Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep residual learning”

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.

119 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations

In the last 50 years there has been a tremendous progress in solving numerically the Navier-Stokes equations using finite differences, finite elements, spectral, and even meshless methods. Yet, in many real cases, we still cannot incorporate seamlessly (multi-fidelity) data into existing algorithms, and for industrial-complexity applications the mesh generation is time consuming and still an art. Moreover, solving ill-posed problems (e.g., lacking boundary conditions) or inverse problems is often prohibitively expensive and requires different formulations and new computer codes. Here, we employ physics-informed neural networks (PINNs), encoding the governing equations directly into the deep neural network via automatic differentiation, to overcome some of the aforementioned limitations for simulating incompressible laminar and turbulent flows. We develop the Navier-Stokes flow nets (NSFnets) by considering two different mathematical formulations of the Navier-Stokes equations: the velocity-pressure (VP) formulation and the vorticity-velocity (VV) formulation. Since this is a new approach, we first select some standard benchmark problems to assess the accuracy, convergence rate, computational cost and flexibility of NSFnets; analytical solutions and direct numerical simulation (DNS) databases provide proper initial and boundary conditions for the NSFnet simulations. The spatial and temporal coordinates are the inputs of the NSFnets, while the instantaneous velocity and pressure fields are the outputs for the VP-NSFnet, and the instantaneous velocity and vorticity fields are the outputs for the VV-NSFnet. This is unsupervised learning and, hence, no labeled data are required beyond boundary and initial conditions and the fluid properties. The residuals of the VP or VV governing equations, together with the initial and boundary conditions, are embedded into the loss function of the NSFnets. No data is provided for the pressure to the VP-NSFnet, which is a hidden state and is obtained via the incompressibility constraint without extra computational cost. Unlike the traditional numerical methods, NSFnets inherit the properties of neural networks (NNs), hence the total error is composed of the approximation, the optimization, and the generalization errors. Here, we empirically attempt to quantify these errors by varying the sampling (“residual”) points, the iterative solvers, and the size of the NN architecture. For the laminar flow solutions, we show that both the VP and the VV formulations are comparable in accuracy but their best performance corresponds to different NN architectures. The initial convergence rate is fast but the error eventually saturates to a plateau due to the dominance of the optimization error. For the turbulent channel flow, we show that NSFnets can sustain turbulence at , but due to expensive training we only consider part of the channel domain and enforce velocity boundary conditions on the subdomain boundaries provided by the DNS data base. We also perform a systematic study on the weights used in the loss function for balancing the data and physics components, and investigate a new way of computing the weights dynamically to accelerate training and enhance accuracy. In the last part, we demonstrate how NSFnets should be used in practice, namely for ill-posed problems with incomplete or noisy boundary conditions as well as for inverse problems. We obtain reasonably accurate solutions for such cases as well without the need to change the NSFnets and at the same computational cost as in the forward well-posed problems. As a result, we also present a simple example of transfer learning that will aid in accelerating the training of NSFnets for different parameter settings.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Analysis of antibodies from HCV elite neutralizers identifies genetic determinants of broad neutralization

The high genetic diversity of hepatitis C virus (HCV) complicates effective vaccine development. We screened a cohort of 435 HCV-infected individuals and found that 2%–5% demonstrated outstanding HCV-neutralizing activity. From four of these patients, we isolated 310 HCV antibodies, including neutralizing antibodies with exceptional breadth and potency. High neutralizing activity was enabled by the use of the VH1-69 heavy-chain gene segment, somatic mutations within CDRH1, and CDRH2 hydrophobicity. Structural and mutational analyses revealed an important role for mutations replacing the serines at positions 30 and 31, as well as the presence of neutral and hydrophobic residues at the tip of the CDRH3. Based on these characteristics, we computationally created a de novo antibody with a fully synthetic VH1-69 heavy chain that efficiently neutralized multiple HCV genotypes. Finally, our findings provide a deep understanding of the generation of broadly HCV-neutralizing antibodies that can guide the design of effective vaccine candidates.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

$\text{GPLaSDI}$: Gaussian Process-based interpretable Latent Space Dynamics Identification through deep autoencoder

Numerically solving partial differential equations (PDEs) can be challenging and computationally expensive. This has led to the development of reduced-order models (ROMs) that are accurate but faster than full order models (FOMs). Recently, machine learning advances have enabled the creation of non-linear projection methods, such as Latent Space Dynamics Identification (LaSDI). LaSDI maps full-order PDE solutions to a latent space using autoencoders and learns the system of ODEs governing the latent space dynamics. By interpolating and solving the ODE system in the reduced latent space, fast and accurate ROM predictions can be made by feeding the predicted latent space dynamics into the decoder. In this paper, we introduce GPLaSDI, a novel LaSDI-based framework that relies on Gaussian process (GP) for latent space ODE interpolations. Using GPs offers two significant advantages. First, it enables the quantification of uncertainty over the ROM predictions. Second, leveraging this prediction uncertainty allows for efficient adaptive training through a greedy selection of additional training data points. This approach does not require prior knowledge of the underlying PDEs. Consequently, GPLaSDI is inherently non-intrusive and can be applied to problems without a known PDE or its residual. Here we demonstrate the effectiveness of our approach on the Burgers equation, Vlasov equation for plasma physics, and a rising thermal bubble problem. Our proposed method achieves between 200 and 100,000 times speed-up, with up to 7% relative error.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

ResNet and CycleGAN for pulse shape discrimination of He-4 detector pulses: Recovering pulses conventional algorithms fail to label unanimously

Pulse shape discrimination (PSD) capable detectors, such as He-4, that respond to neutron and gamma-ray 7 interactions have a threshold deposited energy value below which n/γ discrimination vanishes when using 8 conventional PSD algorithms. Recent attempts in applying supervised learning based artificial neural 9 networks for PSD use the pulses in the separated regions to train the networks so they can be used to classify 10 another set of separated pulses. In doing so, pulses previously indistinguishable are not recovered for 11 classification, which would have increased the number of neutron and gamma-ray pulses that could be used 12 for further analysis. Assuming the reason why conventional PSD algorithms have unseparated regions is 13 because the parameter space of the algorithms fail to capture the intrinsic (but subtle) distinguishing 14 behavior of some of the neutron and gamma-ray pulses, a cycle-consistent generative adversarial network 15 (CycleGAN) was trained to amplify those differences and extract well separated neutron and gamma-ray 16 clusters. Results show that, once the network is trained with pulses from separated and unseparated regions, 17 it was able to transform the pulses in the unseparated region to improve the PSD. Subsequent n/γ 18 classification was performed using deep residual network (ResNet) that takes pulses with 512 data points 19 as an input. Two different ResNets were explored – simple ResNet and modified ResNet which takes 20 segmented pulse inputs in the first layer and the corresponding time axis values in the last hidden layer. 21 The later approach enables the network to extract time correlated pulse features to enhance its ability to 22 capture the pulse behaviors relevant for PSD. Although it achieves slightly lower accuracy, 99.41% versus 23 99.89%, based on simply counting the number of correct n/γ labels assigned, compared to the simple 24 ResNet, the modified ResNets architecture was able to decreases the cross-entropy loss function by half, 25 which implies that the correct n/γ labels assigned are less likely to be accidental. PSD parameter 26 distributions based on n/γ classification by ResNet before and after transforming unseparated pulses using 27 CycleGAN show that by enhancing the separation between neutrons and gamma-rays, the transformation 28 helps improve the performance of classifier networks that are trained using labeled dataset. The 29 enhancement of neutron and gamma-ray separation by the CycleGAN increased the PSD figure of merit 30 (FOM) by up to 70% in some regions. Here, the results show that, if a given detector achieves clear separation 31 between neutron and gamma-ray pulses in any energy region, such neural network approaches can help 32 lower the energy threshold for the separation and increasing the number of neutron and gamma-ray pulses 33 that can be used for further analysis.

4He↗
DOE OSTI2026

Machine learning identifies novel signatures of antifungal drug resistance in Saccharomycotina yeasts

Antifungal drug resistance is a major challenge in fungal infection management. Numerous genomic changes are known to contribute to acquired drug resistance in clinical isolates of specific pathogens, but whether they broadly explain natural resistance across entire lineages is unknown. We leveraged genomic, ecological, and phenotypic trait data from naturally sampled strains from nearly all known species in subphylum Saccharomycotina to examine the evolution of resistance to eight antifungal drugs. The phylogenetic distribution of drug resistance varied by drug; fluconazole resistance was widespread, while 5-fluorocytosine resistance was rare, except in Lipomycetales. A random forest algorithm trained on genomic data predicted drug-resistant yeasts with 54–75% accuracy. Fluconazole resistance was consistently predicted with the highest accuracy (75.2%). Furthermore, fluconazole resistance prediction accuracy was similar between models trained on genome-wide variation in the presence and number of InterPro protein annotations across Saccharomycotina (75.2%) and those trained on amino acid sequence alignment data of Erg11, a protein known to be involved in fluconazole resistance (74.3-74.9%). Interestingly, the top Erg11 residues for predicting fluconazole resistance across Saccharomycotina do not overlap with, are not spatially close to, and are less conserved than those previously linked to resistance in clinical isolates of Candida albicans. In silico deep mutational scanning of the C. albicans Erg11 protein reveals that amino acid variants implicated in clinical cases of resistance are almost universally destabilizing while variants in our most informative residues are energetically more neutral, explaining why the latter are much more common than the former in natural populations. Importantly, previous experimental analyses of C. albicans Erg11 have shown that amino acid variation in our most informative residues, despite having never been directly implicated in clinical cases, can directly contribute to resistance. Our results suggest that studies of natural resistance in yeast species never encountered in the clinic will yield a fuller understanding of antifungal drug resistance.

Harrison, Marie-Claire [Vanderbilt Univ., Nashvill↗
DOE OSTI2023

Neural Network-Based Variational Methods for Solving Quadratic Porous Medium Equations in High Dimensions

Here, in this paper, we propose and study neural network-based methods for solutions of high-dimensional quadratic porous medium equation (QPME). Three variational formulations of this nonlinear PDE are presented: a strong formulation and two weak formulations. For the strong formulation, the solution is directly parameterized with a neural network and optimized by minimizing the PDE residual. It can be proved that the convergence of the optimization problem guarantees the convergence of the approximate solution in the $L^1$ sense. The weak formulations are derived following (Brenier in Examples of hidden convexity in nonlinear PDEs, 2020) which characterizes the very weak solutions of QPME. Specifically speaking, the solutions are represented with intermediate functions who are parameterized with neural networks and are trained to optimize the weak formulations. Extensive numerical tests are further carried out to investigate the pros and cons of each formulation in low and high dimensions. This is an initial exploration made along the line of solving high-dimensional nonlinear PDEs with neural network-based methods, which we hope can provide some useful experience for future investigations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Detecting and Characterizing Fracture Zones Using a Convolutional Neural Network

This project directly supports the Geothermal Technologies Office (GTO) objectives outlined in the Multi-Year Program Plan (MYPP) by advancing two key research areas: “Exploration and Characterization” and “Data, Modeling, and Analysis.” This project has successfully demonstrated a pre-drilling ability to image and characterize the distribution and connectivity of subsurface faults and fractures, key parameters for identifying permeable pathways that enable geothermal fluids to circulate and produce energy. Specifically, we developed and implemented innovative machine learning methodologies to enhance geothermal exploration. Large-scale faults were detected using a Convolutional Neural Network (CNN), while small-scale fractures were characterized using a novel Double-Beam Neural Network (DBNN). These tools have proven both technically effective and cost-efficient by reducing reliance on expensive exploratory drilling. Through collaboration with our geothermal industry partner, this research has significantly advanced techniques for identifying hidden geothermal systems and extending the productive lifespan of existing geothermal fields. We applied our methods to two geothermal fields—Soda Lake (Nevada) and Lightning Dock (New Mexico)—to identify shallow steam-charged fracture zones and characterize deep faults at depths of 1.5-2 km. The steam zone identified at the Soda Lake geothermal field showed excellent agreement with prior drilling data, validating the effectiveness of our approaches. In addition, the analysis revealed three new prospective drilling targets for further development and verification. The outcomes of this project improve our scientific understanding of geothermal reservoir behavior, enhance exploration efficiency, extend the economic life of existing geothermal plants. Ultimately, these advancements contribute to GTO’s goal of achieving more sustainable, affordable, and data-driven geothermal energy development across the United States.

15 GEOTHERMAL ENERGY↗
NASA NTRSDate not supplied

Gateway Program Safety and Mission Assurance Integration - the Future of Safe Deep Space Human Exploration

As a foundational element of the National Aeronautics and Space Administration (NASA) Artemis Campaign, the Gateway is an incrementally built cislunar spacecraft that will serve as a platform for deep space human exploration, science, and technology demonstration. The Gateway will be a unifying catalyst for international partners around the world to establish sustained deep space scientific investigations, lunar surface access, and missions to Mars. As human exploration moves farther away from Earth, spacecraft designs must prioritize and optimize mass and volume allocations, while minimizing human and spacecraft risk. To accomplish this objective, the Gateway Program Safety and Mission Assurance functions develop, implement, and ensure compliance with requirements, in concert with the accurate characterization and transparent communication of residual hazard risks, for integrated safety, reliability and maintainability and quality assurance. Safety and Mission Assurance was a key contributor during Gateway program pre-formulation and formulation activities where safety and reliability analysis was embedded in the Gateway Systems Engineering and Integration team. During these early program stages, a preliminary Gateway Integrated Hazard Analysis and Preliminary Gateway Probabilistic Risk Assessment assisted in Gateway architectural and operational definition as part of a risk-informed design process. As the deep space architecture has matured, the integrated Safety and Mission Assurance analyses have matured, new safety review processes have been developed, and requirements have been refined to ensure compliance with integrated safety and mission assurance objectives. The Gateway Program is currently concluding the preliminary design review informed milestone, where the primary objectives included: - Ensured completeness and consistency of the preliminary design, including the meeting of all requirements within appropriate margins and acceptable risk posture. - Identification of any major issues moving forward to the Critical Design phase. At this milestone, Safety and Mission Assurance provided numerous products, including Gateway Top Risks and Risk Mitigation Plans, updated integrated hazard analyses, updated probabilistic risk assessment, Crew Survival Analysis Report, and updated Safety and Mission Assurance Requirements and Plans. These products provide a many-faceted perspective on the inherent risk and available mitigations involved in flying the current proposed vehicle design and anticipated stack configurations. In addition, Safety and Mission Assurance identified top technical, process and workforce concerns to be addressed as the program progresses toward the critical design phase. This paper will detail the evolution of the Gateway Program Safety and Mission Assurance integration functions, provide its current status and lessons learned for future human spaceflight programs. Throughout this paper the key tenets of the Gateway Program Safety and Mission Assurance will be discussed: - Application of a risk-informed approach to identify and mitigate areas of highest risk. - Leverage of valuable processes and lessons learned from earlier spaceflight programs. - Development of Safety and Mission Assurance products to inform design risk trades. - Utilization of common Safety and Mission Assurance practices to identify safety risks for multiple perspectives: top-down, bottom-up, and across lines of integration. - Approval of safety hazards at the appropriate level of authority, keeping most deliberation closest to design expertise and elevating risks of greatest concern for program-level consideration. - Championing of Safety and Mission Assurance processes and forums to foster a pervasive safety culture that is transparent, inclusive, and collaborative between all partners. These tenets have allowed the Gateway Safety and Mission Assurance function to play a key role in optimized vehicle design evolution, and early identification and mitigation of Gateway program and Artemis mission risk.

Helen Vaccaro↗
DOE OSTI2021

Heat transport in liquid water from first-principles and deep neural network simulations

In this work, we compute the thermal conductivity of water within linear response theory from equilibrium molecular dynamics simulations, by adopting two different approaches. In one, the potential energy surface (PES) is derived on the fly from the electronic ground state of density functional theory (DFT) and the corresponding analytical expression is used for the energy flux. In the other, the PES is represented by a deep neural network (DNN) trained on DFT data, whereby the PES has an explicit local decomposition and the energy flux takes a particularly simple expression. By virtue of a gauge invariance principle, established by Marcolongo, Umari, and Baroni, the two approaches should be equivalent if the PES were reproduced accurately by the DNN model. We test this hypothesis by calculating the thermal conductivity, at the GGA (PBE) level of theory, using the direct formulation and its DNN proxy, finding that both approaches yield the same conductivity, in excess of the experimental value by approximately 60%. Besides being numerically much more efficient than its direct DFT counterpart, the DNN scheme has the advantage of being easily applicable to more sophisticated DFT approximations, such as meta-GGA and hybrid functionals, for which it would be hard to derive analytically the expression of the energy flux. We find in this way that a DNN model, trained on meta-GGA (SCAN) data, reduces the deviation from experiment of the predicted thermal conductivity by about 50%, leaving the question open as to whether the residual error is due to deficiencies of the functional, to a neglect of nuclear quantum effects in the atomic dynamics, or, likely, to a combination of the two.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS2014

Current Ground Test Options for Nuclear Thermal Propulsion (NTP)

About 20 different NTP engines/ reactors were tested from 1959 to 1972 as part of the Rover and Nuclear Engine for Rocket Vehicle Application (NERVA) program. Most were tested in open air at test cell A or test cell C, at the Nevada Test Site (NTS). Even after serious engine breakdowns of the reactor (e.g., Phoebus 1A), the test cells were cleaned up for other engine tests. The engine test stand (ETS) was made for high altitude (approximately 1 psia) testing of an NTP engine with a flight configuration, but still had the exhaust released to open air. The Rover/NERVA program became aware of new environmental regulations which would prohibit the release of any significant quantity of radioactive particulates and noble gases into the open air. The nuclear furnace (NF-1) was the last reactor tested before the program was cancelled in 1973, but successfully demonstrated a scrubber concept on how to filter the NTP exhaust. The NF-1 was demonstrated in the summer of 1972. The NF-1 used a 44MW reactor and operated each run for approximately 90 minutes. The system cooled the hot hydrogen exhaust from the engine with a water spray before entering a particle filter. The exhaust then passed through a series of heat exchangers and water separators to help remove water from the exhaust and further reduce the exhaust temperatures. The exhaust was next prepared for the charcoal trap by passing through a dryer and effluent cooler to bring exhaust temperatures close to liquid nitrogen. At those low temperatures, most of the noble gases (e.g., Xe and Kr made from fission products) get captured in the charcoal trap. The filtered hydrogen is finally passed through a flare stack and released to the air. The concept was overall successful but did show a La plating on some surfaces and had multiple recommendations for improvement. The most recent detailed study on the NTP scrubber concept was performed by the ARES Corporation in 2006. The concept is based on a 50,000 lbf thrust engine (approximately 1 GW) with a maximum burn time of 1 hour. The concept utilized lessons learned from NF-1. The strategy breaks down the exhaust into parallel paths to allow flexibility with engine size and mass flow of exhaust. Similar to NF-1, the exhaust is slowed down, cooled, filtered of particulates, filtered of noble gases, and then the clean hydrogen is flared to open air. Another concept proposed by Steve Howe (currently Director of the Center for Space Nuclear Research) to simplify the NTP exhaust filtering is to run the hydrogen exhaust into boreholes underground to filter the exhaust. The two borehole site locations proposed are at the NTS and at the Idaho National Laboratory (INL). At NTS, the boreholes are 8' diameter and 1200' deep. The permeability of hydrogen through the soil and its buoyancy will allow it to rise up through the soil and allow the filtering of noble gases and radioactive particulates. The exhaust needs to be cooled to 600C before entering the borehole to avoid soil glazing. Preliminary analysis shows a small buildup of back pressure with time which depends on permeability. Noble gases entering the borehole walls deep can take a long time before reaching the surface. Other factors affecting permeability include borehole pressure, water saturation, and turbulence. Also, a possible need to pump out contaminated water collected at the bottom of the borehole. At INL, the borehole concept is slightly different. The underground borehole has openings to the soil at special depths which have impermeable interbeds above the water table and below the surface to allow the exhaust to travel horizontal between the impermeable layers. Preliminary results indicate better permeability than at NTS. The last option is total containment of the exhaust during the test run. The concept involves slowing down the flow to subsonic in a water cooled diffuser. The hydrogen is burned off in an oxygen rich afterburner with the only products being steam, oxygen, and some noble gases. A heat exchanger and water spray pulls heat from the steam and lowers the temperature for condensation. The optimum ratio between the two is being investigated, with a goal to minimize the total volume of the water hold tanks. A water tank farm collects the contaminated water. The amount of water produced from burning the hydrogen is approximately 100,000 gallons (not including cooling water) for a 25k lbf engine operating for 50 minutes. Residual gases (e.g., oxygen and some noble gases) can be captured at cryogenic levels with a liquid nitrogen cooled dewar. After a few weeks post-test, the radiation levels can drop to more favorable levels before slowly draining each capture tank and using existing filters. With today's environmental regulations, the NTP exhaust is filtered to meet 10 mrem/year exposure to the general public (at a DOE site) or 100 mrem/year (via NRC when tested elsewhere), when natural background radiation exposure to the general public is 300- 600 mrem per year. The current society feels more comfortable with filtering even lower to as low as reasonably achievable (ALARA).

Gerrish, Harold P., Jr.↗
NASA NTRS2012

Next Generation Microbiology Requirements

As humans continue to explore deep into space, microorganisms will travel with them. The primary means to mitigate the risk of infectious disease are a combination of prudent spacecraft design and rigorous operational controls. The effectiveness of these methods are evaluated by microbiological monitoring of spacecraft, food, water, and the crew that is performed preflight, in-flight, and post-flight. Current NASA requirements associated with microbiological monitoring are based on culture-based methodology where microorganisms are grown on a semi-solid growth medium and enumerated. Subsequent identification of the organisms requires specialized labor and large equipment, which historically has been performed on Earth. Requirements that rely strictly on culture-based units limit the use of non-culture based monitoring technology. Specifically, the culture-based "measurement criteria" are Colony Forming Units (CFU, representing the growth of one microorganism at a single location on the agar medium) per a given volume, area, or sample size. As the CFU unit by definition is culture-based, these requirements limit alternative technologies for spaceflight applications. As spaceflight missions such as those to Mars extend further into space, culture-based technology will become difficult to implement due to the (a) limited shelf life of the culture media, (b) mass/volume necessary to carry these consumables, and (c) problems associated with the production of biohazardous material in the habitable volume of the spacecraft. In addition, an extensive amount of new knowledge has been obtained during the Space Shuttle, NASA-Mir, and International Space Station Programs, which gave direction for new or modified microbial control requirements for vehicle design and mission operations. The goal of this task is to develop and recommend a new set of requirements for vehicle design and mission operations, including microbiological monitoring, based upon "lessons learned" and new technology. During 2011, this study focused on evaluating potable water requirements by assembling a forum of internal and external experts from NASA, other federal agencies, and academia. Key findings from this forum included: (1) Preventive design and operational strategies should be stringent and the primary focus of NASA's mitigation efforts, as they are cost effective and can be attained with conventional technology. (2) Microbial monitoring hardware should be simple and must be able to measure the viability of microorganisms in a sample. Multiple monitoring technologies can be utilized as long as at the microorganisms being identified can also be confirmed as viable. (3) Evidence showing alterations in the crew immune function and microbial virulence complicates risk assessments and creates the need for very conservative requirements. (4) One key source of infectious agents will always be the crew, and appropriate preventative measures should be taken preflight. (5) Water systems should be thoroughly disinfected (sterilized if possible) preflight and retain a residual biocide throughout the mission. Future forums will cover requirements for other types of samples, specifically spaceflight food and environmental samples, such as vehicle air and vehicle and cargo surfaces. An interim report on the potable water forum has been delivered to the Human Research Program with a final report on the recommendations for all sample types being delivered in September 2013.

Ott, C. M.↗