Neural network based constitutive model for elastomeric foams

Liang, Guanghui; Chandrashekhara, K.

doi:10.1016/j.engstruct.2007.12.021

Cited by 61 publications

(48 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this method the time interval is divided into n gp subintervals and the timedependent constraints are imposed at each time grid point. Let the jth time-dependent constraint be written as: (11) where t i is time interval over which the constraints need to be imposed.…”

Section: Formulation Of the Optimal Seismic Design Problemmentioning

confidence: 99%

See 1 more Smart Citation

Optimal seismic design of steel structures by an efficient soft computing based algorithm

Gholizadeh

Salajegheh

2010

Journal of Constructional Steel Research

View full text Add to dashboard Cite

Section: Formulation Of the Optimal Seismic Design Problemmentioning

confidence: 99%

“…Salajegheh et al [9,10] incorporated wavelet transforms and neural networks into the GA-based optimization processes to predict structural responses for a specific earthquake time history loading. In recent years, neural network techniques have been broadly utilized in civil and structural engineering applications [11][12][13].…”

Section: Introductionmentioning

confidence: 99%

Optimal seismic design of steel structures by an efficient soft computing based algorithm

Gholizadeh

Salajegheh

2010

Journal of Constructional Steel Research

View full text Add to dashboard Cite

“…An ANN constitutive model has also been used to capture the viscoplastic behavior of a carbon-fiber/ polymer matrix composite under thermomechanical loading conditions [18,19]. It has also been used as a constitutive model for elastomeric foams [17]. In this latter example, the ANN was used to implicitly determine the strain energy function by first being given the first and second deviatoric strain invariants and the total volume ratio, and returning the value of the strain energy function.…”

Section: Introductionmentioning

confidence: 99%

“…These networks are able to automatically produce nonlinear mappings between multiple input and output data through learning from a so-called training set, and are able to generalize well for new data not appearing in the training set while requiring minimal computational resources [15]. ANNs are used to either predict the incremental constitutive response directly [16][17][18][19] or provide the values of the parameters of a predefined constitutive function [9,20,21].…”

Section: Introductionmentioning

confidence: 99%

Reduced-Order Constitutive Modeling of Directionally Solidified Ni-Base Superalloys

Neal

Neu

2014

Journal of Engineering Materials and Technology

View full text Add to dashboard Cite

Temperature-dependent crystal viscoplasticity models are ideal for modeling largegrained, directionaiiy solidified Ni-base superalloys but are computationally expensive. This work explores the use of reduced-order models that are potentially more efficient with similar predictive capability of capturing temperature and orientation dependence. First, a transversely isotropic viscoplasticity model is calibrated to a directionaiiy solidifled Ni-base superalloy using the response predicted by a crystal viscoplasticity model. The unifled macroscale model is capable of capturing isothermal and thermomechanical responses in addition to secondary creep behavior over the temperature range of 20-1050°C. A second approach is an extreme reduced-order microstructure-sensitive constitutive model that uses an artificial neural network to provide a set of parameters that depend on orientation, temperature, and strain rate to give a flrst-order approximation of the material response using a simple constitutive model. This simple relationship is then used in a Neuber-type fatigue notch analysis to predict the local response.

show abstract

“…In this sense, Shen et al [36] and Liang and Chandrashekhara [26] trained the strain-energy function directly by the strain invariants to model the isotropic hyper-elastic behavior of rubber and foam materials, respectively. Jung and Ghaboussi [20] incorporated isotropy ad-hoc into their neural network representation of a visco-elastic material.…”

Section: Introductionmentioning

confidence: 99%

A hybrid approach to simulate the homogenized irreversible elastic–plastic deformations and damage of foams by neural networks

Settgast

Hütter

Kuna

et al. 2020

International Journal of Plasticity

View full text Add to dashboard Cite

Classically, the constitutive behavior of materials is described either phenomenologically, or by homogenization approaches. Phenomenological approaches are computationally very efficient, but are limited for complex non-linear and irreversible mechanisms. Such complex mechanisms can be described well by computational homogenization, but respective FE 2 computations are very expensive.As an alternative way, neural networks have been proposed for constitutive modeling, using either experiments or computational homogenization results for training. However, the application of this method to irreversible material behavior is not trivial.The present contribution presents a hybrid methodology to embed neural networks into the established framework of rate-independent plasticity. Both, the yield function and the evolution equations of internal state variables are represented by neural networks. Respective training data for a foam material are generated from RVE-simulations under monotonic loading. It is demonstrated that this hybrid multi-scale neural network approach (HyMNNA) allows to simulate efficiently even the anisotropic elastic-plastic behavior of foam structures with coupled anisotropic evolution of damage and non-associated plastic flow.

show abstract

Neural network based constitutive model for elastomeric foams

Cited by 61 publications

References 17 publications

Optimal seismic design of steel structures by an efficient soft computing based algorithm

Optimal seismic design of steel structures by an efficient soft computing based algorithm

Reduced-Order Constitutive Modeling of Directionally Solidified Ni-Base Superalloys

A hybrid approach to simulate the homogenized irreversible elastic–plastic deformations and damage of foams by neural networks

Contact Info

Product

Resources

About