Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Data Reasoning”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Multi-year assessment of the impact of ship-borne radiosonde observations on polar WRF forecasts in the Arctic

Abstract To compensate for the lack of conventional observations over the Arctic Ocean, ship-borne radiosonde observations have been regularly carried out during summer Arctic expeditions and the observed data have been broadcast via the global telecommunication system since 2017. With these data obtained over the data-sparse Arctic Ocean, observing system experiments were carried out using a polar-optimized version of the Weather Research and Forecasting (WRF) model and the WRF Data Assimilation (WRFDA) system to investigate their effects on analyses and forecasts over the Arctic. The results of verification against reanalysis data reveal: (1) DA effects on analyses and forecasts; (2) the reason for the year-to-year variability of DA effects; and (3) the possible role of upper-level potential vorticity in delayed DA effects. The overall assimilation effects of the extra data on the analyses and forecasts over the Arctic are positive. Initially, the DA effects are the most apparent in the temperature variables in the middle/lower troposphere, which spread to the wind variables in the upper troposphere. The effects decrease with time but reappear after approximately 120 h, even in the 240-h forecasts. The effects on forecasts vary depending on the proximity of the radiosonde observation locations to the high synoptic variability. The upper-level potential vorticity is known to play an important role in the development of Arctic cyclones, and it is suggested as a possible explanation for the delayed DA effects after about 120 h.

Geology↗
DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗
DOE OSTI2023

Dose reconstruction for plutonium-239 intakes at the Rocky Flats Plant

Purpose: The Rocky Flats (RF) Plant was a weapons manufacturing facility that operated from the early 1950s to 1989. Its primary missions were the production of plutonium (Pu) pits for thermonuclear weapons and the processing of retired weapons for Pu recovery. The purpose of this study was to estimate radiation doses to a cohort of 4499 RF workers from an intake of 239 Pu, the primary plutonium isotope handled at the site. Materials and methods: The latest biokinetic models of the International Commission on Radiological Protection, or site-specific variations of those models, were used to estimate 239 Pu intakes for each worker based on model fits to bioassay data often coupled with lung measurements. Results: Urinary excretion and lung retention data for most 239 Pu intakes could be fit reasonably well by a mixture of Pu dioxide and moderately soluble material. For some workers, better fits were obtained by application of other absorption types including Type S, 239 Pu nitrate, or pure 239 Pu dioxide, or by assuming intake via a wound. The lungs typically received the highest tissue doses, with fifty-year committed equivalent doses in the range of 0.5–1 Sv for 275 workers, 1–5 Sv for 115 workers, and greater than 5 Sv for 12 workers. Conclusions: RF was a unique site regarding a large number of lung measurements available for determining the appropriate absorption types for inhaled material. Further, this provided higher confidence in reconstructed 239 Pu doses than is generally gained from urinary data alone.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2020

Molecular Structure of Sphingomyelin in Fluid Phase Bilayers Determined by the Joint Analysis of Small-Angle Neutron and X-ray Scattering Data

In this work, we have determined the fluid bilayer structure of palmitoyl sphingomyelin (PSM) and stearoyl sphingomyelin (SSM) by simultaneously analyzing small-angle neutron and X-ray scattering data. Using a newly developed scattering density profile (SDP) model for sphingomyelin lipids, we report structural parameters including the area per lipid, total bilayer thickness, and hydrocarbon thickness, in addition to lipid volumes determined by densitometry. Unconstrained all-atom simulations of PSM bilayers at 55 °C using the C36 CHARMM force field produced a lipid area of 56 Å 2 , a value that is 10% lower than the one determined experimentally by SDP analysis (61.9 Å2). Furthermore, scattering form factors calculated from the unconstrained simulations were in poor agreement with experimental form factors, even though segmental order parameter (S CD ) profiles calculated from the simulations were in relatively good agreement with S CD profiles obtained from NMR experiments. Conversely, constrained area simulations at 61.9 Å 2 resulted in good agreement between the simulation and experimental scattering form factors, but not with S CD profiles from NMR. We discuss possible reasons for the discrepancies between these two types of data that are frequently used as validation metrics for molecular dynamics force fields.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Evaluation of Oak Ridge National Laboratory Health Physics Research Reactor Operation Data for Critical Benchmark Creation

The Oak Ridge National Laboratory (ORNL) Health Physics Research Reactor (HPRR) was a research reactor designed and built at ORNL in 1961. The critical assembly used a highly enriched uranium and molybdenum alloy as the fuel and could be operated in steady-state or burst modes. The HPRR has recently been the object of an investigation to create a criticality benchmark. Such benchmarks are very important, as they are used primarily to show the accuracy of newly developed modeling codes and to help experimental validation and reactor licensing. The evaluated experiments considered in this paper were carried out between 1974 and 1986 from various HPRR activities such as steady-state subcritical, steady-state critical, and burst prompt super-critical operations of the reactor for dosimetry, irradiation, or training purposes. By using the HPRR experimental logbook information and the as-built drawings of the critical assembly, a highly detailed model of the HPRR was created with SCALE 6.2.4/KENO-VI, and a first version of a critical benchmark of the HPRR was developed following the International Criticality Safety Benchmark Evaluation Project (ICSBEP) guidelines for thorough description and uncertainty/sensitivity quantification. Unfortunately, in most of the evaluated experiments, the obtained difference between calculated and experimental k eff is around 1,000 pcm, corresponding to a relative error of approximately 1%, beyond the quality standards of the ICSBEP recommending a relative error below 0.1%. Moreover, the derived experimental uncertainty is high, around 4% relative, mainly due to the U-Mo fuel density uncertainty, but also from numerous other factors. For these reasons, the creation of a valuable critical benchmark from HPRR operation data is thus far compromised. In this paper, the different steps of the experiments’ evaluation are summarized, and the reasons for the experimental/calculation discrepancies and potential ways to solve them are explored. This paper also aims to remind us always to exercise considerable care when performing experimental work, and to record all the data possible for potential future uses.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2026

ϕ → 3 π and ϕ π 0 transition form factor from Khuri-Treiman equations

This work studies the ϕ → 3 π decay and the ϕ → π 0 γ * transition form factor, utilizing the Khuri-Treiman formalism to account for analyticity, crossing, and unitarity. Using once-subtracted dispersion relations, we perform a simultaneous fit to the ϕ → 3 π Dalitz plot distribution and the ϕ → π 0 γ * measurements from the KLOE collaboration, finding good agreement with these experimental data. These results reaffirm the applicability of the Khuri-Treiman approach in the analysis of three-body decays. An interesting result is that the subtraction constant appearing in the equations is similar to a sum rule expectation, in contrast to analogous studies of ω → 3 π decays and ω → π 0 γ * , which shows significant deviations. Our results also provide a reasonable description of the trend of the transition form factor data from in the e + e − → ϕ π 0 scattering region. These intriguing theoretical differences between the decays of ϕ and ω could encourage further experimental measurements to assess the discrepancies and refine the theoretical predictions.

Lorenzo, A. G. [Instituto de Física Corpuscular; U↗
DOE OSTI2020

PBX9502 PAGOSA/SURF Calibration for 100 µm Grid Size

There are two methods to calibrate the SURF reactive burn explosives model.The first is to calibrate using Pop plot data. Because there are only two main parameters in SURF, called A and B, there is just enough flexibility in SURF to determine the Pop plot line, i.e., equivalent information to a slope and a single point on the line. However, due to the linearity of the Pop plot being based on logarithmic scales, a slight deviation off the Pop plot (though the fit to the Pop plot might be very good overall) can make considerable difference when comparing with the velocity profile data. The second calibration method thusly focuses strongly on matching entire velocity profile data. This is to say that individual velocity profiles extracted from embedded gages (in the case of PBX 9502, we refer to R. Gustavsen’s experimental data) may fit poorly, while the overall fit to the Pop plot may seem reasonable. In addition, there are other factors that could produce slightly different calibrations for SURF, while leading to an overall acceptable fit to data. One of those may be the physics code used and here we use Pagosa. Another is mesh size dependence. A third one stems from fitting a reactive burn model being done in conjunction with an equation of state (EOS) of the reactants and one for the products – different EOS can result in slightly different parameterizations. Finally, there are several secondary material parameters in SURF, and the overall fit will depend on the entire parameterization. The goal of this report is to clarify some of our previously made statements about SURF calibration, as well as to redo a systematic calibration using 100 µm grid resolution.

42 ENGINEERING↗
DOE OSTI2024

GraphAide: Advanced Graph-Assisted Query and Reasoning System

Curating knowledge from multiple siloed sources that contain both structured and unstructured data is a major challenge in many real-world applications. Pattern matching and querying represent fundamental tasks in modern data analytics that leverage this curated knowledge. The development of such applications necessitates overcoming several research challenges, including data extraction, named entity recognition, data modeling, and designing query interfaces. Moreover, the explainability of these functionalities is critical for their broader adoption. The emergence of Large Language Models (LLMs) has accelerated the development lifecycle of new capabilities. Nonetheless, there is an ongoing need for domain-specific tools tailored to user activities. The creation of digital assistants has gained considerable traction in recent years, with LLMs offering a promising avenue to develop such assistants utilizing domain-specific knowledge and assumptions. In this context, we introduce an advanced query and reasoning system, GraphAide, which constructs a knowledge graph (KG) from diverse sources and allows to query and reason over the resulting KG. GraphAide harnesses both the KG and LLMs to rapidly develop domain-specific digital assistants. It integrates design patterns from retrieval augmented generation (RAG) and the semantic web to create an agentic LLM application. GraphAide underscores the potential for streamlined and efficient development of specialized digital assistants, thereby enhancing their applicability across various domains.

Purohit, Sumit [BATTELLE (PACIFIC NW LAB)] (ORCID:↗
DOE OSTI2023

Uncertainty quantification of machine learning models to improve streamflow prediction under changing climate and environmental conditions

Machine learning (ML) models, and Long Short-Term Memory (LSTM) networks in particular, have demonstrated remarkable performance in streamflow prediction and are increasingly being used by the hydrological research community. However, most of these applications do not include uncertainty quantification (UQ). ML models are data driven and can suffer from large extrapolation errors when applied to changing climate/environmental conditions. UQ is required to quantify the influence of data noises on model predictions and avoid overconfident projections in extrapolation. In this work, we integrate a novel UQ method, called PI3NN, with LSTM networks for streamflow prediction. PI3NN calculates Prediction Intervals by training 3 Neural Networks. It can precisely quantify the predictive uncertainty caused by the data noise and identify out-of-distribution (OOD) data in a non-stationary condition to avoid overconfident predictions. We apply the PI3NN-LSTM method in the snow-dominant East River Watershed in the western US and in the rain-driven Walker Branch Watershed in the southeastern US. Results indicate that for the prediction data which have similar features as the training data, PI3NN precisely quantifies the predictive uncertainty with the desired confidence level; and for the OOD data where the LSTM network fails to make accurate predictions, PI3NN produces a reasonably large uncertainty indicating that the results are not trustworthy and should avoid overconfidence. PI3NN is computationally efficient, robust in performance, and generalizable to various network structures and data with no distributional assumptions. It can be broadly applied in ML-based hydrological simulations for credible prediction.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Improving Performance of M-to-N Processing and Data Redistribution in In Transit Analysis and Visualization

In an in transit setting, a parallel data producer, such as a numerical simulation, runs on one set of ranks M, while a data consumer, such as a parallel visualization application, runs on a different set of ranks N. One of the central challenges in this in transit setting is to determine the mapping of data from the set of M producer ranks to the set of N consumer ranks. This is a challenging problem for several reasons, such as the producer and consumer codes potentially having different scaling characteristics and different data models. The resulting mapping from M to N ranks can have a significant impact on aggregate application performance. In this work, we present an approach for performing this M-to-N mapping in a way that has broad applicability across a diversity of data producer and consumer applications. We evaluate its design and performance with a study that runs at high concurrency on a modern HPC platform. By leveraging design characteristics, which facilitate an “intelligent” mapping from M-to-N, we observe significant performance gains are possible in terms of several different metrics, including time-to-solution and amount of data moved.

Loring, Burlen↗
DOE OSTI2020

Improving Performance of M-to-N Processing and Data Redistribution in In Transit Analysis and Visualization

In an in transit setting, a parallel data producer, such as a numerical simulation, runs on one set of ranks M, while a data consumer, such as a parallel visualization application, runs on a different set of ranks N: One of the central challenges in this in transit setting is to determine the mapping of data from the set of M producer ranks to the set of N consumer ranks. This is a challenging problem for several reasons, such as the producer and consumer codes potentially having different scaling characteristics and different data models. The resulting mapping from M to N ranks can have a significant impact on aggregate application performance. In this work, we present an approach for performing this M-to-N mapping in a way that has broad applicability across a diversity of data producer and consumer applications. We evaluate its design and performance with a study that runs at high concurrency on a modern HPC platform. By leveraging design characteristics, which facilitate an ''intelligent'' mapping from M-to-N, we observe significant performance gains are possible in terms of several different metrics, including time-to-solution and amount of data moved.

Loring, Burlen↗
DOE OSTI2022

Parked aeroelastic field rotor response for a 20% scaled demonstrator of a 13‐MW downwind turbine

Abstract Aeroelastic parked testing of a unique downwind two‐bladed subscale rotor was completed to characterize the response of an extreme‐scale 13‐MW turbine in high‐wind parked conditions. A 20% geometric scaling was used resulting in scaled 20‐m‐long blades, whose structural and stiffness properties were designed using aeroelastic scaling to replicate the nondimensional structural aeroelastic deflections and dynamics that would occur for a lightweight, downwind 13‐MW rotor. The subscale rotor was mounted and field tested on the two‐bladed Controls Advanced Research Turbine (CART2) at the National Renewable Energy Laboratory's Flatiron Campus (NREL FC). The parked testing of these highly flexible blades included both pitch‐to‐run and pitch‐to‐feather configurations with the blades in the horizontal braked orientation. The collected experimental data includes the unsteady flapwise root bending moments and tip deflections as a function of inflow wind conditions. The bending moments are based on strain gauges located in the root section, whereas the tip deflections are captured by a video camera on the hub of the turbine pointed toward the tip of the blade. The experimental results are compared against computational predictions generated by FAST, a wind turbine simulation software, for the subscale and full‐scale models with consistent unsteady wind fields. FAST reasonably predicted the bending moments and deflections of the experimental data in terms of both the mean and standard deviations. These results demonstrate the efficacy of the first such aeroelastically scaled turbine test and demonstrate that a highly flexible lightweight downwind coned rotor can be designed to withstand extreme loads in parked conditions.

17 WIND ENERGY↗
DOE OSTI2020

Data-driven materials research enabled by natural language processing and information extraction

Given the emergence of data science and machine learning throughout all aspects of society, but particularly in the scientific domain, there is increased importance placed on obtaining data. Data in materials science are particularly heterogeneous, based on the significant range in materials classes that are explored and the variety of materials properties that are of interest. This leads to data that range many orders of magnitude, and these data may manifest as numerical text or image-based information, which requires quantitative interpretation. The ability to automatically consume and codify the scientific literature across domains - enabled by techniques adapted from the field of natural language processing - therefore has immense potential to unlock and generate the rich datasets necessary for data science and machine learning. This review focuses on the progress and practices of natural language processing and text mining of materials science literature and highlights opportunities for extracting additional information beyond text contained in figures and tables in articles. Here, we discuss and provide examples for several reasons for the pursuit of natural language processing for materials, including data compilation, hypothesis development, and understanding the trends within and across fields. Current and emerging natural language processing methods along with their applications to materials science are detailed. We, then, discuss natural language processing and data challenges within the materials science domain where future directions may prove valuable.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Orographic enhancement of rainfall over the Congo Basin

Abstract The Congo rainforest located in central equatorial Africa is an important, yet understudied part of the globe surrounded by complex orographic features. A primitive understanding of precipitation processes such as mesoscale convective dynamics magnifies uncertainties in the future climate projections of the hydrological cycle over the Congo. Furthermore, the effects of orography, which is an important forcing for convection and precipitation, are poorly resolved by climate models, and ill‐conceptualized over the Congo. To address this knowledge gap, perturbed orographic forcing experiments are conducted using the Weather Research and Forecasting (WRF) mesoscale numerical model in a high‐resolution convection‐permitting model setup. The model simulated selected dates in November 2014. The thunderstorms and rainfall simulated in the control run for the case study analyzed in this article compared reasonably well to satellite‐derived brightness temperature and rainfall data. The results from this case study show that the dynamical impact of increasing the height of the East African Highlands is the blocking of the lower‐tropospheric tropical easterlies. This weakening of the lower‐tropospheric zonal winds increases the windshear over the Congo Basin resulting in slower propagating, more intense mesoscale convective systems with enhanced rainfall.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Measurement of coherent exclusive J/ψ → μ+μ− production in ultraperipheral Pb+Pb collisions at sNN=5.36 TeV with the ATLAS detector

The ATLAS experiment has performed a measurement of coherent exclusive J/ψ → μ+μ− production in ultraperipheral Pb+Pb collisions at sNN=5.36$$ \sqrt{s_{\textrm{NN}}}=5.36 $$ TeV. The data was recorded at the Large Hadron Collider (LHC) during 2023, and corresponds to an integrated luminosity of 79 μb−1. Exclusive J/ψ candidates were selected with a dedicated track-sensitive trigger based on the ATLAS transition radiation tracker. The analysis involves reconstruction of the dimuon invariant mass based on muon tracks from the inner detector, as the muon transverse momentum range of interest precludes the use of the standard muon reconstruction and identification algorithms. Differential cross sections are measured as a function of J/ψ rapidity and are compared with theoretical predictions. After extrapolation to sNN=5.02$$ \sqrt{s_{\textrm{NN}}}=5.02 $$ TeV, they are also compared with previous measurements performed by other experiments using data from LHC Run 2. While the results agree reasonably well with theoretical predictions, they are in tension with previous Run-2 results for the central rapidity region.

Aad, G↗
DOE OSTI2020

Effects of various parameters of different porous transport layers in proton exchange membrane water electrolysis

Porous transport layers (PTLs) play an important role in proton exchange membrane water electrolysis (PEMWE) cells. The PTL facilitates water and gas transport, as well as thermal and electrical conduction, and is required to sustain good contact with adjacent components. It is expected that using PTLs with variations in material properties such as structure, composition, thickness and wettability results in performance changes of the PEMWE. Here, a general mathematical PEMWE model is developed that separates and analyzes the contributions of ohmic, activation, diffusion and Nernst potentials. For model validation, three inherently different anode PTL structures (carbon paper, sintered titanium particles, and titanium felt) are operated over a range of conditions. Additionally, the effects of PTL wettability were used to verify the model using Polytetrafluoroethylene (PTFE) treated Toray papers with PTFE loading ranging from 0% to 20%. The modeling results of both PTFE treated and untreated materials show good agreement with the experimental data. Mass transport or diffusion loss is the primary reason for performance differences between PTFE treated and untreated PTLs. Sintered titanium PTLs with thicknesses above 1 mm suffer from up to 33% increased ohmic losses without indicating any obvious changes in activation and diffusion losses when compared to untreated PTLs. The losses of the cell increase when using PTFE treated Toray paper. Individual contributions are quantified and assigned to increased ohmic, activation, and diffusion losses. In conclusion, the proposed model offers insights into the overpotential contributions of a PEMWE. It is a useful tool for predicting performance of various PTL materials and can be applied for PTL development and optimization efforts.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Benchmarking GEANT4 and PHITS for 14.8-MeV neutron transport in polyethylene and graphite materials

The feasibility of GEANT4 and PHITS with ENDF/B-VIII.0, JEFF-3.3 and JENDL-4.0 evaluated nuclear data libraries in 14.8-MeV neutron transport simulation are verified by comparing with the MCNP calculations and the experimental data. Here the measured data are from neutron integral experiments including graphite and polyethylene samples. For 6 cm-thick polyethylene and 20 cm-thick graphite samples, the results from GEANT4 and PHITS with three evaluated nuclear data libraries agree well with the MCNP calculations. For a 2 cm-thick graphite sample, the simulated results from GEANT4 and PHITS agree with the MCNP calculations, and they all underestimate the experiment data in below 3 MeV and overestimate them in the energy range of 5–7 MeV. The difference between the simulation results and the experimental data depends on the evaluated nuclear data libraries. In overall, GEANT4 and PHITS are performing reasonable jobs for 14.8-MeV neutron transport simulations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗