Accurate Prediction of Microstructure of Composites using Machine Learning

Abstract

In this prospective work, a machine learning (ML) model based on multiple independent random forest models to predict the configuration of binary composite bars is developed. The input variables to the ML model are elastic wave signals collected at one end of the composite bar, while the targets of the ML model are binary vectors representing the configuration of the bars.This study results indicate: First, a short period of elastic wave propagated through a composite bar can collect and carry the detailed information of the entire bar; second, the patterns hidden in the collected signals can be detected, extracted, and used by the ML model; finally, the ML model can be well trained using a relatively small dataset pool (less than 0.1% of all possible samples), and make accurate predictions. For the 30 sections of bars used in this study, the average prediction accuracy for each section of this bar can reach 95% and even higher. This ML guided technique can be modified and used in different functionalities and applications such as composites characterization, structure health monitoring, limestone determination, and archaeological detection.

Architecture Prediction of 3D Composites Using Machine Learning and No‐Destructive Technique

“…To train the model, the study can either use all 25 signals collected by the sensors independently or combine them to form a single vector and use it as input. In the initial study, [ 43,44 ] both approaches achieved similar and high prediction accuracy. Therefore, the second approach is to be adopted, which is simpler and requires less training time than the first one due to fewer classifiers being used.…”

“…[41,42] In this study, a composite cubic uniformly divided into 5×5×5 sub-blocks were used, and elastic wave signals collected at the right surface of the composite were used as the dataset. [43,44] Multi-output classification models were compared using different classifiers. Hyperparameter tuning was not within the scope of this study.…”

Fracture Prediction of Hydrogel Using Machine Learning and Inhomogeneous Multiscale Network

“…In recent years, machine learning (ML) has emerged as a powerful tool for addressing a wide range of scientific challenges, including mechanics of materials research, [22][23][24][25][26][27][28] especially predicting material properties and understanding the underlying relationships between material composition, structure, and performance. [29][30][31][32] By leveraging large datasets and advanced algorithms, ML models can identify and learn complex patterns in the data, providing a new avenue for the study and prediction of hydrogel fracture behavior.…”