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.
We describe in detail a method of construction of a discrete linear dynamical model of a controllable object in the recurrent Cauchy form corresponding to an initial continuous dynamical system. We describe discretization algorithms for free and forced motion of a linear continuous dynamical system and propose methods of construction of the state, control, and perturbation transition matrices.
Drone delivery is widely-researched to alter the current e-commerce delivery convention for providing short delivery lead times. Yet, flight range of drones, constrained by the available battery technology, sets a milestone toward realizing the sole-drone delivery. To tackle with the flight range limitation, locating automated battery swapping machines (ABSM) have been proposed and a few studies modeled the problem. Using the ABSMs, drones can be loaded with fully-charged batteries along the route to a demand location. Here, we introduce a mixed-integer nonlinear program to model the problem. The objective of the program is to optimally select ABSM locations, determine the delivery-mode choices (drone-only, truck-only, and mixed delivery) of demand locations, find drone delivery routes, and approximate the baseline requirements for the number of drones and batteries needed. The program minimizes the overall delivery system costs including: ABSM, delivery, drone ownership, battery inventory, and service congestion. A cutting-plane method is developed to find exact solutions in finite iterations. Computational experiments showed that the method quickly yields the optimal solution to instances with less than 60 ABSM candidates and 20 demand locations. A case study shows that the optimal drone delivery infrastructure can save almost 20% cost compared to the conventional truck-only delivery. Sensitivity analyses were conducted to reveal the impact of key parameters in the decision-making and found that a decrease in the ABSM and drone costs highly affect the system cost.
Large language models (LLMs) can surpass humans in certain forecasting tasks. What role does this leave for humans in the overall decision process? One possibility is that humans, despite performing worse than LLMs, can still add value when teamed with them. A human and machine team can surpass each individual teammate when team members’ confidence is well calibrated and team members diverge in which tasks they find difficult (i.e., calibration and diversity are needed). We simplified and extended a Bayesian approach to combining judgments using a logistic regression framework that integrates confidence-weighted judgments for any number of team members. Using this straightforward method, we demonstrated its effectiveness in both image classification and neuroscience forecasting tasks. Combining human judgments with one or more machines consistently improved overall team performance. Our hope is that this simple and effective strategy for integrating the judgments of humans and machines will lead to productive collaborations.
The necessity to reduce global energy consumption calls for innovative strategies in building thermal management. Passive thermal regulation, particularly through bio-inspired designs, offers a promising avenue by mimicking nature's efficient control of optical properties. This research introduces a novel, climate-responsive coating that integrates optimized bio-inspired surface features with a solid-solid phase change material (SS-PCM) to dynamically manage solar absorptivity without adding additional thickness, enabling both heating and cooling as needed. Drawing on the photonic architectures of the Saharan silver ant and Morpho Didius butterfly, we employed a modeling and multi-objective optimization framework to tailor these surface features. Simulations reveal that surface texture, rather than the intrinsic phase transition of the SS PCM, dominates optical control. Relative to a flat SS PCM coating, optimized isotropic random roughness and broader range features yielded the highest passive heating power increase of about 144 % and 319 % respectively suitable for cold climates. Saharan ant-inspired features enhanced passive cooling for hot climates, achieving a 21.8 % improvement. For moderate climates, Butterfly-wing-inspired surface features provided a balanced enhancement of 19 % for heating and 7 % for cooling. Across all cases, the optimized surface features reduced combined heating and cooling energy demand more effectively than the baseline coating, while preserving material thickness. These findings demonstrate that climate-adaptive, optimized bio-inspired surface features can unlock the full potential of SS PCM coatings, providing a versatile pathway to significant energy savings in buildings and other applications. The methodology establishes a framework for designing next-generation adaptive envelopes that leverage natural photonic principles for high-impact, low-cost thermal regulation.
Aims. In this work we extended an energy-integrated neutrino transport method to facilitate efficient, yet precise, modeling of compact astrophysical objects. We particularly focus on core-collapse supernovae. Methods. We implemented a gray neutrino-transport framework from the literature into FLASH and performed a detailed evaluation of its accuracy in core-collapse supernova simulations. Based on comparisons with results from simulations using energy-dependent neutrino transport, we incorporated several improvements to the original scheme. Results. Our analysis shows that our gray neutrino transport method successfully reproduces key aspects from more complex energy-dependent transport across a variety of progenitors and equations of state. We find both qualitative and reasonable quantitative agreement with multi-group M1 transport simulations. However, the gray scheme tends to slightly favor shock revival. In terms of gravitational wave and neutrino signals, there is a good alignment with the energy-dependent transport, although we find 15–30% discrepancies in the average energy and luminosity of heavy-lepton neutrinos. Simulations using the gray transport are around four times faster than those using energy-dependent transport.
We report simulating and predicting multiscale problems that couple multiple physics and dynamics across many orders of spatiotemporal scales is a great challenge that has not been investigated systematically by deep neural networks (DNNs). Herein, we develop a framework based on operator regression, the so-called deep operator network (DeepONet), with the long-term objective to simplify multiscale modeling by avoiding the fragile and time-consuming “hand-shaking” interface algorithms for stitching together heterogeneous descriptions of multiscale phenomena. To this end, as a first step, we investigate if a DeepONet can learn the dynamics of different scale regimes, one at the deterministic macroscale and the other at the stochastic microscale regime with inherent thermal fluctuations. Specifically, we test the effectiveness and accuracy of the DeepONet in predicting multirate bubble growth dynamics, which is described by a Rayleigh–Plesset (R–P) equation at the macroscale and modeled as a stochastic nucleation and cavitation process at the microscale by dissipative particle dynamics (DPD). First, we generate data using the R–P equation for multirate bubble growth dynamics caused by randomly time-varying liquid pressures drawn from Gaussian random fields (GRFs). Our results show that properly trained DeepONets can accurately predict the macroscale bubble growth dynamics and can outperform long short-term memory networks. We also demonstrate that the DeepONet can extrapolate accurately outside the input distribution using only very few new measurements. Subsequently, we train the DeepONet with DPD data corresponding to stochastic bubble growth dynamics. Although the DPD data are noisy and we only collect sparse data points on the trajectories, the trained DeepONet model is able to predict accurately the mean bubble dynamics for time-varying GRF pressures. Taken together, our findings demonstrate that DeepONets can be employed to unify the macroscale and microscale models of the multirate bubble growth problem, hence providing new insight into the role of operator regression via DNNs in tackling realistic multiscale problems and in simplifying modeling with heterogeneous descriptions.
ABSTRACT We present the results of a proof-of-concept experiment that demonstrates that deep learning can successfully be used for production-scale classification of compact star clusters detected in Hubble Space Telescope(HST) ultraviolet-optical imaging of nearby spiral galaxies ($D\lesssim 20\, \textrm{Mpc}$) in the Physics at High Angular Resolution in Nearby GalaxieS (PHANGS)–HST survey. Given the relatively small nature of existing, human-labelled star cluster samples, we transfer the knowledge of state-of-the-art neural network models for real-object recognition to classify star clusters candidates into four morphological classes. We perform a series of experiments to determine the dependence of classification performance on neural network architecture (ResNet18 and VGG19-BN), training data sets curated by either a single expert or three astronomers, and the size of the images used for training. We find that the overall classification accuracies are not significantly affected by these choices. The networks are used to classify star cluster candidates in the PHANGS–HST galaxy NGC 1559, which was not included in the training samples. The resulting prediction accuracies are 70 per cent, 40 per cent, 40–50 per cent, and 50–70 per cent for class 1, 2, 3 star clusters, and class 4 non-clusters, respectively. This performance is competitive with consistency achieved in previously published human and automated quantitative classification of star cluster candidate samples (70–80 per cent, 40–50 per cent, 40–50 per cent, and 60–70 per cent). The methods introduced herein lay the foundations to automate classification for star clusters at scale, and exhibit the need to prepare a standardized data set of human-labelled star cluster classifications, agreed upon by a full range of experts in the field, to further improve the performance of the networks introduced in this study.
ABSTRACT Relativistic jets launched by rotating black holes are powerful emitters of non-thermal radiation. Extraction of the rotational energy via electromagnetic stresses produces magnetically dominated jets, which may become turbulent. Studies of magnetically dominated plasma turbulence from first principles show that most of the accelerated particles have small pitch angles, i.e. the particle velocity is nearly aligned with the local magnetic field. We examine synchrotron self-Compton radiation from anisotropic particles in the fast cooling regime. The small pitch angles reduce the synchrotron cooling rate and promote the role of inverse Compton (IC) cooling, which can occur in two different regimes. In the Thomson regime, both synchrotron and IC components have soft spectra, νFν ∝ ν1/2. In the Klein–Nishina regime, synchrotron radiation has a hard spectrum, typically νFν ∝ ν, over a broad range of frequencies. Our results have implications for the modelling of BL Lacertae objects (BL Lacs) and gamma-ray bursts (GRBs). BL Lacs produce soft synchrotron and IC spectra, as expected when Klein–Nishina effects are minor. The observed synchrotron and IC luminosities are typically comparable, which indicates a moderate anisotropy with pitch angles θ ≳ 0.1. Rare orphan gamma-ray flares may be produced when θ ≪ 0.1. The hard spectra of GRBs may be consistent with synchrotron radiation when the emitting particles are IC cooling in the Klein–Nishina regime, as expected for pitch angles θ ∼ 0.1. Blazar and GRB spectra can be explained by turbulent jets with a similar electron plasma magnetization parameter, σe ∼ 104, which for electron–proton plasmas corresponds to an overall magnetization σ = (me/mp)σe ∼ 10.
Computing at large scales has become extremely challenging due to increasing heterogeneity in both hardware and software. More and more scientific workflows must tackle a range of scales and use machine learning and AI intertwined with more traditional numerical modeling methods, placing more demands on computational platforms. These constraints indicate a need to fundamentally rethink the way computational science is done and the tools that are needed to enable these complex workflows. The current set of C++-based solutions may not suffice, and relying exclusively upon C++ may not be the best option, especially because several newer languages and boutique solutions offer more robust design features to tackle the challenges of heterogeneity. In June 2023, we held a mini symposium that explored the use of newer languages and heterogeneity solutions that are not tied to C++ and that offer options beyond template metaprogramming and Parallel. For for performance and portability. In conclusion, we describe some of the presentations and discussion from the mini symposium in this article.
Here, we describe an experiment to form and characterize a section of a spherically imploding plasma liner by merging six supersonic plasma jets that are launched by newly designed contoured-gap coaxial plasma guns. This experiment is a prelude to forming a fully spherical imploding plasma liner using many dozens of plasma guns, as a standoff driver for plasma-jet-driven magneto-inertial fusion. The objectives of the six-jet experiments are to assess the evolution and scalings of liner Mach number and uniformity, which are important metrics for spherically imploding plasma liners to compress magnetized target plasmas to fusion conditions. Lastly, this article describes the design of the coaxial plasma guns, experimental characterization of the plasma jets, six-jet experimental setup and diagnostics, initial diagnostic data from three- and six-jet experiments, and the high-level objectives of associated numerical modeling.
With increasing wind power penetration in the electricity grid, system operators are recognizing the need for additional flexibility, and some are implementing new ramping products as a type of ancillary service. However, wind is generally thought of as causing the need for ramping services, not as being a potential source for the service. In this paper, a multi-timescale unit commitment and economic dispatch model is developed to consider the wind power ramping product (WPRP). An optimized swinging door algorithm with dynamic programming is applied to identify and forecast wind power ramps (WPRs). Designed as positive characteristics of WPRs, the WPRP is then integrated into the multi-timescale dispatch model that considers new objective functions, ramping capacity limits, active power limits, and flexible ramping requirements. Numerical simulations on the modified IEEE 118-bus system show the potential effectiveness of WPRP in increasing the economic efficiency of power system operations with high levels of wind power penetration. It is found that WPRP not only reduces the production cost by using less ramping reserves scheduled by conventional generators, but also possibly enhances the reliability of power system operations. Moreover, wind power forecasts play an important role in providing high-quality WPRP service.
The Electricity Markets Investment Suite Approximate Equilibrium (EMIS-AE) package is developed at NREL and provides a solution capability to solve generation expansion equilibrium problems in the electricity markets. EMIS-AE is designed to capture the evolution of the electricity generation portfolio resulting from the interactions of heterogeneous investors under different policy and market designs. The investment problem for each generation company (GENCO) is a bi-level problem with the investment decision made in the upper level and market clearing condition in the lower level, which traditionally is represented as a Mathematical Program with Equilibrium Constraint (MPEC). EMIS-AE provides a predictive model to be trained for estimating the system-wide revenues for each technology type across energy, ancillary services and capacity markets given the amount of installed capacity on the grid. The profit maximization investment problem for each GENCO is solved using a global search algorithm, which uses the predictive model to evaluate the objective function. To solve for the strategic equilibrium, each GENCO’s problem is plugged into a diagonalization algorithm that is generally used in multi-leader, single-follower bi-level problems.
N2A includes an object-oriented language for modeling neural systems as well as a software tool for editing and simulating models. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. SAND2021-2695 O
AMLRO (Active Machine Learning Reaction Optimizer) is an open-source framework designed to accelerate chemical reaction optimization using active learning with classical machine learning regression models. AMLRO integrates space-filling sampling strategies (e.g., Sobol and Latin Hypercube sampling) with iterative model training, prediction, and experiment selection to efficiently navigate complex reaction spaces. The platform supports multiple regression models, flexible multi-objective definitions, and user-defined parameter bounds, enabling data-efficient optimization from small initial datasets. AMLRO is designed for ease of use by experimentalists and can operate as a standalone decision-support tool or be integrated into closed-loop automated experimentation workflows.
Background: The importance of infectious disease epidemic forecasting and prediction research is underscored by decades of communicable disease outbreaks, including COVID-19. Unlike other fields of medical research, such as clinical trials and systematic reviews, no reporting guidelines exist for reporting epidemic forecasting and prediction research despite their utility. We therefore developed the EPIFORGE checklist, a guideline for standardized reporting of epidemic forecasting research. Methods and findings: We developed this checklist using a best-practice process for development of reporting guidelines, involving a Delphi process and broad consultation with an international panel of infectious disease modelers and model end users. The objectives of these guidelines are to improve the consistency, reproducibility, comparability, and quality of epidemic forecasting reporting. The guidelines are not designed to advise scientists on how to perform epidemic forecasting and prediction research, but rather to serve as a standard for reporting critical methodological details of such studies. Conclusions: These guidelines have been submitted to the EQUATOR network, in addition to hosting by other dedicated webpages to facilitate feedback and journal endorsement.
The objective of this first milestone was to calibrate and if necessary, modify a constitutive model predicting the primary and secondary creep response of P91. The overall path to be followed consists of first calibrating this constitutive model against experimental data available from the literature (this milestone). In parallel, data acquired by indentation tests will be used to calibrate the model (this involves transforming the indentation data into equivalent uniaxial tests). An overall comparison of model parameters will allow assessing to what extent these rapid tests can be used to calibrate models predicting the overall mechanical response of the bulk. Given that COVID naturally hindered the gathering of new experimental data, the choice was made to ensure that LANL would have sufficient funds in year two to perform the bulk of the new calibration work. The following report details (i) how and which of the bulk experimental data was selected, (ii) constitutive model formulation, (iii) predictions of the constitutive model: effects of imposed stress, temperature and precipitate content, (iv) future work.
This presentation describes the OCHRE model as part of NREL's Powered By webinar series. OCHRE is a building energy modeling tool designed to model flexible loads in residential buildings. OCHRE includes detailed models and controls for flexible devices including HVAC equipment, water heaters, electric vehicles, solar PV, and batteries. It is designed to run in co-simulation with custom controllers, aggregators, and grid models.