Search NASA⌕ Search

SEARCH · Search NASA

Results for “High dimensional modeling”

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 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Distance preserving machine learning for uncertainty aware accelerator capacitance predictions

Abstract Accurate uncertainty estimations are essential for producing reliable machine learning models, especially in safety-critical applications such as accelerator systems. Gaussian process models are generally regarded as the gold standard for this task; however, they can struggle with large, high-dimensional datasets. Combining deep neural networks with Gaussian process approximation techniques has shown promising results, but dimensionality reduction through standard deep neural network layers is not guaranteed to maintain the distance information necessary for Gaussian process models. We build on previous work by comparing the use of the singular value decomposition against a spectral-normalized dense layer as a feature extractor for a deep neural Gaussian process approximation model and apply it to a capacitance prediction problem for the High Voltage Converter Modulators in the Oak Ridge Spallation Neutron Source. Our model shows improved distance preservation and predicts in-distribution capacitance values with less than 1% error.

43 PARTICLE ACCELERATORS↗
NASA NTRS2003

Towards an Automated Full-Turbofan Engine Numerical Simulation

The objective of this study was to demonstrate the high-fidelity numerical simulation of a modern high-bypass turbofan engine. The simulation utilizes the Numerical Propulsion System Simulation (NPSS) thermodynamic cycle modeling system coupled to a high-fidelity full-engine model represented by a set of coupled three-dimensional computational fluid dynamic (CFD) component models. Boundary conditions from the balanced, steady-state cycle model are used to define component boundary conditions in the full-engine model. Operating characteristics of the three-dimensional component models are integrated into the cycle model via partial performance maps generated automatically from the CFD flow solutions using one-dimensional meanline turbomachinery programs. This paper reports on the progress made towards the full-engine simulation of the GE90-94B engine, highlighting the generation of the high-pressure compressor partial performance map. The ongoing work will provide a system to evaluate the steady and unsteady aerodynamic and mechanical interactions between engine components at design and off-design operating conditions.

Reed, John A.↗
DOE OSTI2024

Reduced-dimension Bayesian optimization for model calibration of transient vapor compression cycles

Development and calibration of first-principles dynamic models of vapor compression cycles (VCCs) is of critical importance for applications that include control design and fault detection and diagnostics. Nevertheless, the inherent complexity of models that are represented by large systems of differential–algebraic equations leads to significant challenges for model calibration processes that utilize classical gradient-based methods. Bayesian optimization (BO) is a sample-efficient and gradient-free approach using a probabilistic surrogate model and optimal search over a feasible parameter space. Despite the benefits of BO in reducing computational costs, challenges remain in dealing with a high-dimensional calibration task resulting from a large set of parameters that have significant impacts on system behavior and need to be calibrated simultaneously. This paper presents a reduced-dimension BO framework for calibrating transient VCCs models where the calibration space is projected to a low-dimensional subspace for accelerating convergence of the solution algorithm and consequently reducing the number of transient simulations. The proposed approach was demonstrated via two case studies associated with different VCC applications where 10 parameters were calibrated in each case using laboratory measurements. The reduced-dimension BO framework only required 1 / 8 th of the iterations associated with a standard BO method that deals with high-dimensional calibration parameters for converged solutions and yielded comparable accuracy. Furthermore, both calibrated models revealed significant accuracy improvements compared to uncalibrated models.

Ma, Jiacheng↗
DOE OSTI2024

Ab Initio-Based Bond Order Potential for Arsenene Polymorphs Developed via Hierarchical Reinforcement Learning

Arsenene, a less-explored two-dimensional material, holds the potential for applications in wearable electronics, memory devices, and quantum systems. This study introduces a bond-order potential model with Tersoff formalism, the ML-Tersoff, which leverages multireward hierarchical reinforcement learning (RL), trained on an ab initio data set. This data set covers a spectrum of properties for arsenene polymorphs, enhancing our understanding of its mechanical and thermal behaviors without the complexities of traditional models requiring multiple parameter sets. Our RL strategy utilizes decision trees coupled with a hierarchical reward strategy to accelerate convergence in high-dimensional continuous search spaces. Unlike the Stillinger-Weber approach, which demands separate formalisms for buckled and puckered forms, the ML-Tersoff model concurrently captures multiple properties of the two polymorphs by effectively representing the local environment, thereby avoiding the need for different atomic types. Here, we apply the ML model to understand the mechanical and thermal properties of the arsenene polymorphs and nanostructures. We observe an inverse relationship between the critical strain and temperature in arsenene. Thermal conductivity calculations in nanosheets show good agreement with ab initio data, reflecting a decrease in thermal conductivity attributable to increased anharmonic effects at higher temperatures. We also apply the model to predict the thermal behavior of arsenene nanotubes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1978

Some characteristics of electric field momentum coupling with the neutral atmosphere

A three-dimensional model has been developed to describe momentum coupling between high-latitude electric fields, neutral winds, temperature, and composition. The Hall drag is found to be the main source for atmospheric winds and the small divergence component of winds is due to the Pedersen drag and the Hall drag. Adiabatic heat transfer is responsible for the back pressure which damps the divergence field and for the reversal in circulation of the divergence field at higher altitudes. Back pressure causes a decrease in total wind velocity of about 10% at exospheric heights and by a factor of about 2 at 120 km. The wind field with the pressure feedback may be simulated by neglecting pressure variations and the Coriolis force. Density variations of Ar, N2, O, and He, induced by the momentum source, are in phase above 120 km and out of phase with the temperature amplitude above 150 km. The electrostatic field momentum source is ineffective for directly inducing density and temperature variations in the upper thermosphere.

Mayr, H. G.↗
NASA NTRS1986

Observations and mechanisms of GATE waterspouts

The present numerical and observational investigation of interacting cumulus processes implicated in the formation of waterspouts, the GATE database for days 261 and 186 is noted to imply that the existence of cumulus-scale parent vortices is a necessary (albeit not sufficient) condition for the production of waterspouts. A high resolution version of the Schlessinger (1975) three-dimensional cumulus model with a Kessler (1969) type precipitation scheme is used to analyze cumulus-scale vorticity organization, which on the two days in question exhibited contrasting thermal stratification and cloud features. The observations from both days suggest that the waterspouts formed ahead of the wind shift, due to the passage of a gust front.

Simpson, J.↗
NASA NTRS1988

Active region coronal loops - Structural and variability

X-ray images of a pair of active region loops are studied which show significant, short time-scale variability in the line fluxes of O VIII, Ne IX, and Mg XI and in the 3.5-11.5 keV soft X-ray bands. Vector magnetograms and high-resolution UV images were used to model the three-dimensional characteristics of the loops. X-ray light curves were generated spanning four consecutive orbits for both loops individually, and light curves of the loop tops and brightest points were also generated. The largest variations involve flux changes of up to several hundred percent on time scales of 10 minutes. No significant H-alpha flare activity is reported, and loop temperatures remain in the four to six million K range. The decay phases of the light curves indicate radiative cooling, inhibition of conduction, and some type of 'continued heating' due to ongoing, underlying activity at the microflare level.

Haisch, Bernhard M.↗
NASA NTRS2000

Effect of Crystal Orientation on Analysis of Single-Crystal, Nickel-Based Turbine Blade Superalloys

High-cycle fatigue-induced failures in turbine and turbopump blades is a pervasive problem. Single-crystal nickel turbine blades are used because of their superior creep, stress rupture, melt resistance, and thermomechanical fatigue capabilities. Single-crystal materials have highly orthotropic properties making the position of the crystal lattice relative to the part geometry a significant and complicating factor. A fatigue failure criterion based on the maximum shear stress amplitude on the 24 octahedral and 6 cube slip systems is presented for single-crystal nickel superalloys (FCC crystal). This criterion greatly reduces the scatter in uniaxial fatigue data for PWA 1493 at 1,200 F in air. Additionally, single-crystal turbine blades used in the Space Shuttle main engine high pressure fuel turbopump/alternate turbopump are modeled using a three-dimensional finite element (FE) model. This model accounts for material orthotrophy and crystal orientation. Fatigue life of the blade tip is computed using FE stress results and the failure criterion that was developed. Stress analysis results in the blade attachment region are also presented. Results demonstrate that control of crystallographic orientation has the potential to significantly increase a component's resistance to fatigue crack growth without adding additional weight or cost.

Swanson, G. R.↗
NASA NTRS2000

A Composite View of Ozone Evolution in the 1995-1996 Northern Winter Polar Vortex Developed from Airborne Lidar and Satellite Observations

The processes which contribute to the ozone evolution in the high latitude northern lower stratosphere are evaluated using a three dimensional model simulation and ozone observations. The model uses winds and temperatures from the Goddard Earth Observing System Data Assimilation System. The simulation results are compared with ozone observations from three platforms: the differential absorption lidar (DIAL) which was flown on the NASA DC-8 as part of the Vortex Ozone Transport Experiment; the Microwave Limb Sounder (MLS); the Polar Ozone and Aerosol Measurement (POAM II) solar occultation instrument. Time series for the different data sets are consistent with each other, and diverge from model time series during December and January. The model ozone in December and January is shown to be much less sensitive to the model photochemistry than to the model vertical transport, which depends on the model vertical motion as well as the model vertical gradient. We evaluate the dependence of model ozone evolution on the model ozone gradient by comparing simulations with different initial conditions for ozone. The modeled ozone throughout December and January most closely resembles observed ozone when the vertical profiles between 12 and 20 km within the polar vortex closely match December DIAL observations. We make a quantitative estimate of the uncertainty in the vertical advection using diabatic trajectory calculations. The net transport uncertainty is significant, and should be accounted for when comparing observations with model ozone. The observed and modeled ozone time series during December and January are consistent when these transport uncertainties are taken into account.

Douglass, A. R.↗
NASA NTRS2007

Method and apparatus for multiple-projection, dual-energy x-ray absorptiometry scanning

Methods and apparatuses for advanced, multiple-projection, dual-energy X-ray absorptiometry scanning systems include combinations of a conical collimator; a high-resolution two-dimensional detector; a portable, power-capped, variable-exposure-time power supply; an exposure-time control element; calibration monitoring; a three-dimensional anti-scatter-grid; and a gantry-gantry base assembly that permits up to seven projection angles for overlapping beams. Such systems are capable of high precision bone structure measurements that can support three dimensional bone modeling and derivations of bone strength, risk of injury, and efficacy of countermeasures among other properties.

Charles, Jr., Harry K.↗
NASA NTRS2013

Integrated Model Reduction and Control of Aircraft with Flexible Wings

This paper presents an integrated approach to the modeling and control of aircraft with exible wings. The coupled aircraft rigid body dynamics with a high-order elastic wing model can be represented in a nite dimensional state-space form. Given a set of desired output covariance, a model reduction process is performed by using the weighted Modal Cost Analysis (MCA). A dynamic output feedback controller, which is designed based on the reduced-order model, is developed by utilizing output covariance constraint (OCC) algorithm, and the resulting OCC design weighting matrix is used for the next iteration of the weighted cost analysis. This controller is then validated for full-order evaluation model to ensure that the aircraft's handling qualities are met and the uttering motion of the wings suppressed. An iterative algorithm is developed in CONDUIT environment to realize the integration of model reduction and controller design. The proposed integrated approach is applied to NASA Generic Transport Model (GTM) for demonstration.

AEROELASTIC↗
DOE OSTI2025

Initial steady-state core simulation capability or thermal and pool-type molten salt reactors, coupling reactor physics, thermal-hydraulics, and evolving chemistry

This report presents the development and validation of an initial steady-state multiphysics capability for molten salt reactors (MSRs) under the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program in Fiscal Year 2025. The framework integrates neutronics, thermal-hydraulics, species transport, and thermochemistry to capture the coupled dynamics of liquid-fueled systems. Implementation and testing were performed on two representative designs: the Molten Salt Reactor Experiment (MSRE), a thermal-spectrum, channeled-flow reactor, and the Lotus Molten Salt Reactor (L-MSR), a fast-spectrum, pool-type reactor. The modeling suite employs Griffin for reactor physics and depletion, Pronghorn and SAM for thermal-hydraulics, Thermochimica for chemistry, and Saline for thermophysical properties, with benchmarking and validation carried out against historical MSRE data, experimental flow-loop measurements, and reference depletion calculations from Monte Carlo codes. The framework demonstrated the ability to reproduce key reactor behaviors including temperature feedback, reactivity losses, delayed neutron precursor transport, xenon poisoning, and redox potential evolution. The results confirm the feasibility and accuracy of the coupled models in predicting steady-state and selected transient MSR behaviors. This latter ones are used in this report as a proxy indicating that the steady-state models from which the transient starts are accurate. For MSRE, validation showed good agreement with pump start-up and natural circulation tests, while for the L-MSR, benchmarking confirmed hydraulic calibration and consistency of neutronics–thermal coupling. The tools also provided new insights into species transport, noble metal deposition, and salt solidification dynamics. On the Xenon transport front, the code is validated against the steady state Xenon poisoining measurement and showed good agreement with the experimental value. Identified areas for future work include advanced void transport modeling, three-dimensional simulations, improved alloy corrosion models, and tighter integration with high-fidelity Monte Carlo codes. These developments provide a foundation for high-fidelity MSR simulations that can support reactor design optimization, safety assessments, and long-term operational strategies.

42 - ENGINEERING↗
NASA NTRS1978

Pressure and velocity measurements in a three dimensional wall jet

A flat plate and a curved wall surface, intended to model a wing-flap combination in a high V/STOL configuration, were used to measure pressure and flow velocity in a freely expanding coflowing jet and a three dimensional wall jet. The effects of increasing the velocity ratio of the jet exit plane velocity to the free stream velocity were investigated. Velocity measurements were made using a two color laser Doppler velocimeter with a phase-locked loop processor. Fluctuating pressures were monitored with condenser-type microphones. Quantities measured include the width of the mixing region, the mean velocity field, turbulent intensities, and time scales. Wall and static pressure-velocity correlations and coherences are presented.

Catalano, G. D.↗
DOE OSTI2024

Physics-coupled data-driven design of high-temperature alloys

We present a materials design loop, which streamlines physics-coupled machine learning (ML) surrogate models to discover new alloy chemistries with improved properties. The efficacy is demonstrated by discovering a high-temperature alumina-forming austenitic (AFA) stainless steel with enhanced creep, followed by experimental validation. The ML models have been trained using a well-curated, highly consistent experimental dataset augmented with synthetic microstructural features from a computational thermodynamic approach. We have populated a large number of hypothetical AFA alloys to explore the high-dimensional composition space and have predicted their creep properties by providing the same synthetic input features obtained from the trained ML models. Uncertainties from the ML training were taken as thresholds for truncating predicted results to identify alloys with improved or deteriorated creep. Individual elemental compositions have been determined via probability density distribution analysis from the group of alloys at the top and bottom of the predicted creep values for further virtual and experimental validations. In conclusion, we anticipate that this workflow can be applied to screen desired conditions, such as chemistry and processing parameters, in high-dimensional space through physics-guided data analytics.

Alloy design↗
DOE OSTI2024

ASGarD: Adaptive Sparse Grid Discretization

Many areas of science exhibit physical processes that are described by high dimensional partial differential equations (PDEs), e.g., the 4D, 5D and 6D models describing magnetized fusion plasmas, models describing quantum chemistry, or derivatives pricing. Such problems are affected by the so-called “curse of dimensionality” where the number of degrees of freedom (or unknowns) required to be solved for scales as N D where N is the number of grid points in any given dimension D. A simple, albeit naive, 6D example is demonstrated in the left panel of Figure 1. With N = 1000 grid points in each dimension, the memory required just to store the solution vector, not to mention forming the matrix required to advance such a system in time, would exceed an exabyte - and also the available memory on the largest of supercomputers available today. The right panel of Figure 1 demonstrates potential savings for a range of problem dimensionalities and grid resolution. While there are methods to simulate such high-dimensional systems, they are mostly based on Monte-Carlo methods, which rely on a statistical sampling such that the resulting solutions include noise. Since the noise in such methods can only be reduced at a rate proportional to $\sqrt{N_p}$ where N p is the number of Monte-Carlo samples, there is a need for continuum, or grid/mesh-based methods for high-dimensional problems, which both do not suffer from noise and bypass the curse of dimensionality. We present a simulation framework that provides such a method using adaptive sparse grids.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2015

Unsteady Thick Airfoil Aerodynamics: Experiments, Computation, and Theory

An experimental, computational and theoretical investigation was carried out to study the aerodynamic loads acting on a relatively thick NACA 0018 airfoil when subjected to pitching and surging, individually and synchronously. Both pre-stall and post-stall angles of attack were considered. Experiments were carried out in a dedicated unsteady wind tunnel, with large surge amplitudes, and airfoil loads were estimated by means of unsteady surface mounted pressure measurements. Theoretical predictions were based on Theodorsen's and Isaacs' results as well as on the relatively recent generalizations of van der Wall. Both two- and three-dimensional computations were performed on structured grids employing unsteady Reynolds-averaged Navier-Stokes (URANS). For pure surging at pre-stall angles of attack, the correspondence between experiments and theory was satisfactory; this served as a validation of Isaacs theory. Discrepancies were traced to dynamic trailing-edge separation, even at low angles of attack. Excellent correspondence was found between experiments and theory for airfoil pitching as well as combined pitching and surging; the latter appears to be the first clear validation of van der Wall's theoretical results. Although qualitatively similar to experiment at low angles of attack, two-dimensional URANS computations yielded notable errors in the unsteady load effects of pitching, surging and their synchronous combination. The main reason is believed to be that the URANS equations do not resolve wake vorticity (explicitly modeled in the theory) or the resulting rolled-up un- steady flow structures because high values of eddy viscosity tend to \smear" the wake. At post-stall angles, three-dimensional computations illustrated the importance of modeling the tunnel side walls.

Strangfeld, C.↗
DOE OSTI2025

Data Quality Monitoring for the Hadron Calorimeters Using Transfer Learning for Anomaly Detection

The proliferation of sensors brings an immense volume of spatio-temporal (ST) data in many domains, including monitoring, diagnostics, and prognostics applications. Data curation is a time-consuming process for a large volume of data, making it challenging and expensive to deploy data analytics platforms in new environments. Transfer learning (TL) mechanisms promise to mitigate data sparsity and model complexity by utilizing pre-trained models for a new task. Despite the triumph of TL in fields like computer vision and natural language processing, efforts on complex ST models for anomaly detection (AD) applications are limited. In this study, we present the potential of TL within the context of high-dimensional ST AD with a hybrid autoencoder architecture, incorporating convolutional, graph, and recurrent neural networks. Motivated by the need for improved model accuracy and robustness, particularly in scenarios with limited training data on systems with thousands of sensors, this research investigates the transferability of models trained on different sections of the Hadron Calorimeter of the Compact Muon Solenoid experiment at CERN. The key contributions of the study include exploring TL’s potential and limitations within the context of encoder and decoder networks, revealing insights into model initialization and training configurations that enhance performance while substantially reducing trainable parameters and mitigating data contamination effects.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Investigation of the Performance and Explainability Tradeoffs for Machine-Learning Models for Predictive Maintenance of Circulating Water Systems in Nuclear Power Plants

Predictive maintenance (PdM) has shown great potential for achieving substantial cost savings and enhancing the economic competitiveness of nuclear power plants (NPPs) in today's energy market. Among the different modeling approaches that exist, machine learning (ML) tools in particular have a demonstrated ability to handle high dimensional and multivariate data and to extract hidden relationships within data in industrial environments. While ML methods show great potential, their lack of explainability---especially for black-box models---is a major hurdle to their adoption. Moreover, considering the supposed trade-off between explainability and performance challenges, careful consideration must be made as to which of these quality aspects takes precedence in light of multiple modeling options, resource availability, and domain characteristics. The present work evaluates the performance of six ML models, each with a different degree of explainability, in classifying the conditions of circulating water pumps (CWPs) by utilizing sensor data from nuclear power plants. To determine the drivers behind the trade-offs presented by this array of models, this work also tests different combinations of CWP units as the training and testing data, degrees of data imbalance, and objective functions for hyperparameter tuning. It was found that black-box models tend to afford superior performance in cases where there are far more instances of one type of labeled data than of any other type. It is recommended that a guided procedure be followed for designing and delivering an ML system that is sufficiently explainable to all involved stakeholders.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗