Search NASA⌕ Search

SEARCH · Search NASA

Results for “Network parameter error”

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

Deep Learning of Dark Energy Spectroscopic Instrument Mock Spectra to Find Damped Lyα Systems

We have updated and applied a convolutional neural network (CNN) machine-learning model to discover and characterize damped Ly α systems (DLAs) based on Dark Energy Spectroscopic Instrument (DESI) mock spectra. We have optimized the training process and constructed a CNN model that yields a DLA classification accuracy above 99% for spectra that have signal-to-noise ratios (S/N) above 5 per pixel. The classification accuracy is the rate of correct classifications. This accuracy remains above 97% for lower S/N ≈1 spectra. This CNN model provides estimations for redshift and H i column density with standard deviations of 0.002 and 0.17 dex for spectra with S/N above 3 pixel -1 . Also, this DLA finder is able to identify overlapping DLAs and sub-DLAs. Further, the impact of different DLA catalogs on the measurement of baryon acoustic oscillations (BAO) is investigated. The cosmological fitting parameter result for BAO has less than 0.61% difference compared to analysis of the mock results with perfect knowledge of DLAs. This difference is lower than the statistical error for the first year estimated from the mock spectra: above 1.7%. We also compared the performances of the CNN and Gaussian Process (GP) models. Our improved CNN model has moderately 14% higher purity and 7% higher completeness than an older version of the GP code, for S/N > 3. Both codes provide good DLA redshift estimates, but the GP produces a better column density estimate by 24% less standard deviation. A credible DLA catalog for the DESI main survey can be provided by combining these two algorithms.

79 ASTRONOMY AND ASTROPHYSICS↗

Modeling Time-Dependent Surrogates of Additive-Manufactured Nuclear Fuels Processes

Additive manufacturing (AM) technology is being increasingly adopted in a wide variety of application areas for its ability to rapidly produce, prototype, and customize designs. Recently, a hybrid AM technique was successfully developed at Idaho National Laboratory (INL) to manufacture nuclear fuels [1]. Despite the advantages, this AM technique needs optimization due to defects from a highly complex melting and sintering process. The complex metallurgical phenomena during AM processes are strongly related to parameters such as applied laser power, traveling speed, and scan style, which could lead to differences in density, residual stress, crystallographic texture, and mechanical properties. In addition, stochastic variations in laser energy interaction and associated multiscale/multiphysics phenomena cause variations in microstructure evolution and mechanical properties. Currently, researchers at INL are focusing on developing a comprehensive modeling framework, leveraging INL’s simulation tools MOOSE/MARMOT/BISON/RAVEN [2-4] to describe all steps of this AM process across multiple length scales. Although this advanced framework plays a critical role in enabling enhancements to traditional trial and error approaches for design and optimization of nuclear fuel materials, it remains computationally intense, limiting its use in sensitivity and optimization analysis. In this case, an accurate and inexpensive surrogate becomes an effective tool for providing a tractable approximation of the underlying underline physics. Surrogate models generally not based on the physics of a system are purely mathematical models used to capture the relationships between specific system inputs and outputs. Popular approaches, including neural networks [5], response surfaces [6], and subspace-based reduced order models [7], have been applied to a wide range of disciplines, such as nuclear reactor design, aerospace design and automotive design. In this summary, we employ advanced time-dependent surrogate models such as high-dimensional model representation (HDMR) [8] and physics-informed deep neural network (PINNs) [9] to accelerate the design and optimization of AM process.

42 ENGINEERING↗

Wasserstein Normalized Autoencoder for Anomaly Detection in ProtoDUNE Vertical-Drift Detector

ProtoDUNE Vertical Drift needs a selective triggering algorithm. The detector sits on Earth's surface, so cosmic activity dominates its data. Our goal in this paper is to trigger on neutrino events more robustly than the current deployed Analog-to-Digital Converter Simple Window (ADCSW) model and, eventually, search for signals of Beyond Standard Model (BSM) physics at DUNE as our ultimate North Star objective. As a step towards this goal, we evaluate a Wasserstein Normalized Autoencoder (WNAE) on simulated collection-plane only windows of shape $1\times10\times10$ where Neutrinos act as our BSM-proxy and Cosmic-ray Muons serve as our learned background. The network parameters are fitted using only cosmic-ray muon events as background in order to maintain an unsupervised pipeline. Training uses finite-step Langevin $x^-$ samples, positive-sample reconstruction energy, and an empirical sliced $2$-Wasserstein objective to learn a normalized Boltzmann energy model. We then calibrate on a nominal $5\,\mathrm{Hz}$ operating threshold calculated from cosmic validation data. Both WNAE and ADCSW accept 311 of 194,083 held-out cosmic background events at this $5\,\mathrm{Hz}$ threshold. We found that WNAE accepts 9,677 of 34,634 neutrino-proxy events $(27.9\pm0.24)\%$, compared with 10,076 $(29.1\pm0.24)\%$ for ADCSW, an observed WNAE-minus-ADCSW difference of $-1.15\%$. At another nominal $2\,\mathrm{Hz}$ target threshold, the corresponding efficiencies are $(20.5\pm0.22)\%$ and $(22.6\pm0.22)\%$, respectively. Of the WNAE-selected neutrino proxies at $5\,\mathrm{Hz}$, $(20.8\pm0.4)\%$ of the classified neutrino-proxy events are unique to WNAE, where the uncertainty is an absolute binomial standard error of $0.4\%$.

Zheng, Jake [U. Chicago (main)] (ORCID:00090002189↗

Wasserstein Normalized Autoencoder for Anomaly Detection in ProtoDUNE Vertical-Drift Detector

ProtoDUNE Vertical Drift needs a selective triggering algorithm. The detector sits on Earth's surface, so cosmic activity dominates its data. Our goal in this paper is to trigger on neutrino events more robustly than the current deployed Analog-to-Digital Converter Simple Window (ADCSW) model and, eventually, search for signals of Beyond Standard Model (BSM) physics at DUNE as our ultimate North Star objective. As a step towards this goal, we evaluate a Wasserstein Normalized Autoencoder (WNAE) on simulated collection-plane only windows of shape $1\times10\times10$ where Neutrinos act as our BSM-proxy and Cosmic-ray Muons serve as our learned background. The network parameters are fitted using only cosmic-ray muon events as background in order to maintain an unsupervised pipeline. Training uses finite-step Langevin $x^-$ samples, positive-sample reconstruction energy, and an empirical sliced $2$-Wasserstein objective to learn a normalized Boltzmann energy model. We then calibrate on a nominal $5\,\mathrm{Hz}$ operating threshold calculated from cosmic validation data. Both WNAE and ADCSW accept 311 of 194,083 held-out cosmic background events at this $5\,\mathrm{Hz}$ threshold. We found that WNAE accepts 9,677 of 34,634 neutrino-proxy events $(27.9\pm0.24)\%$, compared with 10,076 $(29.1\pm0.24)\%$ for ADCSW, an observed WNAE-minus-ADCSW difference of $-1.15\%$. At another nominal $2\,\mathrm{Hz}$ target threshold, the corresponding efficiencies are $(20.5\pm0.22)\%$ and $(22.6\pm0.22)\%$, respectively. Of the WNAE-selected neutrino proxies at $5\,\mathrm{Hz}$, $(20.8\pm0.4)\%$ of the classified neutrino-proxy events are unique to WNAE, where the uncertainty is an absolute binomial standard error of $0.4\%$.

Zheng, Jake [Chicago U.] (ORCID:0009000218901379)↗

A novel conditional generative model for efficient ensemble forecasts of state variables in large-scale geological carbon storage

Integrating monitoring data to efficiently update reservoir pressure and CO 2 plume distribution forecasts presents a significant challenge in geological carbon storage (GCS) applications. Inverse modeling techniques are commonly used to fuse observational data and refine reservoir model parameters, thereby improving state variable forecasts. However, these techniques often rely on linear or Gaussian assumptions, which can limit their effectiveness in accurately predicting state variables. Moreover, simulating large-scale three-dimensional (3D) GCS problems is computationally expensive, making iterative runs in inverse problems prohibitive. To address these challenges, we propose a conditional generative model utilizing the score-based diffusion method for real-time 3D pressure and saturation field distribution predictions. Our approach involves solving the score function with a mini-batch-based Monte Carlo estimator to generate labeled data. This data is subsequently employed to train a fully connected neural network, enabling it to learn the conditional sample generator within a supervised learning framework. This method enables the rapid generation of a large ensemble of predictions, facilitating comprehensive uncertainty quantification of state variables. Here we applied our method to forecast the dynamic 3D distributions of pressure and saturation fields over a 30-year injection period. The statistical assessment with low root mean square error (RMSE) values demonstrates that our method can accurately predict the spatiotemporal distributions of both pressure and saturation fields. Moreover, the developed conditional generative model shows high computational efficiency by generating 100 ensemble forecasts of 3D state variables in less than 10 min. The consistency between ensemble averages and ground truth values further illustrates the model’s capability to capture state variable dynamics during the CO 2 plume injection process. Notably, the ground truth values fall within the ensemble forecasts, indicating that our uncertainty quantification effectively captures variability and potential noise in the observations. Thus, the developed conditional generative model proves to be a more efficient, accurate, and practical tool for GCS applications, facilitating timely risk analysis and informed decision-making.

58 GEOSCIENCES↗

Real-Time High-Accuracy Digital Wireless Time, Frequency, and Phase Calibration for Coherent Distributed Antenna Arrays

his work presents a fully-digital high-accuracy real-time calibration procedure for frequency and time alignment of open-loop wirelessly coordinated coherent distributed antenna array (CDA) modems, enabling radio frequency (RF) phase coherence of spatially separated commercial off-the-shelf (COTS) software-defined radios (SDRs) without cables or external references such as the global navigation satellite system (GNSS). Building on previous work using high-accuracy spectrally-sparse time of arrival (ToA) waveforms and a multistep ToA refinement process, a high-accuracy two-way time transfer (TWTT)-based time–frequency coordination approach is demonstrated. Due to the two-way nature of the high-accuracy TWTT approach, the time and frequency estimates are Doppler and multipath tolerant, so long as the channel is reciprocal over the synchronization epoch. This technique is experimentally verified using COTS SDRs in a lab environment in static and dynamic scenarios and with significant multipath scatterers. Time, frequency, and phase stability were evaluated by beamforming over coaxial cables to an oscilloscope which achieved time and phase precisions of ~60– 70 ps , with median coherent gains above 99% using optimized coordination parameters, and a beamforming frequency root-mean-square error (RMSE) of 3.73 ppb in a dynamic scenario. Finally, experiments were conducted to compare the performance of this technique with previous works using an analog continuous-wave two-tone (CWTT) frequency reference technique in both static and dynamic settings.

Clock synchronization↗

Sustainable Energy Solutions Task 4.1 Intelligent Manufacturing of Hybrid Carbon-Glass Fiber-Reinforced Composite Wind Turbine Blades

In this subtask, the manufacturability of hybrid carbon-glass fiber-reinforced composite wind turbine blades using Vacuum-Assisted Resin Transfer Molding (VARTM) was investigated. The objective of this investigation was to study the VARTM process and its parameters to manufacture cost-effective wind turbine blades with no defects (mainly eliminate dry spots and reduce manufacturing time). A 2.5-dimensional model and a 3-dimensional model were developed to simulate mold filling and part curing under different conditions. These conditions included isothermal and non-isothermal filling, curing of the part during and after filling, and placement of injection gates at different locations. Results from this investigation reveal that the process can be simulated and also that manufacturing parameters can be optimized to eliminate dry spot formation and reduce the manufacturing time. Using computer-based models is a cost-effective way to simulate manufacturing of wind turbine blades. The approach taken herein allows the design of the wind blade manufacturing processes without physically running trial-and-error experiments that are expensive and time-consuming; especially for larger blades needed for more demanding environmental conditions. This will benefit the wind energy industry by reducing initial design and manufacturing costs which can later be passed down to consumers and consequently make the wind energy industry more competitive.

17 WIND ENERGY↗

Accelerating massively parallel hemodynamic models of coarctation of the aorta using neural networks

Comorbidities such as anemia or hypertension and physiological factors related to exertion can influence a patient’s hemodynamics and increase the severity of many cardiovascular diseases. Observing and quantifying associations between these factors and hemodynamics can be difficult due to the multitude of co-existing conditions and blood flow parameters in real patient data. Machine learning-driven, physics-based simulations provide a means to understand how potentially correlated conditions may affect a particular patient. Here, we use a combination of machine learning and massively parallel computing to predict the effects of physiological factors on hemodynamics in patients with coarctation of the aorta. We first validated blood flow simulations against in vitro measurements in 3D-printed phantoms representing the patient’s vasculature. We then investigated the effects of varying the degree of stenosis, blood flow rate, and viscosity on two diagnostic metrics – pressure gradient across the stenosis (ΔP) and wall shear stress (WSS) - by performing the largest simulation study to date of coarctation of the aorta (over 70 million compute hours). Using machine learning models trained on data from the simulations and validated on two independent datasets, we developed a framework to identify the minimal training set required to build a predictive model on a per-patient basis. We then used this model to accurately predict ΔP (mean absolute error within 1.18 mmHg) and WSS (mean absolute error within 0.99 Pa) for patients with this disease.

59 BASIC BIOLOGICAL SCIENCES↗

Surface daytime net radiation estimation using artificial neural networks

Net all-wave surface radiation (R n ) is one of the most important fundamental parameters in various applications. However, conventional R n measurements are difficult to collect because of the high cost and ongoing maintenance of recording instruments. Therefore, various empirical R n estimation models have been developed. This study presents the results of two artificial neural network (ANN) models (general regression neural networks (GRNN) and Neuroet) to estimate R n globally from multi-source data, including remotely sensed products, surface measurements, and meteorological reanalysis products. R n estimates provided by the two ANNs were tested against in-situ radiation measurements obtained from 251 global sites between 1991–2010 both in global mode (all data were used to fit the models) and in conditional mode (the data were divided into four subsets and the models were fitted separately). Based on the results obtained from extensive experiments, it has been proved that the two ANNs were superior to linear-based empirical models in both global and conditional modes and that the GRNN performed better and was more stable than Neuroet. The GRNN estimates had a determination coefficient (R 2 ) of 0.92, a root mean square error (RMSE) of 34.27 W·m –2 , and a bias of –0.61 W·m –2 in global mode based on the validation dataset. In conclusion, ANN methods are a potentially powerful tool for global R n estimation.

14 SOLAR ENERGY↗

Comment on"Sensitivity of the active fracture model parameter to fracture network orientation and injection scenarios" by Basagaoglu et al. (2009)

Basagaoglu et al. (2009) present a study on detailed unsaturated flow behavior in two-dimensional fracture networks using numerical experiments (simulations) based on the lattice-Boltzmann method. Their results are valuable for improving our understanding of unsaturated flow processes and evaluating the active fracture model (AFM) that was developed for capturing large-scale preferential flow in fractured rocks (Liu et al., 1998; 2003). As indicated in Basagaoglu et al. (2009), a previous study was conducted to evaluate the AFM with numerical experiments (Seol et al., 2003). However, the methodology used in that study and the corresponding conclusions are highly questioned for the following two reasons. First, the evaluation relies on a condition that simulated water flow processes in a fracture network are adequately represented with a continuum approach, because they draw their conclusions by comparing simulation results with those obtained from a dual-continuum model based on the AFM. No effort was made by Seol et al. (2003) to justify the validity of the continuum approach for their specific fracture network that includes a small number of fractures only. (The analyses of Basagaoglu et al. (2009) do not need the similar condition.) Second, Seol et al. (2003) use numerical dispersion to represent the matrix diffusion process. This treatment is not valid simply because numerical dispersion results from numerical errors and is not a physical process.

54 ENVIRONMENTAL SCIENCES↗

Fast and accurate reduced-order modeling of a MOOSE-based additive manufacturing model with operator learning

One predominant challenge in additive manufacturing (AM) is to achieve specific material properties by manipulating manufacturing process parameters during the runtime. Such manipulation tends to increase the computational load imposed on existing simulation tools employed in AM. The goal of the present work is to construct a fast and accurate reduced-order model (ROM) for an AM model developed within the Multiphysics Object-Oriented Simulation Environment (MOOSE) framework, ultimately reducing the time/cost of AM control and optimization processes. Our adoption of the operator learning (OL) approach enabled us to learn a family of differential equations produced by altering process variables in the laser’s Gaussian point heat source. More specifically, we used the Fourier neural operator (FNO) and deep operator network (DeepONet) to develop ROMs for time-dependent responses. Furthermore, we benchmarked the performance of these OL methods against a conventional deep neural network (DNN)-based ROM. Ultimately, we found that OL methods offer comparable performance and, in terms of accuracy and generalizability, even outperform DNN at predicting scalar model responses. The DNN-based ROM afforded the fastest training time. Furthermore, all the ROMs were faster than the original MOOSE model yet still provided accurate predictions. FNO had a smaller mean prediction error than DeepONet, with a larger variance for time-dependent responses. Unlike DNN, both FNO and DeepONet were able to simulate time series data without the need for dimensionality reduction techniques. Finally, the present work can help facilitate the AM optimization process by enabling faster execution of simulation tools while still preserving evaluation accuracy.

36 MATERIALS SCIENCE↗

Knowledge-based prediction of three-dimensional dose distributions for external beam radiotherapy

Purpose: To demonstrate knowledge-based 3D dose prediction for external beam radiotherapy. Methods: Using previously treated plans as training data, an artificial neural network (ANN) was trained to predict a dose matrix based on patient-specific geometric and planning parameters, such as the closest distance (r) to planning target volume (PTV) and organ-at-risks (OARs). Twenty-three prostate and 43 stereotactic radiosurgery/radiotherapy (SRS/SRT) cases with at least one nearby OAR were studied. All were planned with volumetric-modulated arc therapy to prescription doses of 81 Gy for prostate and 12–30 Gy for SRS. Using these clinically approved plans, ANNs were trained to predict dose matrix and the predictive accuracy was evaluated using the dose difference between the clinical plan and prediction, δD = D{sub clin} − D{sub pred}. The mean (〈δD{sub r}〉), standard deviation (σ{sub δD{sub r}}), and their interquartile range (IQR) for the training plans were evaluated at a 2–3 mm interval from the PTV boundary (r{sub PTV}) to assess prediction bias and precision. Initially, unfiltered models which were trained using all plans in the cohorts were created for each treatment site. The models predict approximately the average quality of OAR sparing. Emphasizing a subset of plans that exhibited superior to the average OAR sparing during training, refined models were created to predict high-quality rectum sparing for prostate and brainstem sparing for SRS. Using the refined model, potentially suboptimal plans were identified where the model predicted further sparing of the OARs was achievable. Replans were performed to test if the OAR sparing could be improved as predicted by the model. Results: The refined models demonstrated highly accurate dose distribution prediction. For prostate cases, the average prediction bias for all voxels irrespective of organ delineation ranged from −1% to 0% with maximum IQR of 3% over r{sub PTV} ∈ [ − 6, 30] mm. The average prediction error was less than 10% for the same r{sub PTV} range. For SRS cases, the average prediction bias ranged from −0.7% to 1.5% with maximum IQR of 5% over r{sub PTV} ∈ [ − 4, 32] mm. The average prediction error was less than 8%. Four potentially suboptimal plans were identified for each site and subsequent replanning demonstrated improved sparing of rectum and brainstem. Conclusions: The study demonstrates highly accurate knowledge-based 3D dose predictions for radiotherapy plans.

61 RADIATION PROTECTION AND DOSIMETRY↗

AutoSourceID-Classifier: Star-galaxy classification using a convolutional neural network with spatial information

Aims.Traditional star-galaxy classification techniques often rely on feature estimation from catalogs, a process susceptible to introducing inaccuracies, thereby potentially jeopardizing the classification’s reliability. Certain galaxies, especially those not manifesting as extended sources, can be misclassified when their shape parameters and flux solely drive the inference. We aim to create a robust and accurate classification network for identifying stars and galaxies directly from astronomical images. Methods.The AutoSourceID-Classifier (ASID-C) algorithm developed for this work uses 32x32 pixel single filter band source cutouts generated by the previously developed AutoSourceID-Light (ASID-L) code. By leveraging convolutional neural networks (CNN) and additional information about the source position within the full-field image, ASID-C aims to accurately classify all stars and galaxies within a survey. Subsequently, we employed a modified Platt scaling calibration for the output of the CNN, ensuring that the derived probabilities were effectively calibrated, delivering precise and reliable results. Results.We show that ASID-C, trained on MeerLICHT telescope images and using the Dark Energy Camera Legacy Survey (DECaLS) morphological classification, is a robust classifier and outperforms similar codes such as SourceExtractor. To facilitate a rigorous comparison, we also trained an eXtreme Gradient Boosting (XGBoost) model on tabular features extracted by SourceExtractor. While this XGBoost model approaches ASID-C in performance metrics, it does not offer the computational efficiency and reduced error propagation inherent in ASID-C’s direct image-based classification approach. ASID-C excels in low signal-to-noise ratio and crowded scenarios, potentially aiding in transient host identification and advancing deep-sky astronomy.

Astronomy & Astrophysics↗

Characterizing particle-scale equilibrium adsorption and kinetics of uranium(VI) desorption from U-contaminated sediments

Rates of contaminant U(VI) release from individual size fractions of a composite sediment from the seasonally saturated lower vadose zone of the Hanford 300-Area were examined in flow-through batch reactors to maintain quasi-constant chemical conditions. Variability in equilibrium adsorption among the various size fractions was determined in static batch reactors and analyzed using the surface complexation modeling approach. The estimated stoichiometric coefficients of U(VI) surface complexation reactions with respect to pH and carbonate concentrations varied with size fractions. This source of variability significantly increased the uncertainty in U(VI) conditional equilibrium constants over that estimated from experimental errors alone. A minimum difference between conditional equilibrium constants was established in order to evaluate statistically significant differences between sediment adsorption properties. A set of equilibrium and kinetic expressions for cation exchange, calcite dissolution, aerobic respiration, and silica dissolution were incorporated in a reaction-rate model to describe the temporal evolution of solute concentrations observed during the flow-through batch experiments. Parameters in the reaction-rate model, calibrated using experimental data for select size fractions, predicted the changes in solute concentrations for the bulk, <2 mm, sediment sample. Kinetic U(VI) desorption was well described using a multi-rate surface complexation model with an assumed lognormal distribution for the rate constants. The estimated mean and standard deviation were the same for all < 2mm size fractions, but differed in the 2-8mm size fraction. Micropore volumes in the varied size fractions were also similar as assessed using t-plots to analyze N2 desorption data. These findings provide further support for the link between microporosity and particle-scale mass transfer rates controlling kinetic U(VI) adsorption/desorption and for the utility of N2 desorption isotherms for characterizing pore networks that influence mass transfer rates.

Stoliker, Deborah L.↗

Attention-based 3D – convolutional neural network model for mechanical property predictions using visible light images in metal additive manufacturing

Additive manufacturing (AM), while commonly used for rapid prototyping and creating components with complex geometries, has not been widely adopted for critical applications across the aerospace, automotive, defense, energy, and medical industries. This is, in part, due to the challenges of controlling flaws and uncertainty in the mechanical behavior of additively manufactured components. In recent years, there has been an increase in research aimed at predicting the final mechanical properties of additively manufactured components during the printing process. To address these issues, a 3D-CNN model was trained using low-cost in situ visible-light camera data, anomaly classifications, and the chosen process parameters to predict the ultimate tensile strength (UTS), yield strength (YS), total elongation (TE), and uniform elongation (UE). The 3D-CNN layers of the model employed attention mechanisms to prioritize features in the data, thereby improving prediction accuracy. Furthermore, the effect of each process parameter and anomaly class is investigated using attention-based dynamic sigmoid weighted gates to interpret the influence each class has on the final prediction. Different combinations of the in situ data were fed into the 3D-CNN, with varying amounts of image layers, to determine the ideal combination for predicting mechanical properties in situ. Here, the 3D-CNN model achieved mean absolute percentage errors (MAPE) below 5% for both UTS and YS while using only a single camera input and under half of the available image layers.

36 MATERIALS SCIENCE↗

Deep Learning Approaches to Surrogates for Solving the Diffusion Equation for Mechanistic Real-World Simulations

In many mechanistic medical, biological, physical, and engineered spatiotemporal dynamic models the numerical solution of partial differential equations (PDEs), especially for diffusion, fluid flow and mechanical relaxation, can make simulations impractically slow. Biological models of tissues and organs often require the simultaneous calculation of the spatial variation of concentration of dozens of diffusing chemical species. One clinical example where rapid calculation of a diffusing field is of use is the estimation of oxygen gradients in the retina, based on imaging of the retinal vasculature, to guide surgical interventions in diabetic retinopathy. Furthermore, the ability to predict blood perfusion and oxygenation may one day guide clinical interventions in diverse settings, i.e., from stent placement in treating heart disease to BOLD fMRI interpretation in evaluating cognitive function (Xie et al., 2019; Lee et al., 2020). Since the quasi-steady-state solutions required for fast-diffusing chemical species like oxygen are particularly computationally costly, we consider the use of a neural network to provide an approximate solution to the steady-state diffusion equation. Machine learning surrogates, neural networks trained to provide approximate solutions to such complicated numerical problems, can often provide speed-ups of several orders of magnitude compared to direct calculation. Surrogates of PDEs could enable use of larger and more detailed models than are possible with direct calculation and can make including such simulations in real-time or near-real time workflows practical. Creating a surrogate requires running the direct calculation tens of thousands of times to generate training data and then training the neural network, both of which are computationally expensive. Often the practical applications of such models require thousands to millions of replica simulations, for example for parameter identification and uncertainty quantification, each of which gains speed from surrogate use and rapidly recovers the up-front costs of surrogate generation. We use a Convolutional Neural Network to approximate the stationary solution to the diffusion equation in the case of two equal-diameter, circular, constant-value sources located at random positions in a two-dimensional square domain with absorbing boundary conditions. Such a configuration caricatures the chemical concentration field of a fast-diffusing species like oxygen in a tissue with two parallel blood vessels in a cross section perpendicular to the two blood vessels. To improve convergence during training, we apply a training approach that uses roll-back to reject stochastic changes to the network that increase the loss function. The trained neural network approximation is about 1000 times faster than the direct calculation for individual replicas. Because different applications will have different criteria for acceptable approximation accuracy, we discuss a variety of loss functions and accuracy estimators that can help select the best network for a particular application. We briefly discuss some of the issues we encountered with overfitting, mismapping of the field values and the geometrical conditions that lead to large absolute and relative errors in the approximate solution.

60 APPLIED LIFE SCIENCES↗

An Investigation of LES Wall Modeling for Rayleigh–Bénard Convection via Interpretable and Physics-Aware Feedforward Neural Networks with DNS

Abstract The traditional approach of using the Monin–Obukhov similarity theory (MOST) to model near-surface processes in large-eddy simulations (LESs) can lead to significant errors in natural convection. In this study, we propose an alternative approach based on feedforward neural networks (FNNs) trained on output from direct numerical simulation (DNS). To evaluate the performance, we conduct both a priori and a posteriori tests. In the a priori (offline) tests, we compare the statistics of the surface shear stress and heat flux, computed from filtered DNS input variables, to the stress and flux obtained from the filtered DNS. Additionally, we investigate the importance of various input features using the Shapley additive explanations value and the conditional average of the filter grid cells. In the a posteriori (online) tests, we implement the trained models in the System for Atmospheric Modeling (SAM) LES and compare the LES-generated surface shear stress and heat flux with those in the DNS. Our findings reveal that vertical velocity, a traditionally overlooked flow quantity, is one of the most important input features for determining the wall fluxes. Increasing the number of input features improves the a priori test results but does not always improve the model performance in the a posteriori tests because of the differences in input variables between the LES and DNS. Last, we show that physics-aware FNN models trained with logarithmic and scaled parameters can well extrapolate to more intense convection scenarios than in the training dataset, whereas those trained with primitive flow quantities cannot. Significance Statement The traditional near-surface turbulence model, based on a shear-dominated boundary layer flow, does not represent near-surface turbulence in natural convection. Using a feedforward neural network (FNN), we can construct a more accurate model that better represents the near-surface turbulence in various flows and reveals previously overlooked controlling factors and process interactions. Our study shows that the FNN-generated models outperform the traditional model and highlight the importance of the near-surface vertical velocity. Furthermore, the physics-aware FNN models exhibit the potential to extrapolate to convective flows of various intensities beyond the range of the training dataset, suggesting their broader applicability for more accurate modeling of near-surface turbulence.

54 ENVIRONMENTAL SCIENCES↗

A hybrid embedded discrete fracture model for simulating tight porous media with complex fracture systems

Numerical simulation considering complex fracture networks in tight oil reservoir becomes a hot issue during these years. Recently, embedded discrete fracture model (EDFM) shows great advantages over the DFM approaches in both calculation efficiency and simplicity of simulation workflow. EDFM has been widely used in tight oil/gas natural energy or enhanced oil recovery (EOR) development. If a large number of small-scale natural fractures exist in reservoir, the calculation efficiency of EDFM method may decrease. A new hybrid simulation model combining the EDFM and dual porosity (DP) model is proposed. The large-scale fractures are dealt with EDFM method while small scale fractures are dealt with DP model. Considering the low permeability of matrix, transient transfer between matrix and natural fracture is considered. The proposed hybrid model is validated compared with the local grid refinement (LGR) model and EDFM-MINC model, and the difference between transient transfer and pseudo-steady-state transfer effect on production is presented. Finally, two test cases are presented, and the sensitivity of some key parameters are investigated. Result shows that the new simulation model has higher calculation efficiency than LGR and EDFM-MINC models. Meanwhile, it honors a good accuracy. The pseudo-steady-state transfer between matrix and natural fracture leads to big error in tight oil production performance analysis both in single phase and two phases flow. Low matrix permeability and big matrix size increase the production difference between pseudo-steady-state transfer and transient transfer models. The proposed model can simulate reservoirs considering stochastic permeability distribution.

58 GEOSCIENCES↗