Search NASA⌕ Search

SEARCH · Search NASA

Results for “AI”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,117 records · Page 62

DNS of ignition and flame stabilization in a simplified gas turbine premixer

With the increasing need for fuel flexibility, mitigation of auto-ignition (AI) inside gas turbine (GT) premixers becomes crucial. They must be designed to yield a sufficiently homogeneous fuel-air mixture to achieve low emissions while at the same time avoiding the occurrence of AI and subsequent flame stabilization. This challenge requires a detailed understanding of turbulent mixing and chemistry interactions. In the present work, a direct numerical simulation (DNS) of an array of jets in crossflow (JICF), representative of an industrial GT premixer, is reported to shed light on these complex phenomena. It is found that AI kernels form in the aft part of the premixer and coalesce into a flame front that then propagates upstream, mainly through the boundary layer, and successively engulfs the jets. This, therefore, suggests a significant role of the jet array pattern on the flame stabilization. It is noted that AI kernels continue to form independently during the whole time of the simulation. To clarify the contribution of AI and diffusion in the ignition kernels and the main flame, chemical explosive mode analysis (CEMA) is employed jointly with a kernel tracking algorithm. It is found that during the initial formation of the flame, many ignition kernels form in mixtures with low scalar dissipation rate and large contribution from AI mode. As they quickly grow, they merge into a single flame front that becomes increasingly more diffusion-assisted over time, balancing the AI mode. Turbulence is shown to have a significant enhancing effect in lean premixed flames, but further analysis is required to fully characterize it. These findings are relevant for the industrial premixer studied, and also for novel micromix concepts that may be used in the next generation of GT combustion systems.

ADVANCED PROPULSION SYSTEMS↗

Virus-like particles displaying H5, H7, H9 hemagglutinins and N1 neuraminidase elicit protective immunity to heterologous avian influenza viruses in chickens

Avian influenza (AI) viruses circulating in wild birds pose a serious threat to public health. Human and veterinary vaccines against AI subtypes are needed. Here we prepared triple-subtype VLPs that co-localized H5, H7 and H9 antigens derived from H5N1, H7N3 and H9N2 viruses. VLPs also contained influenza N1 neuraminidase and retroviral gag protein. The H5/H7/H9/N1/gag VLPs were prepared using baculovirus expression. Biochemical, functional and antigenic characteristics were determined including hemagglutination and neuraminidase enzyme activities. VLPs were further evaluated in a chicken AI challenge model for safety, immunogenicity and protective efficacy against heterologous AI viruses including H5N2, H7N3 and H9N2 subtypes. All vaccinated birds survived challenges with H5N2 and H7N3 highly pathogenic AI (HPAI) viruses, while all controls died. Immune response was also detectable after challenge with low pathogenicity AI (LPAI) H9N2 virus suggesting that H5/H7/H9/N1/gag VLPs represent a promising approach for the development of broadly protective AI vaccine. - Highlights: •VLPs were prepared that co-localized H5, H7 and H9 subtypes in a VLP envelope. •VLPs were characterized including electron microscopy, HA assay and NA enzyme activity. •Experimental VLP vaccine was evaluated in an avian influenza challenge model. •VLPs induced immune responses against heterologous H5, H7 and H9 virus challenges.

60 APPLIED LIFE SCIENCES↗

Modeling of the phase evolution in Mg1-xAlxB2 (0<x<0.5) and its experimental signatures

Despite the chemical and structural simplicity of MgB{sub 2}, at 39 K this compound has the highest known {Tc} of any binary compound. Electron doping by substituting Al for Mg leads to decreasing Tc and the observed concentration dependent rate of decrease has been proposed to arise from the non-ideal character of MgB{sub 2}-AIB{sub 2} solid solutions, which derives from the existence of an ordered Mg{sub 0.5}Al{sub 0.5}B{sub 2} compound. Heterogeneous nano-scale structure patterns in solid solutions have emerged as an important concept for complex materials, ranging from actinide alloys and oxides to high-temperature cuprate superconductors and mallganite-based materials exhibiting colossal magnetoresistivity. In this work we investigate the formation of structural heterogeneities in Mg{sub 1-x}AI{sub x}B{sub 2}, which take the form of nano-scale AI-AI and AI-Mg domains of different geometry and size, using molecular statics/dynamics simulations and in particular we study the corresponding signatures in diffraction experiments. In order to undertake this task we first derive appropriate Mg-AI-B semi-empirical potentials within the Modified Embedded Atom Method formalism. These potentials are also applied to explore the equilibrium Mg{sub 1-x}AI{sub x}B{sub 2} phase diagram for 0 < x < 0.5. Additionally, density functional theory calculations were utilized to study the influence of heterogeneities on the electronic structure and charge distribution in Mg{sub 1-x}AI{sub x}B{sub 2}.

36 MATERIALS SCIENCE↗

TU-G-BRD-03: IMRT Dosimetry Differences in An Institution with Community and Academic Model

Purpose: Radiation outcome among institutions can be interpreted meaningfully if the dose delivery and prescription to the target volume is documented accurately and consistently. ICRU-83 recommended specific guidelines in IMRT for target volume definitions and dose reporting. This retrospective study evaluates the pattern of IMRT dose prescription and recording in an academic institution (AI) and a community hospital (CH) models in a single institution with reference to ICRU-83 recommendation. Materials & Methods: Dosimetric information of 625 (500 from academic and 125 from community) patients treated with IMRT was collected retrospectively from the AI and a CH. The dose-volume histogram (DVH) for the target volume of each patient was extracted. Standard dose parameters such as D2, D50, D95, D98, D100, as well as the homogeneity index (HI) defined as (D2-D98)/D50 and monitor units (MUs) were collected. Results: Significant dosimetric variations were observed in disease sites and between AI and CH. The variation in the mean value of D95 for AI is 98.48±4.12 and for CH is 96.41±4.13. A similar pattern was noticed for D50 (104.18±6.04 for AI and 101.05±3.49 for CH). Thus, nearly 95% of patients received dosage higher than 100% to the site viewed by D50 and varied between AI and CH models. The average variation of HI is found to be 0.12±0.08 and 0.11±0.08 for AI and CH model, showing better IMRT treatment plans for academic model compared to community. Conclusion: Even with the implementation of ICRU-83 guidelines, there is a large variation in dose prescription and delivery in IMRT. The variation is institution and site specific. For any meaningful comparison of the IMRT outcome, strict guidelines for dose reporting should be maintained in every institution.

60 APPLIED LIFE SCIENCES↗

Characterization of Extreme Hydroclimate Events in Earth System Models using ML/AI

Focal Area(s): (1) We put forward the concepts of data assimilation enabled by machine learning, AI, and advanced methods including experimental/network design/optimization and unsupervised learning applied to downscale information within Earth System Models (ESMs). (2) We discuss predictive modeling through the use of AI techniques and other tools to design a prediction system comprising of a hierarchy of models (e.g., AI-driven model/component/parameterization selection) to improve the characterization of extreme hydroclimate events in ESMs. The AI-based models will run five-six order of magnitude times faster, yet will provide similar accuracy, allowing us to provide range bounds on uncertainty faster and thus enabling faster extreme event identification. (3) Further, we consider the insight gleaned from complex data (observed/simulated) using AI, big data analytics, and other advanced methods, including explainable AI and physics- or knowledge-guided AI to improve the characterization of extreme hydroclimate events in ESMs.

54 ENVIRONMENTAL SCIENCES↗

Representing the Unrepresented Impact of River Ice on Hydrology, Biogeochemistry, Vegetation, and Geomorphology: A Hybrid Physics-Machine Learning Approach

Focal areas include: Predictive modeling through the use of AI techniques and AI-derived model components; the use of AI and other tools to design a prediction system composed of a hierarchy of models (e.g., AI driven model/component/parameterization selection). Insight gleaned from complex data (both observed and simulated) using AI, big data analytics, and other advanced methods, including explainable AI and physics- or knowledge-guided AI.

54 ENVIRONMENTAL SCIENCES↗

Artificial intelligence applications in space and SDI: A survey

The purpose of this paper is to survey existing and planned Artificial Intelligence (AI) applications to show that they are sufficiently advanced for 32 percent of all space applications and SDI (Space Defense Initiative) software to be AI-based software. To best define the needs that AI can fill in space and SDI programs, this paper enumerates primary areas of research and lists generic application areas. Current and planned NASA and military space projects in AI will be reviewed. This review will be largely in the selected area of expert systems. Finally, direct applications of AI to SDI will be treated. The conclusion covers the importance of AI to space and SDI applications, and conversely, their importance to AI.

Fiala, Harvey E.↗

A new activity index for comets

An activity index (AI) is derived from observational data to measure the increase of activity in magnitudes for comets when brightest near perihelion as compared to their inactive reflective brightnesses at great solar distances. In all, AI determinations are made for 32 long-period (L-P) comets and for 14 short-period (S-P). The range of values of AI is the order of 3 to 10 magnitudes with a median about 6. An expected strong correlation with perihelion distance, q, was found to vary as approx. q(-2.3). Residuals from a least-square solution, delta AI, were used for comparing comets of different orbital classes, the standard deviation of a single value of delta AI was only plus or minus 1(m).1 for L-P comets and plus or minus 1(m)2 for S-P comets. Among the L-P comets, 19 of Period, P greater than 10(exp 4) yr yielded delta AI greater than or equal to -0(m).27 compared to +0(m).39 for 13 of 10(exp 4) greater than P greater than 10(exp 2) yr. This denies any effect of aging among the L-P comets. The 14 S-P comets yielded (delta AI) less 0(m).3 plus or minus 0(m).3 than the 32 L-P comets. The results suggest a common actively level, nature and probably origin for comets of all orbits, sizes and ages.

Whipple, Fred L.↗

Harnessing Artificial Intelligence for Medical Diagnosis and Treatment During Space Exploration Missions

BACKGROUND The medical capabilities necessary for long-duration exploration missions (LDEMs) will differ tremendously from those currently available to crew medical officers (CMOs) on the International Space Station (ISS). Ground support will be more challenging due to distance-related communication delays and data transmission, and resource utilization must be optimized given limited ability for resupply. Clinical decision support systems (CDSSs) can help mitigate these limitations. The recent launch of generative artificial intelligence (AI) tools based upon large language models (LLM) support the creation of a smart assistant for onboard triage, diagnosis, and guided treatment of medical conditions during these missions. The Informing Mission Planning via Analysis of Complex Tradespaces (IMPACT) tool can help predict which clinical problems and outcomes are likely to occur for a design reference mission (DRM) and assist Medical Operations and systems engineering teams in creating a medical system that may optimally mitigate the predicted risks. The purpose of this study was to identify AI tools currently available or in development for the assistive diagnosis and care of medical conditions predicted for an extended duration Lunar mission. METHODS The 119 medical conditions currently built into the IMPACT suite were categorized into systems, and these diagnoses were used as keywords for our literature search. Using PubMed and Google Scholar, we performed a literature survey of AI tools applicable to these conditions. Article inclusion criteria included publication between the years 2017-2023, as the sentinel paper discussing the “selective attention” driving ChatGPT and other generative transformer models was published in June 2017. Where applicable, we reviewed only the top 1000 research articles (based on relevance) for each of the keywords/phrases. AI tools whose training sets were exclusive to a pediatric patient population were excluded. We also excluded any medical diagnostic tools (such as CT, MRI, mass spectrometry) or procedures (such as endoscopy, surgery) that are unlikely to be available during LDEMs due to mass and volume constraints, CMO knowledge, skills, and abilities, and/or inherent procedural risks. RESULTS Our survey highlighted several AI-driven tools for the triage, diagnosis, and management of those medical conditions highlighted by IMPACT. Selected publications for each medical condition were then screened for inclusion within ten systems-based categories including: general diagnostic tools (25), tools to diagnose and manage respiratory (40), dermatologic (34), neurologic (28), auditory and vestibular (30), ophthalmic (34), musculoskeletal (104), infection-associated (92), and gynecologic (19) conditions, as well as tools that could be deployed in the setting of trauma and emergency (34). CONCLUSIONS Numerous AI-driven tools were highlighted within this literature survey, ranging from chatbot assistants that triage knee pain to vision transformer models for diagnosis of ophthalmic conditions using ocular surface images captured with a mobile phone. Remaining challenges include optimizing connectivity and integration of existing and developing systems into the vehicles or habitats. Notably, findings from this survey could help guide the initial design of an all-encompassing, onboard medical AI assistant for use during future LDEMs.

R A Lacinski↗

Artificial Intelligence Application to D and D - 20492

As aging facilities across the DOE complex await decommissioning, there is an ongoing need to understand any changes in the structural conditions. Many of these facilities were built over 50 years ago and, in some cases, these facilities have gone beyond the expected operational lifetime. Many facilities have been placed in a state of 'cold and dark,' sitting unused and awaiting decommissioning. Especially challenging are the aging facilities that provide unique operational/production capabilities to support critical DOE missions and cannot be shut down. In any of these scenarios, the structural integrity of these facilities may become compromised as time passes. It is critical that adequate inspections be performed on a continual basis and that the data collected undergoes sufficient analysis to support timely identification of any new or worsening structural issues as well as prompt needed maintenance and repairs to maintain the facilities in a safe condition. In recent days, Artificial Intelligence (AI) [1] and its application to various domains are growing at fast speed. FIU is performing research in this area and exploring the associated technologies to solve nuclear decommissioning problems. Artificial intelligence refers to the capability of a program to autonomously act, react and adapt to the working environment. AI enables the machine to behave like humans and perform the cognitive functions such as 'learning' and 'problem solving'. AI systems gradually moving from traditional approaches (algorithms and expert systems) towards more efficient and advanced technologies (machine learning [1] and deep learning [2] [3]). AI is the study of algorithms and statistical models that is being used by computers to perform specific tasks without using explicit instructions. FIU is working to develop a pilot-scale infrastructure to implement structural health monitoring using AI technologies with focus on machine learning, deep learning. This research is focused on Computer Vision/Image Classification area of AI applications. This can also be expanded to other areas of AI related to Object Recognition and Character Recognition in images. In addition to utilizing existing data sets, FIU will collect and investigate image and video data using FIU test-bed mockups to monitor structural health of the facility. Resulting data will be processed and analyzed using machine learning/deep learning technologies. The proposed pilot system is intended to serve as a starting point to engage the DOE field sites on related data sets and their decision making needs. It is anticipated that proposed machine learning/deep learning technologies can be effectively employed using anomaly detection to solve EM challenges in surveillance and maintenance of the D and D facilities. FIU will work with research stakeholders to identify applications at various sites and other DOE facilities. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Leveraging Artificial Intelligence in Federal Projects

Artificial intelligence (AI) has the potential to transform grid operations. As energy demand rises, weather patterns shift, and foreign threats to critical infrastructure grow, it is essential to harness advanced technology to modernize the grid and increase overall resiliency. This paper examines the integration of AI in federally funded grid infrastructure projects. By analyzing project data, it evaluates the penetration and use cases of AI technologies within key funding initiatives and explores opportunities and challenges associated with their deployment. This analysis categorizes AI adoption in recent federal grid investments, establishing a baseline for measuring near-term impacts and identifying promising AI applications. The research found that 16% of recent federal energy projects included AI integration, and approximately 75% of these projects used AI for more than one primary application.

24 - POWER TRANSMISSION AND DISTRIBUTION↗

Personnel Dose Assessment during Active Interrogation

A leading candidate in the detection of special nuclear material (SNM) is active interrogation (AI). Unlike passive interrogation, AI uses a source to enhance or create a detectable signal from SNM (usually fission), particularly in shielded scenarios or scenarios where the SNM has a low activity. The use of AI thus makes the detection of SNM easier or, in some scenarios, even enables previously impossible detection. During the development of AI sources, significant effort is put into determining the source strength required to detect SNM in specific scenarios. Usually during this process, but not always, an evaluation of personnel dose is also completed. In this instance personnel dose could involve any of the following: (1) personnel performing the AI; (2) unknown stowaways who are inside the object being interrogated; or (3) in clandestine interrogations, personnel who are known to be inside the object being interrogated but are unaware of the interrogation. In most instances, dose to anyone found smuggling SNM will be a secondary issue. However, for the organizations performing the AI, legal if not moral considerations should make dose to the personnel performing the AI, unknown stowaways, or innocent bystanders in clandestine interrogations a serious concern.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A Systems-Level Approach to Address Risks and Ethics in Artificial Intelligence Systems

Artificial intelligence (AI) is rapidly changing the world, from completely controlling routine or mundane tasks like text and image generation, to powering advanced algorithms that control critical systems. The recent advances in generative AI quickly overwhelmed multiple industries from education to finance as first adopters rushed (and continue to rush) to take advantage of the technology. The expanding AI ecosystem presents novel risks and ethical challenges that must be handled to ensure that technology is leveraged fairly and ethically. There are intertwined risks and ethical challenges stemming from the stochastic nature of AI (i.e., intrinsic risks), as well as from specific applications (i.e., extrinsic risks). Appropriately regulating AI requires a systems-approach to develop an integrated solution to these dependent challenges. Thus far, however, questions of risk, ethics and regulation appear to occupy separate spaces. This paper reviews the risks and ethical implications of AI and proposes a system-level approach to integrating ethics and regulation for the nascent industry.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

A primer on artificial intelligence in plant digital phenomics: embarking on the data to insights journey

Artificial intelligence (AI) has emerged as a fundamental component of global agricultural research that is poised to impact on many aspects of plant science. In digital phenomics, AI is capable of learning intricate structure and patterns in large datasets. We provide a perspective and primer on AI applications to phenome research. We propose a novel human-centric explainable AI (X-AI) system architecture consisting of data architecture, technology infrastructure, and AI architecture design. We clarify the difference between post hoc models and 'interpretable by design' models. We include guidance for effectively using an interpretable by design model in phenomic analysis. We also provide directions to sources of tools and resources for making data analytics increasingly accessible. In conclusion, this primer is accompanied by an interactive online tutorial.

60 APPLIED LIFE SCIENCES↗

Vulnerabilities in Artificial Intelligence and Machine Learning Applications and Data

Artificial intelligence (AI) applications driven by machine learning (ML) are transformational technologies within the international nuclear security regime. Advancements realized by AI—faster and improved data insights, more efficient and automated processes, reductions in human error—enable nuclear security applications such as behavior analysis for insider threat mitigation, source tracking of stolen nuclear material, and facial recognition software for physical protection. In addition to the advantages, however, there are also inherent vulnerabilities and threats associated with its use and risk mitigations must be built into any AI/ML-enabled systems. This work provides a background on AI and ML and different data types used in the field, including open-source intelligence information (OSINT) that is discoverable by AI tools and application data that are used by AI tools for decision-making and automation. Current and potential AI applications and vulnerabilities related to their use within the nuclear security regime are also discussed.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Adversarial Artificial Intelligence: State of the Malpractice

Warfare systems are increasingly being supported by artificial intelligence (AI), and its influence is growing. As AI becomes a warfare enabler it has also become a target. For example, AI-enabled machine vision systems can be trivially deceived by specially formulated camouflage. Many approaches to making AI more resilient to protect against specific attacks fail to generalize. However, attacks often do generalize across models trained for the same task, regardless of architecture or training data. This paper will illustrate the state of the practice in offensive and defensive AI and clarify some myths about AI's triumphs and failures in the real world.

Artificial intelligence, AI, cybersecurity, milita↗

Simulated responses of terrestrial aridity to black carbon and sulfate aerosols

Aridity index (AI), defined as the ratio of precipitation to potential evapotranspiration (PET), is a measure of the dryness of terrestrial climate. Global climate models generally project future decreases of AI (drying) associated with global warming scenarios driven by increasing greenhouse gas and declining aerosols. Given their different effects in the climate system, scattering and absorbing aerosols may affect AI differently. In this work, we explore the terrestrial aridity responses to anthropogenic black carbon (BC) and sulfate (SO 4 ) aerosols with Community Earth System Model simulations. Positive BC radiative forcing decreases precipitation averaged over global land at a rate of 0.9%/°C of global mean surface temperature increase (moderate drying), while BC radiative forcing increases PET by 1.0%/°C (also drying). BC leads to a global decrease of 1.9%/°C in AI (drying). SO 4 forcing is negative and causes precipitation a decrease at a rate of 6.7%/°C cooling (strong drying). PET also decreases in response to SO 4 aerosol cooling by 6.3%/°C cooling (contributing to moistening). Thus, SO 4 cooling leads to a small decrease in AI (drying) by 0.4%/°C cooling. Despite the opposite effects on global mean temperature, BC and SO 4 both contribute to the twentieth century drying (AI decrease). Sensitivity test indicates that surface temperature and surface available energy changes dominate BC- and SO 4 -induced PET changes.

54 ENVIRONMENTAL SCIENCES↗

The Geothermal Artificial Intelligence for geothermal exploration

Exploration of geothermal resources involves analysis and management of a large number of uncertainties, which makes investment and operations decisions challenging. Remote Sensing (RS), Machine Learning (ML) and Artificial Intelligence (AI) have potential in managing the challenges of geothermal exploration. In this paper, we present a methodology that integrates RS, ML and AI to create an initial assessment of geothermal potential, by resorting to known indicators of geothermal areas namely mineral markers, surface temperature, faults and deformation. We demonstrated the implementation of the method in two sites (Brady and Desert Peak geothermal sites) that are close to each other but have different characteristics (Brady having clear surface manifestations and Desert Peak being a blind site). Here, we processed various satellite images and geospatial data for mineral markers, temperature, faults and deformation and then implemented ML methods to obtain pattern of surface manifestation of geothermal sites. We developed an AI that uses patterns from surface manifestations to predict geothermal potential of each pixel. We tested the Geothermal AI using independent data sets obtaining accuracy of 92-95%; also tested the Geothermal AI trained on one site by executing it for the other site to predict the geothermal / non-geothermal delineation, the Geothermal AI performed quite well in prediction with 72-76% accuracy.

15 GEOTHERMAL ENERGY↗