Search NASA⌕ Search

SEARCH · Search NASA

Results for “task 4”

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 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Knowledge Oriented Graph Unified Transformer (KOGUT) v0.1

KOGUT — Knowledge Oriented Graph Unified Transformer KOGUT implements the Relational Graph Transformer (RelGT) architecture for knowledge graph link prediction in biological domains, with a primary focus on microbial growth media prediction. While the original RelGT (arXiv:2505.10960) targets relational tables, time series, and multi-table databases, KOGUT adapts this architecture for heterogeneous biological knowledge graphs, providing first-in-class AI predictive models for microbial cultivation. Key Adaptations Beyond Original RelGT: - Knowledge Graph Focus: Applied to biological KGs with semantic node types (taxa, chemicals, media, phenotypes, environments) versus generic relational database tables, trained on the KG-Microbe knowledge graph (1.3M entities, 2.9M edges, 24 relation types). - Multimodal Node Encoding: Integrates node labels, categories, descriptions, and synonyms from KG metadata through learned embedding layers—adapting relational column features to graph node attributes with textual semantics. - Extended K-Hop Subgraph Strategy: Optimized neighborhood sampling (3-hop default, configurable up to 200 nodes) tuned for sparse biological networks, building on the original local-global attention framework with biological relation preservation. - Biolink Predicate Preservation: Type-specific transformations for 24 biological edge semantics (occurs_in, consumes, produces, has_phenotype, subclass_of) beyond standard relational foreign keys, enabling multi-relation link prediction. - Inductive Learning Support: Enables zero-shot predictions for novel taxa through feature-based embeddings (temperature, oxygen requirements, gram stain, cell shape), extending the original transductive relational benchmark scope to uncultured microorganisms. CheapSOTA Performance Optimizations (This Distribution): - VQ-EMA Centroid Attention: Vector quantization with exponential moving average for improved global context modeling (+5-10% MRR improvement). - HDF5 Precomputed Data Loading: One-time preprocessing of k-hop subgraphs to eliminate redundant graph traversals (2-5× training speedup). - Distributed Data Parallel Training: Multi-GPU support for scaling to larger knowledge graphs (tested on 4× NVIDIA A100 GPUs at NERSC Perlmutter). - Mixed Precision Training: Automatic mixed precision (AMP) for memory efficiency and faster training. Advantages Over Standard Knowledge Graph Embedding Models: Combines RelGT's proven multi-element tokenization (features, type, hop, structure) with graph-native biological representations, enabling interpretable link prediction across heterogeneous entities that standard embedding models (TransE, RotatE, ComplEx) and table-based transformers cannot directly model. Achieves near-perfect performance on microbial growth media prediction (MRR: 0.9966, Precision@1: 0.9932, Hit@10: 1.0000) while maintaining explainability through attention-based reasoning over biological pathways. Training Data: - KG-Microbe merged knowledge graph: 1,379,337 nodes, 2,960,472 edges - 24 biological relation types including taxonomic hierarchies, metabolic interactions, phenotype associations, and environmental relationships - Primary prediction task: Growth media suitability for microbial taxa (biolink:occurs_in, 50K edges) - Multi-relation capability: Predicts links for any of the 24 relation types, including chemical consumption/production, phenotype associations, and taxonomic classification Citation: Original RelGT Architecture: Dwivedi et al., "Relational Graph Transformer", arXiv:2505.10960, 2025 KOGUT Implementation: Knowledge Oriented Graph Unified Transformer for Microbial Growth Media Prediction Developed at Lawrence Berkeley National Laboratory (LBNL) Trained on NERSC Perlmutter supercomputer

Joachimiak, Marcin [Lawrence Berkeley National Lab↗
DOE OSTI2024

A new method for quantifying 64 Cu in nuclear debris samples

Quantifying 64 Cu in post-detonation nuclear debris samples can provide important diagnostic information regarding the structural materials used within a nuclear device. However, this task is challenging due to the weak gamma emissions associated with the decay of 64 Cu, its short half-life (12.701 h), and the presence of interfering fission product radioisotopes. Large quantities of debris sample are generally needed to accurately quantify 64 Cu, which can be problematic in sample-limited scenarios where other radiometric analyses are required. Herein, we present a new method for the separation of 64 Cu from solutions of mixed fission products and demonstrate the quantification of its activity through use of gas-flow proportional beta counting. Here, the new method was validated through a series of rigorous tests and was shown to improve the detection limit of 64 Cu by over two orders of magnitude, from 2.5 × 10 6 to 1.3 × 10 4 atoms/sample for 100 min measurements.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗
DOE OSTI2024

Improving Text Classification with Large Language Model-Based Data Augmentation

Large Language Models (LLMs) such as ChatGPT possess advanced capabilities in understanding and generating text. These capabilities enable ChatGPT to create text based on specific instructions, which can serve as augmented data for text classification tasks. Previous studies have approached data augmentation (DA) by either rewriting the existing dataset with ChatGPT or generating entirely new data from scratch. However, it is unclear which method is better without comparing their effectiveness. This study investigates the application of both methods to two datasets: a general-topic dataset (Reuters news data) and a domain-specific dataset (Mitigation dataset). Our findings indicate that: 1. ChatGPT generated new data consistently enhanced model’s classification results for both datasets. 2. Generating new data generally outperforms rewriting existing data, though crafting the prompts carefully is crucial to extract the most valuable information from ChatGPT, particularly for domain-specific data. 3. The augmentation data size affects the effectiveness of DA; however, we observed a plateau after incorporating 10 samples. 4. Combining the rewritten sample with new generated sample can potentially further improve the model’s performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Initial Findings from US Department of Energy’s Better Plants Virtual in-Plant Training on 50001 Ready

Manufacturing facilities use about 35% of the domestic energy in the United States every year. Implementing an effective energy management system (EnMS) is one of the most important approaches to improve energy efficiency. However, the implementation of EnMS is low for many countries (including the US) and even for energy-intensive sectors. The reasons for the low implementation rate of energy management systems had been investigated by multiple researchers, but very few studies have focused on the barriers and challenges of implementing ISO 50001-based energy management systems. To contribute to this understudied area, this paper discusses the implementation and outcomes of the first Better Plants 50001 Ready Virtual In-plant Training. This paper first provides an overview of 50001 Ready and the 50001 Ready Navigator Tool. Then, it provides details on this training event and its outcomes. Finally, it discusses findings from the responses to 40 live polling questions about the status of the 25 tasks of the 50001 Ready Navigator for participating companies, key components of the participating manufacturing companies’ energy management systems, and challenges and barriers that these companies are facing. The findings suggest that although many companies understood the importance of an effective energy management system, about half of them do not understand the required resources for building energy management systems, and most of them have only just begun establishing these systems and need more assistance and resources in multiple areas. More specifically, more assistance is necessary for the following: (1) improving corporate management’s understanding of the time and resources needed to build an EnMS as well as the benefits; (2) creating linear regression models for more accurate energy performance tracking; (3) understanding energy use, collecting and analyzing energy performance data; (4) optimizing equipment operational controls, and creating action plans.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2024

4th Big Data for Nuclear Power Plants Workshop 2023

The Ohio State University and Idaho National Laboratory organized the 4 th Big Data for Nuclear Power Plants Workshop in November, 2023 in Columbus, Ohio. Workshop topics were chosen to understand the challenges and gaps that need to be addressed to maximize the impact of data on the nuclear industry, as well as the associated applications and risks. Discussions were focused around six specific application areas: Operation and Maintenance; Machine Learning in Nuclear Materials and Advanced Manufacturing; Cybersecurity; High-Performance Computing and Massive Computation; Big Data and Digital Twins; and Nuclear Non-Proliferation. The opportunities, challenges, and risks identified in the six focus areas explored in this workshop are diverse, but some common themes emerge, such as the importance of data integrity, quality, coverage, privacy, and traceability. Big data and AI/ML tools can be leveraged to reduce costs, optimize human tasking, and reduce human error across various application areas. In order for the nuclear industry to benefit from big data and advanced analytic capabilities, it is essential to address challenges and risks, such as data privacy, model reliability, and computational resource availability. Learning from other industries that have successfully implemented big data and AI/ML technologies, like the aerospace industry, can help the nuclear industry successfully integrate these technologies.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

A.I. for nuclear physics

Nuclear science is concerned with the understanding of the nature of matter, its basic constituents and their interaction to form the elements and the properties we observe. This includes the forms of matter we see around us and also exotic forms such as those that existed in the first moments after the Big Bang and that exist today inside neutron stars. The techniques, tools, and expertise needed for nuclear physics (NP) research are therefore diverse in nature. State-of-the art accelerators are being developed to illuminate the dynamical basis of the core of the atom in terms of the fundamental constituents called quarks and gluons and to increase the number of isotopes with known properties. This scientific infrastructure is reaching scales and complexities that require computational methods for tasks such as anomaly detection in operational data. New methodologies are needed to detect anomalies and to optimize operating parameters, predict failures as well as to discover new optimization algorithms.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

Dip leg plug formation evaluation and abatement testing in support of the H-Canyon accelerated basin de-inventory

As requested in the Technical Task Request (TTR), the Savannah River National Laboratory (SRNL) performed testing on the current typical bubbler dip leg design used in the H-Area Canyon (HCA). The TTR has a functional classification of General Services. Testing was initiated to ascertain if the plugging described in the TTR could be reproduced and, if so, to obtain general metrics on the resulting plugging dynamics, and evaluate possible mitigating methods, including design changes and cleaning/purging protocols. The bubblers were to be suspended in a simulant typical of H-Canyon solutions containing aluminum nitrate and silica at a temperature of 25°C using a nominal bubbler air flowrate of 7.5 SCFH. Emphasis was placed on minimizing equipment changes necessary to accomplish the mitigations, as equipment changes in the existing facility would be costly. Testing evolved into four (4) phases, concluding that plugging could indeed be reproduced (believed to be attributed to localized evaporation and bridging of crystals in the dip legs). During testing, plug formation information was obtained, including time, pressure, and temperature traces leading to plugging were defined; solution compositions leading to plugging were determined, and various physical characteristics (visual and mechanical) of the crystallized plug were obtained.

42 ENGINEERING↗
DOE OSTI2025

In-Situ Laser Diagnostics to Understand Plasma-Surface Interactions in Titanium Thin Film Deposition

During the current project period the following tasks were completed: (1) We demonstrated the feasibility of a new way of measuring CH3 radicals in a plasma environment using a single, femtosecond (fs) pulsed laser for the first time, in a CH4 + Ar flow. A single fs pulsed Ti:Sa laser was split into two beams and with beam conversion using crystals we generatedtwo beams of 275 nm and 205 nm. The 275 nm was used to predissociate the CH3 radicals to CH2+H, followed by 205 nm Two Photon Laser Induced Fluorescence (TALIF) detection of H atoms, with a 11 ns delay from 275 nm predissociation pulse. (2) We demonstrated preliminary CH3 concentration measurements with spatial (< 100 µm) and time resolution (< 10 ns) by image processing of H and Kr TALIF signal images. (3) The methodology to quantify CH3 concentrations was developed. (4) Preliminary measurements of laser dissociation cross sections for CH4 and CH3 to (CH3+H), and (CH2+H), respectively, were conducted. (5) 275 nm laser dissociation of CH4 is a multiphoton process, and the CH4 dissociation cross section is a more sensitive function of the laser pulse energy than CH4 concentration. In the coming months, we will make more precise measurements of the dissociation cross sections, 𝑘275, 𝐶𝐻4 and 𝑘275, 𝐶𝐻3, by varying the laser energy while keeping the flow composition and conditions constant. This will enable more precise quantification of CH3 radical concentrations. More detailed measurements of CH3 radicals will be conducted in the following months of 2025 (Sept-Dec) in a Titanium Isopropoxide (TIP) + H2 + Ar flow meant for pure phase Ti thin film deposition by Chemical Vapor Deposition (CVD). More precise measurements of 275 nm laser dissociation cross sections for CH4 (to CH3 + H) and CH3 (to CH2 + H) will be conducted in the future. Detailed 275 nm laser predissociation cross sections for CH4 and CH3 dissociation to (CH3+H), and (CH2+H), respectively, as a function of gas heating, plasma power, and laser pulse energy will be evaluated. This will enable quantification of CH3 concentrations in the precursor flow over a substrate with time and spatial resolution. Comparison of the concentrations with 2D numerical simulations can lead to understanding of plasma surface interactions during pure metallic Ti thin film deposition using environmentally safer non-halogen TIP precursor. Flow conditions for pure Ti thin film deposition using non-halogen precursor (TIP) will be identified in the following months.

Uddi, Mruthunjaya [Advanced Cooling Technologies, ↗
DOE OSTI2022

Model-Based Diagnostics and Mitigation of Cyber Threats

The report summarizes key tasks performed to develop a toolkit for detecting cyber-attack events in instrumentation and control (I&C) systems of nuclear power plants. The toolkit connects the state-of-the-art GPWR Simulator with the RELAP5 code providing best-estimate nuclear steam supply system (NSSS) analyses, via the application programming interface (API), and allows users to introduce potential cyber-attack scenarios into power plant operational simulation. This summary for the project reflects topical reports submitted during the project as well as a journal paper published in 2022. The focus areas of the summary include: (1) modeling I&C systems for the AP1000 Generation III+ nuclear plant and GPWR simulator, (2) simulation and monitoring of plant response to cyber-attack events, (3) API structure for the toolkit interfacing the GPWR simulator and RELAP5 code, and (4) restructuring of the three-loop NSSS software of GPWR to model the two-loop AP1000 structure. Discussed in some details are (a) the attack tree analysis assessing the susceptibility of the AP1000 I&C system, resulting in reactor trips, in terms of the attack possibility and component sensitivity and (b) realistic estimation of the time to steam generator trip due to cyber-intrusions in the GPWR Simulator. Finally, sample demonstrations of the cyber-security tool kit, in the form of the GPWR-RELAP5 API, are summarized.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2020

Intermediate-Term Thermal Aging Effect Evaluation for Grade 92 and 316L at The LWR Relevant Temperature

Life extension of the existing nuclear reactors imposes accumulated damages, such as higher fluences and longer period of corrosion, to structural materials, which would result in significant challenges to the traditional reactor materials such as type 304 and 316 stainless steels. Advanced alloys with superior radiation resistance will increase safety margins, design flexibility, and economics for not only the life extension of the existing fleet but also new builds with advanced reactor designs. The Electric Power Research Institute (EPRI) teamed up with Department of Energy (DOE) on the Advanced Radiation Resistant Materials (ARRM) program, aiming to develop and test degradation resistant alloys from current commercial alloy specifications by 2021 to a new advanced alloy with superior degradation resistance in light water reactor (LWR)-relevant environments by 2024. Based on a comprehensive microstructure and property screening performed in Phase-1 of the ARRM program, a total of five alloys, together with 316L and X-750 as references, were down-selected for Phase-2 neutron irradiation studies. Because thermal aging could exert a synergistic effect on neutron irradiation due to the low neutron damage rate on the order of 10 –7 displacements per atom per second (dpa/s), Grade 92 (one of the five down-selected alloys) and two heats of 316L were selected in this task to study the effect of aging at 350°C for 12.5–12.7 kh on microstructure and mechanical properties. The aging time is approximately corresponding to 5 dpa neutron irradiation, which can be served as a reference for the 5-dpa-irradiated samples of the alloys to help understand the independent neutron damage influence on microstructure and mechanical properties. Optical microscopy and scanning electron microscopy were used for microstructural characterization. Hardness, tensile, Charpy impact toughness, and fracture toughness in the ductile regime of the aged samples were examined. The aging of Grade 92 led to the formation of many Laves phase in sizes of ~100–200 nm and resulted in some reduction in hardness and yield/ultimate tensile strength with some increases in uniform and total plastic elongations, which may have helped the enhancement of Charpy impact toughness, e.g., ~4 J increase in upper-shelf energy and 20.5°C reduction in ductile-brittle transition temperature compared with the unaged condition. The fracture toughness of the aged Grade 92 showed decent toughness of ~303 MPa√m (K Jq ) with ~87 tearing modulus at 22°C, which reduced to ~241 MPa√m (~20% reduction) with ~75 tearing modulus at 300°C. The two heats of 316L are differentiated by their amounts of δ-ferrite, i.e., ~1 vol% in heat T1103 and ~4 vol% in heat N5B8. The two heats of 316L were subjected to 15% cold work (CW) as this condition is often used in nuclear reactors to help trapping radiation-induced defects and thus postpone the steady state swelling stage. The aging did not result in any noticeable microstructure changes, except for possible segregations at grain boundaries that need to be further investigated. The aging resulted in slight reduction in hardness of 316L-T1103 and slight increase in hardness of 316L-N5B8, which is consistent with their tensile testing results. The 15%CW might have introduced some inhomogeneity, resulting in large standard deviations in hardness, which were reduced after the aging. The aged 316L followed the same trend as the unaged condition for the strength and elongation results from the tensile testing. Minimum elongations appeared within ~300–600°C, above which the elongations seem to increase. Unlike the negligible or minor changes in microstructure, hardness and tensile results of the two heats of 316L, their fracture toughness showed noticeable difference. The aged 316L-T1103 showed good toughness of ~350 MPa√m with ~80 tearing modulus at 22°C, which reduced to ~271 MPa√m (~23% reduction) with ~83 tearing modulus at 300°C. In contrast, the aged 316L-N5B8 showed lower toughness of ~244 MPa√m with ~53 tearing modulus at 22°C, which reduced to ~180 MPa√m (~26% reduction) with ~31 tearing modulus at 300°C. The preliminary results indicate that the presence of high volume of δ-ferrite would noticeable impair fracture toughness although it may not influence hardness and tensile properties.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

EM-Enhanced HyPOR Loop for Fast Fusion Cycles

Vacuum pumps are the heart of a fusion energy facility – fusion power cannot be generated without them. Nevertheless, the vacuum technology needed to operate a viable fuel cycle for a compact fusion power plant does not exist. Commercial vacuum technology offers the best solution to this challenge, but a pump oil recycling and detritiation system is necessary. Conventional oil detritiation processes have only been developed to deal with legacy waste and are too slow and destructive to the oil. Further, post hoc detritiation strategies are intrinsically inefficient. Our approach is to rethink the challenge holistically by designing the pumping fluid and tritium extraction system in an integrated manner to achieve an innovative solution. By selecting an oil for the specific task of pumping tritium and then designing our catalytic system to selectively target the tritium-bearing functional groups, an effective detritiation system (hydrocarbon pump oil recycling, HyPOR, loop) for fusion power plants can be obtained. The project demonstrated a HyPOR loop process that can selectively remove heavier hydrogen isotopes from pump oil, reaching the target of 99.5 % removal, with an uptake of less than 0.01% of tritium throughput, while also purifying the oil of radiation-induced damage. The recycled oil retained its pumping characteristics over 7 recycles and gamma irradiation over 30 MGy. By meeting these targets, the project enable a reduction in pump operational costs from $\$$14.5M/year to $\$$103k/year (>140× reduction), reducing pump electric power consumption from 2.8 MW to 0.25 MW (>10× reduction), and reducing in-process tritium inventory from 2.03 kg to 0.48 kg (>4× reduction).

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

GenomeFace v1.0

GenomeFace is meta-genome binning software. Metagenomic binning, the process of grouping DNA sequences into taxonomic units, is critical for understanding the functions, interactions, and evolutionary dynamics of microbial communities. We propose a deep learning approach to binning using two neural networks, one based on composition and another on environmental abundance, dynamically weighting the contribution of each based on characteristics of the input data. Trained on over 43,000 prokaryotic genomes, our network for composition-based binning is inspired by metric learning techniques used for facial recognition. Using a task-specific, multi-GPU accelerated algorithm to cluster the embeddings produced by our network, our binner leverages marker genes observed to be universally present in nearly all taxa to grade and select optimal clusters of sequences from a hierarchy of candidates. We evaluate our approach on four simulated datasets with known ground truth. Our linear time integration of marker genes recovers more near complete genomes than state of the art but computationally infeasible solutions using them, while being over an order of magnitude faster. Finally, we demonstrate the scalability and acuity of our approach by testing it on three of the largest metagenome assemblies ever performed. Compared to other binners, we produced 47%-183% more near complete genomes. From these datasets, we find over the genomes of over 3000 new candidate species which have never been previously cataloged, representing a potential 4% expansion of the known bacterial tree of life.

Lettich, Richard [Lawrence Berkeley National Labor↗
DOE OSTI2020

Multi-Gas Sensors for Enhanced Reliability of SOFC Operation

GE Research, in partnership with SUNY Polytechnic Institute (SUNY Poly), designed, built, and tested gas sensors for in situ monitoring of H 2 and CO anode tail gases produced with on-site steam reforming in solid oxide fuel cell (SOFC) systems. The knowledge of the H 2 /CO ratio of these anode tail gases should allow accurate determination and control of the efficiency of the reforming process in the SOFC system and should deliver a lower operating cost for SOFC customers. The project objectives were to achieve multi-gas monitoring capability with a single multivariable sensor, and to sustain this performance in the presence of gaseous interferences and a potential poison for the sensor. The duration of the project was 24 months with the project structure that included three technical tasks such as (1) development of design rules of photonic nanostructures for H 2 and CO gas detection, (2) laboratory validation of photonic nanostructures for selective H 2 and CO gas detection, and (3) validation of photonic nanostructures for initial stability and poison-resistance against H 2 S. To build multivariable sensors for selective H 2 and CO gas detection in the presence of interferences, we expanded our earlier knowledge of multi-gas sensors into new fabrication and functionalization methodologies as well as into new methodologies for the spectral data analysis of multi-gas responses. We have advanced our design rules of the three-dimensional (3D) photonic nanostructures that allowed detection of H 2 and CO at high temperatures as individual gases and as their mixtures and rejection of interferences such as CO 2 , H 2 O, CH 4 , and other hydrocarbons for SOFC applications. Our advanced design rules should be attractive for building the new generation of cost-effective industrial sensors. Stability and poison resistance of our 3D photonic nanostructures was tested in the laboratory conditions. While initially we utilized conventional machine learning data analysis tools, we have found that they were unable to correct for the sensor drift. Thus, we have implemented new methods of machine learning for the analysis of our spectral data. These learnings pave the way to move the future studies into advanced testing of effects of interferences, aging and field tests. In future, our work will continue to advance our sensing designs to operate in conditions with known and unknown interferences by implementing nanostructures with enhanced spectral diversity of responses to gaseous species of interest and interferences. Our systematic reduction of technical risks in this completed project and in future studies will ensure transition of this sensing technology to commercialization.

03 NATURAL GAS↗
DOE OSTI2022

CLARIFYING THE NEXUS BETWEEN LIFE CYCLE ASSESSMENT AND CIRCULARITY INDICATORS: A SETAC/ACLCA INTEREST GROUP

Purpose Improving the circularity of resources is important to the sustainability of consumer goods. Current research has indicated that circularity practices and circular economy (CE) methods do not always reduce environmental impacts. The aim of this research is to investigate the adoption of the life cycle assessment (LCA) methodology to improve the environmental impacts of circularity practices. Methods As part of the Society for Environmental Toxicology And Chemistry (SETAC) forum, an interest group (IG) on Circularity and LCA was formed in partnership with the American Center for Life Cycle Assessment (ACLCA) to tackle methodological and technical issues related to circularity in LCA. The IG’s research approach is summarized in four key steps: defining goals and objectives, literature review and gap analysis, ideation, and experimentation. The twelve active persons within this IG meet monthly and have been divided into four sub-working groups (sub-WGs) so that complementary tasks can be completed concurrently in an effective manner. Each sub-WG meets monthly and reports back to the main group for collaboration and brainstorming to meet the research objectives. Results and discussion First, the sub-WG #1, focusing on the “pool of circularity and LCA-based indicators”, analyzed the complementarity between two of the most used circularity indicators and LCA. Second, the sub-WG #2, working on the “evaluation of CE loops performance through LCA”, built a mind map of pain points that reflect the challenges that the LCA practitioners face when combining LCA with CE approaches. Third, the sub-WG #3, dealing with the “trade-offs between circularity and sustainability”, highlighted key alignments and/or conflicts between circularity and sustainability performance depending on the scope, product, industry, or system of analysis. Fourth, the sub-WG #4, focusing on “business and industrial cases”, plans to leverage the knowledge base developed within this IG to develop use cases documenting the benefits and challenges associated with CE-related loops modeling in LCA. Conclusions The first findings of this SETAC/ACLCA IG provide a state-of-the-art overview of the synergists of LCA methodology and the CE measurement frameworks reported in the literature. To move forward and capitalize on the first findings, one valuable point will be to discuss and provide concrete solutions to the pain points that emerged when considering circularity in LCA. Eventually, the knowledge base and resources created within this IG ultimately aim to support the proper application of LCA for practitioners in CE contexts, and could provide relevant inputs for the ISO Technical Committee ISO/TC 323 working on the upcoming standard for the measurement of CE performance.

Life cycle assessment, circular economy, circulari↗
DOE OSTI2020

Efficient bifurcation and tabulation of multi-dimensional combustion manifolds using deep mixture of experts: An a priori study

This work describes and validates an approach for autonomously bifurcating turbulent combustion man-ifolds to divide regression tasks amongst specialized artificial neural networks (ANNs). This approach relies on the mixture of experts (MoE) framework, where each neural network is trained to be specialized in a given portion of the input space. The assignment of different input regions to the experts is determined by a gating network, which is a neural network classifier. In some previous studies, it has been demonstrated that bifurcation of a complex combustion manifold and fitting different ANNs for each part leads to better fits or faster inference speeds. However, the manner of bifurcation in these studies was based on heuristic approaches or clustering techniques. In contrast, the proposed technique enables automatic bifurcation using non-linear planes in high-dimensional turbulent combustion manifolds that are often associated with complex behavior due to different dominating physics in various zones. Furthermore, the proposed concept is validated using 4-dimensional (4D) and 5D flamelet tables, showing that the errors obtained with a given network size, or conversely the network size required to achieve a given accuracy, is considerably reduced. The effect of the number of experts on inference speed is also investigated, showing that by increasing the number of experts from 1 to 8, the inference time can be approximately reduced by a factor of two. Moreover, it is shown that the MoE approach divides the input manifold in a physically intuitive manner, suggesting that the MoE framework can elucidate high-dimensional datasets in a physically meaningful way.

42 ENGINEERING↗
DOE OSTI2024

ExaWorks software development kit: a robust and scalable collection of interoperable workflows technologies

Scientific discovery increasingly requires executing heterogeneous scientific workflows on high-performance computing (HPC) platforms. Heterogeneous workflows contain different types of tasks (e.g., simulation, analysis, and learning) that need to be mapped, scheduled, and launched on different computing. That requires a software stack that enables users to code their workflows and automate resource management and workflow execution. Currently, there are many workflow technologies with diverse levels of robustness and capabilities, and users face difficult choices of software that can effectively and efficiently support their use cases on HPC machines, especially when considering the latest exascale platforms. We contributed to addressing this issue by developing the ExaWorks Software Development Kit (SDK). The SDK is a curated collection of workflow technologies engineered following current best practices and specifically designed to work on HPC platforms. We present our experience with (1) curating those technologies, (2) integrating them to provide users with new capabilities, (3) developing a continuous integration platform to test the SDK on DOE HPC platforms, (4) designing a dashboard to publish the results of those tests, and (5) devising an innovative documentation platform to help users to use those technologies. Our experience details the requirements and the best practices needed to curate workflow technologies, and it also serves as a blueprint for the capabilities and services that DOE will have to offer to support a variety of scientific heterogeneous workflows on the newly available exascale HPC platforms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

The Myth of Fungible FTE: A Quantitative Assessment of Matrixed Resource Allocation

Matrix organizations allow scientific facilities to share specialized personnel across projects, operations, maintenance, and strategic initiatives. Nominal staffing allocations, however, may not capture the schedule consequences of fragmented individual commitments, limited access to specialist groups, and intermittent availability of key decision makers. We developed a stochas- tic, daily-time-step simulation of a hypothetical medium-sized accelerator-facility project com- prising sequential phases and parallel tasks. Each task requires role-specific work measured in FTE-days. Ordinary personnel may be unavailable because they contribute concurrently to other institutional activities, while designated key roles have independently specified daily un- availability probabilities. An organization-wide priority factor scales the number of people from each functional group who can effectively contribute to the project. It is interpreted as a composite proxy for project access and workforce fragmentation across competing commit- ments. We examined project completion time as a function of this factor and Project Lead unavailability using 100 Monte Carlo runs per condition. Increasing priority factor from 0.1 to 1.0 reduced median completion time from 1708.5 days (interquartile range 1681.5–1735.25) to 390 days (interquartile range 379–399). At priority factor = 0.1, increasing Project Lead unavailability from 0.5 to 0.9 increased median completion time from 1713.5 days (interquartile range 1691–1733.25) to 4,417 days (interquartile range 4271.75–4550.5). The model quantifies the commonly expected sensitivity of project schedules to fragmented resource commitments and limited coordination availability. Within this model, the results also indicate a possible threshold regime in which small increases in workforce availability yield only modest sched- ule improvements until sufficient capacity becomes accessible, after which project performance improves sharply. With further validation and calibration, this quantitative framework could support resource-allocation decisions during initial project planning and subsequent schedule rebaselining.

Bai, Mei [SLAC National Accelerator Laboratory (SL↗
DOE OSTI2022

Towards Compact Neural Networks via End-to-End Training: A Bayesian Tensor Approach with Automatic Rank Determination

Post-training model compression can reduce the inference costs of deep neural networks, but uncompressed training still consumes enormous hardware resources and energy. To enable low-energy training on edge devices, it is highly desirable to directly train a compact neural network from scratch with a low memory cost. Low-rank tensor decomposition is an effective approach to reduce the memory and computing costs of large neural networks. However, directly training low-rank tensorized neural networks is a very challenging task because it is hard to determine a proper tensor rank a priori, and the tensor rank controls both model complexity and accuracy. Here, this paper presents a novel end-to-end framework for low-rank tensorized training. We first develop a Bayesian model that supports various low-rank tensor formats (e.g., CANDECOMP/PARAFAC, Tucker, tensor-train, and tensor-train matrix) and reduces neural network parameters with automatic rank determination during training. Then we develop a customized Bayesian solver to train large-scale tensorized neural networks. Our training methods shows orders-of-magnitude parameter reduction and little accuracy loss (or even better accuracy) in the experiments. On a very large deep learning recommendation system with over 4.2 ×10 9 model parameters, our method can reduce the parameter number to 1.6 ×10 5 automatically in the training process (i.e., by 2.6 ×10 4 times) while achieving almost the same accuracy. Code is available at https://github.com/colehawkins/bayesian-tensor-rank-determination.

compact neural networks↗