Probabilistic physics-guided transfer learning for material property prediction in extrusion deposition additive manufacturing

DOI10.1016/J.CMA.2023.116660MaRDI QIDQ6185217zbMATH OpenOpenAlexFDO

Authors Akshay J. Thomas, Mateusz Jaszczuk, Eduardo Barocio, Gourab Ghosh, I. Bilionis, R. Byron Pipes

Publication date 29 January 2024

Published in Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.cma.2023.116660

zbMATH Keywords

extrusion deposition additive manufacturingfiber orientationGaussian process surrogateheat transfershort fiber compositestransfer learning

Mathematics Subject Classification ID

Mechanics of deformable solids (74-XX) Classical thermodynamics, heat transfer (80-XX)

Recommendations

• Hybrid thermal modeling of additive manufacturing processes using physics-informed neural networks for temperature prediction and parameter identification
• Transfer learning of deep material network for seamless structure-property predictions
• Data-driven prognostic model for temperature field in additive manufacturing based on the high-fidelity thermal-fluid flow simulation
• Physics-integrated neural differentiable (PiNDiff) model for composites manufacturing
• Denoising diffusion-based synthetic generation of three-dimensional (3D) anisotropic microstructures from two-dimensional (2D) micrographs

Cites work

• A neural network-aided Bayesian identification framework for multiscale modeling of nanocomposites
• Algorithms for generating maximin Latin hypercube and orthogonal designs
• Bayesian calibration of computer models. (With discussion)
• Bayesian inference of non-linear multiscale model parameters accelerated by a deep neural network
• scientific article; zbMATH DE number 6378127 (Why is no real title available?)
• Multi-output separable Gaussian process: towards an efficient, fully Bayesian paradigm for uncertainty quantification
• On the condition number of covariance matrices in kriging, estimation, and simulation of random fields
• Pattern recognition and machine learning.
• Problems with likelihood estimation of covariance functions of spatial Gaussian processes
• Solution of inverse problems with limited forward solver evaluations: a Bayesian perspective
• The no-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo
• The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations

Cited in

(1)

• Physics-integrated neural differentiable (PiNDiff) model for composites manufacturing

This page was built for publication: Probabilistic physics-guided transfer learning for material property prediction in extrusion deposition additive manufacturing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6185217)