework for developing the future efficient lead-free photovoltaic systems.