SEARCH · Engineering Papers

Results for “Data Transfer”

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 469 records · Page 26

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Cross-facility science with the Superfacility Project at LBNL

As data sets from DOE user science facilities grow in both size and complexity there is an urgent need for new capabilities to transfer, analyze and manage the data underlying scientific discoveries. LBNL's Superfacility project brings together experimental and observational research instruments with computational and network facilities at the National Energy Research Scientific Computing Center (NERSC) and the Energy Sciences Network (ESnet) with the goal of enabling user science. Here, we report on recent innovations in the Superfacility project, including advanced data management, API-based automation, real-Time interactive user interfaces, and supported infrastructure for 'edge' services.

Enders, B↗

DOE OSTI2022

Sharing is caring: An extensive analysis of parameter-based transfer learning for the prediction of building thermal dynamics

In recent years deep neural networks have been proposed as a lightweight data-driven model to capture high-dimensional, nonlinear physical processes to predict building thermal responses. However, the need of a large amount of data for the training process of deep neural networks clashes with the potential limited data availability in most existing or new buildings. Transfer learning aims to enhance the performance of a target learner exploiting knowledge from related and similar environments. This study conducted a suite of experiments that leveraged 250 data-driven models based on a synthetic dataset of a building archetype to study the influence of data availability, energy efficiency level, occupancy and climate for the transfer process of thermal dynamics. The performance of the transfer learning process was compared against a classical machine learning approach. Here, the results suggest that building thermal dynamics can be effectively transferred under the same climatic conditions, increasing performance when dealing with different occupancy schedules, efficiency levels and low data availability. Furthermore, the paper compares the performance of both transfer learning and machine learning approaches in an online fashion, to support the implementation in real-world deployment.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2026

Predicting roughness effects in additively manufactured coolant channels with helical enhancements

Additive manufacturing (AM) is a promising technique for fabrication of complex geometries such as those expected to be utilized in the blanket, first wall, and divertor. In the case of cooling, metallic AM may be exploited to embed geometric enhancements (ribs, rifling, etc.) to improve cooling performance. However, due to the roughness of these unfinished internal AM surfaces, prediction of thermal hydraulic performance in such channels is difficult. In this work, we consider a methodology for predicting pressure drop and heat transfer in AM channels containing helical enhancements (e.g. helical ribs, twisted tapes) that allows the incorporation of roughness data through conventional pipe flow correlations. This methodology is tested using experimental friction factor and heat transfer coefficient data from high-pressure helium coolant flow measurements in AM stainless steel tubes fabricated by laser powder bed fusion. Both a featureless AM tube and one containing helical ribs were considered alongside a conventionally manufactured smooth tube. The AM surface roughness is obtained by profilometry and used to predict an equivalent sand-grain roughness, with this equivalent roughness confirmed through AM featureless tube measurements. Under the proposed methodology, this roughness information is incorporated into predictions of friction factor and Nusselt number for the rifled tube. Furthermore, these predictions agree well with experimental data across a large range of Reynolds numbers, encouraging the use of this methodology for thermal hydraulic analysis of similar systems and design of future coolant channel geometries.

Additive manufacturing↗

DOE OSTI2020

Energy efficient data center liquid cooling with geothermal enhancement

A data center cooling system is operated in a first mode, has an indoor portion wherein heat is absorbed from components in the data center by a heat transfer fluid, and has an outdoor heat exchanger portion and a geothermal heat exchanger portion. The first mode includes ambient air cooling of the heat transfer fluid in the outdoor heat exchanger portion and/or geothermal cooling of the heat transfer fluid in the geothermal heat exchanger portion. Based on an appropriate metric, a determination is made that a switch should be made from the first mode to a second, different, mode; and, responsive thereto, the data center cooling system is switched to the second mode. The second mode includes at least another of ambient air cooling of the heat transfer fluid in the outdoor heat exchanger portion and geothermal cooling of the heat transfer fluid in the geothermal heat exchanger portion.

15 GEOTHERMAL ENERGY↗

DOE OSTI2022

Energy efficient data center liquid cooling with geothermal enhancement

Chainer, Timothy J.↗

DOE OSTI2020

Hybrid deep learning architecture for general disruption prediction across tokamaks

In this paper, we present a new deep learning disruption prediction algorithm based on important findings from explorative data analysis which effectively allows knowledge transfer from existing devices to new ones, thereby predicting disruptions using very limited disruptive data from the new devices. Here, the explorative data analysis conducted via unsupervised clustering techniques confirms that time-sequence data are much better separators of disruptive and non-disruptive behavior than the instantaneous plasma state data with further advantageous implications for a sequence-based predictor. Based on such important findings, we have designed a new algorithm for multi-machine disruption prediction that achieves high predictive accuracy on the C-Mod (AUC=0.801), DIII-D (AUC=0.947) and EAST (AUC=0.973) tokamaks with limited hyperparameter tuning. Through numerical experiments, we show that boosted accuracy (AUC=0.959) is achieved on EAST predictions by including in the training only 20 disruptive discharges, thousands of non-disruptive discharges from EAST, and combining this with more than a thousand discharges from DIII-D and C-Mod. The improvement of predictive ability obtained by combining disruptive data from other devices is found to be true for all permutations of the three devices. Furthermore, by comparing the predictive performance of each individual numerical experiment, we find that non-disruptive data are machine-specific while disruptive data from multiple devices contain device-independent knowledge that can be used to inform predictions for disruptions occurring on a new device.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2021

Hybrid deep learning architecture for general disruption prediction across tokamaks

In this paper, we present a new deep learning disruption prediction algorithm based on important findings from explorative data analysis which effectively allows knowledge transfer from existing devices to new ones, thereby predicting disruptions using very limited disruptive data from the new devices. The explorative data analysis conducted via unsupervised clustering techniques confirms that time-sequence data are much better separators of disruptive and non-disruptive behavior than the instantaneous plasma state data with further advantageous implications for a sequence-based predictor. Based on such important findings, we have designed a new algorithm for multi-machine disruption prediction that achieves high predictive accuracy on the C-Mod (AUC=0.801), DIII-D (AUC=0.947) and EAST (AUC=0.973). tokamaks with limited hyperparameter tuning. Through numerical experiments, we show that boosted accuracy (AUC=0.959) is achieved on EAST predictions by including in the training only 20 disruptive discharges, thousands of non-disruptive discharges from EAST, and combining this with more than a thousand discharges from DIII-D and C-Mod. The improvement of predictive ability obtained by combining disruptive data from other devices is found to be true for all permutations of the three devices. Furthermore, by comparing the predictive performance of each individual numerical experiment, we find that non-disruptive data are machine-specific while disruptive data from multiple devices contain device-independent knowledge that can be used to inform predictions for disruptions occurring on a new device.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2026

Scientific Data Compression for Large Scale Computational Fluid Dynamics (CFD) Simulations

This Cooperative Research and Development Agreement (CRADA) between Oak Ridge National Laboratory (ORNL) and General Electric (GE) investigated methods for reducing the size of large computational fluid dynamics (CFD) simulation datasets using scientific data compression techniques. The work focused on adapting the MultiGrid Adaptive Reduction of Data (MGARD) compression framework and integrating it with high-performance I/O and visualization tools used in CFD workflows. MGARD uses hierarchical multilevel decomposition to enable error-controlled compression of floating-point scientific data while preserving quantities of interest. During the project, MGARD compression was integrated with the ADIOS I/O framework and visualization tools such as ParaView to enable efficient storage, transfer, and analysis of simulation data. The collaboration also explored approaches for improving compression performance for CFD data defined on unstructured meshes. Results demonstrate that scientific data compression can significantly reduce storage requirements and improve data management for large-scale CFD simulations.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Development and Experimental Validation of a Heat Transfer Model for Spilled Molten Salt Pools

A spill of radionuclide-bearing molten salt is one of the major postulated events that needs to be analyzed for liquid fluorine salt-cooled high-temperature reactor (FHR) or molten salt reactor licensing purposes. In this postulated event, radioactive source term materials (RSTMs) in the molten salt are discharged from the reactor vessel to the reactor building. The release of RSTMs from the spilled salt pool to the gas space in the reactor building is expected to be controlled by the cooling behavior of the spilled salt, including the growth and shrinkage of the solid crust on the surface of the spilled salt pool. This paper presents a simulation model for spilled salt pool heat transfer and validation efforts. The validation data come from two molten salt spill tests that were performed recently: the PELE2 test by the Rapid Experimental Laboratory of Kairos Power LLC (KP) and the Argonne salt cooling test conducted by Argonne National Laboratory. The former was a large-scale test involving kilograms of molten spilled FLiNaK salt, and the latter was a relatively smaller-scale test targeting various processes associated with a salt spill event. Both tests generated valuable data sets that can be used to assess salt cooling and validate evaluation models. This paper provides a new one-dimensional model that can simulate the cooling process of a spilled salt pool as well as the thermal responses of heat structures, such as the stainless steel liner and the concrete below the salt. The model has been implemented as part of KP-SAM code, which is a branch of the systems code SAM specific to KP FHR. In conclusion, the simulation results of the model are compared with the data of the PELE2 and Argonne tests, and reasonable agreements are observed between the model and test data.

heat transfer model↗

DOE OSTI2024

Accelerating data acquisition with FPGA-based edge machine learning: a case study with LCLS-II

New scientific experiments and instruments generate vast amounts of data that need to be transferred for storage or further processing, often overwhelming traditional systems. Edge machine learning (EdgeML) addresses this challenge by integrating machine learning (ML) algorithms with edge computing, enabling real-time data processing directly at the point of data generation. EdgeML is particularly beneficial for environments where immediate decisions are required, or where bandwidth and storage are limited. In this paper, we demonstrate a high-speed configurable ML model in a fully customizable EdgeML system using a field programmable gate array (FPGA). Our demonstration focuses on an angular array of electron spectrometers, referred to as the ‘CookieBox,’ developed for the Linac Coherent Light Source II project. The EdgeML system captures 51.2 Gbps from a 6.4 GS s −1 analog to digital converter and is designed to integrate data pre-processing and ML inside an FPGA. Our implementation achieves an inference latency of 0.2 µs for the ML model, and a total latency of 0.4 µs for the complete EdgeML system, which includes pre-processing, data transmission, digitization, and ML inference. The modular design of the system allows it to be adapted for other instrumentation applications requiring low-latency data processing.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Optimizing Error-Bounded Lossy Compression for Scientific Data With Diverse Constraints

Vast volumes of data are produced by today's scientific simulations and advanced instruments. These data cannot be stored and transferred efficiently because of limited I/O bandwidth, network speed, and storage capacity. Error-bounded lossy compression can be an effective method for addressing these issues: not only can it significantly reduce data size, but it can also control the data distortion based on user-defined error bounds. In practice, many scientific applications have specific requirements or constraints for lossy compression, in order to guarantee that the reconstructed data are valid for post hoc analysis. For example, some datasets contain irrelevant data that should be isolated in particular and users often have intuition regarding value ranges, geospatial regions, and other data subsets that are crucial for subsequent analysis. Existing state-of-the-art error-bounded lossy compressors, however, do not consider these constraints during compression, resulting in inferior compression ratios with respect to user's post hoc analysis, due to the fact that the data itself provides little or no value for post hoc analysis. In this work we address this issue by proposing an optimized framework that can preserve diverse constraints during the error-bounded lossy compression, e.g., cleaning the irrelevant data, efficiently preserving different precision for multiple value intervals, and allowing users to set diverse precision over both regular and irregular regions. We perform our evaluation on a supercomputer with up to 2,100 cores. Experiments with six real-world applications show that our proposed diverse constraints based error-bounded lossy compressor can obtain a higher visual quality or data fidelity on reconstructed data with the same or even higher compression ratios compared with the traditional state-of-the-art compressor SZ. Furthermore, our experiments also demonstrate very good scalability in compression performance compared with the I/O throughput of the parallel file system.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

A Comparison of Three Neodymium Atomic Data Sets for Kilonova Modeling

We examine the impact of input neodymium (Nd) atomic data on the light curves and spectra of kilonovae (KNe), probing the sensitivity of kilonova observables to the atomic physics of this important lanthanide element. We use the SuperNu Monte Carlo radiative transfer code, simulating a simple semianalytic 1D kilonova (KN) with a pure Nd atmosphere, fixing the radiative transfer method while using input atomic data generated by three different codes: the LANL suite of atomic physics codes, HULLAC, and Autostructure. We see that the choice of atomic data significantly shapes the resulting light curves and spectra. Peak bolometric luminosities differ by a ratio of nearly 1.5 between HULLAC/Autostructure and LANL data sets. Moreover, we observe significant near- to mid-IR differences in the structure of the spectra. We specifically attribute these differences to the choice of atomic data for neutral Nd I. Many of the results here have been adapted from a presentation at “Radiative Transfer and Atomic Physics of Kilonovae” in Stockholm, 2023. We additionally present a LANL data set with energies calibrated to available values in the NIST Atomic Spectra Database, and demonstrate that this calibration also significantly affects IR spectral structure at late time. The substantial differences in KN observables that arise from tuning the atomic data of just one lanthanide element highlight the special attention that must be paid to atomic physics uncertainties when modeling KNe, from AT2017gfo to beyond.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2022

Influence of design and operating parameters for additively manufactured intensified packing devices on CO 2 -Absorption column cooling and capture efficiency

Solvent-based CO 2 absorption is technologically a matured CO 2 capture pathway but suffers from: high regeneration energy demand, and solvent temperature rise and decreased capture efficiency caused by the heat of reaction. While research has focused on developing non-aqueous and low-aqueous solvents for decreasing the regeneration energy, the temperature bulge due to exothermic absorption is typically dealt with by cooling the solvent with an external inter-stage heat-exchanger. This approach may increase the overall process footprint, as well as capital and operating costs. Here, the current study explores a process intensification approach by incorporating inside the column an additively manufactured intensified packing device that consists of corrugated plates and internal coolant channels. The corrugated plates provide surface area for mass transfer between gas and liquid, while cooling fluid inside the internal channels removes heat from the exothermic reactive system. Two different intensified devices with specific surface areas of 266 m 2 /m 3 and 359 m 2 /m 3 were designed, manufactured, and tested inside a 2.06-m long and 0.203-m diameter column packed with commercial Mellapak 250Y packing. The device with a lower surface area showed up to 27 % reduction in cooling performance. A steady-state heat transfer model provides good agreement with the experimental column temperature and intra-stage heat removal data. Although a decrease in heat transfer performance was observed for the device with lower surface area, CO 2 capture experiments performed with simulated flue gas and low-aqueous solvent demonstrated that both intensified devices lead to a similar improvement of 12 % in capture efficiency. This study provides an understanding on simplifying the intensified packing device geometry and decreasing the device fabrication cost by 25 % without compromising with the CO 2 absorption performance of the packed column.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2020

Inverted Region in Bimolecular Electron Transfer in Solution Enabled by Delocalization

Rate constants for bimolecular electron transfer (ET) increased with driving force, -ΔG°, reached a plateau, and then decreased in an inverted region. This rate data was described well by electron transfer theory subject to a diffusion-controlled limit. These were for ET from radical anions of polydecylthiophene (P3DT) to a series of acceptors in THF solution. When the donor was the smaller anion of quaterthiophene (T4 •- ) the inverted region was much less prominent, and still less so for when the donor was the anion of bithiophene (T2 •- ). Description of the data using ET theory identifies smaller electronic couplings for the highly-delocalized P3DT anions as enabling the inverted behavior: The presence of a Marcus inverted region is a consequence of delocalized electronic states. The results of this study further imply that electronic couplings smaller than usually found for molecules in contact could boost efficiency of energy storage by electron transfer, and identifies size-mismatch as an important concept in control of electronic couplings.

14 SOLAR ENERGY↗

DOE OSTI2020

Quantal diffusion approach for multinucleon transfer processes in the 58,64 Ni + 208 Pb reactions: Toward the production of unknown neutron-rich nuclei

In recent years, substantial efforts have been made for the study of multinucleon transfer reactions at energies around the Coulomb barrier both experimentally and theoretically, aiming at the production of unknown neutron-rich heavy nuclei. It is crucial to provide reliable theoretical predictions based on microscopic theories with sufficient predictive power. Purpose: This article aims to clarify the applicability of the quantal diffusion approach based on the stochastic mean-field (SMF) theory for multinucleon transfer processes. Isotope production cross sections are evaluated for the reactions of 64 Ni + 208 Pb at E c.m. = 268 MeV and 58 Ni + 208 Pb at E c.m. = 270 MeV and are compared with available experimental data. Methods: Three-dimensional time-dependent Hartree-Fock (TDHF) calculations are carried out for a range of initial orbital angular momenta with Skyrme SLy4d functional. Quantal diffusion equations, derived based on the SMF theory, for variances and covariance of neutron and proton numbers of reaction products are solved, with microscopic drift and diffusion coefficients obtained from time evolution of occupied single-particle orbitals in TDHF. Secondary de-excitation processes, both particle evaporation and fission, are simulated by a statistical compound-nucleus de-excitation model, gemini++. Dynamics of a fast isospin equilibration process followed by a slow drift toward the mass symmetry are commonly observed, as expected. Various reaction outcomes are evaluated, including average mass and charge numbers of reaction products, total kinetic energy loss (TKEL), scattering angle, contact time, and production cross sections for primary and secondary products. By comparing with the experimental data, we find that SMF and TDHF quantitatively reproduce experimental data for few-nucleon-transfer channels around the average values. In contrast, for many-nucleon-transfer channels, we find that the SMF approach provides much better description of the experimentally measured isotopic distributions. The results underline the importance of beyond-mean-field effects, especially one-body (mean-field) fluctuations and correlations, in describing multinucleon transfer processes. Moreover, through a combined analysis of SMF with a statistical model, gemini++, we find a significant contribution of transfer-induced fission, which is consistent with the experimental observation. In some cases, the SMF approach overestimates the isotopic width, requiring further improvements of the theoretical description. Possible ways to improve the description are discussed. The SMF approach is designed to describe the quantum many-body problem according to an ensemble of mean-field trajectories, taking into account part of many-body correlations in the description. As it requires feasible computational costs comparable to the ordinary TDHF approach, together with further model improvements, it will be a promising tool in the search for optimal reaction conditions to produce yet-unknown neutron-rich heavy nuclei through the multinucleon transfer reaction.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2020

Kinetic Parameters and Diffusivity of Uranium in FLiNaK and ClLiK

Anodic stripping voltammetry (ASV) and cyclic voltammetry (CV) measurements at 773, 823, and 873 K were made of uranium trifluoride (UF 3 ) in lithium fluoride-sodium fluoride-potassium fluoride eutectic (FLiNaK) and uranium trichloride (UCl 3 ) in lithium chloride-potassium chloride eutectic (ClLiK). ASV data were used to estimate the charge transfer coefficients, exchange current densities, and activation energies of the uranium reactions. Charge transfer coefficients of both salt systems were within the range of 0.17 to 0.38. Exchange current densities in the fluoride and chloride salts were estimated within the range of 0.060 to 0.12 A cm -2 . Activation energy of uranium exchange current was 16.0 kJ mol -1 in the fluoride salt and 28.9 kJ mol -1 in the chloride salt. Kinetics of charge transfer were found to be faster in FLiNaK. Analyses of the CV data suggest the electrochemical system was diffusion controlled and irreversible in the chloride and fluoride salt mixtures. Diffusion coefficients of uranium in the range of temperatures were on the order of 10-5 cm 2 s -1 in both systems. Greater values of diffusivity in ClLiK are attributed to its lower density compared with FLiNaK. Activation energy of uranium diffusion in the fluoride and chloride salt mixtures were 53.4 and 89.4 kJ mol -1 , respectively.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2024

Bridging the gap between experiments and simulations using machine learning

The physics of inertial confinement fusion is rich and complex. Simulation codes that are used to design experiments are computationally expensive and lack the predictive capability required for extensive parameter exploration in search of a high-performing design for laser direct drive. In this work we use deep learning to build a fast emulator of experiments. To facilitate the development of the deep-learning model, an autoencoder is used to reduce the dimensionality of the input space. Two deep learning models are developed. One model is trained on a vast array of simulation data and is subsequently calibrated to expensive and limited experimental data using a technique known as “transfer learning.” The other model is trained on a statistical model and is subsequently calibrated using experimental data. A comparative study of the two predictive models is carried out. The models potentially reproduce key experimental observables with high accuracy and unprecedented inference times relative to those achieved with simulation codes. These models facilitate rapid exploration of a high dimensional input parameter space.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2020

Data Preservation at MINERvA

Over the past ten years, MINERvA has collected an accelerator neutrino interaction dataset that is uniquely relevant to the energy range of DUNE. These are the only currently available data at intermediate and high momentum transfers for multiple nuclear targets in the same beam. MINERvA is undertaking a campaign to preserve these data and make them publicly available so that they may be analyzed beyond the end of the MINERvA collaboration. We encourage the community to consider the development of centralized resources to enable long-term access to these data and analysis tools for the entire HEP community.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗