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 289 records · Page 16

Graybox and adaptative dynamic neural network identification models to infer the steady state efficiency of solar thermal collectors starting from the transient condition

This paper deals with the development of methods for non steady state test of solar thermal collectors. Our goal is to infer performances in steady-state conditions in terms of the efficiency curve when measures in transient conditions are the only ones available. We take into consideration the method of identification of a system in dynamic conditions by applying a Graybox Identification Model and a Dynamic Adaptative Linear Neural Network (ALNN) model. The study targets the solar collector with evacuated pipes, such as Dewar pipes. The mathematical description that supervises the functioning of the solar collector in transient conditions is developed using the equation of the energy balance, with the aim of determining the order and architecture of the two models. The input and output vectors of the two models are constructed, considering the measures of 4 days of solar radiation, flow mass, environment and heat-transfer fluid temperature in the inlet and outlet from the thermal solar collector. The efficiency curves derived from the two models are detected in correspondence to the test and validation points. The two synthetic simulated efficiency curves are compared with the actual efficiency curve certified by the Swiss Institute Solartechnik Puffung Forschung which tested the solar collector performance in steady-state conditions according to the UNI-EN 12975 standard. An acquisition set of measurements of only 4 days in the transient condition was enough to trace through a Graybox State Space Model the efficiency curve of the tested solar thermal collector, with a relative error of synthetic values with respect to efficiency certified by SPF, lower than 0.5%, while with the ALNN model the error is lower than 2.2% with respect to certified one. (author)

14 SOLAR ENERGY↗

Neuromorphic computing for temporal scientific data classification

In this work, we apply a spiking neural network model and an associated memristive neuromorphic implementation to an application in classifying temporal scientific data. We demonstrate that the spiking neural network model achieves comparable results to a previously reported convolutional neural network model, with significantly fewer neurons and synapses required.

Schuman, Catherine↗

A generalized distillation network synthesis model

In this report we propose a generalized superstructure-based distillation network synthesis model with improved modeling capabilities resulting in an extended solution space. The model can assign multiple mixtures to be separated (e.g., reactor network effluents) to different columns of the network while considering the interactions among different separation steps. In terms of outlets, products with general specifications, including pure components and multi-component mixtures, as well as streams without strict specifications (e.g., reactor recycle streams) can be readily handled. Furthermore, stream bypass is considered to avoid unnecessary separations, and thermal coupling is considered to yield novel solutions. The proposed model enables seamless integration with reactor network synthesis models thereby leading to solutions that are superior to the ones obtained by conventional sequential approaches.

42 ENGINEERING↗

Deterministic Discrete Fracture Network (DFN) Model for the EGS Collab Project on the 4850 Level of the Sanford Underground Research Facility (SURF)

The EGS Collab is conducting hydraulic fracture stimulation and fluid circulation experiments in the Sanford Underground Research Facility (SURF) located in Lead, South Dakota. A total of eight ~60m-long subhorizontal boreholes were drilled from the 4850 Level (~1.5 km below the ground surface) into the crystalline rock of this former mine. Six of these holes are used for geophysical monitoring, one is used for hydraulic fracture stimulation, and the remaining hole was designed as a production borehole that receives water from the injection well via the induced and natural fracture system. The primary goal of creating the discrete fracture network model is to show that these modeling methods are critical for the development of enhanced geothermal systems (EGS). This includes the prediction of rock behavior during fracturing and during an extended period of water flow between the parallel injection and production boreholes. Understanding the results from the induced fracturing and flow is complicated by the presence of significant natural fractures that interact with the stimulation and/or flow pathways. The delineation and characterization of natural fractures is thus an important part of the project, and therefore a model of the Discrete Fracture Network (DFN) was developed on a deterministic basis. The DFN was populated using observations and interpretations integrated from drift (horizontal passageways that allow access in the underground) fracture mapping, analysis of core recovered from the eight boreholes, borehole televiewer logs and videos, and observations of flow between and within boreholes and in the drift. The natural fracture system is dominated by a pervasive northwest-trending, steeply dipping shear system that is identifiable in the drifts and the core. Hydraulic fracture stimulation, flow/tracer circulation tests, and geophysical monitoring revealed that the behavior of the injected water, and perhaps the growth of induced fractures, has been significantly influenced by the existing fractures identified in the DFN.

40 EE - Geothermal Technologies Office (EE-4G)↗

Real Time Effective Impedance Estimation for Power System State Estimation

Here, we introduce two tools-the Effective Impedance State Estimator (EISE), and the Real Time Effective Impedance Estimator (RTEIE). The EISE is novel in that it is an entirely data-driven state estimator, and does not require a network model. Thus, the EISE is well suited for applications in which an accurate network model is not available. Instead of using a network impedance model, the EISE uses an effective impedance network model that is learned online by the RTEIE. The RTEIE recursively estimates the effective impedance network model using phasor measurement unit (PMU) measurements. The mathematical reasoning for using an effective impedance model, rather than a full network impedance model, is given. The recursive estimation framework allows the RTEIE to take the entire history of phasor measurements into account to produce a model estimate, while still adjusting to changes in the network impedance due to external factors such as topology changes and temperature. The merits of the RTEIE and EISE are demonstrated by comparing the accuracy of the EISE with a conventional state estimator on a toy network.

24 POWER TRANSMISSION AND DISTRIBUTION↗

A local structure model for network analysis

The statistical analysis of networks is a popular research topic with ever widening applications. Exponential random graph models (ERGMs), which specify a model through interpretable, global network features, are common for this purpose. In this study we introduce a new class of models for network analysis, called local structure graph models (LSGMs). In contrast to an ERGM, a LSGM specifies a network model through local features and allows for an interpretable and controllable local dependence structure. In particular, LSGMs are formulated by a set of full conditional distributions for each network edge, e.g., the probability of edge presence/absence, depending on neighborhoods of other edges. Additional model features are introduced to aid in specification and to help alleviate a common issue (occurring also with ERGMs) of model degeneracy. Finally, the proposed models are demonstrated on a network of tornadoes in Arkansas where a LSGM is shown to perform significantly better than a model without local dependence.

97 MATHEMATICS AND COMPUTING↗

Neural network surrogate models for equations of state

Equation of state (EOS) data provide necessary information for accurate multiphysics modeling, which is necessary for fields such as inertial confinement fusion. Here, we suggest a neural network surrogate model of energy and entropy and use thermodynamic relationships to derive other necessary thermodynamic EOS quantities. We incorporate phase information into the model by training a phase classifier and using phase-specific regression models, which improves the modal prediction accuracy. Our model predicts energy values to 1% relative error and entropy to 3.5% relative error in a log-transformed space. Although sound speed predictions require further improvement, the derived pressure values are accurate within 10% relative error. Our results suggest that neural network models can effectively model EOS for inertial confinement fusion simulation applications.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A framework for strategic discovery of credible neural network surrogate models under uncertainty

The widespread integration of deep neural networks in developing data-driven surrogate models for high-fidelity simulations of complex physical systems highlights the critical necessity for robust uncertainty quantification techniques and credibility assessment methodologies, ensuring the reliable deployment of surrogate models in consequential decision-making. Here, this study presents the Occam Plausibility Algorithm for surrogate models (OPAL-surrogate), providing a systematic framework to uncover predictive neural network-based surrogate models within the large space of potential models, including various neural network classes and choices of architecture and hyperparameters. The framework is grounded in hierarchical Bayesian inferences and employs model validation tests to evaluate the credibility and prediction reliability of the surrogate models under uncertainty. Leveraging these principles, OPAL-surrogate introduces a systematic and efficient strategy for balancing the trade-off between model complexity, accuracy, and prediction uncertainty. The effectiveness of OPAL-surrogate is demonstrated through two modeling problems, including the deformation of porous materials for building insulation and turbulent combustion flow for ablation of solid fuels within hybrid rocket motors.

42 ENGINEERING↗

Machine Learning Models for Network Traffic Classification in Programmable Logic

Network traffic classification via machine learning on network packet payloads has emerged as an active area of research for network security due to the high accuracy machine learning models have achieved in classifying payloads. For effective deployment as part of network security, these machine learning models must not only classify malicious packet payloads accurately, they must also identify anomalous payloads and perform inference at speeds generally faster than 10,000 packets per second to be effective. This work explores the in- ference speeds and accuracy of several neural network models implemented in programmable logic on various field programmable gate arrays (FPGA) including the Xilinx VC1902 and Xilinx Zynq Ultrascale+. This work also presents the design and performance of both an autoencoder and variational autoencoder programmed on the FPGA for identifying anomalous packet payloads. The performance benefits of the FPGA implementation for this type of packet payload inspection driven by machine learning are compared against graphics processing unit (GPU) inference implementations run on two state-of-the-art datacenter GPU devices, the NVIDIA V100 and A100. The model accuracy difference between the FPGA and GPU implementations was found to be 4% or less while the Xilinx VC1902 outperformed both the NVIDIA V100 and A100 for inference speeds on all the models explored except the variational autoencoder.

97 MATHEMATICS AND COMPUTING↗

Recurrent convolutional neural networks for modeling nonadiabatic dynamics of quantum-classical systems

Recurrent neural networks (RNNs) have recently been extensively applied to model the time evolution in fluid dynamics, weather predictions, and even chaotic systems due to their ability to capture temporal dependencies and sequential patterns in data. Here we present an RNN model based on convolutional neural networks for modeling the nonlinear nonadiabatic dynamics of hybrid quantum-classical systems. The dynamical evolution of the hybrid systems is governed by equations of motion for classical degrees of freedom and von Neumann equation for electrons. The Physics-Aware Recurrent Convolution (PARC) neural network structure incorporates a differentiator-integrator architecture that inductively models the spatiotemporal dynamics of generic physical systems. Here, we apply our RNN approach to learn the space-time evolution of a one-dimensional semiclassical Holstein model after an interaction quench. For shallow quenches (small changes in electron-lattice coupling), the deterministic dynamics can be accurately captured using a single-CNN-based recurrent network. In contrast, deep quenches induce chaotic evolution, making long-term trajectory prediction significantly more challenging. Nonetheless, we demonstrate that the PARC-CNN architecture can effectively learn the statistical climate of the Holstein model under deep-quench conditions.

Holstein model↗

Risk-Averse Bi-Level Stochastic Network Interdiction Model for Cyber-Security Risk Management

Security of cyber networks is crucial; recent severe cyber-attacks have had a devastating effect on many large organizations. The attack graph, which maps the potential attack paths of a cyber network, is a popular tool for analyzing cyber system vulnerability. In this study, we propose a bi-level stochastic network interdiction model on an attack graph to enable a risk-averse, resource constrained cyber network defender to optimally deploy security countermeasures that protect against attackers with an uncertain budget. This risk- averse conditional-value-at-risk (CVaR) model minimizes a weighted sum of the expected maximum loss over all scenarios and the expected maximum loss from the most damaging attack scenarios. We develop a customized constraint and column generation algorithm to solve our model as well as several acceleration techniques to improve the computational efficiency. Numerical experiments demonstrate that the acceleration techniques enable the solution of relatively large problems within a reasonable amount of time: applying all the acceleration techniques also reduces the average computation time of the basic algorithm by 71% for 100-node graphs. Using metrics called mean-risk value of stochastic solution and value of risk-aversion, computational results suggest that our stochastic risk-averse model significantly outperforms deterministic and risk-neutral models when 1) the distribution of attacker budget is heavy-right-tailed and 2) the defender is highly risk-averse.

Bhuiyan, Tanveer H.↗

A generative graph model for electrical infrastructure networks

We propose a generative graph model for electrical infrastructure networks that accounts for heterogeneity in both node and edge type. To inform the design of this model, we analyze the properties of power grid graphs derived from the U.S. Eastern Interconnection, Texas Interconnection, and Poland transmission system power grids. Across these datasets, we find subgraphs induced by nodes of the same voltage level exhibit shared structural properties atypical to small-world networks, including low local clustering, large diameter and large average distance. On the other hand, we find subgraphs induced by transformer edges linking nodes of different voltage types contain a more limited structure, consisting mainly of small, disjoint star graphs. The goal of our proposed model is to match both these inter and intra-network properties by proceeding in two phases: we first generate subgraphs for each voltage level and then generate transformer edges that connect these subgraphs. The first phase of the model adapts the Chung-Lu random graph model, taking desired vertex degrees and desired diameter as inputs, while the second phase of the model is based on a simpler random star graph generation process. We test the performance of our model by comparing its output across many runs to the aforementioned real data. In nearly all categories tested, we find our model is more accurate in reproducing the unusual mixture of properties apparent in the data than the Chung-Lu model. We also include graph visualization comparisons, as well as a brief analysis of edge-deletion resiliency. Ultimately, our model may be used to generate synthetic graph data, test hypotheses and algorithms at different scales, and serve as a baseline model on top of which further electrical network properties, device models, and interdependencies to other networks, may be appended.

24 POWER TRANSMISSION AND DISTRIBUTION↗

A Fundamental Investigation on the Structural Dynamics of Model Siloxane Networks under Extreme Pressure

Poly(dimethylsiloxane) elastomers are ubiquitous in their application as key components in a range of (often extreme) environments. and as such, they often experience pressures in a region where we do not understand their physical response and therefore can offer no predictive estimate of lifetime of failure . We are proposing a fundamental scientific investigation of the pressure induced morphological changes in model PDMS networks. Both NMR and BDS spectroscopies will be used to interrogate pressure induced changes and meaningful phase relations will be sought. The addition of filler to the networks will be used as an attempt to engineer the network to be more resistant to pressure changes. We have the equipment, model networks and expertise to start this project immediately; funding is sought as an investment in our time allocation.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Convolutional Neural Network Surrogate Models for the Mechanical Properties of Periodic Structures

The work in this paper describes neural network surrogate models for calculating the effective mechanical properties of a periodic composites. The models achieve good accuracy even when only provided with training data sampling a small portion of the design space. As an example, the surrogate models are applied to solving the inverse design problem of finding structures with optimal mechanical properties. The surrogate models are sufficiently accurate to recover optimal solutions in general agreement with established topology optimization methods. However, improvements will be required to develop robust, efficient neural network-based surrogate models and several directions for future research are highlighted here.

42 ENGINEERING↗

Model metamers reveal divergent invariances between biological and artificial neural networks

Deep neural network models of sensory systems are often proposed to learn representational transformations with invariances like those in the brain. To reveal these invariances, we generated ‘model metamers’, stimuli whose activations within a model stage are matched to those of a natural stimulus. Metamers for state-of-the-art supervised and unsupervised neural network models of vision and audition were often completely unrecognizable to humans when generated from late model stages, suggesting differences between model and human invariances. Targeted model changes improved human recognizability of model metamers but did not eliminate the overall human–model discrepancy. The human recognizability of a model’s metamers was well predicted by their recognizability by other models, suggesting that models contain idiosyncratic invariances in addition to those required by the task. Metamer recognizability dissociated from both traditional brain-based benchmarks and adversarial vulnerability, revealing a distinct failure mode of existing sensory models and providing a complementary benchmark for model assessment.

59 BASIC BIOLOGICAL SCIENCES↗

pyflowline: a mesh-independent river network generator for hydrologic models

River networks are crucial in hydrologic and Earth system models. Accurately representing river networks in spatially distributed hydrologic models requires considering the model's spatial discretization and computational mesh. However, current methods of generating river networks for hydrologic models do not typically support unstructured meshes. Unstructured meshes offer numerous advantages over traditional, structured meshes. To overcome this limitation, we developed PyFlowline, a Python package that generates mesh-independent river networks. With PyFlowline, hydrologic modelers can generate conceptual river networks and their topological relationships for both structured and unstructured meshes.

47 OTHER INSTRUMENTATION↗

Dissolved Nutrient Retention Dynamics in River Networks: A Modeling Investigation of Transient Flow and Scale Effects

In this paper, we use a dynamic network flow model, coupled with a transient storage zone biogeochemical model, to simulate dissolved nutrient removal processes at the channel network scale. We have explored several scenarios in respect of the combination of rainfall variability, and the biological and geomorphic characteristics of the catchment, to understand the dominant controls on removal and delivery of dissolved nutrients (e.g., nitrate). These model-based theoretical analyses suggested that while nutrient removal efficiency is lower during flood events compared to during baseflow periods, flood events contribute significantly to bulk nutrient removal, whereas bulk removal during baseflow periods is less. This is due to the fact that nutrient supply is larger during flood events; this trend is even stronger in large rivers. However, the efficiency of removal during both periods decreases in larger rivers, however, due to (i) increasing flow velocities and thus decreasing residence time, and (ii) increasing flow depth, and thus decreasing nutrient uptake rates. Besides nutrient removal processes can be divided into two parts: in the main channel and in the hyporheic transient storage zone. When assessing their relative contributions the size of the transient storage zone is a dominant control, followed by uptake rates in the main channel and in the transient storage zone. Increasing size of the transient storage zone with downstream distance affects the relative contributions to nutrient removal of the water column and the transient storage zone, which also impacts the way nutrient removal rates scale with increasing size of rivers. Intra-annual hydrologic variability has a significant impact on removal rates at all scales: the more variable the streamflow is, compared to mean discharge, the less nutrient is removed in the channel network. A scale-independent first order uptake coefficient, ke, estimated from model simulations, is highly dependent on the relative size of the transient storage zone and how it changes in the downstream direction, as well as the nature of hydrologic variability.

54 ENVIRONMENTAL SCIENCES↗

Characteristics of a cost-effective blood test for colorectal cancer screening

Background: Blood-based biomarker tests can potentially change the landscape of colorectal cancer (CRC) screening. We characterize the conditions under which blood test screening would be as effective and cost-effective as annual fecal immunochemical testing or decennial colonoscopy. Methods: We used the 3 Cancer Information and Surveillance Modeling Network–Colon models to compare scenarios of no screening, annual fecal immunochemical testing, decennial colonoscopy, and a blood test meeting Centers for Medicare & Medicaid (CMS) coverage criteria (74% CRC sensitivity and 90% specificity). We varied the sensitivity to detect CRC (74%-92%), advanced adenomas (10%-50%), screening interval (1-3 years), and test cost ($25-$500). Primary outcomes included quality-adjusted life-years (QALY) gained from screening and costs for a US average-risk cohort of individuals aged 45 years. Results: Annual fecal immunochemical testing yielded 125-163 QALY gained per 1000 at a cost of 3811-5384 dollars per person, whereas colonoscopy yielded 132-177 QALY gained at a cost of 5375-7031 dollars per person. A blood test with 92% CRC sensitivity and 50% advanced adenoma sensitivity yielded 117-162 QALY gained if used every 3 years and 133-173 QALY gained if used every year but would not be cost-effective if priced above $$125 per test. If used every 3 years, a $500 blood test only meeting CMS coverage criteria yielded 83-116 QALY gained at a cost of $8559-$9413 per person. Conclusion: Blood tests that only meet CMS coverage requirements should not be recommended to patients who would otherwise undergo screening by colonoscopy or fecal immunochemical testing because of lower benefit. Blood tests need higher advanced adenoma sensitivity (above 40%) and lower costs (below $125) to be cost-effective.

60 APPLIED LIFE SCIENCES↗