When deformation gradients act on the scale of the microstructure of a part due to geometry and loading, spatial correlations and finite-size effects in simulation cells cannot be neglected. We propose a multiscale method that accounts for these effects using a variational autoencoder to encode the structure-property map of the stochastic volume elements making up the statistical description of the part. In this paradigm the autoencoder can be used to directly encode the microstructure or, alternatively, its latent space can be sampled to provide likely realizations. We demonstrate the method on three examples using the common additively manufactured material AlSi10Mg in: (a) a comparison with direct numerical simulation of the part microstructure, (b) a push forward of microstructural uncertainty to performance quantities of interest, and (c) a simulation of functional gradation of a part with stochastic microstructure.
In this work, a multiscale approach based on phase field was developed to simulate the microstructure's evolution under irradiation in binary systems, from atomic to microstructural scale. For that purpose, an efficient numerical scheme was developed. In the case of AgCu alloy under Krypton ions irradiation, phenomenological parameters were computed using atomistic methods, as a function of the temperature and the irradiation flux. As a result, we predicted the influence of the irradiation flux and the temperature on the formation of patterned microstructures. In the case of AgCu, our model wa…
The superite phase and phase transition inducing multiscale solidification microstructures and segregations in steels
Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences（中国科学院金属研究所）；Department of Materials Science and Engineering, University of Science and Technology Liaoning（辽宁科技大学材料科学与工程学院）；Shanghai Synchrotron Radiation Facility (SSRF), Shanghai Advanced Research Institute, Chinese Academy of Sciences（中国科学院上海高等研究院）
In this article, we combine casting experiments and quantitative simulations to present a novel multiscale modeling approach to predict local primary dendritic spacings in metallic alloys solidified in conditions relevant to industrial casting processes. To this end, primary dendritic spacings were measured in instrumented casting experiments in Al-Cu alloys containing 1\,wt\% and 4\,wt\% of Cu, and they were compared to spacing stability ranges and average spacings in dendritic arrays simulated using phase-field (PF) and dendritic needle network (DNN) models. It is first shown that PF and DNN…
Multiscale simulations are demanding in terms of computational resources. In the context of continuum micromechanics, the multiscale problem arises from the need of inferring macroscopic material parameters from the microscale. If the underlying microstructure is explicitly given by means of microCT-scans, convolutional neural networks can be used to learn the microstructure-property mapping, which is usually obtained from computational homogenization. The CNN approach provides a significant speedup, especially in the context of heterogeneous or functionally graded materials. Another applicati…
ICME approaches provide decision support for materials design by establishing quantitative process-structure-property relations. Confidence in the decision support, however, must be achieved by establishing uncertainty bounds in ICME model chains. The quantification and propagation of uncertainty in computational materials science, however, remains a rather unexplored aspect of computational materials science approaches. Moreover, traditional uncertainty propagation frameworks tend to be limited in cases with computationally expensive simulations. A rather common and important model chain is t…
Nantes Université（南特大学）；Ecole Centrale Nantes（南特中央理工学院）；CNRS（法国国家科学研究中心）
The mechanical behavior of inelastic materials with microstructure is very complex and hard to grasp with heuristic, empirical constitutive models. For this purpose, multiscale, homogenization approaches are often used for performing reliable, accurate predictions of the macroscopic mechanical behavior of solids and structures. Nevertheless, the calculation cost of such approaches is extremely high and prohibitive for real-scale applications involving inelastic materials. Here, we propose the so-called Thermodynamics-based Artificial Neural Networks (TANN) for the constitutive modeling of mate…
This study explores the role that the microstructure plays in determining the macroscopic static response of porous elastic continua and exposes the occurrence of position-dependent nonlocal effects that are strictly correlated to the configuration of the microstructure. Then, a nonlocal continuum theory based on variable-order fractional calculus is developed in order to accurately capture the complex spatially distributed nonlocal response. The remarkable potential of the fractional approach is illustrated by simulating the nonlinear thermoelastic response of porous beams. The performance, e…
Consistent machine learning for topology optimization with microstructure-dependent neural network material models
Delft University of Technology（代尔夫特理工大学）；Princeton University（普林斯顿大学）；Princeton Materials Institute (PMI), Princeton University（普林斯顿材料研究所（PMI），普林斯顿大学）；Brown University（布朗大学）
Size effects and idealized dislocation microstructure at small scales: predictions of a phenomenological model of Mesoscopic Field Dislocation Mechanics: Part II
We present a novel machine learning based surrogate modeling method for predicting spatially resolved 3D microstructure evolution of polycrystalline materials under uniaxial tensile loading. Our approach is orders of magnitude faster than the existing crystal plasticity methods enabling the simulation of large volumes that would be otherwise computationally prohibitive. This work is a major step beyond existing ML-based modeling results, which have been limited to either 2D structures or only providing average, rather than local, predictions. We demonstrate the speed and accuracy of our surrog…
Gravity-induced buoyancy, inevitable in most solidification processes, substantially alters the dynamics of crystal growth, such that incorporating fluid flow in solidification models is crucial to understand and predict key aspects of microstructure selection. Here, we present a multi-scale Dendritic Needle Network (DNN) model for directional solidification that includes buoyant flow in the liquid, and apply it to a range of alloys and growth conditions. After a brief presentation of the model, we study the selection of stable primary dendrite arm spacings in Al-4at.%Cu and in Ti-45at.%Al all…
We report a multiscale approach of broad applicability to stochastic reconstruction of multiphase materials, including porous ones. The approach devised uses an optimization method, such as the simulated annealing (SA) and the so-called entropic descriptors (EDs). For a binary pattern, they quantify spatial inhomogeneity or statistical complexity at discrete length-scales. The EDs extract dissimilar structural information to that given by two-point correlation functions (CFs). Within the SA, we use an appropriate cost function consisting of EDs or comprised of EDs and CFs. It was found that th…
Plastic deformation of crystalline and amorphous matter often involves intermittent local strain burst events. To understand the physical background of the phenomenon a minimal stochastic mesoscopic model was introduced, where microstructural details are represented by a fluctuating local yielding threshold. In the present paper, we propose a method for determining this yield stress distribution by lower scale discrete dislocation dynamics simulations and using a weakest link argument. The success of scale-linking is demonstrated on the stress-strain curves obtained by the resulting mesoscopic…
Focused electron beam induced deposition (FEBID) is a powerful technique for 3D-printing of complex nanodevices. However, for resolutions below 10 nm, it struggles to control size, morphology and composition of the structures, due to a lack of molecular-level understanding of the underlying irradiation-driven chemistry (IDC). Computational modelling is a tool to comprehend and further optimise FEBID-related technologies. Here we utilise a novel multiscale methodology which couples Monte Carlo simulations for radiation transport with irradiation-driven molecular dynamics for simulating IDC with…
