Optoelectronic performance and artificial neural networks (ANNs) modeling of n-InSe/p-Si solar cell

Nanostructure thin film of InSe deposited on p-Si single crystal to fabricate n-InSe/p-Si heterojunction. Electrical and photoelectrical have been studied by the current density–voltage (J–V). The fabricated cell exhibited rectifying characteristics. Analyzing the results of dark forward J–V shows that there are differrent conduction mechanisms. At low voltages, the current density is controlled by a Schottky emission mechanism. While at a relatively high voltage, a space charge-limited-conduction mechanism is observed with a single trap level. The cell also exhibited a photovoltaic characteristic with a power conversion efficiency of 3.42%. Moreover, artificial neural networks (ANNs) are adopted to model the J–V through the obtained functions. Different network configurations and many runs were trying to achieve good performance and finally obtained the current density, J, as a function of the junction temperature, T, and applied voltage, V. In all cases studied, we compared our obtained functions produced by the ANN technique with the corresponding experimental data and the excellent matching was so clear.