Intelligent Material Characterization: A ML Approach for Predicting Microstructure of Nanomaterials

Abstract

Nanomaterials are important for many businesses today, such as computer chips and cloud storage devices. A lot of study is being done to make new nanomaterials at the same time, machine learning (ML) is being used more to solve problems in fields like physics, chemical engineering, and manufacturing sector. Because ML is capable of developing in both controlled and unstructured ways, it could help solve many problems in the real world. Using ML techniques for examining at images of nanomaterials is necessary to find out more about them and characterize and analyze their architecture and spectral data, according to the current state of the art. In order to achieve this, researchers presented in this study a ML based approach for analyzing Scanning transmission electron microscopy (STEM) images and spectral data from STEM images of nanomaterials. To analyze STEM images of a nanomaterial, researchers suggested an approach called Machine Learning for STEM Image Analysis (ML-SIA). In order to analyze the spectrum data of a STEM image of a nanomaterial, researchers have introduced an approach called ML for STEM Image spectrum Data Analysis (ML-SISDA). To execute the algorithms into practice and assess the suggested methods, researchers created a prototype ML application. According to experimental findings, ML-based methods are effective for characterizing nanomaterials. Therefore, by spurring more research in the field of material analysis using AI, this research assists in moving this into the future.

Keywords
Machine Learning, Nanomaterials, STEM Images, Spectral Data, Chemical Engineering, Manufacturing Sector