SEARCH · Engineering Papers
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.
The goal of this mini report is to give opinions on the current state of the art in large language model (LLM) assisted code translation, which aims to replicate the behavior of a program written in one programming language to another. The first section provides our overall sense of literature and suggested directions of research, and the second section summarizes a few of the most relevant papers in greater depth.
This report reviews regulations, codes, and standards to be considered for FCEB operations on airports governing vehicle safety, hydrogen infrastructure, and airport operations. Overall, the review found the existing regulatory framework largely suitable for FCEB operations on airports, with only minor gaps identified pertaining to heavy-duty vehicle fueling, operation and fire safety on airport aprons, and authorization for emergency power use.
Artificial intelligence (AI) has quickly become an important tool in scientific research, where significant efforts are underway to develop tools that will expedite the research process. One area of particular impact is scientific software, which can be particularly complex, and therefore time consuming to learn and use effectively. AI assistants are increasingly helping to streamline the process by performing tasks such as interactively answering user questions or suggesting solutions. Los Alamos National Laboratory (LANL) develops several advanced scientific codes, such as FLAG, which can be used to run multiphysics simulations. With this study, our goal was to develop an AI assistant for FLAG that could help make the process of understanding the software and running physics simulations more efficient. To develop an AI assistant for FLAG, we used a method called retrieval-augmented generation (RAG), which is a technique that uses information from relevant data sources to enhance the accuracy of large language models (LLMs). We used the FLAG user manual and other FLAG documentation as the knowledge base for the RAG system. When a user provides a query, RAG retrieves relevant sections from the knowledge base in response, then uses those excerpts to generate grounded and contextually rich answers. We found that our AI assistant was able to provide context aware answers and source references to user queries. To evaluate performance, we developed a set of 40 benchmark questions and compared the accuracy of the responses to those of two standard LLMs without retrieval. Our AI assistant significantly outperformed the standard LLMs at answering FLAG-related questions, with an 82.5% accuracy rate, compared to 47.5% for both of the standard LLMs. This has the potential to make the process of learning and using FLAG much easier, especially for new users. Ultimately, it supports LANL’s broader mission by empowering scientists and engineers to focus more on discovery and analysis rather than on navigating complex software systems.
This report demonstrates calculations of equilibrium isotope fractionation with thermochemical code magpie developed and maintained under the ASC-PEM-HE program at LANL. All the necessary background is provided, and the results of our calculations are compared with literature data.
The ENDF/B-VIII.1 evaluation library has seen a great growth in the thermal neutron scattering sub-library. The SCALE code system has traditionally approached CE transport by assuming that the CE library on disk represented the fully expanded cumulative probability distributions, conditional on exiting angle and marginal on exiting energy. While this is a complete description of the data, it comes at the potential cost of large amounts of on-disk storage. This approach was strained by several TSL files in ENDF/B-VIII.1, such as graphite, which contained data for a large number of Bragg edges. In the fully expanded probability distributions, this was found to be a disproportionately large fraction of the SCALE CE library.
BISON is a finite element-based nuclear fuel performance code applicable to a variety of fuel forms including light water reactor fuel rods, TRISO particle fuel, and metallic rod and plate fuel. It is a multiphysics fuel analysis tool that solves fully-coupled thermomechanical problems. BISON is based on MOOSE and can efficiently solve problems using standard workstations or very large high-performance computers in a variety of different dimensions, including full 3D, 2D-RZ axisymmetric, layered axisymmetric 1D, and spherically symmetric 1D systems. It is developed by a team of scientists and engineers at Idaho National Laboratory and by collaborators. The development of BISON is supported by various funding agencies, principally the United States Department of Energy.
HSSHTC, the safety basis steady state TH code for HFIR, uses a highly conservative approach in which all input and calculation uncertainties are resolved simultaneously at their most limiting setting. This results in excessive conservatism which does not account for the high unlikelihood of such simultaneous worst-case conditions. The present study explores an alternative approach, BEPU, in which reasonable working assumptions for the probability distribution of each input uncertainty are used to determine a relationship between burnout power margin and core fuel failure probability. This was performed under a philosophy of perturbing uncertainty parameters already defined within the HSSHTC methodology while preserving the HSSHTC calculation approach and solution methodology itself. Based on the assumptions employed in this study, the BEPU approach resulted in a 0.29 increase in burnout power ratio (25 MW increase in burnout power) compared to the latest HSSHTC calculations of C-HFIR-2026-004. The study can be refined in the future by employing fuel fabrication data to provide more realistic input distributions. Future changes to the HSSHTC methodology would potentially allow a more comprehensive treatment of uncertainties which may further increase the burnout power ratio.
[ZP] Self-assembling DNA crystal with expanded genetic code using C,A,T,G, Z and P nucleotides
There has been an increasing need for forecasting power generation at the sub-seasonal to seasonal (S2S) timescales to support the operation, management, and planning of the wind-energy system. At the S2S timescales, atmospheric variability is largely related to recurrent and persistent weather patterns, referred to as weather regimes (WRs). In the study "How do the weather regimes drive wind speed and power production at the sub-seasonal to seasonal timescales over the CONUS?", we identify four WRs that influence wind resources over North America using a self-organizing map (SOM) algorithm. These WRs are responsible for large-scale wind and power production anomalies over the CONUS at the S2S timescales. The WR-based reconstruction explains up to 50% of the monthly variance of power production over the western United States, and the explanatory power generally increases with the increase of timescales. The identified relationship between WRs and power production reveals the potential and limitations of the regional WR-based wind resource assessment over different regions of the CONUS across multiple timescales. This repository includes all the data and codes we use for analyses in this study. Users may use them to reproduce the results of this study on their end.
This repository contains code for the experiments in the paper "Understanding Generative AI Content with Embedding Models". Constructing high-quality features is critical to any quantitative data analysis. While feature engineering was historically addressed by carefully hand-crafting data representations based on domain expertise, deep neural networks (DNNs) now offer a radically different approach. DNNs implicitly engineer features by transforming their input data into hidden feature vectors called embeddings. For embedding vectors produced by foundation models -- which are trained to be useful across many contexts -- we demonstrate that simple and well-studied dimensionality-reduction techniques such as Principal Component Analysis uncover inherent heterogeneity in input data concordant with human-understandable explanations. Of the many applications for this framework, we find empirical evidence that there is intrinsic separability between real samples and those generated by artificial intelligence (AI).
Motivation: Nutrient and contaminant behavior in the subsurface are governed by multiple coupled hydrobiogeochemical processes which occur across different temporal and spatial scales. Accurate description of macroscopic system behavior requires accounting for the effects of microscopic and especially microbial processes. Microbial processes mediate precipitation and dissolution and change aqueous geochemistry, all of which impacts macroscopic system behavior. As `omics data describing microbial processes is increasingly affordable and available, novel methods for using this data quickly and effectively for improved ecosystem models are needed. Results: We propose a workflow (`Omics to Reactive Transport – ORT) for utilizing metagenomic and environmental data to describe the effect of microbiological processes in macroscopic reactive transport models. This workflow utilizes and couples two open-source software packages: KBase (a software platform for systems biology) and PFLOTRAN (a reactive transport modeling code). We describe the architecture of ORT and demonstrate an implementation using metagenomic and geochemical data from a river system. Our demonstration uses microbiological drivers of nitrification and denitrification to predict nitrogen cycling patterns which agree with those provided with generalized stoichiometries. While our example uses data from a single measurement, our workflow can be applied to spatiotemporal metagenomic datasets to allow for iterative coupling between KBASE and PFLOTRAN. Live, interactive models, which incorporate the results from this narrative into a PFLOTRAN simulation, are available (without login) at https://pflotranmodeling.paf.subsurfaceinsights.com/pflotran-simple-model/.
In the future, it will be possible to build high-quality models of building interiors based on data from a dense fleet of sensors reporting on air volumes much smaller than a room or zone. To enable such models, we are creating technologies that allow a fleet of sensors to be commissioned quickly at low cost. Our sensor commissioning process builds a 3D model of each building interior that includes sensor positions and sensor networking information such as sensor MAC addresses. It employs multiple technologies, including augmented reality, LiDAR, QR codes, and computer vision. Sensors can be commissioned at more than 10x the speed and at less than one tenth the cost of traditional approaches.
The High Energy Physics (HEP) experiments, such as those at the Large Hadron Collider (LHC), traditionally consume large amounts of CPU cycles for detector simulations and data analysis, but rarely use compute accelerators such as GPUs. As the LHC is upgraded to allow for higher luminosity, resulting in much higher data rates, purely relying on CPUs may not provide enough computing power to support the simulation and data analysis needs. As a proof of concept, we investigate the feasibility of porting a HEP parameterized calorimeter simulation code to GPUs. We have chosen to use FastCaloSim, the ATLAS fast parametrized calorimeter simulation. While FastCaloSim is sufficiently fast such that it does not impose a bottleneck in detector simulations overall, significant speed-ups in the processing of large samples can be achieved from GPU parallelization at both the particle (intra-event) and event levels; this is especially beneficial in conditions expected at the high-luminosity LHC, where extremely high per-event particle multiplicities will result from the many simultaneous proton-proton collisions. We report our experience with porting FastCaloSim to NVIDIA GPUs using CUDA. A preliminary Kokkos implementation of FastCaloSim for portability to other parallel architectures is also described.
During spaceflight, astronauts are exposed to multiple unique environmental factors, particularly microgravity and ionizing radiation, that can cause a range of harmful health consequences. Over the past decades, increasing evidence demonstrates that the space environment can induce changes in gene expression and RNA processing. Long non-coding RNA (lncRNA) represent an emerging area of focus in molecular biology as they modulate chromatin structure and function, the transcription of neighboring genes, and affect RNA splicing, stability, and translation. They have been implicated in cancer development and associated with diverse cardiovascular conditions and associated risk factors. However, their role on astronauts’ health after spaceflight remains poorly understood. In this perspective article, we provide new insights into the potential role of exosomal lncRNA after spaceflight. We analyzed the transcriptional profile of exosomes isolated from peripheral blood plasma of three astronauts who flew on various Shuttle missions between 1998–2001 by RNA-sequencing. Computational analysis of the transcriptome of these exosomes identified 27 differentially expressed lncRNAs with a Log 2 fold change, with molecular, cellular, and clinical implications.
A series of experiments involving the detonation of PBX 9501 encased in a copper cylinder are modeled with the objective of evaluating a proposed set of phenomenological parameters for the Wescott–Stewart–Davis reactive burn model. The numerical analysis is conducted using the Los Alamos continuum mechanics code FLAG. Numerical considerations pertaining to various aspects of modeling the experiments using FLAG are discussed. It is shown that use of the proposed set of phenomenological parameters results in predictions of free-surface velocity that match empirically measured velocities reasonably well.
In this note, I review a recent approach to quantum gravity that “gravitizes” quantum mechanics by emerging geometry and gravity from complex quantum states. Drawing further insights from tensor network toy models in AdS/CFT, I propose that approximate quantum error correction codes, when re-adapted into the aforementioned framework, also have promise in emerging gravity in near-flat geometries.
Single-dose, immunogenic DNA (iDNA) vaccines coding for whole live-attenuated viruses are reviewed. This platform, sometimes called immunization DNA, has been used for vaccine development for flavi- and alphaviruses. An iDNA vaccine uses plasmid DNA to launch live-attenuated virus vaccines in vitro or in vivo. When iDNA is injected into mammalian cells in vitro or in vivo, the RNA genome of an attenuated virus is transcribed, which starts replication of a defined, live-attenuated vaccine virus in cell culture or the cells of a vaccine recipient. In the latter case, an immune response to the live virus vaccine is elicited, which protects against the pathogenic virus. Unlike other nucleic acid vaccines, such as mRNA and standard DNA vaccines, iDNA vaccines elicit protection with a single dose, thus providing major improvement to epidemic preparedness. Still, iDNA vaccines retain the advantages of other nucleic acid vaccines. In summary, the iDNA platform combines the advantages of reverse genetics and DNA immunization with the high immunogenicity of live-attenuated vaccines, resulting in enhanced safety and immunogenicity. This vaccine platform has expanded the field of genetic DNA and RNA vaccines with a novel type of immunogenic DNA vaccines that encode entire live-attenuated viruses.
We present a new Milky Way microlensing simulation code, dubbed PopSyCLE (Population Synthesis for Compact object Lensing Events). PopSyCLE is the first resolved microlensing simulation to include a compact object distribution derived from numerical supernovae explosion models and both astrometric and photometric microlensing effects. We demonstrate the capabilities of PopSyCLE by investigating the optimal way to find black holes (BHs) with microlensing. Candidate BHs have typically been selected from wide-field photometric microlensing surveys, such as OGLE, by selecting events with long Einstein crossing times (tE > 120 days). These events can be selected at closest approach and monitored astrometrically in order to constrain the mass of each lens; PopSyCLE predicts a BH detection rate of ~40% for such a program. We find that the detection rate can be enhanced to ~ 85% by selecting events with both tE > 120 days and a microlensing parallax of πE < 0.08. Unfortunately, such a selection criterion cannot be applied during the event as πE requires both pre- and post-peak photometry. However, historical microlensing events from photometric surveys can be revisited using this new selection criteria in order to statistically constrain the abundance of BHs in the Milky Way. The future WFIRST microlensing survey provides both precise photometry and astrometry and will yield individual masses of O(100 - 1000) black holes, which is at least an order of magnitude more than is possible with individual candidate follow-up with current facilities. The resulting sample of BH masses from WFIRST will begin to constrain the shape of the black hole present-day mass function, BH multiplicity, and BH kick velocity distributions