Search NASA⌕ Search

SEARCH · Search NASA

Results for “Network Models”

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 577 records · Page 32

Compressing deep neural networks on FPGAs to binary and ternary precision with hls4ml

We present the implementation of binary and ternary neural networks in the hls4ml library, designed to automatically convert deep neural network models to digital circuits with FPGA firmware. Starting from benchmark models trained with floating point precision, we investigate different strategies to reduce the network's resource consumption by reducing the numerical precision of the network parameters to binary or ternary. We discuss the trade-off between model accuracy and resource consumption. In addition, we show how to balance between latency and accuracy by retaining full precision on a selected subset of network components. As an example, we consider two multiclass classification tasks: handwritten digit recognition with the MNIST data set and jet identification with simulated proton-proton collisions at the CERN Large Hadron Collider. The binary and ternary implementation has similar performance to the higher precision implementation while using drastically fewer FPGA resources.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Robustness of cluster synchronous patterns in small-world networks with inter-cluster co-competition balance

All edges in the classical Watts and Strogatz's small-world network model are unweighted and cooperative (positive). By introducing competitive (negative) inter-cluster edges and assigning edge weights to mimic more realistic networks, this paper develops a modified model which possesses co-competitive weighted couplings and cluster structures while maintaining the common small-world network properties of small average shortest path lengths and large clustering coefficients. Based on theoretical analysis, it is proved that the new model with inter-cluster co-competition balance has an important dynamical property of robust cluster synchronous pattern formation. More precisely, clusters will neither merge nor split regardless of adding or deleting nodes and edges, under the condition of inter-cluster co-competition balance. Numerical simulations demonstrate the robustness of the model against the increase of the coupling strength and several topological variations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Automating Traffic Microsimulation from SYNCHRO UTDF to SUMO

Modern transportation research relies on seamlessly integrating traffic signal data with robust network representation and simulation tools. This study presents utdf2gmns, an open-source Python tool that automates conversion of the Universal Traffic Data Format, including network representation, signalized intersections, and turning volumes into the General Modeling Network Specification (GMNS) Standard. The resulting GMNS-compliant network can be converted for microsimulation in SUMO. By automatically extracting intersection control parameters and aligning them with GMNS conventions, utdf2gmns minimizes manual preprocessing and data loss. utdf2gmns also integrates with the Sigma-X engine to extract and visualize key traffic control metrics, such as phasing diagrams, turning volumes, volume-tocapacity ratios, and control delays. This streamlined workflow enables efficient scenario testing, accurate model building, and consistent data management. Validated through case studies, utdf2gmns reliably models complex urban corridors, promoting reproducibility and standardization. Documentation is available on GitHub and PyPI, supporting easy integration and community engagement.

Luo, Roy [ORNL] (ORCID:0009000312909983)↗

Electronic structures and physical properties of Na 2 O doped silicate glass

Ab initio molecular dynamics has been applied to construct seven sodium silicate glass models with Na2O concentration ranging from 0 to 50 mol. %. The structures of the simulated (Na 2 O) x (SiO 2 ) 1-x glasses are critically analyzed and validated by comparing with available experimental data. Because the initial seed model is based on a near-perfect continuous random network model for amorphous SiO 2 with periodic boundaries, the structures of these silicate glasses are highly reliable. The electronic structure, interatomic bonding, and the mechanical and optical properties of seven models are calculated using the first-principles density functional method. In particular, a single quantum mechanical metric, the total bond order density (TBOD), is used to characterize the internal cohesion of sodium silicate glass. This is a significant step beyond the traditional analysis of glasses based purely on the geometric parameters. The TBOD value is found to decrease with increasing Na content, indicating the destruction of silica network connectivity. The calculated mass density and refractive index increase with x are in good agreement with experiment. The elastic coefficients and bulk mechanical properties exhibit a nonlinear variation in the series and depend greatly on the internal bonding and cohesion of the glass. The calculated Poisson's ratio indicates that the glass becomes more ductile with the addition of Na 2 O. Our results indicate that sodium silicate glass tends to be unstable for x greater than 0.4 due to the total destruction of the SiO 2 network.

36 MATERIALS SCIENCE↗

Chemical composition based machine learning model to predict defect formation in additive manufacturing

With a goal of exploiting additive manufacturing to improve the manufacturing of existing reactor materials, we developed a chemical composition-based machine learning model to predict the printability of any given alloy in laser powder bed fusion (L-PBF) using experimental data from peer-reviewed literature. We defined printability as the ability to avoid defects like cracking, balling, porosity, and lack of fusion, that are caused by thermal stresses (during solidification or liquation), molten pool disintegration into disconnected small beads or lack of heat input respectively. Our models predict the tendency of balling defect formation and porosity percentage for a given composition, under a given set of processing conditions. To predict the likelihood of balling defect, three models: a random forest classifier, a gradient boost regressor and a neural network were trained on a dataset containing both traditional alloys and high entropy alloys. The neural network model showed the highest accuracy of 92.3 % in predicting the balling defect formation. A random forest regressor, gradient boost regressor and neural network were trained and tested on a dataset of various alloys to predict porosity. The random forest regressor showed the best predictions with an R 2 score of 0.97. The models also revealed the relative importance of the input descriptors on defect-formation tendency. Of particular significance was the identification of carbon as an important element in determining the occurrence of balling and percent porosity in alloys like steel, as well as being moderately important to the percentage porosity in other alloys as well as steel. Manganese was also identified as a key descriptor for the percentage of porosity in steel and other alloys. Manganese’s low thermal conductivity and consistent presence in the dataset is the likely cause for its contribution. Carbon’s role is attributable to its relatively high specific heat and high melting temperature. In conclusion, our model serves as a swift, chemistry-based tool to design experiments and find modified compositions better suited for additive manufacturing.

36 MATERIALS SCIENCE↗

Classification of bacterial plasmid and chromosome derived sequences using machine learning

Plasmids are important genetic elements that facilitate horizonal gene transfer between bacteria and contribute to the spread of virulence and antimicrobial resistance. Most bacterial genome sequences in the public archives exist in draft form with many contigs, making it difficult to determine if a contig is of chromosomal or plasmid origin. Using a training set of contigs comprising 10,584 chromosomes and 10,654 plasmids from the PATRIC database, we evaluated several machine learning models including random forest, logistic regression, XGBoost, and a neural network for their ability to classify chromosomal and plasmid sequences using nucleotide k-mers as features. Based on the methods tested, a neural network model that used nucleotide 6-mers as features that was trained on randomly selected chromosomal and plasmid subsequences 5kb in length achieved the best performance, outperforming existing out-of-the-box methods, with an average accuracy of 89.38% ± 2.16% over a 10-fold cross validation. The model accuracy can be improved to 92.08% by using a voting strategy when classifying holdout sequences. In both plasmids and chromosomes, subsequences encoding functions involved in horizontal gene transfer—including hypothetical proteins, transporters, phage, mobile elements, and CRISPR elements—were most likely to be misclassified by the model. This study provides a straightforward approach for identifying plasmid-encoding sequences in short read assemblies without the need for sequence alignment-based tools.

59 BASIC BIOLOGICAL SCIENCES↗

Gray-Box Modeling for Distribution Systems With Inverter-Based Resources: Integrating Physics-Based and Data-Driven Approaches

Here, in this paper, we develop a novel gray-box modeling approach for distribution systems with inverter-based resources (IBRs). The proposed gray-box modeling method aims to improve estimation accuracy by taking advantages of both physics-based (white-box) and data-driven (black-box) modeling approaches. To this end, we utilize partial physical knowledge of the system, including the inverters’ structures and control diagrams, as well as the equivalent network model simplified through Kron reduction. The white-box model containing unknown parameters is then constructed with mathematical equations and an optimization-based method is subsequently employed to identify these unknown parameters within the white-box model. Next, the graybox modeling framework is then constructed by embedding the output variables of the white-box model into the input vector of a black-box model (represented using a neural network). Finally, the black-box section is trained using the collected input-output datasets and the gray-box model is then obtained. Furthermore, case studies demonstrate that our gray-box modeling approach effectively improves estimation accuracy compared to purely physics-based or data-driven methods.

42 ENGINEERING↗

Air Quality Impacts of Liquefied Natural Gas In the South Coast Air Basin Of California

The effects of liquefied natural gas (LNG) on pollutant emission inventories and air quality in the South Coast Air Basin of California were evaluated using recent LNG emission measurements by Lawrence Berkeley National Laboratory and the Southern California Gas Company (SoCalGas), and with a state-of-the-art air quality model. Pollutant emissions can be affected by LNG owing to differences in composition and physical properties, including the Wobbe index, a measure of energy delivery rate. This analysis uses LNG distribution scenarios developed by modeling Southern California gas flows, including supplies from the LNG receiving terminal in Baja California, Mexico. Based on these scenarios, the projected penetratino of LNG in the South Coast Air Basin is expected to be limited. In addition, the increased Wobbe index of delivered gas (resulting from mixtures of LNG and conventional gas supplies) is expected to cause increases smaller than 0.05 percent in overall (area-wide) emissions of nitrogen oxides (NOx). BAsed on the photochemical state of the South Coast Air Basin, any increase in NOx is expected to cause an increase in the highest local ozone concentrations, and this is reflected in model results. However, the magnitude of the increase is well below the generally accepted accuracy of the model and would not be discernible with the existing monitoring network. Modeling of hypothetical scenarios indicates that discernible changes to ambient ozone and particulate matter concentrations would occur only at LNG distribution rates that are not achievable with current or planned infrastructure and with Wobbe index vlaues that exceed current gas quality tariffs. Results of these hypothetical scenarios are presented for consideration of any proposed substantial expansion of LNG supply infrastructure in Southern California.

03 NATURAL GAS↗

Surrogate Model of Flexible Research Platform EnergyPlus Models to Enable Sensitivity Analysis

This letter report describes the surrogate models developed from the EnergyPlus model of Oak Ridge National Laboratory’s Flexible Research Platform. Two data-driven black-box models were developed, and the outputs of the surrogate models were compared with the EnergyPlus model. The two models developed are a multilayer perceptron deep learning model, and a long short-term memory (LSTM) neural network model. The three factors for selecting the black-box models are scalability, computation time, and accuracy. A total of 107 input variables were the dominant variables in determining the outputs of building energy consumptions and thermal comfort. A total of 54 output variables were identified as the prediction targets, including the system- and zone-level outputs. The large set of the simulation cases were generated by integrating sensor errors into an emulator based on EnergyPlus and Python EMS, which includes advanced control sequences from ASHRAE Guideline 36-2018: High-Performance Sequences of Operation. The surrogate models were developed based on a set of large-scale simulation runs (i.e., 4,000 runs) on a cloud platform. The comparison analysis shows that the two black-box models had good accuracy for predicting new outputs for sensitivity analysis using the root mean square error metric. As a next step, the developed surrogate models will be used to perform sensitivity analysis for different sensor impacts (e.g., sensor types, sensor locations).

42 ENGINEERING↗

1D Modeling for the Development of a Digital Twin for Hydropower Systems with Applications for a Testing Rig

Modeling and simulation constitute two important parts in constructing a sensible digital twin to mimic the dynamics of hydropower systems. Because of the physical nature of a hydropower system, the basic modeling should cover the water flow systems from the reservoir to the penstock (inlet water pipes), penstock to hydro turbine, and hydro turbine to generator and from the linkage of the hydropower systems to the grid. This report describes initial attempts to model these dynamic components, including the formulation of the linearized state space model for hydro turbine systems using the well-known six-coefficients linearization method and the formulation of the voltage and power dynamical models of a synchronous generator. To validate the accuracy of the proposed model, data collected from Norwegian University of Science and Technology were used to obtain relevant parameters for the model, and the desired simulation results were obtained. Additionally, neural network modeling and learning were also developed and applied to model the generation torque and water flow rate subsystems to demonstrate a potential learning scheme, which can be used in the development of a digital twin.

13 HYDRO ENERGY↗

Estimating vehicle fuel economy from overhead camera imagery and application for traffic control

In this work, we explore the ability to estimate vehicle fuel consumption using imagery from overhead fisheye lens cameras deployed as traffic sensors. We utilize this information to simulate vision-based control of a traffic intersection, with a goal of improving fuel economy with minimal impact to mobility. We introduce the ORNL Overhead Vehicle Dataset (OOVD), consisting of a data set of paired, labeled vehicle images from a ground-based camera and an overhead fisheye lens traffic camera. The data set includes segmentation masks based on Gaussian mixture models for vehicle detection. We show the dataset utility through three applications: the estimate of fuel consumption based on segmentation bounding boxes, vehicle discrimination for those vehicles with largest bounding boxes, and a fine-grained classification on a limited number of vehicle makes and models using a pre-trained set of convolutional neural network models. We compare these results with estimates based on a large open-source data set based on web-scraped imagery. Finally, we show the utility of the approach using reinforcement learning in a traffic simulator using the open source Simulation of Urban Mobility (SUMO) package. Our results show the feasibility of the approach for controlling traffic lights for better fuel efficiency based solely on visual vehicle estimates from commercial, fisheye lens cameras.

Karnowski, Thomas↗

CyBERT: Cybersecurity Claim Classification by Fine-Tuning the BERT Language Model

We introduce CyBERT, a cybersecurity feature claims classifier based on bidirectional encoder representations from transformers and a key component in our semi-automated cybersecurity vetting for industrial control systems (ICS). To train CyBERT, we created a corpus of labeled sequences from ICS device documentation collected across a wide range of vendors and devices. This corpus provides the foundation for fine-tuning BERT’s language model, including a prediction-guided relabeling process. We propose an approach to obtain optimal hyperparameters, including the learning rate, the number of dense layers, and their configuration, to increase the accuracy of our classifier. Fine-tuning all hyperparameters of the resulting model led to an increase in classification accuracy from 76% obtained with BertForSequenceClassification’s original architecture to 94.4% obtained with CyBERT. Furthermore, we evaluated CyBERT for the impact of randomness in the initialization, training, and data-sampling phases. CyBERT demonstrated a standard deviation of ±0.6% during validation across 100 random seed values. Finally, we also compared the performance of CyBERT to other well-established language models including GPT2, ULMFiT, and ELMo, as well as neural network models such as CNN, LSTM, and BiLSTM. The results showed that CyBERT outperforms these models on the validation accuracy and the F1 score, validating CyBERT’s robustness and accuracy as a cybersecurity feature claims classifier.

97 MATHEMATICS AND COMPUTING↗

Modeling and Simulation of Pore Scale Multiphase Fluid Flow and Reactive Transport in Fractured and Porous Media

In the subsurface fluids play a critical role by transporting dissolved minerals, colloids and contaminants (sometimes over long distances), by mediating dissolution and precipitation processes and enabling chemical transformations in solution and at mineral surfaces. Although the complex geometries of fracture apertures, fracture networks and pore spaces may make it difficult to accurately predict fluid flow in saturated (single-phase) subsurface systems, well developed methods are available. The simulation of multiphase fluid flow in the subsurface is much more challenging because of the large density and/or viscosity ratios found in important applications (water/air in the vadose zone, water/oil, water/gas, gas/oil and water/oil/gas in oil reservoirs, water/air/non-aqueous phase liquids (NAPL) in contaminated vadose zone systems and gas/molten rock in volcanic systems, for example). In addition, the complex behavior of fluid-fluid-solid contact lines, and its impact on dynamic contact angles, must also be taken into account, and coupled with the fluid flow. Pore network models and simple statistical physics based models such as the invasion percolation and diffusion-limited aggregation models have been used quite extensively. However, these models for multiphase fluid flow are based on simplified models for pore space geometries and simplified physics. Other methods such a lattice Boltzmann and lattice gas models, molecular dynamics, Monte Carlo methods, and particle methods such as dissipative particle dynamics and smoothed particle hydrodynamics are based more firmly on first principles, and they do not require simplified pore and/or fracture geometries. However, they are less (in some cases very much less) computationally efficient that pore network and statistical physics models. Recently a combination of continuum computation fluid dynamics, fluid-fluid interface tracking or capturing and simple models for the dependence of contact angles on fluid velocity at the contact line has been used to simulate multiphase fluid flow in fracture apertures, fracture networks and pore spaces. Fundamental conservation principles - conservation of momentum, and conservation of mass (or conservation of volume for incompressible fluids) and conservation of energy, as well as symmetries (Galilean invariance and isotropy) are central to the physics of fluids and the models used to simulate them. In molecular and mesoscale models observance of these conservation principles and symmetries at the microscopic level leads to macroscopic fluid dynamics that can be represented by the Navier Stokes equation. The remarkable fact that the flow of all simpe fluids, irrespective of their chemical nature, can be described by the Navier-Stokes equation is a result of these conservation principles and symmetries acting on the molecular level.

99 GENERAL AND MISCELLANEOUS↗

Modeling and simulation of pore-scale multiphase fluid flow and reactive transport in fractured and porous media

In the subsurface fluids play a critical role by transporting dissolved minerals, colloids and contaminants (sometimes over long distances), by mediating dissolution and precipitation processes and enabling chemical transformations in solution and at mineral surfaces. Although the complex geometries of fracture apertures, fracture networks and pore spaces may make it difficult to accurately predict fluid flow in saturated (single-phase) subsurface systems, well developed methods are available. The simulation of multiphase fluid flow in the subsurface is much more challenging because of the large density and/or viscosity ratios found in important applications (water/air in the vadose zone, water/oil, water/gas, gas/oil and water/oil/gas in oil reservoirs, water/air/non-aqueous phase liquids (NAPL) in contaminated vadose zone systems and gas/molten rock in volcanic systems, for example). In addition, the complex behavior of fluid-fluid-solid contact lines, and its impact on dynamic contact angles, must also be taken into account, and coupled with the fluid flow. Pore network models and simple statistical physics based models such as the invasion percolation and diffusion-limited aggregation models have been used quite extensively. However, these models for multiphase fluid flow are based on simplified models for pore space geometries and simplified physics. Other methods such a lattice Boltzmann and lattice gas models, molecular dynamics, Monte Carlo methods, and particle methods such as dissipative particle dynamics and smoothed particle hydrodynamics are based more firmly on first principles, and they do not require simplified pore and/or fracture geometries. However, they are less (in some cases very much less) computationally efficient that pore network and statistical physics models. Recently a combination of continuum computation fluid dynamics, fluid-fluid interface tracking or capturing and simple models for the dependence of contact angles on fluid velocity at the contact line has been used to simulate multiphase fluid flow in fracture apertures, fracture networks and pore spaces. Fundamental conservation principles - conservation of momentum, and conservation of mass (or conservation of volume for incompressible fluids) and conservation of energy, as well as symmetries (Galilean invariance and isotropy) are central to the physics of fluids and the models used to simulate them. In molecular and mesoscale models observance of these conservation principles and symmetries at the microscopic level leads to macroscopic fluid dynamics that can be represented by the Navier Stokes equation. The remarkable fact that the flow of all simpe fluids, irrespective of their chemical nature, can be described by the Navier-Stokes equation is a result of these conservation principles and symmetries acting on the molecular level.

pore-scale, multiphase flow, reactive transport, n↗

Reinforced dynamics for enhanced sampling in large atomic and molecular systems

A new approach for efficiently exploring the configuration space and computing the free energy of large atomic and molecular systems is proposed, motivated by an analogy with reinforcement learning. There are two major components in this new approach. Like metadynamics, it allows for an efficient exploration of the configuration space by adding an adaptively computed biasing potential to the original dynamics. Like deep reinforcement learning, this biasing potential is trained on the fly using deep neural networks, with data collected judiciously from the exploration and an uncertainty indicator from the neural network model playing the role of the reward function. Parameterization using neural networks makes it feasible to handle cases with a large set of collective variables. This has the potential advantage that selecting precisely the right set of collective variables has now become less critical for capturing the structural transformations of the system. Finally, the method is illustrated by studying the full-atom explicit solvent models of alanine dipeptide and tripeptide, as well as the system of a polyalanine-10 molecule with 20 collective variables.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning

Accurately predicting small molecule partitioning and hydrophobicity is critical in the drug discovery process. There are many heterogeneous chemical environments within a cell and entire human body. For example, drugs must be able to cross the hydrophobic cellular membrane to reach their intracellular targets, and hydrophobicity is an important driving force for drug–protein binding. Atomistic molecular dynamics (MD) simulations are routinely used to calculate free energies of small molecules binding to proteins, crossing lipid membranes, and solvation but are computationally expensive. Machine learning (ML) and empirical methods are also used throughout drug discovery but rely on experimental data, limiting the domain of applicability. We present atomistic MD simulations calculating 15,000 small molecule free energies of transfer from water to cyclohexane. This large data set is used to train ML models that predict the free energies of transfer. We show that a spatial graph neural network model achieves the highest accuracy, followed closely by a 3D-convolutional neural network, and shallow learning based on the chemical fingerprint is significantly less accurate. A mean absolute error of ~4 kJ/mol compared to the MD calculations was achieved for our best ML model. We also show that including data from the MD simulation improves the predictions, tests the transferability of each model to a diverse set of molecules, and show multitask learning improves the predictions. This work provides insight into the hydrophobicity of small molecules and ML cheminformatics modeling, and our data set will be useful for designing and testing future ML cheminformatics methods.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A Parallel Alternative for Energy-Efficient Neural Network Training and Inferencing

Energy efficiency of training and inferencing with large neural network models is a critical challenge facing the future of sustainable large-scale machine learning workloads. This paper introduces an alternative strategy, called phantom parallelism, to minimize the net energy consumption of traditional tensor (model) parallelism, the most energy-inefficient component of large neural network training. The approach is presented in the context of feed-forward network architectures as a preliminary, but comprehensive, proof-of-principle study of the proposed methodology. We derive new forward and backward propagation operators for phantom parallelism, implement them as custom autograd operations within an end-to-end phantom parallel training pipeline and compare its parallel performance and energy-efficiency against those of conventional tensor parallel training pipelines. Formal analyses that predict lower bandwidth and FLOP counts are presented with supporting empirical results on up to 256 GPUs that corroborate these gains. Experiments are shown to deliver ∼50% reduction in the energy consumed to train FFNs using the proposed phantom parallel approach when compared with conventional tensor parallel methods. Additionally, the proposed approach is shown to train smaller phantom models to the same model loss on smaller GPU counts as larger tensor parallel models on larger GPU counts offering the possibility for even greater energy savings.

Seal, Sudip [ORNL] (ORCID:0000000332330656)↗

Metabolic network reconstruction and flux variability analysis of storage synthesis in developing oilseed rape (Brassica napus L.) embryos

Computational simulation of large-scale biochemical networks can be used to analyze and predict the metabolic behavior of an organism, such as a developing seed. Based on the biochemical literature, pathways databases and decision rules defining reaction directionality we reconstructed bna572, a stoichiometric metabolic network model representing Brassica napus seed storage metabolism. In the highly compartmentalized network about 25% of the 572 reactions are transport reactions interconnecting nine subcellular compartments and the environment. According to known physiological capabilities of developing B. napus embryos, four nutritional conditions were defined to simulate heterotrophy or photoheterotrophy, each in combination with the availability of inorganic nitrogen (ammonia, nitrate) or amino acids as nitrogen sources. Based on mathematical linear optimization the optimal solution space was comprehensively explored by flux variability analysis, thereby identifying for each reaction the range of flux values allowable under optimality. The range and variability of flux values was then categorized into flux variability types. Across the four nutritional conditions, approximately 13% of the reactions have variable flux values and 10-11% are substitutable (can be inactive), both indicating metabolic redundancy given, for example, by isoenzymes, subcellular compartmentalization or the presence of alternative pathways. About one-third of the reactions are never used and are associated with pathways that are suboptimal for storage synthesis. Fifty-seven reactions change flux variability type among the different nutritional conditions, indicating their function in metabolic adjustments. This predictive modeling framework allows analysis and quantitative exploration of storage metabolism of a developing B. napus oilseed.

59 BASIC BIOLOGICAL SCIENCES↗