Machine Learning Guided Strategies to Develop High Efficiency Indoor Perovskite Solar Cells

Abstract

AbstractIndoor Perovskite Solar Cells (IPSCs) have recently gathered massive research attention, driven by their promising role in powering the continuously expanding Internet of Things (IoT) devices and simultaneous advancements in the Perovskite solar field. To further accelerate the development of IPSCs, a machine learning (ML) approach to assist the advancement of IPSCs is proposed in the current study. Here, a ML model to predict the most important performance parameters such as short circuit current (JSC), open circuit voltage (VOC), fill factor (FF), and power conversion efficiency (PCE) of IPSCs under various light sources and intensities is presented. This developed model can effectively predict the performances of Perovskite Solar Cells (PSCs) operated under indoor illumination close to the true/experimental values. The factors affecting the IPSC performance by Correlation matrix and SHAPley analysis are also analyzed. These findings demonstrate that the proposed ML model provides accurate predictions of VOC, JSC, FF, and PCE of IPSCs, ultimately contributing to the optimization of solar cell performance under indoor environments and the advancement of renewable energy technology.