Estimation for iron contamination in Si solar cell by ideality factor: Deep neural network approach

Abstract

Defect‐assisted recombination often restricts the performance of photovoltaic devices, and in order to mass‐produce reliable solar cells, low‐cost express methods are in demand, which could monitor contamination during the process of manufacture. In our work, we applied the deep learning‐based approach for estimating iron concentration in silicon solar cells by using ideality factor. The simulation of solar cells with the back surface field design for generating labeled training and test datasets was performed using SCAPS‐1D software. Our results demonstrate that deep neural networks can predict iron concentration using the ideality factor, temperature, base‐thickness, and doping level of solar cells. Our simulation showed smaller prediction errors at high doping level, low temperature, and the two values of ideality factor: the first one for structures containing only iron interstitial atoms and the second for structures where Fei and iron–boron pairs coexist. The proposed method was tested on real silicon structures.

Deep Learning Extraction of the Temperature-Dependent Parameters of Bulk Defects

“…The method was validated with experimental results, where the parameters obtained by the ML-based and DPSS approaches were shown to agree within an acceptable uncertainty range . ML-based regression methods for solar cell parameters to evaluate the impact of the thickness of different layers on the efficiency, , understand material properties, , and understand current–voltage curve analysis ,− were recently published. Regression tasks have also been performed on luminescence images. , Classification methods have mainly revolved around automated image analysis using deep learning algorithms such as convolutional neural networks (CNNs), where the objective is to classify defects − or identify their position in luminescence images. − …”