Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “scratch”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

SmileyLlama: modifying large language models for directed chemical space exploration

Here we show that large language models (LLMs) can be transformed via supervised fine-tuning of engineered prompts into SmileyLlama for exploring the chemical space of drug molecules. We benchmark SmileyLlama against pretrained LLMs and chemical language models trained from scratch for generating valid and novel drug-like molecules, and use direct preference optimization to both improve SmileyLlama’s adherence to a prompt and as part of the iMiner reinforcement learning framework to predict molecules with optimized three-dimensional conformations and high binding affinity to drug targets. By training an LLM to speak directly as a chemical language model, while retaining most of its natural language capabilities, we show that SmileyLlama can reliably generate molecules with user-specified properties rather than acting only as a chatbot with knowledge of chemistry or as a virtual assistant. While SmileyLlama is geared toward drug discovery, the supervised fine-tuning/direct preference optimization/LLM framework can be extended to other chemical, biological and materials applications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Enhancing generative molecular design via uncertainty-guided fine-tuning of variational autoencoders

In recent years, deep generative models have been successfully applied to various molecular design tasks, particularly in the life and materials sciences. One critical challenge for pre-trained generative molecular design (GMD) models is to fine-tune them to be better suited for downstream design tasks that aim at optimizing specific molecular properties. However, redesigning and training an existing effective generative model from scratch for each new design task are impractical. Furthermore, the black-box nature of typical downstream tasks that involve property prediction makes it nontrivial to optimize the generative model in a task-specific manner. In this work, we propose an uncertainty-guided fine-tuning strategy that can effectively enhance a pre-trained variational autoencoder (VAE) for GMD through performance feedback in an active learning setting. The strategy begins by quantifying the model uncertainty of the generative model using an efficient active subspace-based UQ (uncertainty quantification) scheme. Next, the decoder diversity within the characterized model uncertainty class is explored to expand the viable space of molecular generation. The low-dimensionality of the active subspace makes this exploration tractable using a black-box optimization scheme, which in turn enables us to identify and leverage a diverse set of high-performing models to generate enhanced molecules. Empirical results across six target molecular properties using multiple VAE-based generative models demonstrate that our uncertainty-guided fine-tuning strategy consistently leads to improved models that outperform the original pre-trained models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Charting the chemical space of Zintl phases with graph neural networks and bonding insights

A large number of Zintl phases have been discovered by solid-state chemists driven by empirical knowledge, chemical intuition and in some cases, through serendipitous accidents. These discoveries have only scratched the surface, given the vast compositional and structural diversity that Zintl phases can accommodate. The large chemical space of Zintl phases, as well as intermetallic compounds in general, remain under-explored. Here, we use graph neural networks and the upper bound energy minimization approach to efficiently scan a large chemical space of >90 000 hypothetical Zintl phases and accurately discover 1810 new thermodynamically stable phases with 90% precision, as validated with first-principles calculations. We show that our approach is more than 2× more accurate in predicting DFT stability than M3GNet (40% precision) on the same dataset. Using a random forest model and SHAP analysis, we demonstrate the critical role of ionic bonding in the thermodynamic stability of Zintl phases. Our results not only expand the known chemical landscape of Zintl phases but also highlight the efficacy of machine learning frameworks combined with domain knowledge in uncovering chemically meaningful insights across complex intermetallics.

36 MATERIALS SCIENCE↗
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 OSTI2026

MATEY: multiscale adaptive transformer models for spatiotemporal physical systems

Accurate representation of the multiscale features in spatiotemporal physical systems using vision transformer architectures requires extremely long, computationally prohibitive token sequences. To address this issue, we propose two novel adaptive tokenization schemes that dynamically adjust patch sizes based on local features: one ensures convergent behavior to uniform patch refinement, while the other offers better computational efficiency. Moreover, we present a set of spatiotemporal attention schemes, where the temporal or axial spatial dimensions are decoupled, to evaluate their baseline computational and data efficiencies and to determine whether adaptive tokenization can improve this performance. We assess the performance of the proposed multiscale adaptive model, MATEY, in a sequence of experiments. Compared to a full spatiotemporal attention scheme or a scheme that decouples only the temporal dimension, we find that fully decoupled axial attention is less efficient and expressive, requiring more training time and model parameters to achieve the same accuracy. The experiments on the adaptive tokenization schemes show that, compared to a uniformly refined model, the proposed schemes achieve comparable or improved accuracy at a much lower cost in the tested two-dimensional settings. While the asymptotic analysis suggests the potential for favorable scaling, empirical validation at substantially longer sequence lengths remains to be performed in future work. Finally, we demonstrate in two fine-tuning tasks featuring different physics that models pretrained on PDEBench data outperform the ones trained from scratch, especially in the low data regime with frozen attention.

adaptive tokenization↗
DOE OSTI2026

From natural language to control signals: a conceptual framework for semantic channel finding in complex experimental infrastructure

Modern experimental platforms such as particle accelerators, fusion devices, telescopes, and industrial process control systems expose tens to hundreds of thousands of control and diagnostic channels, accumulated over decades of hardware evolution. Operators and AI systems alike depend on informal expert knowledge, inconsistent naming conventions, and scattered documentation to locate the signals required for monitoring, troubleshooting, and automated control, creating a persistent bottleneck for reliability, scalability, and emerging language-model-driven interfaces. We formalize semantic channel finding, the task of mapping natural-language intent to concrete control-system signals, as a general problem in complex experimental infrastructure, and introduce a four-paradigm conceptual framework to guide architecture selection based on facility-specific data regimes. The paradigms span (i) direct in-context lookup over small, curated channel dictionaries, (ii) constrained hierarchical navigation through structured trees, (iii) interactive agent exploration using iterative reasoning and tool-based database queries, and (iv) ontology-grounded semantic search that decouples channel meaning from facility-specific naming conventions. We demonstrate the practical feasibility of each paradigm through proof-of-concept implementations at four operational facilities spanning two orders of magnitude in scale: from compact free-electron lasers to large synchrotron light sources, operating under diverse control-system architectures ranging from clean hierarchical naming schemes to legacy environments with decades of heterogeneous conventions. Where evaluated against expert-curated operational queries, these instantiations achieve 90%–97% accuracy, validating the framework’s applicability across real-world deployment scenarios. To accelerate adoption across the broader scientific and industrial control-system community, we release open-source, plug-and-play implementations of all three interactive paradigms-direct lookup, hierarchical navigation, and middle-layer exploration-within the Osprey framework, together with tools for channel database generation, interactive testing, and minimal-configuration deployment. This work establishes semantic channel finding as a foundational capability for human-centric and agentic AI interfaces at large-scale facilities, providing both a systematic framework for architecture design and practical resources to enable adoption without building custom infrastructure from scratch.

channel finding↗
DOE OSTI2025

Large-scale deep learning for metastasis detection in pathology reports

Objectives No existing algorithm can reliably identify metastasis from pathology reports across multiple cancer types and the entire US population. In this study, we develop a deep learning model that automatically detects patients with metastatic cancer by using pathology reports from many laboratories and of multiple cancer types. Materials and Methods We use 60 471 unstructured pathology reports from 4 Surveillance, Epidemiology, and End Results (SEER) registries. The reports were coded into 1 of 3 labels: metastasis negative, metastases positive, or metastasis undetermined. We utilize a task-specific deep neural network trained from scratch and compare its performance with a widely used large language model (LLM). Results Our deep learning architecture trained on task-specific data outperforms a general-purpose LLM, with a recall of 0.894 compared to 0.824. We quantified model uncertainty and used it to defer reports for human review. We found that retaining 72.9% of reports increased recall from 0.894 to 0.969. Discussion A smaller deep learning architecture trained on task-specific data outperforms a general LLM. Equally critical to model performance is the incorporation of uncertainty quantification, achieved here through an abstention mechanism. Conclusions This study’s finding demonstrate the feasibility of developing algorithms to automatically identify metastatic cancer cases from unstructured pathology reports.

machine learning↗
DOE OSTI2025

Coefficient-to-Basis Network: a fine-tunable operator learning framework for inverse problems with adaptive discretizations and theoretical guarantees

We propose a Coefficient-to-Basis Network (C2BNet), a novel framework for solving inverse problems within the operator learning paradigm. C2BNet efficiently adapts to different discretizations through fine-tuning, using a pre-trained model to significantly reduce computational cost while maintaining high accuracy. Unlike traditional approaches that require retraining from scratch for new discretizations, our method enables seamless adaptation without sacrificing predictive performance. Furthermore, we establish theoretical approximation and generalization error bounds for C2BNet by exploiting low-dimensional structures in the underlying datasets. Our analysis demonstrates that C2BNet adapts to low-dimensional structures without relying on explicit encoding mechanisms, highlighting its robustness and efficiency. To validate our theoretical findings, we conducted extensive numerical experiments that showcase the superior performance of C2BNet on several inverse problems. The results confirm that C2BNet effectively balances computational efficiency and accuracy, making it a promising tool to solve inverse problems in scientific computing and engineering applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Efficient six-dimensional phase space reconstructions from experimental measurements using generative machine learning

Next-generation accelerator concepts, which hinge on the precise shaping of beam distributions, demand equally precise diagnostic methods capable of reconstructing beam distributions within six-dimensional position-momentum spaces. However, the characterization of intricate features within six-dimensional beam distributions using current diagnostic techniques necessitates a substantial number of measurements, using many hours of valuable beam time. Novel phase space reconstruction techniques are needed to reduce the number of measurements required to reconstruct detailed, high-dimensional beam features in order to resolve complex beam phenomena and as a feedback in precision beam shaping applications. In this study, we present a novel approach to reconstructing detailed six-dimensional phase space distributions from experimental measurements using generative machine learning and differentiable beam dynamics simulations. We demonstrate that this approach can be used to resolve six-dimensional phase space distributions from scratch, using basic beam manipulations and as few as 20 two-dimensional measurements of the beam profile. We also demonstrate an application of the reconstruction method in an experimental setting at the Argonne Wakefield Accelerator, where it is able to reconstruct the beam distribution and accurately predict previously unseen measurements 75× faster than previous methods.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

DeepAndes: A Self-Supervised Vision Foundation Model for Multispectral Remote Sensing Imagery of the Andes

By mapping sites at large scales usingremotely sensed data, archaeologists can generate unique insights into long-term demographic trends, interregional social networks, and human adaptations in the past. Remote sensing surveys complement field-based approaches, and their reach can be especially great when combined with deep learning and computer vision techniques. However, conventional supervised deep learning methods face challenges in annotating fine-grained archaeological features at scale. In addition, while recent vision foundation models have shown remarkable success in learning large-scale remote sensing data with minimal annotations, most off-the-shelf solutions are designed for RGB images rather than multispectral satellite imagery, such as the eight-band data used in our study. In this article, we introduce DeepAndes, a transformer-based vision foundation model trained on three million multispectral satellite images, specifically tailored for Andean archaeology. DeepAndes incorporates a customized DINOv2 self-supervised learning algorithm optimized for eight-band multispectral imagery, marking the first foundation model designed explicitly for the Andes region. We evaluate its image understanding performance through imbalanced image classification, image instance retrieval, and pixel-level semantic segmentation tasks. Our experiments show that DeepAndes achieves superior F1 scores, mean average precision, and Dice scores in few-shot learning scenarios, significantly outperforming models trained from scratch or pretrained on smaller datasets. This underscores the effectiveness of large-scale self-supervised pretraining in archaeological remote sensing.

Guo, Junlin [Vanderbilt Univ., Nashville, TN (Unit↗
DOE OSTI2025

A Unifying Framework to Enable Artificial Intelligence in High-Performance Computing Workflows

Current trends point to a future where large-scale scientific applications are tightly coupled high-performance computing/artificial intelligence (HPC/AI) hybrids. Hence, we urgently need to invest in creating a seamless, scalable framework where HPC and AI/machine learning can efficiently work together and adapt to novel hardware and vendor libraries without starting from scratch every few years. Finally, the current ecosystem and sparsely connected community are not sufficient to tackle these challenges, and we require a breakthrough catalyst for science similar to what PyTorch enabled for AI.

high-performance computing↗
DOE OSTI2025

Using a Large Language Model as a Building Block to Generate Usable Validation and Verification Suite for OpenMP

In the HPC area, both hardware and software move quickly. Often new hardware is developed and deployed, the corresponding software stack, including compilers and other tools, are under active development while leading edge software developers are working to port and tune their applications, all at the same time. While the software ecosystem is in flux, one of the key challenges for users is obtaining insight into the state of implementation of key features in the programming languages and models their applications are using – whether they have been implemented, and whether the implementation conforms to the specification, especially for newly implemented features (less tested by widespread use). OpenMP is one of the most prominent shared memory programming models used for on-node programming in HPC. With the shift towards accelerators (such as GPUs and FPGAs) and heterogeneous programming OpenMP features are getting more complex. It is natural to ask whether generative AI approaches, and large language models (LLMs) in particular, can help in producing validation and verification test suites to allow users better and faster insights into the availability and correctness of OpenMP features of interest. In this work, we explore the use of ChatGPT-4 to generate a suite of tests for OpenMP features. We have chosen a set of directives and clauses, a total of 78 combinations, which first appeared in OpenMP 3.0 (released in May 2008) but are also relevant for accelerators. We prompted ChatGPT to generate tests in the C and Fortran languages, for both host (CPU) and device (accelerator). On the Summit super-computer using the GNU implementation, we found that, of the 78 generated tests 67 C tests and 43 Fortran tests compiled successfully and fewer than those executed to completion. On further analysis we show that not all generated tests are valid. We document the process, results, and provide detailed analysis regarding the quality of tests generated. With the aim of providing input to a production quality validation and verification suite, we manually implement the corrections required to make the tests valid according to the current OpenMP specification. We quantify this effort as small, medium, or large, and record the lines of code changed to correct the invalid tests. With the corrected tests we validate recent implementations from HPE, AMD, and GNU on the Frontier supercomputer. Our experiment and subsequent analysis show that although LLMs are capable of producing HPC specific codes, they are limited by their understanding of the deeper semantics and restrictions of programming models such as OpenMP. Unsurprisingly more commonly used features have better support, while some OpenMP 3.0 directives such as sections and tasking are not universally supported on accelerators. We demonstrate that successful compilation and execution to completion are inadequate metrics for evaluating generated code and that, at this time, commodity LLMs require expert intervention for code verification. This points to gaps in the training data that is currently available for HPC. We demonstrate that with "small" effort 37% of generated invalid C tests and 63% of generated invalid Fortran tests could be corrected. This improves productivity of test generation as we circumvent writing from scratch and the common programming errors associated with it.

Pophale, Swaroop [ORNL] (ORCID:0000000185446367)↗
DOE OSTI2024

xesn: Echo state networks powered by Xarray and Dask

Xesn is a Python package that allows scientists to easily design Echo State Networks (ESNs) for forecasting problems. ESNs are a Recurrent Neural Network architecture introduced by Jaeger (2001) that are part of a class of techniques termed Reservoir Computing. One defining characteristic of these techniques is that all internal weights are determined by a handful of global, scalar parameters, thereby avoiding problems during backpropagation and reducing training time significantly. Because this architecture is conceptually simple, many scientists implement ESNs from scratch, leading to questions about computational performance. Xesn offers a straightforward, standard implementation of ESNs that operates efficiently on CPU and GPU hardware. The package leverages optimization tools to automate the parameter selection process, so that scientists can reduce the time finding a good architecture and focus on using ESNs for their domain application. Importantly, the package flexibly handles forecasting tasks for out-of-core, multi-dimensional datasets, eliminating the need to write parallel programming code. Xesn was initially developed to handle the problem of forecasting weather dynamics, and so it integrates naturally with Python packages that have become familiar to weather and climate scientists such as Xarray (Hoyer & Hamman, 2017). However, the software is ultimately general enough to be utilized in other domains where ESNs have been useful, such as in signal processing (Jaeger & Haas, 2004).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Bobfinchite, Na[(UO2)8O3(OH)11]·10H2O, a new Na-bearing member of the schoepite family

Abstract The new mineral bobfinchite (IMA2020-082), Na[(UO2)8O3(OH)11]·10H2O, was found in the Burro mine, Slick Rock district, San Miguel County, Colorado, U.S.A., where it occurs as an oxidation product of uraninite on asphaltite matrix in intimate association with gypsum, natrozippeite, metaschoepite, and uranopilite. Bobfinchite crystals are transparent to translucent, yellow, lozenge-shaped disks up to 0.3 mm wide. Crystals are flattened on [100] and exhibit the forms {100}, {011}, {021}, {021}, and {011}. Bobfinchite has a pale-yellow streak and emits very dim yellow fluorescence under 365 nm ultraviolet illumination. The crystals are brittle with very good {100} cleavage and irregular, stepped fracture. The Mohs hardness is ca. 2 based on scratch tests. The calculated density is 5.044 g/cm3 based on the empirical formula and 5.036 g/cm3 for the ideal formula. Bobfinchite is optically biaxial (–), with α = 1.690(5), β = 1.7205(5), and γ = 1.730(5) (white light). The measured 2V, estimated from the interference figure, is 55(5)° and the calculated value is 59.1°. Dispersion is moderate, r > v; orientation: X = a, Y = b, Z = c; pleochroism: X nearly colorless, Y yellow, Z yellow; X < Y ≈Z. Electron microprobe analysis provided the empirical formula (Na0.99Pb0.02)[(UO2)7.99O3(OH)11]·10H2O. The five strongest X-ray powder diffraction lines are [dobs in Å(I)(hkl)]: 7.34(100)(200), 3.59(50)(024), 3.23(60)(224), 3.18(36)(240), and 2.01(23)(624,551,208,640,346). Bobfinchite is orthorhombic, Pbcn, a = 14.6249(9), b = 14.0389(10), c = 16.6923(10) Å, V = 3427.2(4) Å3, and Z = 4. The structure of bobfinchite (R1 = 0.0330 for 3770 I > 4σI) is built from uranyl oxide-hydroxide sheets that adopt the fourmarierite topology, with interlayer Na+ and H2O groups. Both the sheet and interlayer topology mimic those observed in natural and synthetic Na-metaschoepites studied previously, and as seen in other uranyl oxide hydrate minerals, charge balance is achieved at specific sites in the sheet through the substitution O2– ↔ (OH)–.

Geochemistry & Geophysics↗
DOE OSTI2024

"Soiling, Cleaning, and Abrasion: The Results of the Five-Year Photovoltaic Glass Coating Field Study" [Slides]

External contamination ("soiling") of the incident surface is a major limiting factor for solar technologies. A 5-year field glass coupon study was conducted to better understand external contamination and its effects; compare cleaning methods and the use of preventative coatings; and explore the abrasion resulting from cleaning to advise on accelerated abrasion testing. Test sites included the cities of Dubai (UAE), Kuwait City (Kuwait), Mesa (AZ), Mumbai (India), and Sacramento (CA). Through the 5-year cumulative study, dry brush, water spray, and wet sponge and squeegee cleaning methods were compared to no cleaning. Optical microscopy was used to obtain images, including representative color images, grayscale images for object analysis, and oblique images for coating integrity assessment. A thresholding protocol was developed to analyze and distinguish specimens using the ImageJ software. Optical performance was quantified using a spectrophotometer, including comprehensive optical characterization (transmittance, reflectance, and absorptance in addition to forward- and back-scattering). Atomic force microscopy was used to verify the abrasion damage morphology, including the width and depth of surface scratches. Analysis of the results included correlation of optical performance and particle area coverage, rank order (by coating or location), and the acceleration factor for abrasion damage. The efficacy of external cleaning was more readily distinguished from the effectiveness of antisoiling coatings. The acceleration factor for dry brush cleaning of a porous silica coating was found to be on the order of unity.

14 SOLAR ENERGY↗
DOE OSTI2024

Study of Mechanical Grinding Effects on Niobium Surface

Mechanical grinding is commonly employed to eliminate surface defects such as scratches and pits from niobium cavity surfaces or sheets before cavity fabrication. Subsequently, chemically buffered polishing or electropolishing is often utilized to completely remove residues of the polishing media and any defects induced by mechanical grinding, ensuring a pristine surface. In this study, we conducted a systematic investigation to assess the influence of mechanical grinding using silicon carbide (SiC) and aluminium oxide (Al2O3) polishing media on niobium surfaces. Additionally, the study examines the effects of post-mechanical grinding chemical treatments on surface quality.

Chouhan, V.↗
DOE OSTI2025

DistOPF: Advanced Solutions for Distribution Optimal Power Flow Analysis - DistOPF v0.2 Documentation

To achieve an affordable and reliable energy system, research on power distribution system is often focused on integration of distributed generators, energy storage solution, EV charging, smart meters, and other advanced assets that may benefit from or require more advanced control and optimization techniques. Despite this focus on advanced distribution system topics, early researchers and grid scientists often start from scratch when developing optimization programs for power distribution systems. This report introduces DistOPF, a Python package that consolidates years of research into a versatile and modular tool. DistOPF provides researchers with essential capabilities to solve distribution system Optimal Power Flow (OPF) problems using standard network models. Additionally, it offers a platform to benchmark both new and existing algorithms against established test systems.

24 POWER TRANSMISSION AND DISTRIBUTION↗