Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “semantic”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

chatHPC: Empowering HPC users with large language models

The ever-growing number of pre-trained large language models (LLMs) across scientific domains presents a challenge for application developers. While these models offer vast potential, fine-tuning them with custom data, aligning them for specific tasks, and evaluating their performance remain crucial steps for effective utilization. However, applying these techniques to models with tens of billions of parameters can take days or even weeks on modern workstations, making the cumulative cost of model comparison and evaluation a significant barrier to LLM-based application development. To address this challenge, we introduce an end-to-end pipeline specifically designed for building conversational and programmable AI agents on high performance computing (HPC) platforms. Our comprehensive pipeline encompasses: model pre-training, fine-tuning, web and API service deployment, along with crucial evaluations for lexical coherence, semantic accuracy, hallucination detection, and privacy considerations. Here, we demonstrate our pipeline through the development of chatHPC, a chatbot for HPC question answering and script generation. Leveraging our scalable pipeline, we achieve end-to-end LLM alignment in under an hour on the Frontier supercomputer. We propose a novel self-improved, self-instruction method for instruction set generation, investigate scaling and fine-tuning strategies, and conduct a systematic evaluation of model performance. The established practices within chatHPC will serve as a valuable guidance for future LLM-based application development on HPC platforms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Machine learning-driven descriptions of protein dynamics at solid-liquid interfaces

This chapter has described how ML has enabled quantitative analysis of HS-AFM data to discover the physical phenomena governing protein dynamics and ordering at solid-liquid interfaces. The research detailed in this chapter modeled the rotation models of protein nanorods, the discovery of which would otherwise not be possible. By tracking the trajectories of individual protein rods from frame to frame, it was possible to model Brownian type motion and behaviors and Levy-flight dynamics that had not previously been shown. We also described the application of the Python package AtomAI, which has been developed specifically to analyze and extract physical phenomena, providing exemplar code for training an ensemble of deep neural networks to produce the semantic segmentation of AFM data and functions for encoding and decoding local environments. We last described a combinatorial approach to analyze very noisy data with a densely covered substrate where the emergence of order for the protein liquid crystals could be elucidated. By combining the methods from Case 1 and 2, it was possible to obtain the center of mass and angle for each rod in the images and track the assembly of the rods over time into a 2D liquid crystal array on the surface of mica.

protein dynamics, solid-liquid interfaces, atomic ↗
DOE OSTI2024

Multi deep learning-based stochastic microstructure reconstruction and high-fidelity micromechanics simulation of time-dependent ceramic matrix composite response

A multi deep learning-based framework is developed for efficient, automated microstructure reconstruction and generation of stochastic representative volume elements (SRVEs) with periodic boundary conditions (PBCs) for accurate modeling of ceramic matrix composite (CMC) response. The methodology comprises a convolutional neural network coupled with regression layers to act as a vanilla regression network for semantic segmentation of the microstructure, allowing accurate characterization of the phases and their distributions at the microscale. Scanning electron microscope and confocal microscope are used to obtain C/SiNC and SiC/SiNC CMCs micrographs for vanilla regression testing. Microstructure variability in terms of fiber volume fraction and porosity are quantified through the output regression layer, ensuring accurate representation of material variability in SRVE construction. Generative adversarial network (GAN) and its variants are designed to produce high-fidelity SRVE, spanning CMCs microstructure variability space. A circular padding algorithm is developed to generate SRVEs with PBCs during training of GANs. The accuracy of the generated SRVEs is established through micromechanics simulations, where an efficient formulation of the high-fidelity generalized methods of cells (HFGMC) approach is used to compute the effective mechanical properties. Furthermore, an iterative algorithm is implemented in the HFGMC solver to simulate time-dependent deformation of SiC/SiNC subjected to creep loading conditions.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Leveraging large language models to address data scarcity in machine learning for graphene synthesis

Machine learning in experimental materials science faces significant challenges due to the scarcity of data, which are costly and time-consuming to generate, particularly when relying on in-house experiments. Literature data mining offers a potential solution but introduces issues like mixed data quality, inconsistent formats, and non-uniform reporting of synthesis parameters, resulting in partially missing and heterogeneous features across the dataset. Here, we propose data imputation and feature engineering methods that employ pre-trained large language models (LLMs) to enhance machine learning performance on scarce, heterogeneous datasets, demonstrated on graphene CVD synthesis data and the ML-HydPARK hydrogen storage dataset. GPT models perform data imputation via tailored prompting and semantic normalization of inconsistently reported features through embeddings, for example, to harmonize the complex nomenclature of CVD substrates. Beyond yielding more diverse and richer feature representations than traditional methods such as K-nearest neighbors (KNN) and Multivariate Imputation by Chained Equations (MICE), LLM-based data imputation is evaluated against dataset characteristics and prompting strategies. We vary the level of autonomy granted to the LLM, from generic prompting that leverages pre-trained knowledge for autonomous data generation to data-informed prompting that constrains outputs using target-specific information, and demonstrate which level of autonomy yields superior imputation performance across datasets and feature types. The proposed data engineering methods markedly improve downstream performance; for example, in graphene layer number classification using a support vector machine (SVM), binary accuracy increases from 39% to 65% and ternary accuracy from 52% to 72%. Fine-tuning experiments on both datasets show that combining our proposed LLM-based data imputation and feature encoding methods with numerical machine learning predictors outperforms standalone fine-tuned LLM predictors in data-scarce settings. The proposed strategies emphasize data enhancement techniques rather than refining learning architectures or regularizing loss functions, offering a broadly applicable framework for improving machine learning performance on scarce, inhomogeneous datasets.

Chemical vapor deposition↗
DOE OSTI2025

Parallel sorting algorithm classification: is manual instrumentation necessary?

Understanding parallel algorithms is crucial for accelerating scientific simulations on complex, distributed memory, high-performance computers. Modern algorithm classification approaches learn semantics directly from source code to differentiate between algorithms, however, accessing source code is not always possible. We can learn about parallel algorithms from observing their performance, as programs running the same algorithms and using the same hardware should exhibit similar performance characteristics. We present an approach to learn algorithm classes from parallel performance data directly in order to classify algorithms without access to the source code. We extend previous work to enable classifying parallel sorting algorithms using automatic instrumentation instead of requiring manual region annotations in the source code. In this work, we design and demonstrate a study for classification of parallel sorting algorithms using parallel performance data collected from automatic instrumentation, and evaluate the performance of our new methodology on classification. We leverage Caliper to collect the performance data, Thicket for our exploratory data analysis (EDA), and PyTorch and Scikit-learn to evaluate the effectiveness of random forests, support vector machines (SVMs), decision trees, neural networks, and logistic regressions on parallel performance data. Additionally, we study noise in parallel performance data, whether the removal of noise and pre-processing of the data is necessary to accurately classify parallel sorting algorithms, and determine the effectiveness of features created from performance data. In conclusion, we demonstrate classification accuracy for these five different models of up to 97.7% across four different parallel algorithm classes.

Algorithm Classification↗
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↗
DOE OSTI2025

DOME: Directional medical embedding vectors from Electronic Health Records

Motivation: The increasing availability of Electronic Health Record (EHR) systems has created enormous potential for translational research. Recent developments in representation learning techniques have led to effective large-scale representations of EHR concepts along with knowledge graphs that empower downstream EHR studies. However, most existing methods require training with patient-level data, limiting their abilities to expand the training with multi-institutional EHR data. On the other hand, scalable approaches that only require summary-level data do not incorporate temporal dependencies between concepts. Methods: We introduce a DirectiOnal Medical Embedding (DOME) algorithm to encode temporally directional relationships between medical concepts, using summary-level EHR data. Specifically, DOME first aggregates patient-level EHR data into an asymmetric co-occurrence matrix. Then it computes two Positive Pointwise Mutual Information (PPMI) matrices to correspondingly encode the pairwise prior and posterior dependencies between medical concepts. Following that, a joint matrix factorization is performed on the two PPMI matrices, which results in three vectors for each concept: a semantic embedding and two directional context embeddings. They collectively provide a comprehensive depiction of the temporal relationship between EHR concepts. Results: We highlight the advantages and translational potential of DOME through three sets of validation studies. First, DOME consistently improves existing direction-agnostic embedding vectors for disease risk prediction in several diseases, for example achieving a relative gain of 5.5% in the area under the receiver operating characteristic (AUROC) for lung cancer. Second, DOME excels in directional drug-disease relationship inference by successfully differentiating between drug side effects and indications, correspondingly achieving relative AUROC gain over the state-of-the-art methods by 10.8% and 6.6%. Finally, DOME effectively constructs directional knowledge graphs, which distinguish disease risk factors from comorbidities, thereby revealing disease progression trajectories. The source codes are provided at https://github.com/celehs/Directional-EHRembedding.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Autogenerating a Domain-Specific Question-Answering Data Set from a Thermoelectric Materials Database to Enable High-Performing BERT Models

We present a method for autogenerating a large domain-specific question-answering (QA) dataset from a thermoelectric materials database. We show that a small language model, BERT, once fine-tuned on this automatically generated dataset of 99,757 QA pairs about thermoelectric materials, affords better performance in the field of thermoelectric materials compared to a BERT model fine-tuned on the generic English-language QA data set, SQuAD-v2. We further show that mixing the two data sets (ours and SQuAD-v2), which have significantly different syntactic and semantic scopes, allows the BERT model to achieve even better performance. The best-performing BERT model fine-tuned on the mixed data set outperforms the models fine-tuned on the other two data sets by scoring an exact match of 67.93% and an F1 score of 72.29% when evaluated on our test data set. This has important implications as it demonstrates the ability to realize high-performing small language models, with modest computational resources, empowered by domain-specific materials data sets which can be generated according to our method.

biological databases↗
DOE OSTI2025

Generalizable Image Segmentation for Microstructure Characterization Through Integrated SEM and EBSD Analysis

We demonstrate generalizable semantic segmentation using minimal ground truth data. Correlated scanning electron microscopy (SEM) images and electron backscatter diffraction (EBSD) measurements of frictionstir processed 316L stainless steel plates were used to train deep learning models for grain boundary segmentation. Secondary electron (SE) imaging taken at an accelerating voltage of 10 keV correlated to EBSD-derived grain boundaries produced the best performing model. Notably, an ensemble of three models trained on a single SE image produced accurate segmentation over a series of BSE images of samples manufactured under different processing parameters, with a resultant mean absolute error in grain size of 0.34 µm. The striking generalizability of the models likely results from the similar escape depths of the SE training input and the EBSD training output and the reduced probability of dislocation artifacts appearing in the image. This finding highlights the importance of considering the physical principles behind imaging in the development of robust segmentation models for microstructure characterization.

Taufique, Mohammad Fuad Nur↗
DOE OSTI2026

Hierarchical Multi-agent Large Language Model Reasoning for Autonomous Heterogeneous Catalyst Discovery

Artificial intelligence is reshaping scientific exploration, but most methods automate procedural tasks without engaging in scientific reasoning, limiting autonomy in discovery. We demonstrate that hierarchical agentic large language model reasoning can efficiently drive simulation and scientific exploration. Across two chemical applications, CO adsorption on Cu surface transition metal adatoms and on M–N–C catalysts, reasoning-guided exploration reduces required atomistic simulations by up to 90% relative to heuristic or random selection. Comparisons across single-agent, multi-agent, and stochastic baselines show that hierarchical strategies yield more coherent and information-efficient search trajectories. Reasoning traces reveal chemically grounded decisions that cannot be explained by semantic bias or stochastic sampling. We realize these agentic reasoning strategies in Materials Agents for Simulation and Theory in Electronic-structure Reasoning (MASTER), a multimodal system that translates natural language into density functional theory workflows. Altogether, multi-agent collaboration accelerates heterogeneous catalyst discovery and marks a step toward more autonomous, reasoning-guided scientific exploration.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2024

RhizoNet segments plant roots to assess biomass and growth for enabling self-driving labs

Abstract Flatbed scanners are commonly used for root analysis, but typical manual segmentation methods are time-consuming and prone to errors, especially in large-scale, multi-plant studies. Furthermore, the complex nature of root structures combined with noisy backgrounds in images complicates automated analysis. Addressing these challenges, this article introduces RhizoNet, a deep learning-based workflow to semantically segment plant root scans. Utilizing a sophisticated Residual U-Net architecture, RhizoNet enhances prediction accuracy and employs a convex hull operation for delineation of the primary root component. Its main objective is to accurately segment root biomass and monitor its growth over time. RhizoNet processes color scans of plants grown in a hydroponic system known as EcoFAB, subjected to specific nutritional treatments. The root detection model using RhizoNet demonstrates strong generalization in the validation tests of all experiments despite variable treatments. The main contributions are the standardization of root segmentation and phenotyping, systematic and accelerated analysis of thousands of images, significantly aiding in the precise assessment of root growth dynamics under varying plant conditions, and offering a path toward self-driving labs.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Patch-Based Convolutional Neural Networks for Multiple Microstructural Features Detection in FIB-SEM Micrographs of Irradiated Nuclear Fuel

Focused ion beam scanning electron microscopy (FIB-SEM) tomography has increasingly been utilized for acquiring three-dimensional (3D) microstructure features at the sub-micron scale in irradiated nuclear materials. This technique involves sequential ion beam slicing followed by electron beam imaging and compositional mapping using energy dispersive spectroscopy (EDS). Despite its growing use, several challenges persist. These include the time-intensive nature of data collection of EDS data, difficulties in distinguishing between various microstructures, and issues with image alignment. These challenges currently limit the broader application of FIB-SEM tomography in the field. To overcome these limitations, we propose using convolutional neural networks (CNNs) to automate microstructure identification in SEM images. Our study introduces a new framework for identifying microstructures in irradiated U-10Zr (wt. %) metallic fuel with limited annotated data. The framework includes the creation of a reliable annotated dataset with paired SEM and ground truth data from EDS maps, the applications of CNNs for microstructure identification, and the validation of model performance. Specifically, we employed the Segment Anything Model (SAM) to align SEM images with corresponding EDS maps and focused ion beam (FIB) tomography SEM data. We evaluate several models, including Patch-based U-Net, Attention U-Net, and Residual U-Net, finding that patch-based U-Net exhibits superior segmentation performance and consistency. This approach reduces reliance on EDS detectors and aids in accelerating nuclear material analysis process, highlighting the potential of advanced deep learning techniques to improve microstructural understanding in nuclear material. This is the first framework to integrate SAM and Patch-based CNN models for semantic segmentation of irradiated nuclear materials, with potential applicability to other tomography datasets.

36 - MATERIALS SCIENCE↗
DOE OSTI2026

Machine learning inversion of interatomic force constants from single-crystal inelastic neutron scattering

Atomic vibrations govern many macroscopic properties of materials, but experiments to comprehensively probe them remain challenging. Inelastic neutron scattering (INS) is a powerful technique to map phonon dispersions in crystals, especially when leveraging modern time-of-flight (ToF) spectrometers with large detectors. However, efficiently and robustly extracting interatomic force constants (FCs) parameterizing phonon dynamics from experimental spectra remains a bottleneck due to the complexity and high dimensionality of ToF INS datasets. Here, we present a machine learning approach for the direct inversion of FCs from single-crystal INS measurements. The framework leverages synthetic training data generated using universal machine-learned force fields and an efficient physics-based forward model. We benchmark two neural architectures–one emphasizing structured latent representation learning and the other direct, supervised spectral regression–across simulated datasets for two materials under idealized and noisy conditions. The latent-representation model is subsequently applied to experimental single-crystal INS data on germanium. The model is shown to reproduce FCs derived from both first-principles simulations and from iterative optimization, and furthermore achieves reliable inference even from sparse, single-orientation measurements representing short data acquisitions. Analysis of the learned latent space reveals semantically continuous and physically interpretable encodings that support strong cross-domain generalization. By bridging theoretical and experimental domains, we establish a path toward rapid inversion of experimental spectra and data-driven interpretation of temperature-dependent lattice dynamics.

42 ENGINEERING↗
DOE OSTI2026

Automatic Lane-Level Road Network Extraction from Aerial Imagery for Transportation Digital Twins

Accurate road networks are essential for credible traffic microsimulation and transportation digital twins, yet high-definition maps are often difficult to obtain due to limited availability, high cost, or proprietary restrictions. Some build networks from crowdsourced data, such as OpenStreetMap, but these sources often contain geometric and semantic inconsistencies. Others create networks manually, a process that is labor-intensive and difficult to scale. To address these limitations, this work presents an end-to-end pipeline that automatically extracts georeferenced, lane-level road networks from publicly available high-resolution satellite imagery and converts them into simulation-ready assets. The developed end-to-end pipeline has three primary modules: (1) A computer-vision-based module first detects directed lane geometries and intersection layouts. (2) A heuristic-based topology construction module then identifies approach and exit legs and establishes conflict-free lane-to-lane connections. (3) Finally, an automatic simulation-building module converts the extracted network into standard formats, e.g., OpenDRIVE, and generates routable SUMO networks. The framework supports both complete network construction from scratch and local-scale refinement of existing networks through lane-count correction, transition recovery, and geometric regularization. The proposed pipeline provides a practical pathway to generate traffic simulation networks from satellite imagery, significantly reducing manual reconstruction effort and enabling scalable, continuously updated transportation digital twins.

Guo, Hetian [University of Georgia, Athens] (ORCID↗
DOE OSTI2024

PhysBERT: A text embedding model for physics scientific literature

The specialized language and complex concepts in physics pose significant challenges for information extraction through Natural Language Processing (NLP). Central to effective NLP applications is the text embedding model, which converts text into dense vector representations for efficient information retrieval and semantic analysis. In this work, we introduce PhysBERT, the first physics-specific text embedding model. Pre-trained on a curated corpus of 1.2 × 106 arXiv physics papers and fine-tuned with supervised data, PhysBERT outperforms leading general-purpose models on physics-specific tasks, including the effectiveness in fine-tuning for specific physics subdomains.

Hellert, Thorsten (ORCID:0000000227970926)↗
DOE OSTI2025

The CanBikeCO Full Pilot: Long-Term Results and Analysis From an E-Bike Program in Colorado, USA

Personal micromobility devices like bicycles, e-bikes, and scooters are low- or zero-energy alternatives to single-occupancy vehicles. However, a lack of data has led to a dearth of data-driven research on personally owned e-bike usage. We present longitudinal findings from the CanBikeCO program, focused on e-bike adoption and use across demographics, trip characteristics, and geographies in the state of Colorado. CanBikeCO recorded travel survey data from low-income individuals provided with personal e-bikes by the Colorado Energy Office in six communities across Colorado from July 2021 to December 2022. The data were collected using a custom instance of the National Renewable Energy Laboratory OpenPATH platform, which combines passive data collection with semantic information such as trip mode and purpose labels. To our knowledge, there are no prior travel survey data on personally owned e-bikes with this range and scope. Insights from this unique dataset include: (i) work trips were 17% more likely than average trips to be taken on an e-bike, (ii) e-bikes were most often reported to replace cars (34% of e-bike trips) and other personal micromobility devices (22%), and (iii) participants favored walking for trips less than 1 mile, e-bikes for trips of 1-3 miles, and e-bikes, cars, or shared rides for trips of 3-20 miles. The data used to generate these results have been made available in the Transportation Secure Data Center. We find e-bike use is appealing across age groups and may be related to characteristics of land use, urban form, occupation, income, and car ownership. We conclude for this population that the energy demand added by e-bike use (induced demand and replacing non-motorized modes) is outweighed by the reduction in energy demand from replacement of single-occupancy vehicle trips with e-bike trips. Our findings suggest considerable potential for energy savings from personal e-bike ownership.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2024

Unpaired image translation to mitigate domain shift in liquid argon time projection chamber detector responses

Deep learning algorithms often are developed and trained on a training dataset and deployed on test datasets. Any systematic difference between the training and a test dataset may severely degrade the final algorithm performance on the test dataset—what is known as the domain shift problem . This issue is prevalent in many scientific domains where algorithms are trained on simulated data but applied to real-world datasets. Typically, the domain shift problem is solved through various domain adaptation (DA) methods. However, these methods are often tailored for a specific downstream task, such as classification or semantic segmentation, and may not easily generalize to different tasks. This work explores the feasibility of using an alternative way to solve the domain shift problem that is not specific to any downstream algorithm. The proposed approach relies on modern Unpaired Image-to-Image (UI2I) translation techniques, designed to find translations between different image domains in a fully unsupervised fashion. In this study, the approach is applied to a domain shift problem commonly encountered in Liquid Argon Time Projection Chamber (LArTPC) detector research when seeking a way to translate samples between two differently distributed LArTPC detector datasets deterministically. This translation allows for mapping real-world data into the simulated data domain where the downstream algorithms can be run with much less domain-shift-related performance degradation. Conversely, using the translation from the simulated data to a real-world domain can increase the realism of the simulated dataset and reduce the magnitude of any systematic uncertainties. To evaluate the quality of the translations, we use both pixel-wise metrics and a downstream task to measure the effectiveness of UI2I methods for mitigating the domain shift problem. We adapted several popular UI2I translation algorithms to work on scientific data and demonstrated the viability of these techniques for solving the domain shift problem with LArTPC detector data. To facilitate further development of DA techniques for scientific datasets, the ‘Simple Liquid-Argon Track Samples’ dataset used in this study is also published.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Describing Point Defect Topology in 2D Energy Materials through Computer Vision

Point defects such as vacancies and impurity atoms strongly impact the performance of 2D materials. Traditional efforts often rely on manual detection, a process that is time-intensive, prone to human error, and challenging to scale. Here we leverage machine learning (ML) methods to identify and quantify vacancies within 2D transition metal carbides (Ti3C2, MXenes), aiming to expedite detection while improving accuracy. MXenes exhibit valuable defect-defined electrochemical properties, but we currently lack statistical understanding of defect topology needed to fully harness these materials. Here we employ a convolutional neural network for semantic segmentation of experimental MXene images, opening an opportunity to conduct a rigorous statistical study on defect hierarchy while investigating local relaxation in the lattice. We show how the integration of ML can yield fundamental insight into point defects, providing a powerful tool that will play an increasingly crucial role in the future of materials science.

2d materials↗