Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “task tuning”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Graphitic Aza-Fused π-Conjugated Networks: Construction, Engineering, and Task-Specific Applications

2D π-conjugated networks linked by aza-fused units represent a pivotal category of graphitic materials with stacked nanosheet architectures. Extensive efforts have been directed at their fabrication and application since the discovery of covalent triazine frameworks (CTFs). Besides the triazine cores, tricycloquinazoline and hexaazatriphenylene linkages are further introduced to tailor the structures and properties. Diverse related materials have been developed rapidly, and a thorough outlook is necessitated to unveil the structure–property–application relationships across multiple subcategories, which is pivotal to guide the design and fabrication toward enhanced task-specific performance. Herein, the structure types and development of related materials including CTFs, covalent quinazoline networks, and hexaazatriphenylene networks, are introduced. Advanced synthetic strategies coupled with characterization techniques provide powerful tools to engineer the properties and tune the associated behaviors in corresponding applications. Case studies in the areas of gas adsorption, membrane-based separation, thermo-/electro-/photocatalysis, and energy storage are then addressed, focusing on the correlation between structure/property engineering and optimization of the corresponding performance, particularly the preferred features and strategies in each specific field. In the last section, the underlying challenges and opportunities in construction and application of this emerging and promising material category are discussed.

36 MATERIALS SCIENCE↗
DOE OSTI2025

JuTrack: A Julia package for auto-differentiable accelerator modeling and particle tracking

Efficient accelerator modeling and particle tracking are key for the design and configuration of modern particle accelerators. In this work, we present JuTrack, a nested accelerator modeling package developed in the Julia programming language and enhanced with compiler-level automatic differentiation (AD). With the aid of AD, JuTrack enables rapid derivative calculations in accelerator modeling, facilitating sensitivity analyses and optimization tasks. Here we demonstrate the effectiveness of AD-derived derivatives through several practical applications, including sensitivity analysis of space-charge-induced emittance growth, nonlinear beam dynamics analysis for a synchrotron light source, and lattice parameter tuning of the future Electron-Ion Collider (EIC). Through the incorporation of automatic differentiation, this package opens up new possibilities for accelerator physicists in beam physics studies and accelerator design optimization.

43 PARTICLE ACCELERATORS↗
DOE OSTI2020

Hierarchical supramolecular assembly of a single peptoid polymer into a planar nanobrush with two distinct molecular packing motifs

Significance The preparation of hierarchical biomimetic nanostructures remains a challenging task. Bioinspired peptoid polymers offer a promising opportunity to encode assembly information within the component chains. Here, we report a one-pot preparation of planar nanobrush structures from poly(ethyleneglycol)- block -poly( N -octylglycine) based on tuning the interplay of crystallization and solvophobicity. Each nanobrush assembly is composed of two distinct structural domains, which form sequentially by control of the solution temperature, and contain distinct molecular packing motifs. Micellar intermediates first assemble into linear arrays at elevated temperature in a mesophase state, followed by lateral, crystallization-driven growth in a living manner upon cooling. Precise control over the nanobrush dimensions could be achieved. The structural tunability and excellent biocompatibility of polypeptoids offer great potential for nanomaterials inaccessible previously.

36 MATERIALS SCIENCE↗
NASA NTRS2017

Filter Tuning Using the Chi-Squared Statistic

The Goddard Space Flight Center (GSFC) Flight Dynamics Facility (FDF) performs orbit determination (OD) for the Aqua and Aura satellites. Both satellites are located in low Earth orbit (LEO), and are part of what is considered the A-Train satellite constellation. Both spacecraft are currently in the science phase of their respective missions. The FDF has recently been tasked with delivering definitive covariance for each satellite.The main source of orbit determination used for these missions is the Orbit Determination Toolkit developed by Analytical Graphics Inc. (AGI). This software uses an Extended Kalman Filter (EKF) to estimate the states of both spacecraft. The filter incorporates force modelling, ground station and space network measurements to determine spacecraft states. It also generates a covariance at each measurement. This covariance can be useful for evaluating the overall performance of the tracking data measurements and the filter itself. An accurate covariance is also useful for covariance propagation which is utilized in collision avoidance operations. It is also valuable when attempting to determine if the current orbital solution will meet mission requirements in the future.This paper examines the use of the Chi-square statistic as a means of evaluating filter performance. The Chi-square statistic is calculated to determine the realism of a covariance based on the prediction accuracy and the covariance values at a given point in time. Once calculated, it is the distribution of this statistic that provides insight on the accuracy of the covariance.For the EKF to correctly calculate the covariance, error models associated with tracking data measurements must be accurately tuned. Over estimating or under estimating these error values can have detrimental effects on the overall filter performance. The filter incorporates ground station measurements, which can be tuned based on the accuracy of the individual ground stations. It also includes measurements from the NASA space network (SN), which can be affected by the assumed accuracy of the TDRS satellite state at the time of the measurement.The force modelling in the EKF is also an important factor that affects the propagation accuracy and covariance sizing. The dominant force in the LEO orbit regime is the drag force caused by atmospheric drag. Accurate accounting of the drag force is especially important for the accuracy of the propagated state. The implementation of a box and wing model to improve drag estimation accuracy, and its overall effect on the covariance state is explored.The process of tuning the EKF for Aqua and Aura support is described, including examination of the measurement errors of available observation types (Doppler and range), and methods of dealing with potentially volatile atmospheric drag modeling. Predictive accuracy and the distribution of the Chi-square statistic, calculated based of the ODTK EKF solutions, are assessed versus accepted norms for the orbit regime.

Covariance Analysis↗
DOE OSTI2024

A safe reinforcement learning algorithm for supervisory control of power plants

Traditional control theory-based methods require tailored engineering for each system and constant fine-tuning. In power plant control, one often needs to obtain a precise representation of the system dynamics and carefully design the control scheme accordingly. Model-free Reinforcement learning (RL) has emerged as a promising solution for control tasks due to its ability to learn from trial-and-error interactions with the environment. It eliminates the need for explicitly modeling the environment’s dynamics, which is potentially inaccurate. However, the direct imposition of state constraints in power plant control raises challenges for standard RL methods. To address this, we propose a chance-constrained RL algorithm based on Proximal Policy Optimization for supervisory control. Our method employs Lagrangian relaxation to convert the constrained optimization problem into an unconstrained objective, where trainable Lagrange multipliers enforce the state constraints. In conclusion, our approach achieves the smallest distance of violation and violation rate in a load-follow maneuver for an advanced Nuclear Power Plant design.

constrained optimization↗
DOE OSTI2023

Combining multitask and transfer learning with deep Gaussian processes for autotuning-based performance engineering

We combine deep Gaussian processes (DGPs) with multitask and transfer learning for the performance modeling and optimization of HPC applications. Deep Gaussian processes merge the uncertainty quantification advantage of Gaussian processes (GPs) with the predictive power of deep learning. Multitask and transfer learning allow for improved learning efficiency when several similar tasks are to be learned simultaneously and when previous learned models are sought to help in the learning of new tasks, respectively. A comparison with state-of-the-art autotuners shows the advantage of our approach on two application problems. In this article, we combine DGPs with multitask and transfer learning to allow for both an improved tuning of an application parameters on problems of interest but also the prediction of parameters on any potential problem the application might encounter.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Towards an Aviation Large Language Model by Fine-tuning and Evaluating Transformers

In the aviation domain, there are many applications for machine learning and artificial intelligence tools that utilize natural language. For example, there is a desire to know the commonalities in written safety reports such as voluntary post incidents reports or create more accurate transcripts of air traffic management conversations. Another use-case is the possibility of extracting airspace procedures and constraints currently written in documents such as Letters of Agreement (LOA) which is used as the evaluation case in this paper. These applications can benefit from the use of state-of-the-art Natural Language Processing (NLP) techniques when adapted to the language/phraseology specific to the aviation domain. This paper evaluates the viability of transferring pre-trained large language models to the aviation domain by adapting transformer based models using aviation datasets. This paper utilized two datasets to adapt a ‘Robustly Optimized Bidirectional Encoder Representations from Transformers Approach’ (RoBERTa) model and two down-stream classification tasks to assess its performance. These datasets are all built upon Letters of Agreement which are Federal Aviation Administration (FAA) documents that formalize airspace operations across the national airspace system. The first two datasets are used for the adaptation of RoBERTa to the aviation domain and were of different sizes to assess the number of documents needed to adapt to the aviation domain. They contain many examples of ‘aviation English’ using domain specific terminology and phrasing which serves as a representative basis to perform the unsupervised adaptation. The second dataset is a separate set of LOA documents with two sets of classification labels to be used for evaluation; one at the document level and one at the line level. These down-stream evaluations allowed the measurement of improvement by adapting RoBERTa. The accuracy increased by 4-6% on both tasks and the F1 score on the class of interest increased by 4-8% from the adaptation.

Air Traffic Management↗
NASA NTRSDate not supplied

Towards an Aviation Large Language Model by Fine-tuning and Evaluating Transformers

In the aviation domain, there are many applications for machine learning and artificial intelligence tools that utilize natural language. For example, there is a desire to know the commonalities in written safety reports such as voluntary post incidents reports or create more accurate transcripts of air traffic management conversations. Another use-case is the possibility of extracting airspace procedures and constraints currently written in documents such as Letters of Agreement (LOA) which is used as the evaluation case in this paper. These applications can benefit from the use of state-of-the-art Natural Language Processing (NLP) techniques when adapted to the language/phraseology specific to the aviation domain. This paper evaluates the viability of transferring pre-trained large language models to the aviation domain by adapting transformer based models using aviation datasets. This paper utilized two datasets to adapt a ‘Robustly Optimized Bidirectional Encoder Representations from Transformers Approach’ (RoBERTa) model and two down-stream classification tasks to assess its performance. These datasets are all built upon Letters of Agreement which are Federal Aviation Administration (FAA) documents that formalize airspace operations across the national airspace system. The first two datasets are used for the adaptation of RoBERTa to the aviation domain and were of different sizes to assess the number of documents needed to adapt to the aviation domain. They contain many examples of ‘aviation English’ using domain specific terminology and phrasing which serves as a representative basis to perform the unsupervised adaptation. The second dataset is a separate set of LOA documents with two sets of classification labels to be used for evaluation; one at the document level and one at the line level. These down-stream evaluations allowed the measurement of improvement by adapting RoBERTa. The accuracy increased by 4-6% on both tasks and the F1 score on the class of interest increased by 4-8% from the adaptation.

Air Traffic Management↗
DOE OSTI2025

Tracing terahertz plasmon polaritons with a tunable-by-design dispersion in topological insulator metaelements

Abstract Collective oscillations of massless charge carriers in two-dimensional materials—Dirac plasmon polaritons (DPPs)—are of paramount importance for engineering nanophotonic devices with tunable optical response. However, tailoring the optical properties of DPPs in a nanomaterial is a very challenging task, particularly at terahertz (THz) frequencies, where the DPP momentum is more than one order of magnitude larger than that of the free-space photons, and DDP attenuation is high. Here, we conceive and demonstrate a strategy to tune the DPP dispersion in topological insulator metamaterials. We engineer laterally coupled linear metaelements, fabricated from epitaxial Bi 2 Se 3, with selected coupling distances with the purpose to tune their wavevector, by geometry. We launch and directly map the propagation of DPPs confined within coupled meta-atoms via phase-sensitive scattering-type scanning near-field nanoscopy. We demonstrate that the DPP wavelength can be tuned by varying the metaelements coupling distance, resulting in up to a 20% increase of the polariton wavevector Re(k p ) in dimers and triplets with a 1 μm spacing, with reduced losses and a >50% increase of the polariton attenuation length.

Optics↗
NASA NTRS2015

Spin-Up and Tuning of the Global Carbon Cycle Model Inside the GISS ModelE2 GCM

Planetary carbon cycle involves multiple phenomena, acting at variety of temporal and spacial scales. The typical times range from minutes for leaf stomata physiology to centuries for passive soil carbon pools and deep ocean layers. So, finding a satisfactory equilibrium state becomes a challenging and computationally expensive task. Here we present the spin-up processes for different configurations of the GISS Carbon Cycle model from the model forced with MODIS observed Leaf Area Index (LAI) and prescribed ocean to the prognostic LAI and to the model fully coupled to the dynamic ocean and ocean biology. We investigate the time it takes the model to reach the equilibrium and discuss the ways to speed up this process. NASA Goddard Institute for Space Studies General Circulation Model (GISS ModelE2) is currently equipped with all major algorithms necessary for the simulation of the Global Carbon Cycle. The terrestrial part is presented by Ent Terrestrial Biosphere Model (Ent TBM), which includes leaf biophysics, prognostic phenology and soil biogeochemistry module (based on Carnegie-Ames-Stanford model). The ocean part is based on the NASA Ocean Biogeochemistry Model (NOBM). The transport of atmospheric CO2 is performed by the atmospheric part of ModelE2, which employs quadratic upstream algorithm for this purpose.

Climate model↗
NASA NTRS2009

Latency and User Performance in Virtual Environments and Augmented Reality

System rendering latency has been recognized by senior researchers, such as Professor Fredrick Brooks of UNC (Turing Award 1999), as a major factor limiting the realism and utility of head-referenced displays systems. Latency has been shown to reduce the user's sense of immersion within a virtual environment, disturb user interaction with virtual objects, and to contribute to motion sickness during some simulation tasks. Latency, however, is not just an issue for external display systems since finite nerve conduction rates and variation in transduction times in the human body's sensors also pose problems for latency management within the nervous system. Some of the phenomena arising from the brain's handling of sensory asynchrony due to latency will be discussed as a prelude to consideration of the effects of latency in interactive displays. The causes and consequences of the erroneous movement that appears in displays due to latency will be illustrated with examples of the user performance impact provided by several experiments. These experiments will review the generality of user sensitivity to latency when users judge either object or environment stability. Hardware and signal processing countermeasures will also be discussed. In particular the tuning of a simple extrapolative predictive filter not using a dynamic movement model will be presented. Results show that it is possible to adjust this filter so that the appearance of some latencies may be hidden without the introduction of perceptual artifacts such as overshoot. Several examples of the effects of user performance will be illustrated by three-dimensional tracking and tracing tasks executed in virtual environments. These experiments demonstrate classic phenomena known from work on manual control and show the need for very responsive systems if they are indented to support precise manipulation. The practical benefits of removing interfering latencies from interactive systems will be emphasized with some classic final examples from surgical telerobotics, and human-computer interaction.

Ellis, Stephen R.↗
NASA NTRS2013

Scheduling Operations for Massive Heterogeneous Clusters

High-performance computing (HPC) programming has become increasingly difficult with the advent of hybrid supercomputers consisting of multicore CPUs and accelerator boards such as the GPU. Manual tuning of software to achieve high performance on this type of machine has been performed by programmers. This is needlessly difficult and prone to being invalidated by new hardware, new software, or changes in the underlying code. A system was developed for task-based representation of programs, which when coupled with a scheduler and runtime system, allows for many benefits, including higher performance and utilization of computational resources, easier programming and porting, and adaptations of code during runtime. The system consists of a method of representing computer algorithms as a series of data-dependent tasks. The series forms a graph, which can be scheduled for execution on many nodes of a supercomputer efficiently by a computer algorithm. The schedule is executed by a dispatch component, which is tailored to understand all of the hardware types that may be available within the system. The scheduler is informed by a cluster mapping tool, which generates a topology of available resources and their strengths and communication costs. Software is decoupled from its hardware, which aids in porting to future architectures. A computer algorithm schedules all operations, which for systems of high complexity (i.e., most NASA codes), cannot be performed optimally by a human. The system aids in reducing repetitive code, such as communication code, and aids in the reduction of redundant code across projects. It adds new features to code automatically, such as recovering from a lost node or the ability to modify the code while running. In this project, the innovators at the time of this reporting intend to develop two distinct technologies that build upon each other and both of which serve as building blocks for more efficient HPC usage. First is the scheduling and dynamic execution framework, and the second is scalable linear algebra libraries that are built directly on the former.

Humphrey, John↗
DOE OSTI2025

RINO: Renormalization Group Invariance with No Labels

A common challenge with supervised machine learning (ML) in high energy physics (HEP) is the reliance on simulations for labeled data, which can often mismodel the underlying collision or detector response. To help mitigate this problem of domain shift, we propose RINO (Renormalization Group Invariance with No Labels), a self-supervised learning approach that can instead pretrain models directly on collision data, learning embeddings invariant to renormalization group flow scales. In this work, we pretrain a transformer-based model on jets originating from quantum chromodynamic (QCD) interactions from the JetClass dataset, emulating real QCD-dominated experimental data, and then finetune on the JetNet dataset -- emulating simulations -- for the task of identifying jets originating from top quark decays. RINO demonstrates improved generalization from the JetNet training data to JetClass data compared to supervised training on JetNet from scratch, demonstrating the potential for RINO pretraining on real collision data followed by fine-tuning on small, high-quality MC datasets, to improve the robustness of ML models in HEP.

Hao, Zichun [Caltech] (ORCID:0000000256244907)↗
DOE OSTI2025

ARCH: Large-scale knowledge graph via aggregated narrative codified health records analysis

Objective: Electronic health record (EHR) systems contain a wealth of clinical data stored as both codified data and free-text narrative notes (NLP). The complexity of EHR presents challenges in feature representation, information extraction, and uncertainty quantification. Here, to address these challenges, we proposed an efficient Aggregated naRrative Codified Health (ARCH) records analysis to generate a large-scale knowledge graph (KG) for a comprehensive set of EHR codified and narrative features. Methods: Using data from 12.5 million Veterans Affairs patients, ARCH first derives embedding vectors and generates similarities along with associated p-values to measure the strength of relatedness between clinical features with statistical certainty quantification. Next, ARCH performs a sparse embedding regression to remove indirect linkage between features to build a sparse KG. Finally, ARCH was validated on various clinical tasks, including detecting known relationships between entity pairs, predicting drug side effects, disease phenotyping, as well as sub-typing Alzheimer’s disease patients. Results: ARCH produces high-quality clinical embeddings and KG for over 60,000 codified and narrative EHR concepts. The KG and embeddings are visualized in the R-shiny powered web-API.3 ARCH achieved high accuracy in detecting EHR concept relationships, with AUCs of 0.926 (codified) and 0.861 (NLP) for similar EHR concepts, and 0.810 (codified) and 0.843 (NLP) for related pairs. It detected drug side effects with a 0.723 AUC, which improved to 0.826 after fine-tuning. Using both codified and NLP features, the detection power increased significantly. Compared to other methods, ARCH has superior accuracy and enhances weakly supervised phenotyping algorithms’ performance. Notably, it successfully categorized Alzheimer’s patients into two subgroups with varying mortality rates. Conclusion: The proposed ARCH algorithm generates large-scale high-quality semantic representations and knowledge graph for both codified and NLP EHR features, useful for a wide range of predictive modeling tasks.

Electronic health records↗
NASA NTRS2001

Activation of the E1 Ultra High Pressure Propulsion Test Facility at Stennis Space Center

After a decade of construction and a year of activation the El Ultra High Pressure Propulsion Test Facility at NASA's Stennis Space Center is fully operational. The El UHP Propulsion Test Facility is a multi-cell, multi-purpose component and engine test facility . The facility is capable of delivering cryogenic propellants at low, high, and ultra high pressures with flow rates ranging from a few pounds per second up to two thousand pounds per second. Facility activation is defined as a series of tasks required to transition between completion of construction and facility operational readiness. Activating the El UHP Propulsion Test Facility involved independent system checkouts, propellant system leak checks, fluid and gas sampling, gaseous system blow downs, pressurization and vent system checkouts, valve stability testing, valve tuning cryogenic cold flows, and functional readiness tests.

Messer, Bradley↗
DOE OSTI2025

A Preferences Corpus and Annotation Scheme for Human-Guided Alignment of Time-Series GPTs

The process of time-series forecasting such as predicting trajectories of silicon content in blast furnaces is a difficult task. Most time-series approaches today focus on scalar-type MSE loss optimization. This optimization approach, while widely common, could benefit from the use of human expert or process-level preferences. In this paper, we introduce a novel alignment and fine-tuning approach that involves learning from a corpus of preferred and dis-preferred time-series prediction trajectories. Our contributions include (1) a preference annotation pipeline for time-series forecasts, (2) the application of Score-based Preference Optimization (SPO) to train decoder-only transformers from preferences, and (3) results showing improvements in forecast quality. The approach is validated on both proprietary blast furnace data and the UCI Appliances Energy dataset. The proposed preference corpus and training strategy offer a new option for fine-tuning sequence models in industrial settings.

DPO↗
DOE OSTI2024

Towards Next-Generation Urban Decision Support Systems through AI-Powered Construction of Scientific Ontology Using Large Language Models—A Case in Optimizing Intermodal Freight Transportation

The incorporation of Artificial Intelligence (AI) models into various optimization systems is on the rise. However, addressing complex urban and environmental management challenges often demands deep expertise in domain science and informatics. This expertise is essential for deriving data and simulation-driven insights that support informed decision-making. In this context, we investigate the potential of leveraging the pre-trained Large Language Models (LLMs) to create knowledge representations for supporting operations research. By adopting ChatGPT-4 API as the reasoning core, we outline an applied workflow that encompasses natural language processing, Methontology-based prompt tuning, and Generative Pre-trained Transformer (GPT), to automate the construction of scenario-based ontologies using existing research articles and technical manuals of urban datasets and simulations. From these ontologies, knowledge graphs can be derived using widely adopted formats and protocols, guiding various tasks towards data-informed decision support. The performance of our methodology is evaluated through a comparative analysis that contrasts our AI-generated ontology with the widely recognized pizza ontology, commonly used in tutorials for popular ontology software. We conclude with a real-world case study on optimizing the complex system of multi-modal freight transportation. Our approach advances urban decision support systems by enhancing data and metadata modeling, improving data integration and simulation coupling, and guiding the development of decision support strategies and essential software components.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2023

GenSLMs: Genome-scale language models reveal SARS-CoV-2 evolutionary dynamics

We seek to transform how new and emergent variants of pandemic-causing viruses, specifically SARS-CoV-2, are identified and classified. By adapting large language models (LLMs) for genomic data, we build genome-scale language models (GenSLMs) which can learn the evolutionary landscape of SARS-CoV-2 genomes. By pre-training on over 110 million prokaryotic gene sequences and fine-tuning a SARS-CoV-2-specific model on 1.5 million genomes, we show that GenSLMs can accurately and rapidly identify variants of concern. Thus, to our knowledge, GenSLMs represents one of the first whole-genome scale foundation models which can generalize to other prediction tasks. We demonstrate scaling of GenSLMs on GPU-based supercomputers and AI-hardware accelerators utilizing 1.63 Zettaflops in training runs with a sustained performance of 121 PFLOPS in mixed precision and peak of 850 PFLOPS. We present initial scientific insights from examining GenSLMs in tracking evolutionary dynamics of SARS-CoV-2, paving the path to realizing this on large biological data.

Zvyagin, Maxim↗