SEARCH · Engineering Papers

Results for “High dimensional data,”

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.

SourceResearch connectionFrom yearThrough year

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

At least 541 records · Page 30

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Attend and Decode: 4D fMRI Task State Decoding Using Attention Models

Source code for Brain Attend and Decode paper. Functional magnetic resonance imaging (fMRI) is a neuroimaging modality that captures the blood oxygen level in a subject's brain while the subject either rests or performs a variety of functional tasks under different conditions. Given fMRI data, the problem of inferring the task, known as task state decoding, is challenging due to the high dimensionality (hundreds of million sampling points per datum) and complex spatio-temporal blood flow patterns inherent in the data. In this work, we propose to tackle the fMRI task state decoding problem by casting it as a 4D spatiotemporal classification problem. We present a novel architecture called Brain Attend and Decode (BAnD), that uses residual convolutional neural networks for spatial feature extraction and self-attention mechanisms for temporal modeling. We achieve significant performance gain compared to previous works on a 7-task benchmark from the large-scale Human Connectome Project-Young Adult (HCP-YA) dataset. We also investigate the transferability of BAnD's extracted features on unseen HCP tasks, either by freezing the spatial feature extraction layers and retraining the temporal model, or finetuning the entire model. The pre-trained features from BAnD are useful on similar tasks while finetuning them yields competitive results on unseen tasks/conditions.

Ng, BrendaM.↗

DOE OSTI2025

Personalized and uncertainty-aware coronary hemodynamics simulations: From Bayesian estimation to improved multi-fidelity uncertainty quantification

Non-invasive simulations of coronary hemodynamics have improved clinical risk stratification and treatment outcomes for coronary artery disease, compared to relying on anatomical imaging alone. However, simulations typically use empirical approaches to distribute total coronary flow amongst the arteries in the coronary tree, which ignores patient variability, the presence of disease, and other clinical factors. Further, uncertainty in the clinical data often remains unaccounted for in the modeling pipeline. We present an end-to-end uncertainty-aware pipeline to (1) personalize coronary flow simulations by incorporating vessel-specific coronary flows as well as cardiac function; and (2) predict clinical and biomechanical quantities of interest with improved precision, while accounting for uncertainty in the clinical data. We assimilate patient-specific measurements of myocardial blood flow from clinical CT myocardial perfusion imaging to estimate branch-specific coronary artery flows. Simulated noise in the clinical data is used to estimate the joint posterior distributions of the model parameters using adaptive Markov Chain Monte Carlo sampling. Additionally, the posterior predictive distribution for the relevant quantities of interest is determined using a new approach combining multi-fidelity Monte Carlo estimation with non-linear, data-driven dimensionality reduction. This leads to improved correlations between high- and low-fidelity model outputs. Our framework accurately recapitulates clinically measured cardiac function as well as branch-specific coronary flows under measurement noise uncertainty. We observe substantial reductions in confidence intervals for estimated quantities of interest compared to single-fidelity Monte Carlo estimation and state-of-the-art multi-fidelity Monte Carlo methods. This holds especially true for quantities of interest that showed limited correlation between the low- and high-fidelity model predictions. In addition, the proposed multi-fidelity Monte Carlo estimators are significantly cheaper to compute than traditional estimators, under a specified confidence level or variance. The proposed pipeline for personalized and uncertainty-aware predictions of coronary hemodynamics is based on routine clinical measurements and recently developed techniques for CT myocardial perfusion imaging. The proposed pipeline offers significant improvements in precision and reduction in computational cost.

Bayesian parameter estimation↗

DOE OSTI2015

GAMMA-RAY BURSTS ARE OBSERVED OFF-AXIS

We constrain the jet opening angle and, for the first time, the off-axis observer angle for gamma-ray bursts in the Swift-XRT catalog by using the ScaleFit package to fit afterglow light curves directly to hydrodynamic simulations. The ScaleFit model uses scaling relations in the hydrodynamic and radiation equations to compute synthetic light curves directly from a set of high-resolution two-dimensional relativistic blast wave simulations. The data sample consists of all Swift-XRT afterglows from 2005 to 2012 with sufficient coverage and a known redshift, 226 bursts in total. We find that the jet half-opening angle varies widely but is commonly less than 0.1 rad. The distribution of the electron spectral index is also broad, with a median at 2.30. We find the observer angle to have a median value of 0.57 of the jet opening angle over our sample, which has profound consequences for the predicted rate of observed jet breaks and affects the beaming-corrected total energies of gamma-ray bursts.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2018

Data-Driven Machine Learning for Wind Plant Flow Modeling

In this study, we introduce a data-driven machine learning framework for improving the accuracy of wind plant flow models by learning turbulence model corrections based on data from higher-fidelity simulations. First, a high-dimensional PDE-constrained optimization problem is solved using gradient-based optimization with adjoints to determine optimal eddy viscosity fields that improve the agreement of a medium-fidelity Reynolds-Averaged Navier Stokes (RANS) model with large eddy simulations (LES). A supervised learning problem is then constructed to find general, predictive representations of the optimal turbulence closure. A machine learning technique using Gaussian process regression is trained to predict the eddy viscosity field based on local RANS flow field information like velocities, pressures, and their gradients. The Gaussian process is trained on LES simulations of a single turbine and implemented in a wind plant simulation with 36 turbines. We show improvement over the baseline RANS model with the machine learning correction, and demonstrate the ability to provide accurate confidence levels for the corrections that enable future uncertainty quantification studies.

17 WIND ENERGY↗

DOE OSTI2025

Two-dimensional hydrodynamic viscous electron flow in annular Corbino rings

The concept of fluidic viscosity is ubiquitous in condensed-matter systems hosting a continuum where macroscopic properties can emerge. While an important property of liquids and some solids, only recently was the viscosity of an electron shown to play a role in electronic transport experiments. In this Letter, we present nonlocal electronic transport measurements in concentric annular rings formed in high-mobility two-dimensional electron gases, and the resulting data show that viscous hydrodynamic flow can occur far away from the source-drain current region. Our conclusion of viscous electronic transport is further corroborated by simulations of the Navier-Stokes equations that are found to be in agreement with our measurements below T = 1 K . Finally, this work emphasizes the key role played by viscosity via electron-electron ( e − e ) interaction even when the electronic transport is restricted radially, and for which it should have played no major role. Published by the American Physical Society 2025

Vijayakrishnan, Sujatha (ORCID:0009000080933182)↗

DOE OSTI2016

TU-D-207B-05: Intra-Tumor Partitioning and Texture Analysis of DCE-MRI Identifies Relevant Tumor Subregions to Predict Early Pathological Response of Breast Cancer to Neoadjuvant Chemotherapy

Purpose: To predict early pathological response of breast cancer to neoadjuvant chemotherapy (NAC) based on quantitative, multi-region analysis of dynamic contrast enhancement magnetic resonance imaging (DCE-MRI). Methods: In this institution review board-approved study, 35 patients diagnosed with stage II/III breast cancer were retrospectively investigated using DCE-MR images acquired before and after the first cycle of NAC. First, principal component analysis (PCA) was used to reduce the dimensionality of the DCE-MRI data with a high-temporal resolution. We then partitioned the whole tumor into multiple subregions using k-means clustering based on the PCA-defined eigenmaps. Within each tumor subregion, we extracted four quantitative Haralick texture features based on the gray-level co-occurrence matrix (GLCM). The change in texture features in each tumor subregion between pre- and during-NAC was used to predict pathological complete response after NAC. Results: Three tumor subregions were identified through clustering, each with distinct enhancement characteristics. In univariate analysis, all imaging predictors except one extracted from the tumor subregion associated with fast wash-out were statistically significant (p< 0.05) after correcting for multiple testing, with area under the ROC curve or AUCs between 0.75 and 0.80. In multivariate analysis, the proposed imaging predictors achieved an AUC of 0.79 (p = 0.002) in leave-one-out cross validation. This improved upon conventional imaging predictors such as tumor volume (AUC=0.53) and texture features based on whole-tumor analysis (AUC=0.65). Conclusion: The heterogeneity of the tumor subregion associated with fast wash-out on DCE-MRI predicted early pathological response to neoadjuvant chemotherapy in breast cancer.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2021

Application of an automated machine learning-genetic algorithm (AutoML-GA) coupled with computational fluid dynamics simulations for rapid engine design optimization

In recent years, the use of machine learning-based surrogate models for computational fluid dynamics (CFD) simulations has emerged as a promising technique for reducing the computational cost associated with engine design optimization. However, such methods still suffer from drawbacks. One main disadvantage is that the default machine learning (ML) hyperparameters are often severely suboptimal for a given problem. This has often been addressed by manually trying out different hyperparameter settings, but this solution is ineffective in case of a high-dimensional hyperparameter space. Besides this problem, the amount of data needed for training is also not known a priori. In response to these issues that need to be addressed, the present work describes and validates an automated active learning approach, AutoML-GA, for surrogate-based optimization of internal combustion engines. In this approach, a Bayesian optimization technique is used to find the best machine learning hyperparameters based on an initial dataset obtained from a small number of CFD simulations. Subsequently, a genetic algorithm is employed to locate the design optimum on the ML surrogate surface. In the vicinity of the design optimum, the solution is refined by repeatedly running CFD simulations at the projected optima and adding the newly obtained data to the training dataset. It is demonstrated that AutoML-GA leads to a better optimum with a lower number of CFD simulations, compared to the use of default hyperparameters. The proposed framework offers the advantage of being a more hands-off approach that can be readily utilized by researchers and engineers in industry who do not have extensive machine learning expertise.

Owoyele, Opeoluwa↗

DOE OSTI2020

Identifying Entangled Physics Relationships through Sparse Matrix Decomposition to Inform Plasma Fusion Design [Slides]

The National Ignition Facility (NIF), is a large laser-based inertial confinement fusion (ICF) research device and various input variables in the experimental data are described. The overview included sections on: high-dimensional experimental dataset; surrogate model selection; ML to interpret complex coupling between inputs; importance of variables; Surrogate performance; and, future work.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2019

An Experimental Investigation of Flow Regimes in Imbibition and Drainage Using a Microfluidic Platform

Instabilities in immiscible displacement along fluid−fluid displacement fronts in porous media are undesirable in many natural and engineered displacement processes such as geological carbon sequestration and enhanced oil recovery. In this study, a series of immiscible displacement experiments are conducted using a microfluidic platform across a wide range of capillary numbers and viscosity ratios. The microfluidic device features a water-wet porous medium, which is a two-dimensional representation of a Berea sandstone. Data is captured using a high-resolution camera, enabling visualization of the entire domain, while being able to resolve features as small as 10 µm. The study reports a correlation between fractal dimensions of displacement fronts and displacement front patterns in the medium. Results are mapped on a two-dimensional parameter space of log M and log Ca, and stability diagrams proposed in literature for drainage processes are superimposed for comparison. Compared to recent reports in the literature, the results in this work suggest that transition regimes may constitute a slightly larger portion of the overall flow regime diagram. This two-phase immiscible displacement study helps elucidate macroscopic processes at the continuum scale and provides insights relevant to enhanced oil recovery processes and the design of engineered porous media such as exchange columns and membranes.

42 ENGINEERING↗

DOE OSTI2020

A systematic feature extraction and selection framework for data-driven whole-building automated fault detection and diagnostics in commercial buildings

In data-driven automated fault detection and diagnostics (AFDD) modeling for building energy systems, feature engineering is a critical process of extracting information from high-dimensional and noisy sensor measurement and turning it into informative and representative inputs or features for data-driven modeling. However, few studies specifically discuss the feature engineering, especially the interactions between feature extraction and feature selection in whole-building AFDD. We developed a systematic feature extraction and selection framework for whole-building AFDD. In this framework, features are aggressively extracted from raw sensor data using statistical feature extraction techniques with various window sizes and statistics. With many features extracted, a hybrid feature selection algorithm that combines the filter and wrapper method then selects the best feature set. The framework considers diversity in the duration of fault behavior among fault types in whole-building AFDD, thus achieving high model generalization. We implemented our developed framework in a virtual testbed calibrated with measured data from Oak Ridge National Laboratory's Flexible Research Platform designed to mimic the operation of a typical small commercial building. The AFDD model is trained by the simulation data generated from the virtual testbed. The results show that (1) the developed framework improves the generalization of the AFDD model by 10.7% compared with literature-reported feature extraction and selection methods and (2) features with diverse window sizes and statistics are selected, providing insight into physical systems beyond the current understanding of buildings and faults and improving the detection and diagnostics of multiple fault types.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2024

Data Analytics for Catalysis Predictions: Are We Ready Yet?

Catalysis informatics has received tremendous attention in recent years as a tool to design catalysts and discover unique descriptors that capture the relationships between chemical properties and catalytic performance. One of the stop-gaps in understanding catalytic effects, which is often ignored and limits the deployment of data science tools, relates to the lack of uniform data. The catalytic cleavage of C–X (X= H, C, N, and O) bonds is relevant to many fundamental catalytic processes. In this Perspective, we performed data analytics on four groups of C–X cleavage reactions that are common in production, upcycling, or reactive separation: the C–C cleavage in cyclopropyl alcohol, the C–H cleavage in hydroacylation reactions, the C–O cleavage in β-O-4 linkages, and the C–N cleavage in amides, using experimental data collected from the literature to understand their underlying correlations. Experimental variables of high impact are identified for each reaction by dimensionality reduction methods. We highlight the urgent need for experimental data sets that include full details on the reaction conditions, such as reagent concentration, reaction temperature, or time in machine-readable forms. We discuss the potential improvement of the data of these reactions and promising approaches such as autonomous experiments to fill the gaps in unbiased experimental data. Finally, we also address the early stage consideration of separation aspects in the experimental design of efficient catalytic systems for these fundamental examples of chemical reactivity.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

ML-based Dimension Reduction Strategies

Deep learning (DL)--based surrogate models have achieved success in various applications in carbon capture and storage (CCS). However, the model training on high-dimensional spaces is computationally expensive and impractical for large-scale and complex geological models, because the models usually contain hundreds of thousands to millions of grid cells, each with a set of parameters. Furthermore, the high cost of generating training data with sufficient variation is another limitation of model training on high-dimensional spaces, which may result in overfitting and reduce the model efficiency and prediction performance. We proposed the workflow incorporating dimension reduction methods and deep learning models, which aim to extract the latent variables of input parameters and output state variables, and then build the mapping function at the latent spaces. The proposed workflow can significantly reduce the computational complexity in solving both forward and inverse problems compared to models trained on high-dimensional spaces. Dimensionality reduction models showed great potential in workflows for fast reservoir simulation, history matching, prior model generation, visualization, and more, ultimately enhancing DL model performance in related SMART Work Packages.

Hosseini, Seyyed↗

NASA NTRS1996

Simplified Micromechanics of Plain Weave Composites

A micromechanics based methodology to simulate the complete hygro-thermomechanical behavior of plain weave composites is developed. This methodology is based on micromechanics and the classical laminate theory. The methodology predicts a complete set of thermal, hygral and mechanical properties of plain woven composites, generates necessary data for use in a finite element structural analysis, and predicts stresses all the way from the laminate to the constituent level. This methodology is used in conjunction with a composite mechanics code to analyze and predict the properties/response of a generic graphite/epoxy woven textile composite and a plain weave ceramic composite. The fiber architecture, including the fiber waviness and fiber end distributions through the thickness, is properly accounted for. Predicted results compare reasonably well with those from detailed three-dimensional finite element analyses as well as available experimental data. However, the main advantage of the proposed methodology is its high computational efficiency as compared with three-dimensional finite element analyses.

Mital, Subodh K.↗

NASA NTRS1996

Endwall Heat Transfer Measurements in a Transonic Turbine Cascade

Turbine blade endwall heat transfer measurements are given for a range of Reynolds and Mach numbers. Data were obtained for Reynolds numbers based on inlet conditions of 0.5 and 1.0 x 106, for isentropic exit Mach numbers of 1.0 and 1.3, and for freestream turbulence intensities of 0.25% and 7.0%. Tests were conducted in a linear cascade at the NASA Lewis Transonic Turbine Blade Cascade Facility. The test article was a turbine rotor with 136' of turning and an axial chord of 12.7 cm. The large scale allowed for very detailed measurements of both flow field and surface phenomena. The intent of the work is to provide benchmark quality data for computational fluid dynamics (CFD) code and model verification. The flow field in the cascade is highly three-dimensional as a result of thick boundary layers at the test section inlet. Endwall heat transfer data were obtained using a steady-state liquid crystal technique.

Giel, P. W.↗

DOE OSTI2022

Autodifferentiable Ensemble Kalman Filters

Data assimilation is concerned with sequentially estimating a temporally evolving state. This task, which arises in a wide range of scientific and engineering applications, is particularly challenging when the state is high-dimensional and the state-space dynamics are unknown. This paper introduces a machine learning framework for learning dynamical systems in data assimilation. Here, our auto-differentiable ensemble Kalman filters (AD-EnKFs) blend ensemble Kalman filters for state recovery with machine learning tools for learning the dynamics. In doing so, AD-EnKFs leverage the ability of ensemble Kalman filters to scale to high-dimensional states and the power of automatic differentiation to train high-dimensional surrogate models for the dynamics. Numerical results using the Lorenz-96 model show that AD-EnKFs outperform existing methods that use expectation-maximization or particle filters to merge data assimilation and machine learning. In addition, AD-EnKFs are easy to implement and require minimal tuning.

autodifferentiation↗

NASA NTRS1997

Chaos: Understanding and Controlling Laser Instability

In order to characterize the behavior of tunable diode lasers (TDL), the first step in the project involved the redesign of the TDL system here at the University of Tennessee Molecular Systems Laboratory (UTMSL). Having made these changes it was next necessary to optimize the new optical system. This involved the fine adjustments to the optical components, particularly in the monochromator, to minimize the aberrations of coma and astigmatism and to assure that the energy from the beam is focused properly on the detector element. The next step involved the taking of preliminary data. We were then ready for the analysis of the preliminary data. This required the development of computer programs that use mathematical techniques to look for signatures of chaos. Commercial programs were also employed. We discovered some indication of high dimensional chaos, but were hampered by the low sample rate of 200 KSPS (kilosamples/sec) and even more by our sample size of 1024 (1K) data points. These limitations were expected and we added a high speed data acquisition board. We incorporated into the system a computer with a 40 MSPS (million samples/sec) data acquisition board. This board can also capture 64K of data points so that were then able to perform the more accurate tests for chaos. The results were dramatic and compelling, we had demonstrated that the lead salt diode laser had a chaotic frequency output. Having identified the chaotic character in our TDL data, we proceeded to stage two as outlined in our original proposal. This required the use of an Occasional Proportional Feedback (OPF) controller to facilitate the control and stabilization of the TDL system output. The controller was designed and fabricated at GSFC and debugged in our laboratories. After some trial and error efforts, we achieved chaos control of the frequency emissions of the laser. The two publications appended to this introduction detail the entire project and its results.

Blass, William E.↗

DOE OSTI2023

An Ensemble Score Filter for Tracking High-Dimensional Nonlinear Dynamical Systems

We propose an ensemble score filter (EnSF) for solving high-dimensional nonlinear filtering problems with superior accuracy. A major drawback of existing filtering methods, e.g., particle filters or ensemble Kalman filters, is the low accuracy in handling high-dimensional and highly nonlinear problems. EnSF attacks this challenge by exploiting the score-based diffusion model, defined in a pseudo-temporal domain, to characterizing the evolution of the filtering density. EnSF stores the information of the recursively updated filtering density function in the score function, in stead of storing the information in a set of finite Monte Carlo samples (used in particle filters and ensemble Kalman filters). Unlike existing diffusion models that train neural networks to approximate the score function, we develop a training-free score estimation that uses mini-batch-based Monte Carlo estimator to directly approximate the score function at any pseudo-spatial-temporal location, which provides sufficient accuracy in solving high-dimensional nonlinear problems as well as saves tremendous amount of time spent on training neural networks. Another essential aspect of EnSF is its analytical update step, gradually incorporating data information into the score function, which is crucial in mitigating the degeneracy issue faced when dealing with very high-dimensional nonlinear filtering problems. High-dimensional Lorenz systems are used to demonstrate the performance of our method. EnSF provides surprisingly impressive performance in reliably tracking extremely high-dimensional Lorenz systems (up to 1,000,000 dimension) with highly nonlinear observation processes, which is a well-known challenging problem for existing filtering methods.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

A comparison of two- and three-dimensional single-mode reshocked Richtmyer–Meshkov instability growth

The growth dynamics of two- and three-dimensional single-mode reshocked Richtmyer–Meshkov instability are compared systematically using data from high-resolution implicit large-eddy simulations of a model of the Mach 1.3 air(acetone) and sulfur hexafluoride (Jacobs and Krivets, 2005) shock tube experiment. The vorticity deposition by the incident shock and the dynamics of interface evolution are examined quantitatively and qualitatively. The perturbation amplitudes from the two- and three-dimensional simulations are compared to the experimental data and to the predictions of several nonlinear instability growth models. It is shown that the perturbation amplitudes from the two- and three-dimensional simulations with matching initial Richtmyer velocity are in excellent agreement with the experimental data. In addition, the dynamics of reshock (not considered in the experiment) are described in detail, and the post-reshock mixing layer amplitude growth rate is compared to the predictions of several reshock models. It is shown that using two-dimensional simulations to understand three-dimensional dynamics is valid only at early-to-intermediate times before reshock; at intermediate-to-late times after reshock the three-dimensional growth is generally larger than the corresponding two-dimensional growth. Furthermore, the reshock dynamics are also different between two and three dimensions. The quantitative results, together with visualizations of the flow field, were also used to contrast the difference between two- and three-dimensional vorticity and enstrophy dynamics.

42 ENGINEERING↗