Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Embedded computing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,513 records · Page 84
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

XRF Characterization of MMOD Impact Damage

Introduction The Hypervelocity Impact Technology (HVIT) Group at the NASA Johnson Space Center (JSC) has performed over 150 post flight MMOD inspections on returned space-ex-posed hardware and collected over 1,600 samples. Determination of the impactor source typically requires an analysis of the projectile residue in the damage feature. At this time, about 20% of the samples have been processed. Types of Samples Samples in the HVIT group collection can be classified into two types. Indirect (non-destructive) samples are produced when a pliable material (e.g., adhesive tape or dental mold) is inserted into an impact feature. With this type of sampling technique, crater material and (hopefully) projectile residue material is embedded in the sample media and can detected using further chemical characterizing techniques. Direct (destructive) sampling involves the in-tact extraction of material from the spacecraft and is done in a manner that the entire impact feature can be excised intact. This technique provides the greatest chance of successfully determining impactor source due to the entirety of the projectile remnants being present within the collected sample. Updated Projectile Residue Detection Techniques Previous assessments of projectile provenance have involved the use of samples that are first mounted in epoxy, sectioned to expose the crater interior, imaged with a scanning electron microscope (SEM), then any observed impactor residue was then analyzed using an energy dispersive spectrometer (EDS). This technique, while proven, may be supplemented with other types of elemental analysis techniques that can observe major element chemistry, in order to improve throughput of sample analysis. The Astromaterials Research and Exploration Science (ARES) Division at JSC, for example, operates a Bruker M4 Tornado micro-X-Ray fluorescence (XRF) spectrometer, capable of creating elemental maps of areas around these small impact features in a couple of hours, with a spot size <20 μm. The potential advantage with this technique lies in sample preparation, as no polishing or conductive coatings are needed. Early trials attempted to deter-mine impactor composition by collecting spectra from intact samples by aiming the X-Rays directly into the impact feature, but the signal:noise ratio was too low for confident material identification. Sample preparation for XRF then relies on sectioning the samples parallel to the impactor trajectory in order to expose the largest area of subsurface damage possible, thus maximizing the signal from any small impactor remnants. Complementary to the use of XRF for sample analysis, ARES also houses a Nikon XTH 320 micro-X-Ray computed tomography (XCT). This instrument can be used to create 3D visualizations of impact damage for the determination of crater volumes and dimensions, as well as for the determination of the best place to section each sample for the subsequent XRF analysis.

J L Hyde↗
DOE OSTI2025

CIE Curriculum Guide (V.2.0)

The Cyber-Informed Engineering (CIE) Curriculum Guide offers a comprehensive framework, guidance, and resources for integrating CIE into university-level engineering programs and related educational activities. The primary goal is to help educators adopt CIE principles into their teaching to produce future engineers and technicians who understand digital risks in modern engineered systems, thereby addressing the nation’s infrastructure resilience needs. This guide outlines practical integration examples, links to resources to accelerate CIE adoption, and shares insights from partner academic institutions on various implementation strategies. CIE is a framework for embedding engineered controls that mitigate the impact of cyber-attacks in any cyber-physical system used in critical energy infrastructure and other sectors. Developed by the U.S. Department of Energy’s Office of Cybersecurity, Energy Security, and Emergency Response (CESER), the National Cyber-Informed Engineering Strategy emphasizes embedding CIE into formal education, training, and credentialing. This guide supports this strategic objective by providing examples of integrating CIE concepts into engineering curricula, from class activities to new courses and certificate programs. The importance of educating cyber-informed engineers is underscored by the evolving cybersecurity threats facing engineered systems. As industrial control systems (ICS) increasingly incorporate digital technologies, the responsibility for security extends to both cyber professionals and engineers. CIE addresses critical gaps in designing and protecting physical systems with digital components against cyber risks, ensuring engineers consider digital risk throughout the engineering design lifecycle. Currently, engineering education does not routinely include cyber-informed principles, highlighting a gap in addressing modern engineering system risks. This guide advocates for updating engineering curricula to include digital risk management as a fundamental element. By doing so, future engineers will be equipped to design resilient systems that mitigate digital risks from the outset. Through this guide, engineering faculty can integrate CIE into their curricula, bridging the gap between digital risk and engineering. This approach prepares a cyber-informed workforce capable of safeguarding the cyber-physical systems crucial to national security and public welfare. By embedding CIE into education and training, institutions can produce engineers and technicians who can effectively mitigate cyber impacts throughout the engineering design lifecycle, resulting in more secure critical infrastructures.

42 - ENGINEERING↗
DOE OSTI2022

Design-Technology Co-Optimization for NVM-based Neuromorphic Processing Elements

An emerging use-case of machine learning (ML) is to train a model on a high-performance system and deploy the trained model on energy-constrained embedded systems. Neuromorphic hardware platforms, which operate on principles of the biological brain, can significantly lower the energy overhead of a machine learning inference task, making these platforms an attractive solution for embedded ML systems. In this paper, we present a design-technology tradeoff analysis to implement such inference tasks on the processing elements (PEs) of a Non-Volatile Memory (NVM)-based neuromorphic hardware. Through detailed circuit-level simulations at scaled process technology nodes, we show the negative impact of technology scaling on the information-processing latency, which impacts the quality-of-service (QoS) of an embedded ML system. At a finer granularity, the latency inside a PE depends on 1) the delay introduced by parasitic components on its current paths, and 2) the varying delay to sense different resistance states of its NVM cells. Based on these two observations, we make the following three contributions. First, on the technology front, we propose an optimization scheme where the NVM resistance state that takes the longest time to sense is set on current paths having the least delay, and vice versa, reducing the average PE latency, which improves the QoS. Second, on the architecture front, we introduce isolation transistors within each PE to partition it into regions that can be individually power-gated, reducing both latency and energy. Finally, on the system-software front, we propose a mechanism to leverage the proposed technological and architectural enhancements when implementing a machine-learning inference task on neuromorphic PEs of the hardware. Evaluations with a recent neuromorphic hardware architecture show that our proposed design-technology co-optimization approach improves both performance and energy efficiency of machine-learning inference tasks without incurring high cost-per-bit.

42 ENGINEERING↗
DOE OSTI2023

Extended isogeometric analysis of multi-material and multi-physics problems using hierarchical B-splines

Here, this paper presents an immersed, isogeometric finite element framework to predict the response of multi-material, multi-physics problems with complex geometries using locally refined discretizations. To circumvent the need to generate conformal meshes, this work uses an extended finite element method (XFEM) to discretize the governing equations on non-conforming, embedding meshes. A flexible approach to create truncated hierarchical B-splines discretizations is presented. This approach enables the refinement of each state variable field individually to meet field-specific accuracy requirements. To obtain an immersed geometry representation that is consistent across all hierarchically refined B-spline discretizations, the geometry is immersed into a single mesh, the XFEM background mesh, which is constructed from the union of all hierarchical B-spline meshes. An extraction operator is introduced to represent the truncated hierarchical B-spline bases in terms of Lagrange shape functions on the XFEM background mesh without loss of accuracy. The truncated hierarchical B-spline bases are enriched using a generalized Heaviside enrichment strategy to accommodate small geometric features and multi-material problems. The governing equations are augmented by a formulation of the face-oriented ghost stabilization enhanced for locally refined B-spline bases. We present examples for two- and three-dimensional linear elastic and thermo-elastic problems. The numerical results validate the accuracy of our framework. The results also demonstrate the applicability of the proposed framework to large, geometrically complex problems.

42 ENGINEERING↗
DOE OSTI2023

On the influence of over-parameterization in manifold based surrogates and deep neural operators

Constructing accurate and generalizable approximators (surrogate models) for complex physico-chemical processes exhibiting highly non-smooth dynamics is challenging. The main question is what type of surrogate models we should construct and should these models be under-parameterized or over-parameterized. In this work, we propose new developments and perform comparisons for two promising approaches: manifold-based polynomial chaos expansion (m-PCE) and the deep neural operator (DeepONet), and we examine the effect of over-parameterization on generalization. While m-PCE enables the construction of a mapping by first identifying low-dimensional embeddings of the input functions, parameters, and quantities of interest (QoIs), a neural operator learns the nonlinear mapping via the use of deep neural networks. Here, we demonstrate the performance of these methods in terms of generalization accuracy by solving the 2D time-dependent Brusselator reaction-diffusion system with uncertainty sources, modeling an autocatalytic chemical reaction between two species. We first propose an extension of the m-PCE by constructing a mapping between latent spaces formed by two separate embeddings of the input functions and the output QoIs. To further enhance the accuracy of the DeepONet, we introduce weight self-adaptivity in the loss function. We demonstrate that the performance of m-PCE and DeepONet is comparable for cases of relatively smooth input-output mappings. However, when highly non-smooth dynamics is considered, DeepONet shows higher approximation accuracy. We also find that for m-PCE, modest over-parameterization leads to better generalization, both within and outside of distribution, whereas aggressive over-parameterization leads to over-fitting. In contrast, an even highly over-parameterized DeepONet leads to better generalization for both smooth and non-smooth dynamics. Furthermore, we compare the performance of the above models with another recently proposed operator learning model, the Fourier Neural Operator, and show that its over-parameterization also leads to better generalization. Taken together, our studies show that m-PCE can provide very good accuracy at very low training cost, whereas a highly over-parameterized DeepONet can provide better accuracy and robustness to noise but at higher training cost. In both methods, the inference cost is negligible.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1997

Object detection in natural backgrounds predicted by discrimination performance and models

Many models of visual performance predict image discriminability, the visibility of the difference between a pair of images. We compared the ability of three image discrimination models to predict the detectability of objects embedded in natural backgrounds. The three models were: a multiple channel Cortex transform model with within-channel masking; a single channel contrast sensitivity filter model; and a digital image difference metric. Each model used a Minkowski distance metric (generalized vector magnitude) to summate absolute differences between the background and object plus background images. For each model, this summation was implemented with three different exponents: 2, 4 and infinity. In addition, each combination of model and summation exponent was implemented with and without a simple contrast gain factor. The model outputs were compared to measures of object detectability obtained from 19 observers. Among the models without the contrast gain factor, the multiple channel model with a summation exponent of 4 performed best, predicting the pattern of observer d's with an RMS error of 2.3 dB. The contrast gain factor improved the predictions of all three models for all three exponents. With the factor, the best exponent was 4 for all three models, and their prediction errors were near 1 dB. These results demonstrate that image discrimination models can predict the relative detectability of objects in natural scenes.

NASA Center ARC↗
DOE OSTI2024

Sequential Kalman tuning of the t -preconditioned Crank-Nicolson algorithm: efficient, adaptive and gradient-free inference for Bayesian inverse problems

Ensemble Kalman Inversion (EKI) has been proposed as an efficient method for the approximate solution of Bayesian inverse problems with expensive forward models. However, when applied to the Bayesian inverse problem EKI is only exact in the regime of Gaussian target measures and linear forward models. Here, in this work we propose embedding EKI and Flow Annealed Kalman Inversion, its normalizing flow (NF) preconditioned variant, within a Bayesian annealing scheme as part of an adaptive implementation of the t-preconditioned Crank-Nicolson (tpCN) sampler. The tpCN sampler differs from standard pCN in that its proposal is reversible with respect to the multivariate t-distribution. The more flexible tail behaviour allows for better adaptation to sampling from non-Gaussian targets. Within our Sequential Kalman Tuning (SKT) adaptation scheme, EKI is used to initialize and precondition the tpCN sampler for each annealed target. The subsequent tpCN iterations ensure particles are correctly distributed according to each annealed target, avoiding the accumulation of errors that would otherwise impact EKI. We demonstrate the performance of SKT for tpCN on three challenging numerical benchmarks, showing significant improvements in the rate of convergence compared to adaptation within standard SMC with importance weighted resampling at each temperature level, and compared to similar adaptive implementations of standard pCN. The SKT scheme applied to tpCN offers an efficient, practical solution for solving the Bayesian inverse problem when gradients of the forward model are not available. Code implementing the SKT schemes for tpCN is available at https://github.com/RichardGrumitt/KalmanMC.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Group-Invariant Quantum Machine Learning

Quantum machine learning (QML) models are aimed at learning from data encoded in quantum states. Recently, it has been shown that models with little to no inductive biases (i.e., with no assumptions about the problem embedded in the model) are likely to have trainability and generalization issues, especially for large problem sizes. As such, it is fundamental to develop schemes that encode as much information as available about the problem at hand. In this work we present a simple, yet powerful, framework where the underlying invariances in the data are used to build QML models that, by construction, respect those symmetries. These so-called group-invariant models produce outputs that remain invariant under the action of any element of the symmetry group $\mathfrak{G}$ associated with the dataset. We present theoretical results underpinning the design of $\mathfrak{G}$-invariant models, and exemplify their application through several paradigmatic QML classification tasks, including cases when $\mathfrak{G}$ is a continuous Lie group and also when it is a discrete symmetry group. Notably, our framework allows us to recover, in an elegant way, several well-known algorithms for the literature, as well as to discover new ones. Taken together, we expect that our results will help pave the way towards a more geometric and group-theoretic approach to QML model design.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Exploiting correlations in multi-coincidence Coulomb explosion patterns for differentiating molecular structures using machine learning

Coulomb explosion imaging (CEI) is a powerful technique for capturing the real-time motion of individual atoms during ultrafast photochemical reactions. CEI generates high-dimensional data with naturally embedded correlations that allow mapping the coordinated motion of nuclei in molecules. This enables reliable separation of competing reaction pathways and makes this approach uniquely suited for characterizing weak reaction channels. However, rich information contained in experimental CEI patterns remains largely underexploited due to challenges in visualizing correlations between multiple observables in multi-dimensional parameter space. Here we present a new approach to CEI of intermediate-sized polyatomic molecules, detecting up to eight ionic fragments in coincidence and leveraging machine-learning-based analysis to identify patterns and correlations in the resulting high-dimensional momentum-space data, enabling robust molecular structure identification and differentiation. Our approach provides high-dimensional background-free data encoding exceptionally rich structural information and establishes an automated, scalable framework for extracting insightful information from the data. As a demonstration, we apply this method to image and distinguish dichloroethylene isomers, showcasing its potential for broader applications in molecular imaging. Our results pave the way for channel-specific analysis of ultrafast structural dynamics in chemically relevant systems, particularly for disentangling mixed reaction pathways and detecting contributions from weak channels and minority species.

Chemical Physics (physics.chem-ph)↗
NASA NTRSDate not supplied

A Machine Learning Framework for Error Compensation in Radiative Transfer Calculations

Radiative heat transfer influences the amount of heat flux transferred to the surface of the hypersonic vehicle, which is essential to evaluate the performance of thermal protection systems. The radiative heat flux is found to be computationally prohibitive while accounting for the variation in spatial, angular, and spectral domains. A new methodology has been recently developed to alleviate the cost of computation in the spectral domain by constructing flow-agnostic reduced-order models (ROMs). The developed spectral ROM databases provide grouping strategies that account for non-equilibrium absorption and emission as well as interaction between disparate species due to spectral overlap in associated radiative processes. However, the developed ROMs need to be optimized for a specific combination of interacting gas species and would need to re-calibrated in case individual species are added/omitted. In this work, we use various machine learning (ML) techniques to approximate the radiative intensities determined by a ROM optimized for a specific gas mixture. The ML model relies on the ROM databases developed for a single species which ignores any spectral overlap. Thus, radiation evaluation starts with a simple summation of radiative intensities predicted using these non-calibrated ROMs for the contributing species. The ML framework then provides a correction to account for the interplay in the frequency, i.e., emission of photons by one species and absorption by another, and yields mixture-specific radiation fields. Once trained on the individual ROM databases, the ML framework offers instantaneous corrections that serves as a time/cost effective alternative to the optimization of ROMs for a specific gas mixture. The ML framework is trained on both the high fidelity and ROM evaluated line of sight (LOS) data from Orion, Stardust, and FIRE II cases to obtain a general purpose correction model for earth re-entry scenarios when radiation contributions from both atomic nitrogen and atomic oxygen are considered. A geometric length scale parameter is used in the training process to account for errors introduced in the ROM databases as a consequence of high optical thickness. The efficacy of the ML framework is underscored through extensive analysis of train and test errors with respect to all the re-entry scenarios. The applicability of such an ML framework was further corroborated by embedding it in a state-of-the-art US3D - NERO system for determining the radiative heat flux transferred to the hypersonic vehicle surface.

Radiation↗
DOE OSTI2021

Converting tabular data into images for deep learning with convolutional neural networks

Abstract Convolutional neural networks (CNNs) have been successfully used in many applications where important information about data is embedded in the order of features, such as speech and imaging. However, most tabular data do not assume a spatial relationship between features, and thus are unsuitable for modeling using CNNs. To meet this challenge, we develop a novel algorithm, image generator for tabular data (IGTD), to transform tabular data into images by assigning features to pixel positions so that similar features are close to each other in the image. The algorithm searches for an optimized assignment by minimizing the difference between the ranking of distances between features and the ranking of distances between their assigned pixels in the image. We apply IGTD to transform gene expression profiles of cancer cell lines (CCLs) and molecular descriptors of drugs into their respective image representations. Compared with existing transformation methods, IGTD generates compact image representations with better preservation of feature neighborhood structure. Evaluated on benchmark drug screening datasets, CNNs trained on IGTD image representations of CCLs and drugs exhibit a better performance of predicting anti-cancer drug response than both CNNs trained on alternative image representations and prediction models trained on the original tabular data.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Solving the Orszag–Tang vortex magnetohydrodynamics problem with physics-constrained convolutional neural networks

We study the 2D Orszag–Tang vortex magnetohydrodynamics (MHD) problem through the use of physics-constrained convolutional neural networks (PCNNs) for forecasting the density, ρ, and the magnetic field, B, as well as the prediction of B given the velocity field v of the fluid. In addition to translation equivariance from the convolutional architecture, other physics constraints were embedded: absence of magnetic monopoles, non-negativity of ρ, use of only relevant variables, and the periodic boundary conditions of the problem. The use of only relevant variables and the hard constraint of non-negative ρ were found to facilitate learning greatly. The divergenceless condition ∇·B=0 was implemented as a hard constraint up to machine precision through the use of a magnetic potential to define B=∇×A. Residual networks and data augmentation were also used to improve performance. This allowed for some of the residual models to function as surrogate models and provide reasonably accurate simulations. For the prediction task, the PCNNs were evaluated against a physics-informed neural network, which had the ideal MHD induction equation as a soft constraint. Several models were able to generate highly accurate fields, which are visually almost indistinguishable and have low mean squared error. Only methods with built-in hard constraints produced physical fields with ∇·B=0. The use of PCNNs for MHD has the potential to produce physically consistent real-time simulations to serve as virtual diagnostics in cases where inferences must be made with limited observables.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

CoolPINNs: A physics-informed neural network modeling of active cooling in vascular systems

Emerging technologies like hypersonic aircraft, space exploration vehicles, and batteries avail fluid circulation in embedded microvasculatures for efficient thermal regulation. Modeling is vital during the design and operational phases of these engineered systems. However, many challenges exist in developing a modeling framework. What is lacking is an accurate framework that (i) captures sharp jumps in the thermal flux across complex vasculature layouts, (ii) deals with oblique derivatives (involving tangential and normal components), (iii) handles nonlinearity because of radiative heat transfer, (iv) provides a high-speed forecast for real-time monitoring, and (v) facilitates robust inverse modeling. Here, this paper addresses these challenges by availing the power of physics-informed neural networks (PINNs). We develop a fast, reliable, and accurate Scientific Machine Learning (SciML) framework for vascular-based thermal regulation—called CoolPINNs: a PINNs-based modeling framework for active cooling. The proposed mesh-less framework elegantly overcomes all the mentioned challenges. The significance of the reported research is multi-fold. First, the framework is valuable for real-time monitoring of thermal regulatory systems because of rapid forecasting. Second, researchers can address complex thermoregulation designs since the approach is meshless. Finally, the framework facilitates systematic parameter identification and inverse modeling studies, perhaps the most significant utility of the current framework.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Reduced order modeling for flow and transport problems with Barlow Twins self-supervised learning

Abstract We propose a unified data-driven reduced order model (ROM) that bridges the performance gap between linear and nonlinear manifold approaches. Deep learning ROM (DL-ROM) using deep-convolutional autoencoders (DC–AE) has been shown to capture nonlinear solution manifolds but fails to perform adequately when linear subspace approaches such as proper orthogonal decomposition (POD) would be optimal. Besides, most DL-ROM models rely on convolutional layers, which might limit its application to only a structured mesh. The proposed framework in this study relies on the combination of an autoencoder (AE) and Barlow Twins (BT) self-supervised learning, where BT maximizes the information content of the embedding with the latent space through a joint embedding architecture. Through a series of benchmark problems of natural convection in porous media, BT–AE performs better than the previous DL-ROM framework by providing comparable results to POD-based approaches for problems where the solution lies within a linear subspace as well as DL-ROM autoencoder-based techniques where the solution lies on a nonlinear manifold; consequently, bridges the gap between linear and nonlinear reduced manifolds. We illustrate that a proficient construction of the latent space is key to achieving these results, enabling us to map these latent spaces using regression models. The proposed framework achieves a relative error of 2% on average and 12% in the worst-case scenario (i.e., the training data is small, but the parameter space is large.). We also show that our framework provides a speed-up of $$7 \times 10^{6}$$ 7 × 10 6 times, in the best case, and $$7 \times 10^{3}$$ 7 × 10 3 times on average compared to a finite element solver. Furthermore, this BT–AE framework can operate on unstructured meshes, which provides flexibility in its application to standard numerical solvers, on-site measurements, experimental data, or a combination of these sources.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

A new biogeochemical modelling framework (FLaMe-v1.0) for lake methane emissions on the regional scale: development and application to the European domain

This study presents a new physical-biogeochemical modelling framework for simulating lake methane (CH 4 ) emissions at regional scales. The new model, FLaMe-v1.0 (Fluxes of Lake Methane), rests on an innovative, computationally efficient lake clustering approach that enables the simulation of CH 4 emissions across a large number of lakes. Building on the Canadian Small Lake Model (CSLM) that simulates the lake physics, we develop a suite of biogeochemical modules to simulate transient dynamics of organic Carbon (C), Oxygen (O 2 ), and CH 4 . We first test the performance of FLaMe-v1.0 by analyzing physical and biogeochemical processes in two theoretical lakes with characteristics that can be considered representative for many lakes (an oligotrophic, deep lake driven by cold climate versus a eutrophic, shallow lake driven by warm climate). Next, we evaluate the model by comparing simulated and observed timeseries of CH 4 emissions in four well-surveyed lakes. We then apply FLaMe-v1.0 at the European scale to evaluate simulated diffusive and ebullitive lake CH 4 fluxes against in-situ measurements in both boreal and central European regions. Finally, we provide a first assessment of the spatio-temporal variability in CH 4 emissions from European lakes with a surface area comprised between 0.1–1000 km 2 (n= 108 407, total area = 1.33 × 105 km 2 ), indicating a total emission of 0.97 ± 0.23 Tg CH 4 yr −1 , with the uncertainty constrained by combining FLaMe-v1.0 and machine learning techniques. Moreover, 30 % and 70 % of these CH 4 emissions are through diffusive and ebullitive pathways, respectively. Annually averaged CH 4 emission rates per unit lake area during 2010–2016 have a South-to-North decreasing gradient, resulting in a mean over the European domain as 7.39 g CH 4 m −2 yr −1 . Our simulations reveal a strong seasonality (with ice-blocking effects accounted for) in European lake CH 4 emissions, with nearly ten times higher emissions during late summer than during winter. This pronounced seasonal variation highlights the importance of accounting for the sub-annual variability in CH 4 emissions to accurately constrain regional CH 4 budgets. In the future, FLaMe-v1.0 could be embedded into Earth System Models to investigate the feedback between climate warming and global lake CH 4 emissions.

Maisonnier, Manon [Free Univ. of Brussels (Belgium↗
DOE OSTI2020

Critical Infrastructure Classification via CNN-based Modeling and Image Analysis

With recent advances in the fields of satellite imagery and machine learning we now have the ability to develop explainable deep learning models that enhance critical infrastructure analysis. Funded through Idaho National Laboratory’s (INL) Laboratory Directed Research and Development (LDRD) office we are in the process of developing a deep learning model capable of identifying critical infrastructure facilities and embedded features within those facilities. Utilizing current limit of practice techniques in the machine learning areas of explainability and transfer learning our model, once complete, will have the capacity to be used on multiple different imagery data sets and produce results that not only classify critical infrastructure facilities, but also explain why a critical infrastructure facility was classified as a certain type of facility. These advancements eliminate the ‘black box’ approach deep learning models have had in the past, where a user will have to trust the conclusion of a model without understanding what reasoning when into the model’s classification process. They also expand a model’s usefulness, traditionally a deep learning model will have to use the same data set it was originally trained on. Given the long training times of deep learning models this is impractical in a number of scenarios. By utilizing transfer learning advancements, we are eliminating the need to train our model on the same data set it is then run on to classify critical infrastructure facilities. We are also enabling the analysis and classification of data sets that are potentially too small to be divided into a training and testing data set. Once completed this model can provide a foundation to enhanced critical infrastructure analysis, dependency analysis, and potential disaster relief efforts.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Embedding physics domain knowledge into a Bayesian network enables layer-by-layer process innovation for photovoltaics

Process optimization of photovoltaic devices is a time-intensive, trial-and-error endeavor, which lacks full transparency of the underlying physics and relies on user-imposed constraints that may or may not lead to a global optimum. Herein, we demonstrate that embedding physics domain knowledge into a Bayesian network enables an optimization approach for gallium arsenide (GaAs) solar cells that identifies the root cause(s) of underperformance with layer-by-layer resolution and reveals alternative optimal process windows beyond traditional black-box optimization. Our Bayesian network approach links a key GaAs process variable (growth temperature) to material descriptors (bulk and interface properties, e.g., bulk lifetime, doping, and surface recombination) and device performance parameters (e.g., cell efficiency). For this purpose, we combine a Bayesian inference framework with a neural network surrogate device-physics model that is 100× faster than numerical solvers. With the trained surrogate model and only a small number of experimental samples, our approach reduces significantly the time-consuming intervention and characterization required by the experimentalist. As a demonstration of our method, in only five metal organic chemical vapor depositions, we identify a superior growth temperature profile for the window, bulk, and back surface field layer of a GaAs solar cell, without any secondary measurements, and demonstrate a 6.5% relative AM1.5G efficiency improvement above traditional grid search methods.

14 SOLAR ENERGY↗
DOE OSTI2023

Rheological Properties of Small-Molecular Liquids at High Shear Strain Rates

Molecular-scale understanding of rheological properties of small-molecular liquids and polymers is critical to optimizing their performance in practical applications such as lubrication and hydraulic fracking. We combine nonequilibrium molecular dynamics simulations with two unsupervised machine learning methods: principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE), to extract the correlation between the rheological properties and molecular structure of squalane sheared at high strain rates (10 6 –10 10 s -1 ) for which substantial shear thinning is observed under pressures P ϵ 0.1–955 MPa at 293 K. Intramolecular atom pair orientation tensors of 435 × 6 dimensions and the intermolecular atom pair orientation tensors of 61 × 6 dimensions are reduced and visualized using PCA and t-SNE to assess the changes in the orientation order during the shear thinning of squalane. Dimension reduction of intramolecular orientation tensors at low pressures P = 0.1,100 MPa reveals a strong correlation between changes in strain rate and the orientation of the side-backbone atom pairs, end-backbone atom pairs, short backbone-backbone atom pairs, and long backbone-backbone atom pairs associated with a squalane molecule. At high pressures P ≥ 400 MPa, the orientation tensors are better classified by these different pair types rather than strain rate, signaling an overall limited evolution of intramolecular orientation with changes in strain rate. Dimension reduction also finds no clear evidence of the link between shear thinning at high pressures and changes in the intermolecular orientation. The alignment of squalane molecules is found to be saturated over the entire range of rates during which squalane exhibits substantial shear thinning at high pressures.

36 MATERIALS SCIENCE↗