Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Knowledge Graph”

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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Using spatio-temporal graph neural networks to estimate fleet-wide photovoltaic performance degradation patterns

Accurate estimation of photovoltaic (PV) system performance is crucial for determining its feasibility as a power generation technology and financial asset. PV-based energy solutions offer a viable alternative to traditional energy resources due to their superior Levelized Cost of Energy (LCOE). A significant challenge in assessing the LCOE of PV systems lies in understanding the Performance Loss Rate (PLR) for large fleets of PV systems. Estimating the PLR of PV systems becomes increasingly important in the rapidly growing PV industry. Precise PLR estimation benefits PV users by providing real-time monitoring of PV module performance, while explainable PLR estimation assists PV manufacturers in studying and enhancing the performance of their products. However, traditional PLR estimation methods based on statistical models have notable drawbacks. Firstly, they require user knowledge and decision-making. Secondly, they fail to leverage spatial coherence for fleet-level analysis. Additionally, these methods inherently assume the linearity of degradation, which is not representative of real world degradation. To overcome these challenges, we propose a novel graph deep learning-based decomposition method called the Spatio-Temporal Graph Neural Network for fleet-level PLR estimation (PV-stGNN-PLR). PV-stGNN-PLR decomposes the power timeseries data into aging and fluctuation components, utilizing the aging component to estimate PLR. PV-stGNN-PLR exploits spatial and temporal coherence to derive PLR estimation for all systems in a fleet and imposes flatness and smoothness regularization in loss function to ensure the successful disentanglement between aging and fluctuation. We have evaluated PV-stGNN-PLR on three simulated PV datasets consisting of 100 inverters from 5 sites. Experimental results show that PV-stGNN-PLR obtains a reduction of 33.9% and 35.1% on average in Mean Absolute Percent Error (MAPE) and Euclidean Distance (ED) in PLR degradation pattern estimation compared to the state-of-the-art PLR estimation methods.

14 SOLAR ENERGY↗
DOE OSTI2026

Improving the Performance of NEML2 with Modern Graph Compilation Backends

NEML2 vectorizes constitutive-model evaluation for large-scale multiphysics simulation, using PyTorch as its tensor backend so that a batch of material-point updates runs on CPU or GPU through a single implementation. In the two prior reports in this series it was a C++-native library, deployed through TorchScript tracing and just-in-time (JIT) compilation; it has since been rewritten from the ground up into a Python-native library deployed through Ahead-of-Time Inductor (AOTInductor), a modern PyTorch graph-compilation backend. The rewrite is driven by a persistent tension, not a language preference: NEML2 composes constitutive models at runtime from a registry of small, independently-authored pieces, and that flexibility is difficult to reconcile with the compile-time knowledge an efficient GPU kernel needs. This report documents the rewrite and the investment that accompanied it: the AOTInductor export pipeline that turns a Python-authored model into a portable, Python-free compiled artifact loadable from pure C++; the eager and compiled runtimes and the new implicit solver layer built on them; a head-to-head benchmark of legacy JIT against AOTInductor; the physics-model catalog and its worked examples; the developer tooling; and the corresponding overhaul of MOOSE’s NEML2 integration that lets MOOSE consume it. A central objective is to examine whether modern PyTorch graph-compilation backends are effective for MOOSE GPU integration. The benchmark answers directly: AOTInductor outperforms legacy JIT on every GPU scenario measured, by 1.0–4.5×. Modern graph-compilation backends are effective for MOOSE GPU integration, and AOTInductor specifically – not compilation in the abstract – is why.

Hu, Gary (Tianchen) [Argonne National Laboratory (↗
DOE OSTI2025

Integrated Framework of Multisource Data Fusion for Outage Location in Looped Distribution Systems

Accurate outage location is essential for expediting post-outage power restoration, minimizing outage duration, and enhancing the resilience of distribution networks. With the advent of advanced metering infrastructure, data-driven outage location methods have significantly advanced beyond traditional approaches that rely on manual inspections. However, existing methods still face critical challenges, like reliance on single-source data, limited ability to handle partially observable systems or difficulties with loop networks. To the best of our knowledge, no single approach has comprehensively addressed all of these challenges at once. To this end, this paper proposes a comprehensive multisource data fusion framework for outage locations via probabilistic graph networks. The framework consists of three key phases. First, a novel method for reconstituting distribution networks with loops is developed, transforming looped networks into multiple radial subnetworks that retain all outage causalities of the original network. Second, Bayesian network (BN) models are established for each subnetwork, integrating multiple data sources and network structures. Finally, a joint Gibbs sampling mechanism, featuring forward and backward information flow, is designed to merge data from separate BN models and maximize the utilization of limited evidence, ensuring accurate outage location identification. In conclusion, the framework was validated on two modified public test systems, and comparative studies confirmed its effectiveness.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

HDBind: encoding of molecular structure with hyperdimensional binary representations

Traditional methods for identifying “hit” molecules from a large collection of potential drug-like candidates rely on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have a significant limitation in that they require exceptional computing capabilities for even relatively small collections of molecules. Increasingly large and complex state-of-the-art deep learning approaches have gained popularity with the promise to improve the productivity of drug design, notorious for its numerous failures. However, as deep learning models increase in their size and complexity, their acceleration at the hardware level becomes more challenging. Hyperdimensional Computing (HDC) has recently gained attention in the computer hardware community due to its algorithmic simplicity relative to deep learning approaches. The HDC learning paradigm, which represents data with high-dimension binary vectors, allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas (computer vision, bioinformatics, mass spectrometery, remote sensing, edge devices, etc.). To the best of our knowledge, our work is the first to consider HDC for the task of fast and efficient screening of modern drug-like compound libraries. We also propose the first HDC graph-based encoding methods for molecular data, demonstrating consistent and substantial improvement over previous work. We compare our approaches to alternative approaches on the well-studied MoleculeNet dataset and the recently proposed LIT-PCBA dataset derived from high quality PubChem assays. We demonstrate our methods on multiple target hardware platforms, including Graphics Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs), showing at least an order of magnitude improvement in energy efficiency versus even our smallest neural network baseline model with a single hidden layer. Our work thus motivates further investigation into molecular representation learning to develop ultra-efficient pre-screening tools. We make our code publicly available at https://github.com/LLNL/hdbind.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

py-boomer v0.1.0

Py-BOOMER (Python Bayesian OWL Ontology MErgER in Python) is a probabilistic reasoning system for knowledge representation and ontological reasoning with uncertainty. Itnables reasoning over probabilistic facts and taxonomic relationships, finding the most likely consistent interpretation of potentially conflicting assertions. It uses a combination of graph-based reasoning and Bayesian probabilistic inference. Key features: Represent probabilistic ontological statements Reason over class subsumption hierarchies Evaluate class equivalence relationships Detect and resolve logical inconsistencies Calculate posterior probabilities for each assertion

Mungall, Chris [Lawrence Berkeley National Laborat↗
NASA NTRS2016

Toward a Unified Routing Framework for Delay-Tolerant Networking

Routing in Delay-/Disruption-Tolerant Networking (DTN) has long been recognized as a challenging research topic. The difficulty lies in the fact that link intermittency and network partitioning, possibly coupled with long delays, prevent the use of Internet solutions based on an up-to-date comprehensive knowledge of network topology, as communicated by routing protocols. In the literature on DTN routing, there is a dichotomy between solutions designed for deterministic (e.g., space flight) networks, such as Contact Graph Routing (CGR), and the wide variety of protocols designed for opportunistic terrestrial networks. After a discussion of the origin and motivations of this duality, the paper presents an opportunistic extension of CGR (OCGR). The aim is to try to resolve the DTN routing dichotomy by providing a unified approach suitable for all DTN environments.

Routing↗
NASA NTRS2013

Autonomous Cryogenic Load Operations: Knowledge-Based Autonomous Test Engineer

The Knowledge-Based Autonomous Test Engineer (KATE) program has a long history at KSC. Now a part of the Autonomous Cryogenic Load Operations (ACLO) mission, this software system has been sporadically developed over the past 20 years. Originally designed to provide health and status monitoring for a simple water-based fluid system, it was proven to be a capable autonomous test engineer for determining sources of failure in the system. As part of a new goal to provide this same anomaly-detection capability for a complicated cryogenic fluid system, software engineers, physicists, interns and KATE experts are working to upgrade the software capabilities and graphical user interface. Much progress was made during this effort to improve KATE. A display of the entire cryogenic system's graph, with nodes for components and edges for their connections, was added to the KATE software. A searching functionality was added to the new graph display, so that users could easily center their screen on specific components. The GUI was also modified so that it displayed information relevant to the new project goals. In addition, work began on adding new pneumatic and electronic subsystems into the KATE knowledge base, so that it could provide health and status monitoring for those systems. Finally, many fixes for bugs, memory leaks, and memory errors were implemented and the system was moved into a state in which it could be presented to stakeholders. Overall, the KATE system was improved and necessary additional features were added so that a presentation of the program and its functionality in the next few months would be a success.

Schrading, J. Nicolas↗
NASA NTRS1987

Off-line programming motion and process commands for robotic welding of Space Shuttle main engines

The off-line-programming software and hardware being developed for robotic welding of the Space Shuttle main engine are described and illustrated with diagrams, drawings, graphs, and photographs. The menu-driven workstation-based interactive programming system is designed to permit generation of both motion and process commands for the robotic workcell by weld engineers (with only limited knowledge of programming or CAD systems) on the production floor. Consideration is given to the user interface, geometric-sources interfaces, overall menu structure, weld-parameter data base, and displays of run time and archived data. Ongoing efforts to address limitations related to automatic-downhand-configuration coordinated motion, a lack of source codes for the motion-control software, CAD data incompatibility, interfacing with the robotic workcell, and definition of the welding data base are discussed.

Ruokangas, C. C.↗
DOE OSTI2022

Projecting the Thermal Response in a HTGR-Type System during Conduction Cooldown Using Graph-Laplacian Based Machine Learning

Accurate prediction of an off-normal event in a nuclear reactor is dependent upon the availability of sensory data, reactor core physical condition, and understanding of the underlying phenomenon. This work presents a method to project the data from some discrete sensory locations to the overall reactor domain during conduction cooldown scenarios similar to High Temperature Gas-cooled Reactors (HTGRs). The existing models for conductive cooldown in a heterogeneous multi-body system, such as an assembly of prismatic blocks or pebble beds relies on knowledge of the thermal contact conductance, requiring significant knowledge of local thermal contacts and heat transport possibilities across those contacts. With a priori knowledge of bulk geometry features and some discrete sensors, a machine learning approach was devised. The presented work uses an experimental facility to mimic conduction cooldown with an assembly of 68 cylindrical rods initially heated to 1200 K. High-fidelity temperature data were collected using an infrared (IR) camera to provide training data to the model and validate the predicted temperature data. The machine learning approach used here first converts the macroscopic bulk geometry information into Graph-Laplacian, and then uses the eigenvectors of the Graph-Laplacian to develop Kernel functions. Support vector regression (SVR) was implemented on the obtained Kernels and used to predict the thermal response in a packed rod assembly during a conduction cooldown experiment. The usage of SVR modeling differs from most models today because of its representation of thermal coupling between rods in the core. When trained with thermographic data, the average normalized error is less than 2% over 400 s, during which temperatures of the assembly have dropped by more than 500 K. The rod temperature prediction performance was significantly better for rods in the interior of the assembly compared to those near the exterior, likely due to the model simplification of the surroundings.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2023

A unified understanding of minimum lattice thermal conductivity

Here, we propose a first-principles model of minimum lattice thermal conductivity ($κ^{min}_L$) based on a unified theoretical treatment of thermal transport in crystals and glasses. We apply this model to thousands of inorganic compounds and find a universal behavior of $κ^{min}_L$ in crystals in the high-temperature limit: The isotropically averaged $κ^{min}_L$ is independent of structural complexity and bounded within a range from ~0.1 to ~2.6 W/(m K), in striking contrast to the conventional phonon gas model which predicts no lower bound. We unveil the underlying physics by showing that for a given parent compound, $κ^{min}_L$ is bounded from below by a value that is approximately insensitive to disorder, but the relative importance of different heat transport channels (phonon gas versus diffuson) depends strongly on the degree of disorder. Moreover, we propose that the diffuson-dominated $κ^{min}_L$ in complex and disordered compounds might be effectively approximated by the phonon gas model for an ordered compound by averaging out disorder and applying phonon unfolding. With these insights, we further bridge the knowledge gap between our model and the well-known Cahill–Watson–Pohl (CWP) model, rationalizing the successes and limitations of the CWP model in the absence of heat transfer mediated by diffusons. Finally, we construct graph network and random forest machine learning models to extend our predictions to all compounds within the Inorganic Crystal Structure Database (ICSD), which were validated against thermoelectric materials possessing experimentally measured ultralow κ L . Our work offers a unified understanding of $κ^{min}_L$, which can guide the rational engineering of materials to achieve .

42 ENGINEERING↗
NASA NTRS1972

Self-contamination and environment of an orbiting spacecraft

The flux of molecules emitted by a spacecraft and subsequently reflected to its surface was investigated. The reflection occurs upon collision of the outgassed molecules with ambient molecules. Evaluation of the flux was based on a knowledge of the spacecraft outgassing rate, the spacecraft dimensions, and the orbit parameters. Condensation rates and adsorption layers on critical surfaces were calculated from the knowledge of this flux and the nature and temperature of the gas and the surface. Based on estimated and measured emission rates, calculation of these parameters was performed for a number of spacecraft. The relationships and graphs developed allow an estimate of several important parameters for an orbiting spacecraft to be made. The pressures and densities at various distances from the spacecraft, as produced by the surrounding ambient molecules and by the spacecraft's own outgassing, are presented. The pressure and density produced by the outgassing can be obtained as a function of time if the behavior of the outgassing with time is known. The number of desorbed molecules ionized by impact with ambient charged particles and the effect of the spacecraft's electric field on polarized desorbed molecules were considered.

Scialdone, J. J.↗
NASA NTRS1973

Path-length distributions of photons diffusely reflected from a semi-infinite atmosphere.

Knowledge of the probability distribution of photon path lengths for diffuse reflection in the continuum from a planetary atmosphere greatly simplifies computation of the shape and strength of absorption lines formed under conditions of multiple scattering. Graphs of this probability distribution are presented for diffuse reflection from a semiinfinite conservative atmosphere for both isotropic and forward-directed scattering, and for several angles of incidence and reflection. Graphs showing distributions for the case of absorption in the continuum are also discussed. The distributions are obtained from an analytic expression for the path-length distribution for nth-order scattering and a tabulation of nth-order intensities. It is found that te shape of the probability distribution depends markedly on the degree of forward scattering due to the phase function. The results are compared with those obtained by a Monte Carlo computation of the sort which might be used if tabulated nth-order intensities were not available.

Appleby, J. F.↗
NASA NTRS1991

System monitoring and diagnosis with qualitative models

A substantial foundation of tools for model-based reasoning with incomplete knowledge was developed: QSIM (a qualitative simulation program) and its extensions for qualitative simulation; Q2, Q3 and their successors for quantitative reasoning on a qualitative framework; and the CC (component-connection) and QPC (Qualitative Process Theory) model compilers for building QSIM QDE (qualitative differential equation) models starting from different ontological assumptions. Other model-compilers for QDE's, e.g., using bond graphs or compartmental models, have been developed elsewhere. These model-building tools will support automatic construction of qualitative models from physical specifications, and further research into selection of appropriate modeling viewpoints. For monitoring and diagnosis, plausible hypotheses are unified against observations to strengthen or refute the predicted behaviors. In MIMIC (Model Integration via Mesh Interpolation Coefficients), multiple hypothesized models of the system are tracked in parallel in order to reduce the 'missing model' problem. Each model begins as a qualitative model, and is unified with a priori quantitative knowledge and with the stream of incoming observational data. When the model/data unification yields a contradiction, the model is refuted. When there is no contradiction, the predictions of the model are progressively strengthened, for use in procedure planning and differential diagnosis. Only under a qualitative level of description can a finite set of models guarantee the complete coverage necessary for this performance. The results of this research are presented in several publications. Abstracts of these published papers are presented along with abtracts of papers representing work that was synergistic with the NASA grant but funded otherwise. These 28 papers include but are not limited to: 'Combined qualitative and numerical simulation with Q3'; 'Comparative analysis and qualitative integral representations'; 'Model-based monitoring of dynamic systems'; 'Numerical behavior envelopes for qualitative models'; 'Higher-order derivative constraints in qualitative simulation'; and 'Non-intersection of trajectories in qualitative phase space: a global constraint for qualitative simulation.'

Kuipers, Benjamin↗
DOE OSTI2025

Meta-Learning Enhanced Physics-Informed Graph Attention Convolutional Network for Distribution Power System State Estimation

Promptly perceiving distribution system states is challenged by frequent topology changes and uncertain power injections. To address these issues, a Meta-learning enhanced physics-informed graph attention convolutional network (Meta-PIGACN) model is proposed to handle topological variability in distribution system state estimation (DSSE). Specifically, physics information is integrated into the graph convolutional network, enabling a physics-informed edge-weighting process that incorporates physical information to control the aggregation of neighboring nodes. Besides, the graph attention mechanism automatically adjusts the importance of different neighboring nodes, allowing the capture and preservation of inherent system features across varying topologies, thereby improving state estimation accuracy. Furthermore, meta-learning is proposed to acquire empirical knowledge across multiple topologies so that the model can rapidly adapt to new configurations through iterative gradient descent updates even in large-scale systems. In conclusion, the simulation results based on the 33/118/1746-node distribution systems show the high accuracy and efficiency of the proposed model.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Learning Distribution Grid Topologies: A Tutorial

Unveiling feeder topologies from data is of paramount importance to advance situational awareness and proper utilization of smart resources in power distribution grids. This tutorial summarizes, contrasts, and establishes useful links between recent works on topology identification and detection schemes that have been proposed for power distribution grids. The primary focus is to highlight methods that overcome the limited availability of measurement devices in distribution grids, while enhancing topology estimates using conservation laws of power-flow physics and structural properties of feeders. Grid data from phasor measurement units or smart meters can be collected either passively in the traditional way, or actively, upon actuating grid resources and measuring the feeder's voltage response. Analytical claims on feeder identifiability and detectability are reviewed under disparate meter placement scenarios. Such topology learning claims can be attained exactly or approximately so via algorithmic solutions with various levels of computational complexity, ranging from least-squares fits to convex optimization problems, and from polynomial-time searches over graphs to mixed-integer programs. Although the emphasis is on radial single-phase feeders, extensions to meshed and/or multiphase circuits are sometimes possible and discussed. Here this tutorial aspires to provide researchers and engineers with knowledge of the current state-of-the-art in tractable distribution grid learning and insights into future directions of work.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

graphenv: a Python library for reinforcement learning on graph search spaces

Many important and challenging problems in combinatorial optimization (CO) can be expressed as graph search problems, in which graph vertices represent full or partial solutions and edges represent decisions that connect them. Graph structure not only introduces strong relational inductive biases for learning (Battaglia et al., 2018) - in this context, by providing a way to explicitly model the value of transitioning (along edges) between one search state (vertex) and the next - but lends itself to problems both with and without clearly defined algebraic structure. For example, classic CO problems on graphs such as the Traveling Salesman Problem (TSP) can be expressed as either pure graph search or integer programs. Other problems, however, such as molecular optimization, do no have concise algebraic formulations and yet are readily implemented as a graph search (V. et al., 2022; Zhou et al., 2019). Such "model-free" problems constitute a large fraction of modern reinforcement learning (RL) research owing to the fact that it is often much easier to write a forward simulation that expresses all of the state transitions and rewards, than to write down the precise mathematical expression of the full optimization problem. In the case of molecular optimization, for example, one can use domain knowledge alongside existing software libraries to model the effect of adding a single bond or atom to an existing but incomplete molecule, and let the RL algorithm build a model of how good a given decision is by "experiencing" the simulated environment many times through. In contrast, a model-based mathematical formulation that fully expresses all the chemical and physical constraints is intractable. In recent years, RL has emerged as an effective paradigm for optimizing searches over graphs and led to state-of-the-art heuristics for games like Go and chess, as well as for classical CO problems such as the TSP. This combination of graph search and RL, while powerful, requires non-trivial software to execute, especially when combining advanced state representations such as Graph Neural Networks (GNN) with scalable RL algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Track Seeding and Labelling with Embedded-space Graph Neural Networks

To address the unprecedented scale of HL-LHC data, the Exa.TrkX project is investigating a variety of machine learning approaches to particle track reconstruction. The most promising of these solutions, graph neural networks (GNN), process the event as a graph that connects track measurements (detector hits corresponding to nodes) with candidate line segments between the hits (corresponding to edges). Detector information can be associated with nodes and edges, enabling a GNN to propagate the embedded parameters around the graph and predict node-, edge- and graph-level observables. Previously, message-passing GNNs have shown success in predicting doublet likelihood, and we here report updates on the state-of-the-art architectures for this task. In addition, the Exa.TrkX project has investigated innovations in both graph construction, and embedded representations, in an effort to achieve fully learned end-to-end track finding. Hence, we present a suite of extensions to the original model, with encouraging results for hitgraph classification. In addition, we explore increased performance by constructing graphs from learned representations which contain non-linear metric structure, allowing for efficient clustering and neighborhood queries of data points. We demonstrate how this framework fits in with both traditional clustering pipelines, and GNN approaches. The embedded graphs feed into high-accuracy doublet and triplet classifiers, or can be used as an end-to-end track classifier by clustering in an embedded space. A set of post-processing methods improve performance with knowledge of the detector physics. Finally, we present numerical results on the TrackML particle tracking challenge dataset, where our framework shows favorable results in both seeding and track finding.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗