Submit

Login Login

Register Register

Home / Journals / CMC / Online First / doi:10.32604/cmc.2026.088359
Journal Menu
Special Issues
Table of Content

All issues

Online First

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Open Access

ARTICLE

A Hybrid Constitutive Model and Machine Learning Approach for Predicting the Dynamic Mechanical Behavior of SiCp/Al Composites

Lijie Wang, Xiaoming Du*
School of Materials Science and Engineering, Shenyang Ligong University, Shenyang, China
* Corresponding Author: Xiaoming Du. Email: email

Computers, Materials & Continua https://doi.org/10.32604/cmc.2026.088359

Received 02 July 2026; Accepted 25 August 2026; Published online 14 September 2026

Abstract

SiCp/Al particle-reinforced aluminum matrix composites exhibit significant strain-rate sensitivity under high-speed impact and dynamic forming conditions, and their mechanical response is further affected by the volume fraction and particle size of the reinforcement phase. To describe these coupled effects, this study developed a hybrid constitutive modeling framework combining microstructure-based finite element simulation, the Johnson–Cook constitutive model, and machine learning. First, three-dimensional microstructural models of SiCp/7075Al composites were established using DIGIMAT FE and ABAQUS, and dynamic compression simulations were conducted under different strain rates, SiC volume fractions, and particle sizes. The simulated stress–strain curves were processed to construct the numerical database. A macroscopic Johnson–Cook model was subsequently calibrated using the low-volume-fraction composite data to provide a physically interpretable baseline response. On this basis, a neural network was employed to learn the stress discrepancy between the finite element simulation response and the Johnson–Cook baseline. The results demonstrate that, compared with the Johnson–Cook model alone, the hybrid model more accurately describes the dynamic stress–strain behavior of SiCp/Al composites under typical conditions. Additionally, strict condition-level holdout testing showed that the model retained a reasonable degree of numerical generalization capability for unseen strain rates, volume fractions, and particle sizes within or at the boundaries of the investigated simulation domain.

Keywords

SiCp/Al composites; dynamic mechanical behavior; Johnson–Cook constitutive model; machine learning; hybrid modeling

  • 58

    View

  • 12

    Download

  • 0

    Like

Related articles
Copyright© 2026 Tech Science Press
© 1997-2026 TSP (Henderson, USA) unless otherwise stated

Share Link