Residual fourier neural operator for thermochemical curing of composites
- URL: http://arxiv.org/abs/2111.10262v1
- Date: Mon, 15 Nov 2021 14:28:11 GMT
- Title: Residual fourier neural operator for thermochemical curing of composites
- Authors: Gengxiang Chen, Yingguang Li, Xu liu, Qinglu Meng, Jing Zhou,
Xiaozhong Hao
- Abstract summary: This paper proposes a Residual Fourier Neural Operator (ResFNO) to establish the direct high-dimensional mapping from any given cure cycle to the corresponding temperature histories.
By integrating domain knowledge into a time-resolution independent parameterized neural network, the mapping between cure cycles to temperature histories can be learned using limited number of labelled data.
- Score: 9.236600710244478
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: During the curing process of composites, the temperature history heavily
determines the evolutions of the field of degree of cure as well as the
residual stress, which will further influence the mechanical properties of
composite, thus it is important to simulate the real temperature history to
optimize the curing process of composites. Since thermochemical analysis using
Finite Element (FE) simulations requires heavy computational loads and
data-driven approaches suffer from the complexity of highdimensional mapping.
This paper proposes a Residual Fourier Neural Operator (ResFNO) to establish
the direct high-dimensional mapping from any given cure cycle to the
corresponding temperature histories. By integrating domain knowledge into a
time-resolution independent parameterized neural network, the mapping between
cure cycles to temperature histories can be learned using limited number of
labelled data. Besides, a novel Fourier residual mapping is designed based on
mode decomposition to accelerate the training and boost the performance
significantly. Several cases are carried out to evaluate the superior
performance and generalizability of the proposed method comprehensively.
Related papers
- Discovering symbolic expressions with parallelized tree search [59.92040079807524]
Symbolic regression plays a crucial role in scientific research thanks to its capability of discovering concise and interpretable mathematical expressions from data.
Existing algorithms have faced a critical bottleneck of accuracy and efficiency over a decade when handling problems of complexity.
We introduce a parallelized tree search (PTS) model to efficiently distill generic mathematical expressions from limited data.
arXiv Detail & Related papers (2024-07-05T10:41:15Z) - A finite element-based physics-informed operator learning framework for spatiotemporal partial differential equations on arbitrary domains [33.7054351451505]
We propose a novel finite element-based physics operator learning framework that allows for predicting dynamics governed by partial differential equations (PDEs)
The proposed operator learning framework takes a temperature field at the current time step as input and predicts a temperature field at the next time step.
Networks successfully predict the temperature evolution over time for any initial temperature field at high accuracy compared to the FEM solution.
arXiv Detail & Related papers (2024-05-21T02:41:40Z) - Online Variational Sequential Monte Carlo [49.97673761305336]
We build upon the variational sequential Monte Carlo (VSMC) method, which provides computationally efficient and accurate model parameter estimation and Bayesian latent-state inference.
Online VSMC is capable of performing efficiently, entirely on-the-fly, both parameter estimation and particle proposal adaptation.
arXiv Detail & Related papers (2023-12-19T21:45:38Z) - Probabilistic Physics-integrated Neural Differentiable Modeling for
Isothermal Chemical Vapor Infiltration Process [3.878427803346315]
Chemical vapor infiltration (CVI) is a widely adopted manufacturing technique used in producing carbon-carbon and carbon-silicon carbide composites.
The densification process during CVI critically influences the final performance, quality, and consistency of these composite materials.
We have developed a data-driven predictive model using the physics-integrated neural differentiable (PiNDiff) modeling framework.
arXiv Detail & Related papers (2023-11-13T23:25:18Z) - Bayesian inference of composition-dependent phase diagrams [47.79947989845143]
We develop a method in which Bayesian inference is employed to combine thermodynamic data from molecular dynamics (MD), melting point simulations, and phonon calculations, process these data, and yield a temperature-concentration phase diagram.
The developed algorithm was successfully tested on two binary systems, Ge-Si and K-Na, in the full range of concentrations and temperatures.
arXiv Detail & Related papers (2023-09-03T20:57:10Z) - Monte Carlo Neural PDE Solver for Learning PDEs via Probabilistic Representation [59.45669299295436]
We propose a Monte Carlo PDE solver for training unsupervised neural solvers.
We use the PDEs' probabilistic representation, which regards macroscopic phenomena as ensembles of random particles.
Our experiments on convection-diffusion, Allen-Cahn, and Navier-Stokes equations demonstrate significant improvements in accuracy and efficiency.
arXiv Detail & Related papers (2023-02-10T08:05:19Z) - Applying Physics-Informed Enhanced Super-Resolution Generative
Adversarial Networks to Turbulent Premixed Combustion and Engine-like Flame
Kernel Direct Numerical Simulation Data [0.0]
This work advances the recently developed PIESRGAN modeling approach to turbulent premixed combustion.
The resulting model provides good results for a priori and a posteriori tests on direct numerical simulation data of a fully turbulent premixed flame kernel.
arXiv Detail & Related papers (2022-10-28T15:27:46Z) - Physics-based Learning of Parameterized Thermodynamics from Real-time
Thermography [0.0]
We present a novel physics-based approach to learning a thermal process's dynamics from real-time thermographic data.
We show that our approach is robust against noise, and can be used to further refine parameter estimates.
arXiv Detail & Related papers (2022-03-24T16:06:31Z) - Analytical Modelling of Exoplanet Transit Specroscopy with Dimensional
Analysis and Symbolic Regression [68.8204255655161]
The deep learning revolution has opened the door for deriving such analytical results directly with a computer algorithm fitting to the data.
We successfully demonstrate the use of symbolic regression on synthetic data for the transit radii of generic hot Jupiter exoplanets.
As a preprocessing step, we use dimensional analysis to identify the relevant dimensionless combinations of variables.
arXiv Detail & Related papers (2021-12-22T00:52:56Z) - Physics-Informed Neural Network for Modelling the Thermochemical Curing
Process of Composite-Tool Systems During Manufacture [11.252083314920108]
We present a PINN to simulate thermochemical evolution of a composite material on a tool undergoing cure in an autoclave.
We train the PINN with a technique that automatically adapts the weights on the loss terms corresponding to PDE, boundary, interface, and initial conditions.
The performance of the proposed PINN is demonstrated in multiple scenarios with different material thicknesses and thermal boundary conditions.
arXiv Detail & Related papers (2020-11-27T00:56:15Z) - A Near-Optimal Gradient Flow for Learning Neural Energy-Based Models [93.24030378630175]
We propose a novel numerical scheme to optimize the gradient flows for learning energy-based models (EBMs)
We derive a second-order Wasserstein gradient flow of the global relative entropy from Fokker-Planck equation.
Compared with existing schemes, Wasserstein gradient flow is a smoother and near-optimal numerical scheme to approximate real data densities.
arXiv Detail & Related papers (2019-10-31T02:26:20Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.