A Mechanical Property Prediction Model for Nanoindentation Data Based on Artificial Neural Networks

View at publisher Summarise with VIRAT
Authors: Jingchun Wang, Cong Li, Junhao Fu, Shoubin Zhang
Cited by: 0
References: 10

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper proposes an artificial neural networkbased method for analyzing nanoindentation data to predict the mechanical properties of metal matrix composites. A dataset of indentation curves for various materials is constructed through finite element simulation, from which features such as contact stiffness and elastic/plastic deformation quantities are extracted. The performance of Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), and Residual Network (ResNet) is compared, with the MLP ultimately selected and optimized. Through data preprocessing, model structure tuning, and a dualloss function mechanism, the prediction accuracy is significantly improved. Experimental results demonstrate that for predicting elastic modulus and yield strength, the model achieves mean relative errors of 8.10% and 8.11%, respectively, validating its effectiveness for the rapid assessment of mechanical properties in materials.

Fields of Study & Subjects

Machine Learning in Materials Science Metal and Thin Film Mechanics Mechanical and Thermal Properties Analysis

Publication details

Journal / source
International Conference on Electrical Automation and Artificial Intelligence (ICEAAI)
Pages
1–5
Publication date
16 January 2026
Language
EN