SIMULATION OF NEURAL NETWORK IN NANO SCIENCE WITH DEEP LEARNING

Abstract

Artificial intelligence (AI) together with machine learning (ML) revolutionize scientific research and studies in physics. The creation of neural network and machine learning algorithms powered vibrant visual simulations of various  physical phenomena. This study aims to demonstrate the use of computational models to bridge the gap between the theoretical understanding of turning bulk material into nanoparticles. The simulation used in this study focuses on the conversion of bulk materials, such as silver, into nanoparticles. Extreme changes in physical, chemical, and optical properties at the nanoscale are its defining characteristics. Using a Python-based framework, it provides a highly comprehensive visual representation of nanoscale phenomena. Neural networks were used in this simulation to track physical changes as the substance was reduced to nanoparticles by analyzing transformation data of bulk materials. This illustrates variations in characteristics like as the surface area-to-volume ratio, which are crucial for nanotechnology applications. Intended for the accurate visualization physics informed neural networks PINNs is incorporated.