SEARCH · Search NASA
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.
Gravitational wave data sourced from the the time-dependent fluid quadrupole motion, in the Chimera D-Series three-dimensional core collapse supernova simulations. Data from three models initiated from three different progenitors are presented: D9.6-3D, D15-3D, and D25-3D. Please see the README for more information about the data structure and progenitors.
Proto-neutron star property evolution data. Data from three models initiated from three different progenitors are presented: D9.6-3D, D15-3D, and D25-3D. Please see the README for more information about the data structure and progenitors.
This dataset contains gravitational wave data sourced from the the time-dependent fluid quadrupole motion as well as neutrino anisotropy, in the Chimera F-Series two-dimensional core collapse supernova simulations. Data from two models initiated from two different progenitors are presented: F15.78 and F15.79. Please see the README for more information about the data structure and progenitors.
Detection of radioxenon in the atmosphere is considered the most specific method to determine whether a nuclear test has occurred. It is therefore an important component of the Comprehensive Test Ban Treaty (CTBT) verification regime. Radioxenon collection is primarily performed by means of a network of remote monitoring stations at fixed locations around the world. This arrangement is effective but often limited in terms of the response time and sensitivity, two factors that could be improved if radioxenon collection devices were portable. By employing recent advancements in porous materials and adsorption technologies, it is possible to overcome some of the limitations of current radioxenon collection systems. This work established the initial modeling and scoping study towards developing an efficient and portable radioxenon collection system for nuclear test monitoring.
Focal Area: (2) Predictive modeling through the use of AI techniques and AI-derived model components and the use of AI and other tools to design a prediction system comprising a hierarchy of models. Science Challenge: While both climate and weather forecast systems have continued to improve due to substantial efforts to improve computational capabilities, observations, and numerical models, the atmosphere is a chaotic system, and this puts a fundamental limit on our ability to make predictions. Forecasts made by high-resolution models initialized with only slightly different atmospheric states can quickly diverge. Quantifying uncertainty in forecasts is essential to adequately understand them and to make the best-informed policy decisions particularly when it comes to hydrology, extreme weather (including extreme precipitation events), and climate.
Construction and deployment of the experimental testbed encountered scheduling delays, and completion was subsequently further delayed by curtailment of LANL operations due to the COVID–19 pandemic. Between March and October, LANL was in a state of “Limited Operations” such that only certain “mission–critical” work was performed with approximately 25% on–site staffing. In spite of this, much of the lab space preparation was executed and the five initial model house” units were constructed and installed during this period. Still, the pace of construction was impacted by difficulty in coordinating personnel and minimizing contact between workers.
This work delineates the work accomplished in the 2nd quarter of CY 2021, which consists of the first quarter of funding for LLNL in the FERMI project (Fusion Energy Reactors Modeling Initiative), a project of the GAMOW (Galvanizing Advances in Market-Aligned Fusion for an Overabundance of Watts) program under ARPA-E. LLNL is responsible for the structural mechanics component of FERMI, concentrating on fusion first wall and blankets.
Brunswick is experiencing unprecedented volume for engines and fully produced boats. Supply Chain (parts) has difficulty accurately forecasting demand needs which presents problems in the manufacturing process. Streamline ordering process enabling the Brunswick Boat Group (BBG) to more efficiently forecast part demand for the Mercury Engine Group by setting up an AI/ML model. Initial focus will be 15 SKUs with inaccurate demand forecasting based on current manual processes.
Construction and deployment of the experimental testbed encountered scheduling delays, and completion was subsequently further delayed by curtailment of LANL operations due to the COVID-19 pandemic. Between March and October, LANL was in a state of “Limited Operations” such that only certain “mission-critical” work was performed with approximately 25% on-site staffing. In spite of this, much of the lab space preparation was executed and the five initial model house” units were constructed and installed during this period. Still, the pace of construction was impacted by difficulty in coordinating personnel and minimizing contact between workers. Currently, LANL is in the mode of “Normal Operations with Maximum Telework,” which continues to limit the availability of on-site personnel. Nevertheless, Milestone 3 is now complete. This document describes the experiment as it is deployed and provides status for each task for this milestone.
Multi-fidelity approaches are emerging as effective strategies in computational science to handle otherwise intractable tasks like Uncertainty Quantification (UQ), training of Machine Learning (ML) models, and optimization, for expensive high-fidelity applications in which the amount of available simulations or data is limited. The main idea is simple: large datasets generated for low-fidelity approximations of the problem at hand are fused with a much sparser dataset for the target (high-fidelity) system. In this paper, we build on our recent success in designing random Fourier Neural Networks (rFNNs) [1] to target problems arising in wind energy applications and in particular problems of interest for hurricane modeling. In this context, data for the high-fidelity models are limited and lower fidelity alternatives are needed. In this work, we introduce a novel multi-fidelity training approach for our rFNNs and demonstrate its use on a simple verification problem and on a hurricane modeling problem in which high-fidelity data are generated via Large-Eddy Simulations (LES), while low-fidelity data are given by a mesoscale model. Initial results demonstrate how the multi-fidelity training approach can improve the quality of the resulting surrogate.
Boltzmann machines have useful roles in deep learning applications, such as generative data modeling, initializing weights for other types of networks, or extracting efficient representations from high-dimensional data. Most Boltzmann machines use restricted topologies that exclude looping connectivity, as such connectivity creates complex distributions that are difficult to sample. We have used an open-system quantum annealer to sample from complex distributions and implement Boltzmann machines with looping connectivity. Further, we have created policies mapping Boltzmann machine variables to the quantum bits of an annealer. These policies, based on correlation and entropy metrics, dynamically reconfigure the topology of Boltzmann machines during training and improve performance.
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.
We combine photometry of Eris from a 6 month campaign on the Palomar 60 inch telescope in 2015, a 1 month Hubble Space Telescope WFC3 campaign in 2018, and Dark Energy Survey data spanning 2013–2018 to determine a light curve of definitive period 15.771 ± 0.008 days (1σ formal uncertainties), with nearly sinusoidal shape and peak-to-peak flux variation of 3%. This is consistent at part-per-thousand precision with the P = 15.785 90 ± 0.00005 day sidereal period of Dysnomia's orbit around Eris, strengthening the recent detection of synchronous rotation of Eris by Szakáts et al. with independent data. Photometry from Gaia are consistent with the same light curve. We detect a slope of 0.05 ± 0.01 mag per degree of Eris's brightness with respect to illumination phase averaged across g, V, and r bands, intermediate between Pluto's and Charon's values. Variations of 0.3 mag are detected in Dysnomia's brightness, plausibly consistent with a double-peaked light curve at the synchronous period. The synchronous rotation of Eris is consistent with simple tidal models initiated with a giant-impact origin of the binary, but is difficult to reconcile with gravitational capture of Dysnomia by Eris. The high albedo contrast between Eris and Dysnomia remains unexplained in the giant-impact scenario.
Abstract. Stratocumulus clouds in the marine boundary layer cover a large fraction of ocean surface and play an important role in the radiative energy balance of the Earth system. Simulating these clouds in Earth system models (ESMs) has proven to be extremely challenging, in part because cloud microphysical processes such as the autoconversion of cloud water into precipitation occur at scales much smaller than typical ESM grid sizes. An accurate autoconversion parameterization needs to account for not only the local microphysical process (e.g., the dependence on cloud water content qc and cloud droplet number concentration Nc) but also the subgrid-scale variability of the cloud properties that determine the process rate. Accounting for subgrid-scale variability is often achieved by the introduction of a so-called enhancement factor E. Previous studies of E for autoconversion have focused more on its dependence on cloud regime and ESM grid size, but they have largely overlooked the vertical dependence of E within the cloud. In this study, we use a large-eddy simulation (LES) model, initialized and constrained with in situ and surface-based measurements from a recent airborne field campaign, to characterize the vertical dependence of the horizontal variation of qc in stratocumulus clouds and the implications for E. Similar to our recent observational study (Zhang et al., 2021), we found that the inverse relative variance of qc, an index of horizontal homogeneity, generally increases from cloud base upward through the lower two-thirds of the cloud and then decreases in the uppermost one-third of the cloud. As a result, E decreases from cloud base upward and then increases towards the cloud top. We apply a decomposition analysis to the LES cloud water field to understand the relative roles of the mean and variances of qc in determining the vertical dependence of E. Our analysis reveals that the vertical dependence of the horizontal qc variability and enhancement factor E is a combined result of condensational growth throughout the lower portion of the cloud and entrainment mixing at cloud top. The findings of this study indicate that a vertically dependent E should be used in ESM autoconversion parameterizations.
Assessing the availability of key wind resources requires augmenting observations to support the implementation of wind energy infrastructure. However, observations are limited, necessitating the development of high-resolution, long-term gridded datasets. This study presents a robust, dynamically downscaled climatological dataset, offering 20 years of hourly wind data at a 4 km spatial resolution across North America, and evaluates its performance against observations, including meteorological towers and automated surface-observing system (ASOS) stations, as well as coarse-resolution reanalysis data (the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis version 5 (ERA5)). Results demonstrate that the downscaled high-resolution wind data outperform ERA5 in regions of complex terrain and coastal areas, with improved overlap coefficients for wind data distributions and reduced root mean square errors (RMSEs) for hub-height and near-surface diurnal wind patterns. The downscaled simulation also captures the synoptic drivers of seasonal wind direction patterns reasonably well, indicated by high wind rose similarity indices. This study also provides an analysis of interannual variability, utilizing the dataset's full 20-year period, and model uncertainty, generated by varying model initial conditions and physics parameterizations across 1-year ensemble members, which are key considerations for wind resource assessment in wind farm development.
Extension of the predictive power of dynamic weather forecasting to scales below the conventional synoptic or cyclonic scales in the near future is assessed. Lower costs per computation, more powerful computers, and a 100 km mesh over the North American area (with coarser mesh extending beyond it) are noted at present. Doubling the resolution even locally (to 50 km mesh) would entail a 16-fold increase in costs (including vertical resolution and halving the time interval), and constraints on domain size and length of forecast. Boundary conditions would be provided by the surrounding 100 km mesh, and time-varying lateral boundary conditions can be considered to handle moving phenomena. More physical processes to treat, more efficient numerical techniques, and faster computers (improved software and hardware) backing up satellite and radar data could produce further improvements in forecasting in the 1980s. Boundary layer modeling, initialization techniques, and quantitative precipitation forecasting are singled out among key tasks.
The paper examines the feasibility of using the H2 rotational Raman spectrum to study the physical structure of the atmospheres of the outer planets. Computations of the strengths of the S(0) and S(1) lines are made for a wide range of physical conditions on the basis of a semi-infinite homogeneous pure H2 atmospheric model. Initial applications of the Raman probe technique to Jupiter and Uranus are reported.
The mechanisms by which orbital resonance locks are maintained among Io, Europa and Ganymede are analyzed, and the effects of the dissipative tides in Jupiter and Io on their establishment and evolution are considered. According to the proposed model, initially all three satellites are in orbits far from the 2:1 commensurabilities or the three body lock. The tide raised on Io damps down the free eccentricity; only modest tidal heating occurs. Subsequently the dissipative tide raised on Jupiter by Io causes Io's orbit to spiral outwards; Io approaches the 2:1 commensurability with Europa. Io's forced eccentricity increases rapidly to a critical value, and thereafter the resonant interaction causes Europa's orbit to expand at half that of Io's orbit. A fluid core is probably formed as the result of tidal heating. Finally Europa approaches the 2:1 commensurability, angular momentum is transferred from Europa's orbit to Ganymede's, and a steady state is attained. On the basis of the observed three-body resonance amplitude (0.066 deg), it is concluded that three-body resonance may have formed less than 500 million years ago; a measurement of the rate of tidal heating in Io may determine the Q value of both Io and Jupiter. Upper and lower limits for the Jovian Q value of 2 million and 20,000 are suggested.