Engineering Papers⌕ Search

Engineering topics

Bennett, Kane

Publications and source records attributed to Bennett, Kane.

labquake_future_prediction

The labquake_future_prediction code is a collection of python modules and scripts that serves as supporting information for the article “Predicting future laboratory fault friction through deep learning” for publication in the journal of “Geophysical Research Letters”. It is designed to predict laboratory fault slips in the immediate future by scanning continuous acoustic emission (AE) waveforms recorded in laboratory biaxial shear experiments. The predictions are made with a deep learning model based on convolutional encoder-decoder (CED) models and the Transformer model primarily developed for Natural Language Processing (NLP). The deep learning model is trained with the tensorflow package using publicly available laboratory data sets in standard binary file format in numpy. The utility functions for reading data files, configuring model hyperparameters, constructing the CED and Transformer models, training and testing of the models are defined in python module files. The workflow of training the models for labquake future predictions and the multiple GPU’s rapid model hyperparameter optimization as described in the journal article, are demonstrated in accompanying python script files and Jupyter notebooks.

Wang, Kun↗

Multiscale investigation of the microstructural mechanisms driving ratchet growth in PBX 9502

The high explosive PBX 9502 undergoes irreversible expansion during thermal cycling (“ratchet growth”). Recent innovations in thermomechanical modeling via homogenization strategies are beginning to incorporate mesoscale information such as grain size, total porosity, and spatial distribution of voids and cracks. To generate a complete experimental data set to challenge and inform these models, PBX 9502 pellets were thermally cycled, cross-sectioned using ion polishing, and imaged in high resolution with scanning electron microscopy. Ratchet growth was found to drive expansion through microcracking. Microcracks were affected by agglomeration of crystals within the PBX. Virgin material showed greater ratchet growth than recycled material.

36 MATERIALS SCIENCE↗

A physics-informed and hierarchically regularized data-driven model for predicting fluid flow through porous media

This paper presents a new deep learning data-driven model for predicting structure dependent pore-fluid velocity fields in rock. The model is based on a Convolutional Auto-Encoder (CAE) artificial neural network capable of learning from image data generated by direct numerical simulations of fluid flow through pore-structures, such as by Lattice Boltzmann or molecular dynamics methods. The main novelty of the model in comparison to previous CAE-based data-driven approaches consists of three parts. The first is a methodology for decomposing the full-domain of the porous media into sub-regions, or “sub-domains”, in order to reduce the overall size of the CAE, batch process the sub-domains in parallel, and enable the CAE to learn local and generalizable nonlinear mappings of pore-fluid velocities. The second consists of embedding the finite difference solutions of the incompressible Navier-Stokes and continuity equations into convolutional layers prior to the CAE in order to provide the CAE with knowledge of fluid dynamics physics (PhyFlow). The third main novelty is that the training of the CAE is regularized with a hierarchical loss function that encourages the learning of fluid flow patterns (in a way similar to ranked modes in principal component analysis), ranking from most to least important. This is shown to increase the stability in learning, reduce over-fitting, and promote interpretability of the CAE neural network layers (HierCAE). The comprehensive new data-driven model, which we call the PhyFlow-HierCAE model, is shown to exhibit improved accuracy and generalizability of flow field predictions over conventional CAE models, attributable to the embedded physical knowledge and the hierarchical regularization, as well as realize orders of magnitude speed-ups in computation times as a surrogate for the direct numerical simulations. Examples of training and forward predictions on unseen pore-structures are provided and evaluated for data from Lattice Boltzmann and molecular dynamics simulations of pore-fluid flow. The model is shown to be a fast and accurate emulator (or “surrogate”) for predicting effective permeability of unseen pore-structures based on learning from relatively small direct numerical simulation datasets.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

New self-consistent homogenization for thermo-elastic polycrystals with imperfect interfaces

Here we present a new thermo-elastic self-consistent formulation to calculate the thermo-mechanical response of polycrystalline aggregates with imperfect interfaces. The new model is based on an improved treatment of the effect of imperfect interfaces on the micromechanical response of the grains compared to pre-existing models. New expressions/algorithms are derived for the effective elastic and thermal properties, localization relations, and calculation of intragranular field fluctuations. In addition, derivation of a novel integral equation for the strain field in a heterogeneous elastic medium subjected to arbitrary thermal strain field and displacement jumps is presented. The new model is applied to study the effective thermo-elastic properties and average field fluctuations as a function of the interface properties for imperfectly-bonded triaminotrinitrobenzene (TATB) polycrystalline granular aggregates. The proposed formulation is general in the sense that it can be extended to other material systems and deformation regimes that require a proper consideration of the effect of imperfect interface properties on the material's thermo-mechanical response.

36 MATERIALS SCIENCE↗