爆炸与冲击2026，Vol.46Issue(3)：37-50,14.DOI:10.11883/bzycj-2025-0339
基于人工神经网络的金属材料本构模型在显式有限元中的实现
Implementation of metallic material constitutive models based on artificial neural networks in explicit finite element analysis
摘要
Abstract
Machine learning techniques have been increasingly applied to the prediction of material behavior and have demonstrated clear advantages over conventional constitutive modeling approaches. The objective of this study was to develop an accurate and computationally efficient data-driven constitutive description for metallic materials under coupled temperature and strain-rate loading conditions. A CoCrFeNiMn high-entropy alloy was selected as the representative material system.Compression experiments were performed over a wide range of temperatures and strain rates to obtain true stress-strain data. Based on the experimental results,a modified Johnson-Cook constitutive model was calibrated to describe strain hardening,strain-rate sensitivity,and thermal softening effects. The calibrated model was then implemented in finite element simulations to generate a large,physically consistent dataset spanning broad thermo-mechanical conditions. This simulation-assisted data generation strategy expanded the training domain while ensuring continuity and stability of the dataset. Using the generated data,an artificial neural network (ANN) model was constructed to learn the nonlinear relationship between strain,strain rate,temperature,and flow stress. The network architecture and training strategy were optimized to improve prediction accuracy and generalization performance. To enable efficient application of the trained ANN within an explicit finite element framework,an automatic FORTRAN code generation tool was developed. The trained ANN parameters were converted into a user-defined material subroutine and embedded into the Abaqus/Explicit platform,allowing direct numerical implementation without external dependencies.The results indicate that the ANN-based constitutive model predicts flow stress with high accuracy,with relative errors remaining below one percent across the investigated loading conditions. In addition,the ANN implementation exhibits higher computational efficiency than the conventional constitutive model in explicit finite element simulations.It is concluded that the data-driven neural network approach can effectively replace traditional phenomenological constitutive models in finite element analysis. The proposed framework provides an efficient and reliable pathway for numerical modeling and simulation of metallic materials under complex thermo-mechanical conditions.关键词
机器学习/人工神经网络/本构模型/CoCrFeNiMn高熵合金/有限元方法/数值实现Key words
machine learning/artificial neural network/constitutive model/CoCrFeNiMn high-entropy alloy/finite element method/numerical implementation分类
数理科学引用本文复制引用
康正东,王少喆,苏步云,康佳鑫,邱吉,树学峰..基于人工神经网络的金属材料本构模型在显式有限元中的实现[J].爆炸与冲击,2026,46(3):37-50,14.基金项目
国家自然科学基金(13202477,12272256) （13202477,12272256）
