Hybrid Numerical Modeling of Ballistic Clay under Low-Speed Impact using Artificial Neural Networks

by   YeonSu Kim, et al.

Roma Plastilina No. 1 clay has been widely used as a conservative boundary condition in bulletproof vests, namely to play the role of a human body. Interestingly, the effect of this boundary condition on the ballistic performance of the vests is indiscernible. Moreover, back face deformation should be characterized by measuring the indentation in the deformed clay, which is important for determining the lethality of gunshots. Therefore, several studies have focused on modeling not only bulletproof vests but also the clay backing material. Despite various attempts to develop a suitable numerical model, determining the appropriate physical parameters that can capture the high-strain-rate behavior of clay is still challenging. In this study, we predicted indentation depth in clay using an artificial neural network (ANN) and determined the optimal material parameters required for a finite element method (FEM)-based model using an inverse tracking method. Our ANN-FEM hybrid model successfully optimized high-strain-rate material parameters without the need for any independent mechanical tests. The proposed novel model achieved a high prediction accuracy of over 98 cases.


page 25

page 27


Forecasting the Colorado River Discharge Using an Artificial Neural Network (ANN) Approach

Artificial Neural Network (ANN) based model is a computational approach ...

Efficient Implementation of Non-linear Flow Law Using Neural Network into the Abaqus Explicit FEM code

Machine learning techniques are increasingly used to predict material be...

A Machine Learning approach to enhance the SUPG stabilization method for advection-dominated differential problems

We propose using Machine Learning and Artificial Neural Networks (ANN) t...

Bayesian Modeling of Nonlinear Poisson Regression with Artificial Neural Networks

Being in the era of big data, modeling and prediction of count data have...

A moving boundary approach of capturing diffusants penetration into rubber: FEM approximation and comparison with laboratory measurements

We propose a moving-boundary scenario to model the penetration of diffus...

Polyconvex anisotropic hyperelasticity with neural networks

In the present work, two machine learning based constitutive models for ...

Micromechanical model for sintering and damage in viscoelastic porous ice and snow. Part I: Theory

Ice and snow have sometime been classified as a viscoelastic or viscopla...

Please sign up or login with your details

Forgot password? Click here to reset