Entity

Time filter

Source Type

Starkville, MS, United States

Islam M.R.,MSU CAVS | Rohbrecht J.,Simufact Engineering GMbh | Buijk A.,Simufact Americas LLC | Namazi E.,University of Winsor | And 2 more authors.
ASME International Mechanical Engineering Congress and Exposition, Proceedings (IMECE) | Year: 2013

An effective and rigorous approach to determine optimum welding process parameters is implementation of advanced computer aided engineering (CAE) tool that integrates efficient optimization techniques and numerical welding simulation. In this paper, an automated computational methodology to determine optimum arc welding process control parameters is proposed. It is a coupled Genetic Algorithms (GA) and Finite Element (FE) based optimization method where GA directly utilizes output responses of FE based welding simulations for iterative optimization. Effectiveness of the method has been demonstrated by predicting optimum parameters of a lap joint specimen of two thin steel plates for minimum distortion. Three dimensional FE model has been developed to simulate the arc welding process and validated by experimental results. Subsequently, it is used by GA as the evaluation model for optimization. The optimization results show that such a CAE based method can predict optimum parameters successfully with limited effort and cost. Copyright © 2013 by ASME. Source


Enakoutsa K.,MSU CAVS | Ahad F.R.,MSU CAVS | Solanki K.N.,Arizona State University | Tjiptowidjojo Y.,MSU CAVS | Bammann D.J.,MSU CAVS
Journal of Engineering Materials and Technology, Transactions of the ASME | Year: 2012

The Bammann, Chiesa, and Johnson (BCJ) material model predicts unlimited localization of strain and damage, resulting in a zero dissipation energy at failure. This difficulty resolves when the BCJ model is modified to incorporate a nonlocal evolution equation for the damage, as proposed by Pijaudier-Cabot and Bazant (1987, Nonlocal Damage Theory, ASCE J. Eng. Mech., 113, pp. 1512-1533.). In this work, we theoretically assess the ability of such a modified BCJ model to prevent unlimited localization of strain and damage. To that end, we investigate two localization problems in nonlocal BCJ metals: appearance of a spatial discontinuity of the velocity gradient in any finite, inhomogeneous body, and localization of the dissipation energy into finite bands. We show that in spite of the softening arising from the damage, no spatial discontinuity occurs in the velocity gradient. Also, we find that the dissipation energy is continuously distributed in nonlocal BCJ metals and therefore cannot localize into zones of vanishing volume. As a result, the appearance of any vanishing width adiabatic shear band is impossible in a nonlocal BCJ metal. Finally, we study the finite element (FE) solution of shear banding in a rectangular plate, deformed in plane strain tension and containing an imperfection, thereby illustrating the effects of imperfections and finite size on the localization of strain and damage. © 2012 American Society of Mechanical Engineers. Source


Ahad F.R.,MSU CAVS | Enakoutsa K.,MSU CAVS | Solanki K.N.,MSU CAVS | Tjipowidjojo Y.,MSU CAVS | Bammann D.J.,MSU CAVS
2011 Joint Rail Conference, JRC 2011 | Year: 2011

In this study, we use a physically-motivated internal state variable plasticity/damage model containing a mathematical length scale to represent the material behavior in finite element (FE) simulations of a large scale boundary value problem. This problem consists of a moving striker colliding against a stationary hazmat tank car. The motivations are (1) to reproduce with high fidelity finite deformation and temperature histories, damage, and high rate phenomena which arise during the impact and (2) to address the pathological mesh size dependence of the FE solution in the post-bifurcation regime. We introduce the mathematical length scale in the model by adopting a nonlocal evolution equation for the damage, as suggested by Pijaudier-Cabot and Bazant (1987) in the context of concrete. We implement this evolution equation into existing implicit and explicit versions of the FE subroutines of the plasticity/failure model. The results of the FE simulations, carried out with the aid of Abaqus/Explicit FE code, show that the material model, accounting for temperature histories and nonlocal damage effects, satisfactorily predicts the damage progression during the tank car impact accident and significantly reduces the pathological mesh size effects. © 2011 by ASME. Source

Discover hidden collaborations