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 973 records · Page 54

Animal movement estimation and network-based epidemic modeling: Illustration for the swine industry in Iowa (US)

Animal movement plays a critical role in disease transmission between farms. However, in the United States, the lack of available animal shipment data, sometimes coupled with a lack of detailed information about farm demographics and characteristics, presents great challenges for epidemic modeling and prediction. In this study, we proposed a new method based on the maximum entropy to generate “synthetic” animal movement networks, considering available statistics about the premises operation type, operation size, and the distance between premises. We illustrated our method for the swine movement networks in Iowa and performed network analyses to gain insights into the swine industry. We then applied the generated networks to a network-based epidemic model to identify potential system vulnerabilities in terms of disease transmission. The model was parameterized for African Swine Fever (ASF) as the US swine industry is quite concerned about this disease. Results show that premises with a central role in the network are more vulnerable to disease outbreaks and play an important role in disease spread. Simulations with outbreaks starting from random farms reveal no significant large outbreaks, indicating the system’s relative robustness against arbitrary disease introductions. However, outbreaks originating from high out-degree farms can lead to large epidemic sizes. This underscores the importance for stakeholders and policymakers to continue improving animal movement records and traceability programs in the US and the value of making that data available to epidemiologists and modelers to better understand risk and inform strategies aimed to cost-effectively prevent and control disease transmission. Our approach could be easily adapted to estimate movement networks in other animal production systems and to inform disease spread models for various infectious diseases.

60 APPLIED LIFE SCIENCES↗

Efficient data acquisition and training of collisional-radiative model artificial neural network surrogates through adaptive parameter space sampling

Abstract Effective plasma transport modeling of magnetically confined fusion devices relies on having an accurate understanding of the ion composition and radiative power losses of the plasma. Generally, these quantities can be obtained from solutions of a collisional-radiative (CR) model at each time step within a plasma transport simulation. However, even compact, approximate CR models can be computationally onerous to evaluate, and in-situ evaluation of these models within a larger plasma transport code can lead to a rigid bottleneck. As a way to bypass this bottleneck, we propose deploying artificial neural network (ANN) surrogates to allow rapid evaluation of the necessary plasma quantities. However, one issue with training an accurate ANN surrogate is the reliance on a sufficiently large and representative training and validation data set, which can be time-consuming to generate. In this work we explore a data-driven active learning and training routine to allow autonomous adaptive sampling of the problem parameter space to ensure a sufficiently large and meaningful set of training data is assembled for the network training. As a result, we can demonstrate approximately order-of-magnitude savings in required training data samples to produce an accurate surrogate.

97 MATHEMATICS AND COMPUTING↗

Catalyst: Fast and flexible modeling of reaction networks

We introduce Catalyst.jl, a flexible and feature-filled Julia library for modeling and high-performance simulation of chemical reaction networks (CRNs). Catalyst supports simulating stochastic chemical kinetics (jump process), chemical Langevin equation (stochastic differential equation), and reaction rate equation (ordinary differential equation) representations for CRNs. Through comprehensive benchmarks, we demonstrate that Catalyst simulation runtimes are often one to two orders of magnitude faster than other popular tools. More broadly, Catalyst acts as both a domain-specific language and an intermediate representation for symbolically encoding CRN models as Julia-native objects. This enables a pipeline of symbolically specifying, analyzing, and modifying CRNs; converting Catalyst models to symbolic representations of concrete mathematical models; and generating compiled code for numerical solvers. Leveraging ModelingToolkit.jl and Symbolics.jl, Catalyst models can be analyzed, simplified, and compiled into optimized representations for use in numerical solvers. Finally, we demonstrate Catalyst’s broad extensibility and composability by highlighting how it can compose with a variety of Julia libraries, and how existing open-source biological modeling projects have extended its intermediate representation.

59 BASIC BIOLOGICAL SCIENCES↗

Emergent Nonthermal Fluid from Jets in the Massive Schwinger Model Using Tensor Networks

We analyze the correlation between the energy, momentum, and spatial entanglement produced by two luminal jets in the massive Schwinger model. Using tensor network methods, we show that for m/g > 1/π, in the vicinity of the strong- to weak-coupling transition, a nearly perfect and chargeless effective fluid behavior appears around the midrapidity region with a universal energy-pressure relationship. The evolution of energy and pressure is strongly correlated with the rise of the spatial entanglement entropy, indicating a key role of quantum dynamics. Some of these observations may be used to analyze high multiplicity jet fragmentation events, energy-energy and energy-charge correlators at current collider energies.

Ab initio calculations↗

Modeling the AC Power Flow Equations with Optimally Compact Neural Networks: Application to Unit Commitment

Nonlinear power flow constraints render a variety of power system optimization problems computationally intractable. Emerging research shows, however, that the nonlinear AC power flow equations can be successfully modeled using neural networks. These neural networks can be exactly transformed into mixed integer linear programs and embedded inside challenging optimization problems, thus replacing nonlinearities that are intractable for many applications with tractable piecewise linear approximations. Such approaches, though, suffer from an explosion of the number of binary variables needed to represent the neural network. Accordingly, this paper develops a technique for training an "optimally compact'' neural network, i.e., one that can represent the power flow equations with a sufficiently high degree of accuracy while still maintaining a tractable number of binary variables. We demonstrate the use of this neural network as an approximator of the nonlinear power flow equations by embedding it in the AC unit commitment problem, transforming the problem from a mixed integer nonlinear program into a more manageable mixed integer linear program. We use the 14-, 57-, and 89-bus networks as test cases and compare the AC-feasibility of commitment decisions resulting from the neural network, DC, and linearized power flow approximations. Our results show that the neural network model outperforms both the DC and linearized power flow approximations when embedded in the unit commitment problem. The neural network formulation most often selects a feasible unit commitment schedule, and furthermore, it only s

AC power flow↗

Controlling the self-organizing dynamics in a sandpile model on complex networks by failure tolerance

In this paper, we propose a strategy to control the self-organizing dynamics of the Bak-Tang-Wiesenfeld (BTW) sandpile model on complex networks by allowing some degree of failure tolerance for the nodes and introducing additional active dissipation while taking the risk of possible node damage. We show that the probability for large cascades significantly increases or decreases respectively when the risk for node damage outweighs the active dissipation and when the active dissipation outweighs the risk for node damage. By considering the potential additional risk from node damage, a non-trivial optimal active dissipation control strategy which minimizes the total cost in the system can be obtained. Under some conditions the introduced control strategy can decrease the total cost in the system compared to the uncontrolled model. Moreover, when the probability of damaging a node experiencing failure tolerance is greater than the critical value, then no matter how successful the active dissipation control is, the total cost of the system will have to increase. This critical damage probability can be used as an indicator of the robustness of a network or system. Copyright (C) EPLA, 2015

Qi, Junjian↗

Effect of Network Structure on Characterization and Flow Modeling Using X-ray Micro-Tomography Images of Granular and Fibrous Porous Media

Image-based network modeling has become a powerful tool for modeling transport in real materials that have been imaged using X-ray computed micro-tomography (XCT) or other three-dimensional imaging techniques. Network generation is an essential part of image-based network modeling, but little quantitative work has been done to understand the influence of different network structures on modeling. We use XCT images of three different porous materials (disordered packings of spheres, sand, and cylinders) to create a series of four networks for each material. Despite originating from the same data, the networks can be made to vary over two orders of magnitude in pore density, which in turn affects network properties such as pore-size distribution and pore connectivity. Despite the orders-of-magnitude difference in pore density, single-phase permeability predictions remain remarkably consistent for a given material, even for the simplest throat conductance formulas. Detailed explanations for this beneficial attribute are given in the article; in general, it is a consequence of using physically representative network models. The capillary pressure curve generated from quasi-static drainage is more sensitive to network structure than permeability. However, using the capillary pressure curve to extract pore-size distributions gives reasonably consistent results even though the networks vary significantly. These results provide encouraging evidence that robust network modeling algorithms are not overly sensitive to the specific structure of the underlying physically representative network, which is important given the variety image-based network-generation strategies that have been developed in recent years.

42 ENGINEERING↗

Anomalous transport in disordered fracture networks: Spatial Markov model for dispersion with variable injection modes

Here we investigate tracer transport on random discrete fracture networks that are characterized by the statistics of the fracture geometry and hydraulic conductivity. While it is well known that tracer transport through fractured media can be anomalous and particle injection modes can have major impact on dispersion, the incorporation of injection modes into effective transport modeling has remained an open issue. The fundamental reason behind this challenge is that—even if the Eulerian fluid velocity is steady—the Lagrangian velocity distribution experienced by tracer particles evolves with time from its initial distribution, which is dictated by the injection mode, to a stationary velocity distribution. We quantify this evolution by a Markov model for particle velocities that are equidistantly sampled along trajectories. This stochastic approach allows for the systematic incorporation of the initial velocity distribution and quantifies the interplay between velocity distribution and spatial and temporal correlation. The proposed spatial Markov model is characterized by the initial velocity distribution, which is determined by the particle injection mode, the stationary Lagrangian velocity distribution, which is derived from the Eulerian velocity distribution, and the spatial velocity correlation length, which is related to the characteristic fracture length. This effective model leads to a time-domain random walk for the evolution of particle positions and velocities, whose joint distribution follows a Boltzmann equation. Finally, we demonstrate that the proposed model can successfully predict anomalous transport through discrete fracture networks with different levels of heterogeneity and arbitrary tracer injection modes.

54 ENVIRONMENTAL SCIENCES↗

Surrogate Model for Distribution Networks Influenced by Weather

Here, we propose a method for generating reduced representations of time series and for constructing low dimensional surrogate models for time dependent calculations of power and voltage in distribution networks. We employ Fourier polynomials. The surrogate model strategy is aimed at reducing the computational cost of time dependent simulations, albeit, at the expense of fidelity. The reduced representation is achieved by identifying a small and most consequential subset of degrees of freedom. In power and voltage distribution networks dynamics that are heavily influenced by strong cyclic weather events, e.g., the hourly, diurnal and seasonal cycles, the weather/climate time series spectrum exposes these most energetic components. Once the degrees of freedom are identified their amplitudes are optimized using training data. The key challenge in using spectral methods in power network surrogates is addressing the computation of quotients. For this we propose a numerically-stable deconvolution strategy.

Weather events↗

Large-scale Gravitational Lens Modeling with Bayesian Neural Networks for Accurate and Precise Inference of the Hubble Constant

We investigate the use of approximate Bayesian neural networks (BNNs) in modeling hundreds of time delay gravitational lenses for Hubble constant (H 0 ) determination. Here, our BNN was trained on synthetic Hubble Space Telescope quality images of strongly lensed active galactic nuclei with lens galaxy light included. The BNN can accurately characterize the posterior probability density functions (PDFs) of model parameters governing the elliptical power-law mass profile in an external shear field. We then propagate the BNN-inferred posterior PDFs into an ensemble H 0 inference, using simulated time delay measurements from a plausible dedicated monitoring campaign. Assuming well-measured time delays and a reasonable set of priors on the environment of the lens, we achieve a median precision of 9.3% per lens in the inferred H 0 . A simple combination of a set of 200 test lenses results in a precision of 0.5 km s –1 Mpc –1 (0.7%), with no detectable bias in this H 0 recovery test. The computation time for the entire pipeline—including the generation of the training set, BNN training and H0 inference—translates to 9 minutes per lens on average for 200 lenses and converges to 6 minutes per lens as the sample size is increased. Being fully automated and efficient, our pipeline is a promising tool for exploring ensemble-level systematics in lens modeling for H 0 inference.

79 ASTRONOMY AND ASTROPHYSICS↗

Exploring Randomly Wired Neural Networks for Climate Model Emulation

Exploring the climate impacts of various anthropogenic emissions scenarios is key to making informed decisions for climate change mitigation and adaptation. State-of-the-art Earth system models can provide detailed insight into these impacts but have a large associated computational cost on a per-scenario basis. This large computational burden has driven recent interest in developing cheap machine learning models for the task of climate model emulation. In this paper, we explore the efficacy of randomly wired neural networks for this task. We describe how they can be constructed and compare them with their standard feedforward counterparts using the ClimateBench dataset. Specifically, we replace the serially connected dense layers in multilayer perceptrons, convolutional neural networks, and convolutional long short-term memory networks with randomly wired dense layers and assess the impact on model performance for models with 1 million and 10 million parameters. We find that models with less-complex architectures see the greatest performance improvement with the addition of random wiring (up to 30.4% for multilayer perceptrons). Furthermore, of 24 different model architecture, parameter count, and prediction task combinations, only one had a statistically significant performance deficit in randomly wired networks relative to their standard counterparts, with 14 cases showing statistically significant improvement. We also find no significant difference in prediction speed between networks with standard feedforward dense layers and those with randomly wired layers. These findings indicate that randomly wired neural networks may be suitable direct replacements for traditional dense layers in many standard models.

54 ENVIRONMENTAL SCIENCES↗

An a priori evaluation of a principal component and artificial neural network based combustion model in diesel engine conditions

A principal component analysis (PCA) and artificial neural network (ANN) based chemistry tabulation approach is presented. ANNs are used to map the thermochemical state onto a low-dimensional manifold consisting of five control variables that have been identified using PCA. Three canonical configurations are considered to train the PCA-ANN model: a series of homogeneous reactors, a nonpremixed flamelet, and a two-dimensional lifted flame. The performance of the model in predicting the thermochemical manifold of a spatially-developing turbulent jet flame in diesel engine thermochemical conditions is a priori evaluated using direct numerical simulation (DNS) data. The PCA-ANN approach is compared with a conventional tabulation approach (tabulation using ad hoc defined control variables and linear interpolation). The PCA-ANN model provides higher accuracy and requires several orders of magnitude less memory. Here, these observations indicate that the PCA-ANN model is superior for chemistry tabulation, especially for modelling complex chemistries that present multiple combustion modes as observed in diesel combustion. The performance of the PCA-ANN model is then compared to the optimal estimator, i.e. the conditional mean from the DNS. The results indicate that the PCA-ANN model gives high prediction accuracy, comparable to the optimal estimator, especially for major species and the thermophysical properties. Higher errors are observed for the minor species and reaction rate predictions when compared to the optimal estimator. It is shown that the prediction of minor species and reaction rates can be improved by using training data that exhibits a variation of parameters as observed in the turbulent flame. The output of the ANN is analysed to assess mass conservation. It is observed that the ANN incurs a mean absolute error of 0.05% in mass conservation. Furthermore, it is demonstrated that this error can be reduced by modifying the cost function of the ANN to penalise for deviation from mass conservation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

On the Effectiveness of Recurrent Neural Networks for Live Modeling of Cyber-Physical Systems

Attention to cyber security of cyber-physical systems (CPS) has led to the development of innovative cyber-resilient methodologies to ensure early detection and mitigation of cyber anomalies and threats. The concept of Digital Twin (DT) has recently emerged as one of the approaches to achieve the objective of resilience. In the approach using DT, a software-based live model of a target CPS is used to continuously monitor, surveil and verify the correctness of the target CPS operation. In this paper, we empirically study the effectiveness of Recurrent Neural Network (RNN)-based models as the basis of DT-based resilience. We uncover the important characteristics of an RNN-based solution with experimentation on a lab-scale Canal Lock CPS emulator with live validations and attack scenarios. For the first time, we demonstrate actual, real-time use of a RNN-based model as a DT for performing live analysis on an operational CPS. Based on the observed results, we highlight the importance of a DT model's training interval, prediction interval and CPS polling interval in the process of anomaly detection. We uncover the limitations in anomaly detection due to real-time synchronization needs of the RNN-based DT. We highlight this uncovered tug of war between synchronization and anomaly detection is inherent in any complex CPS that is monitored and synchronized by relying on the same sensor streams of ground truth for both synchronization as well as anomaly detection.

Yoginath, Srikanth↗

Thermal conductivity model for nanofiber networks

Understanding thermal transport in nanofiber networks is essential for their applications in thermal management, which are used extensively as mechanically sturdy thermal insulation or high thermal conductivity materials. In this study, using the statistical theory and Fourier's law of heat conduction while accounting for both the inter-fiber contact thermal resistance and the intrinsic thermal resistance of nanofibers, an analytical model is developed to predict the thermal conductivity of nanofiber networks as a function of their geometric and thermal properties. A scaling relation between the thermal conductivity and the geometric properties including volume fraction and nanofiber length of the network is revealed. This model agrees well with both numerical simulations and experimental measurements found in the literature. This model may prove useful in analyzing the experimental results and designing nanofiber networks for both high and low thermal conductivity applications.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Intelligent-based Structural Damage Detection Model

This paper presents the application of a novel Artificial Neural Network (ANN) model for the diagnosis of structural damage. The ANN model, denoted as the GRNNFA, is a hybrid model combining the General Regression Neural Network Model (GRNN) and the Fuzzy ART (FA) model. It not only retains the important features of the GRNN and FA models (i.e. fast and stable network training and incremental growth of network structure) but also facilitates the removal of the noise embedded in the training samples. Structural damage alters the stiffness distribution of the structure and so as to change the natural frequencies and mode shapes of the system. The measured modal parameter changes due to a particular damage are treated as patterns for that damage. The proposed GRNNFA model was trained to learn those patterns in order to detect the possible damage location of the structure. Simulated data is employed to verify and illustrate the procedures of the proposed ANN-based damage diagnosis methodology. The results of this study have demonstrated the feasibility of applying the GRNNFA model to structural damage diagnosis even when the training samples were noise contaminated.

97 MATHEMATICS AND COMPUTING↗

Enhancing Drinking Water Quality Modeling: Leveraging Physics Informed Neural Networks for Learning with Imperfect Reaction Models and Partial Data

Chemical kinetics models, typically formulated as systems of ordinary or partial differential equations, are valuable tools for simulating drinking water quality. However, these models often face inaccuracies due to discrepancies between the laboratory and the real-world conditions, as well as limitations in experimental analytical methods, hindering the accurate representation of the true underlying chemical mechanisms. In this study, we propose a Physics Informed Neural Network (PINN), using the eXtreme Theory of Functional Connections, to improve the prediction of chemical concentrations over time. The PINN method accounts for imperfect chemical models and incorporates partial data to improve predictions. Focusing on reactions describing water disinfection residual and disinfectant byproduct formation, which are crucial for public health and regulatory compliance, we demonstrate that the PINN model is able to accurately predict the concentrations of chemical species across various pH values. Notably, the model extends its accuracy to predict concentrations of chemical species not originally included in its training data. The developed method can be extended to a variety of chemical systems, offering a wide array of potential applications.

13 HYDRO ENERGY↗

Modelling a recovery network for WEEE: A case study in Portugal

The European Union directive for electric and electronic waste, published in 2003, enforced all European countries to meet some targets concerning the recycling and recovery of these products. This directive was transposed to the Portuguese legislation in 2004. Following this, a group of EEE producers set up an organization (Amb3e) whose mission was to design and manage a nationwide recovery network for WEEE, which will be the subject matter of this work. A generic MILP model is proposed to represent this network, which is applied to its design and planning, where the best locations for collection and sorting centres are chosen simultaneously with the definition of a tactical network planning. Several analyses are performed to provide further insights regarding the selection of these alternative locations. The results gave support to the company strategic expansion plans for a high number of centres to be opened and to their location near the major WEEE sources.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Acceleration of Graph Neural Network-Based Prediction Models in Chemistry via Co-Design Optimization on Intelligence Processing Units

Atomic structure prediction and associated property calculations are the bedrock of chemical physics. Since high-fidelity ab initio modeling techniques for computing the structure and properties can be prohibitively expensive, this motivates the development of machine-learning (ML) models that make these predictions more efficiently. Training graph neural networks over large atomistic databases introduces unique computational challenges such as the need to process millions of small graphs with variable size and support communication patterns that are distinct from learning over large graphs such as social networks. We demonstrate a novel hardware-software co-design approach to scale up the training of atomistic graph neural networks (GNN) for structure and property prediction. First, to eliminate redundant computation and memory associated with alternative padding techniques and to improve throughput via minimizing communication, we formulate the effective coalescing of the batches of variable-size atomistic graphs as the bin packing problem and introduce a hardware-agnostic algorithm to pack these batches. In addition, we propose hardware-specific optimizations including a planner and vectorization for the gather-scatter operations targeted for Graphcore’s Intelligence Processing Unit (IPU), as well as model-specific optimizations such as merged communication collectives and optimized softplus. Putting these all together, we demonstrate the effectiveness of the proposed co-design approach by providing an implementation of a well-established atomistic GNN on the Graphcore IPUs. We evaluate the training performance on multiple atomistic graph databases with varying degrees of graph counts, sizes and sparsity. Here, we demonstrate that such a co-design approach can reduce the training time of atomistic GNNs and can improve the performance by up to 1.5× compared to the baseline implementation of the model on the IPUs. Additionally, we compare our IPU implementation with a Nvidia GPU-based implementation and show that our atomistic GNN implementation on the IPUs can run 1.8× faster on average compared to the execution time on the GPUs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗