A5014555327- Numerical methods in engineering
- Advanced Numerical Methods in Computational Mathematics
- Composite Material Mechanics
- Model Reduction and Neural Networks
- Advanced Mathematical Modeling in Engineering
- Probabilistic and Robust Engineering Design
- Advanced Numerical Analysis Techniques
- Electromagnetic Simulation and Numerical Methods
- Metal Forming Simulation Techniques
- Advanced machining processes and optimization
- Structural Health Monitoring Techniques
- Fatigue and fracture mechanics
- Rock Mechanics and Modeling
- Drilling and Well Engineering
- Manufacturing Process and Optimization
- Metallurgy and Material Forming
- Composite Structure Analysis and Optimization
- Ultrasonics and Acoustic Wave Propagation
- Computational Geometry and Mesh Generation
- Non-Destructive Testing Techniques
- Elasticity and Material Modeling
- Topology Optimization in Engineering
- Tunneling and Rock Mechanics
- Mineral Processing and Grinding
- Image Processing and 3D Reconstruction
École Nationale Supérieure des Mines de Paris
2013-2024
Centre des Matériaux
2013-2024
Université Paris Sciences et Lettres
2013-2024
Centre National de la Recherche Scientifique
2010-2024
Cardiff University
2014-2023
ParisTech
2019-2021
Delhi Integrated Multi-Modal Transit System
2013
Imam Mohammad ibn Saud Islamic University
2012
Royal Academy of Engineering
2011
École Normale Supérieure Paris-Saclay
2009-2011
Abstract By using the strain smoothing technique proposed by Chen et al. ( Comput. Mech. 2000; 25 :137–156) for meshless methods in context of finite element method (FEM), Liu 2007; 39 (6):859–877) developed Smoothed FEM (SFEM). Although SFEM is not yet well understood mathematically, numerical experiments point to potentially useful features this particularly simple modification FEM. To date, has only been investigated bilinear and Wachspress approximations limited linear reproducing...
Summary The present work addresses shape sensitivity analysis and optimization in two‐dimensional elasticity with a regularized isogeometric boundary element method (IGABEM). Non‐uniform rational B‐splines are used both for the geometry basis functions to discretize integral equations. With advantage of tight integration design analysis, application IGABEM reduces mesh generation/regeneration burden greatly. is distinct from previous literatures mainly two aspects: (1) structural takes form...
Multiscale computational modelling is challenging due to the high cost of direct numerical simulation by finite elements. To address this issue, concurrent multiscale methods use solution cheaper macroscale surrogates as boundary conditions microscale sliding windows. The problems remain a numerically operation both in terms implementation and cost. In work we propose replace local an Encoder-Decoder Convolutional Neural Network that will generate fine-scale stress corrections coarse...
Driven by the need to accelerate numerical simulations, use of machine learning techniques is rapidly growing in field computational solid mechanics. Their application especially advantageous concurrent multiscale finite element analysis (FE2) due exceedingly high costs often associated with it and number similar micromechanical analyses involved. To tackle issue, using surrogate models approximate microscopic behavior simulations a promising increasingly popular strategy. However, several...
This paper proposes a novel technique to reduce the computational burden associated with simulation of localised failure. The proposed methodology affords damage initiation and propagation whilst concentrating effort where it is most needed, i.e. in localisation zones. To do so, local/global devised global (slave) problem (far from zones undergoing severe cracking) solved for reduced space computed by classical Proper Orthogonal Decomposition, while local (master) degrees freedom (associated...
This paper aims to incorporate the node-based smoothed finite element method (NS-FEM) into extended (XFEM) form a novel numerical (NS-XFEM) for analyzing fracture problems of 2D elasticity. NS-FEM uses strain smoothing technique over domains associated with nodes compute system stiffness matrix, which leads line integrations using directly shape function values along boundaries domains. As result, we avoid integration stress singularity at crack tip. It is not necessary divide elements cut...
We propose a method to couple three-dimensional continuum domain molecular dynamics simulate propagating cracks in dynamics. The is treated by an extended finite element handle the discontinuities. coupling based on bridging method, which blends and atomistic energies. Lennard-Jones potential used model interactions domain, Cauchy-Born rule compute material behavior domain. To our knowledge, it first time that three dimensional reported. show suitability of proposed crack problem with region...
In this paper, we present new reliable model order reduction strategies for computational micromechanics. The difficulties rely mainly upon the high dimensionality of parameter space represented by any load path applied onto representative volume element. We take special care challenge selecting an exhaustive snapshot set. This is treated first using a random sampling energy dissipating paths and then in more advanced way Bayesian optimization associated with interlocked division space....
Although being a popular approach for the modeling of laminated composites, mesoscale constitutive models often struggle to represent material response arbitrary load cases. A better alternative in terms accuracy is use FE2 technique upscale microscopic behavior without loss generality, but associated computational effort can be extreme. It therefore interesting explore surrogate strategies that maintain as much fidelity possible while still computationally efficient. In this work, three...
In this work we present a hybrid physics-based and data-driven learning approach to construct surrogate models for concurrent multiscale simulations of complex material behavior. We start from robust but inflexible constitutive increase their expressivity by allowing subset parameters change in time according an evolution operator learned data. This leads flexible model combining encoder decoder. Apart introducing physics-motivated bias the resulting surrogate, internal variables decoder act...