Efficient modeling of double absorber layered structure in perovskite solar cells using machine learning techniques

Abstract

This research article investigates the perovskite solar cells double absorber layer structure with 13 different absorber layer combinations. Our primary objective is to explore the potential for achieving high efficiency in perovskite solar cells. We present a double-layer structure for high efficiency, leveraging double absorber layers in perovskite solar cells expands spectral absorption and reduces thermalization losses, thereby enhancing power conversion efficiency and stability. Double absorber layer perovskite solar cells adaptability and integration potential further promise efficiency and reliability advancements. Through a comprehensive examination of all possible combinations of absorber layers we have identified a double absorber layer structure that exhibits an impressive efficiency exceeding 35%. To further enhance our understanding of the underlying factors influencing efficiency, we generated a dataset of 3490 samples. These samples were characterized by 20 input features and four attributes, including the efficiency of the perovskite structure. Leveraging machine learning approach, we developed model using the Random Forest algorithm to predict the efficiency and classify the structures into four distinct classes. Our findings contribute valuable insights into the optimized design of perovskite solar cells, offering a pathway towards achieving higher efficiencies and advancing the field of renewable energy.

A machine learning approach for optimizing and accurate prediction of performance parameters for stacked nanosheet transistor

“…This is the ideal approach considering the shortage of experimental data. Device failure analysis [8], reverse engineering [9], efficient modeling of solar cells [10], morphology study of thin-films [11], and defect identification including variation analysis [12] uses ML algorithm to train the machine from data generated from TCAD simulation. ML model is used to study the defects and structural characteristics of the 1-D Pin diode through I-V and C-V measurements.…”