Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “empirical deep learning”

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.

103 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Evaluating Physics-Informed Neural Network Performance for Seismic Discrimination between Earthquakes and Explosions

In this article, we evaluate adding a weak physics constraint, that is, a physics‐based empirical relationship, to the loss function with a physics‐informed manner in local distance explosion discrimination in the hope of improving the generalization capability of the machine learning (ML) model. We compare the proposed model with the two‐branch model we previously developed, as well as with a pure data‐driven model. Unexpectedly, the proposed model did not consistently outperform the pure data‐driven model. By varying the level of inconsistency in the training data, we find this approach is modulated by the strength of the physics relationship. In conclusion, this result has important implications for how to best incorporate physical constraints in ML models.

58 GEOSCIENCES↗
DOE OSTI2024

Generalization error guaranteed auto-encoder-based nonlinear model reduction for operator learning

Many physical processes in science and engineering are naturally represented by operators between infinite-dimensional function spaces. The problem of operator learning, in this context, seeks to extract these physical processes from empirical data, which is challenging due to the infinite or high dimensionality of data. An integral component in addressing this challenge is model reduction, which reduces both the data dimensionality and problem size. In this paper, we utilize low-dimensional nonlinear structures in model reduction by investigating Auto-Encoder-based Neural Network (AENet). AENet first learns the latent variables of the input data and then learns the transformation from these latent variables to corresponding output data. Our numerical experiments validate the ability of AENet to accurately learn the solution operator of nonlinear partial differential equations. Furthermore, we establish a mathematical and statistical estimation theory that analyzes the generalization error of AENet. Finally, our theoretical framework shows that the sample complexity of training AENet is intricately tied to the intrinsic dimension of the modeled process, while also demonstrating the robustness of AENet to noise.

Auto-encoder↗
DOE OSTI2023

Using Data-Driven Prediction of Downstream 1D River Flow to Overcome the Challenges of Hydrologic River Modeling

Methods for downstream river flow prediction can be categorized into physics-based and empirical approaches. Although based on well-studied physical relationships, physics-based models rely on numerous hydrologic variables characteristic of the specific river system that can be costly to acquire. Moreover, simulation is often computationally intensive. Conversely, empirical models require less information about the system being modeled and can capture a system’s interactions based on a smaller set of observed data. This article introduces two empirical methods to predict downstream hydraulic variables based on observed stream data: a linear programming (LP) model, and a convolutional neural network (CNN). We apply both empirical models within the Colorado River system to a site located on the Green River, downstream of the Yampa River confluence and Flaming Gorge Dam, and compare it to the physics-based model Streamflow Synthesis and Reservoir Regulation (SSARR) currently used by federal agencies. Results show that both proposed models significantly outperform the SSARR model. Moreover, the CNN model outperforms the LP model for hourly predictions whereas both perform similarly for daily predictions. Although less accurate than the CNN model at finer temporal resolution, the LP model is ideal for linear water scheduling tools.

13 HYDRO ENERGY↗
DOE OSTI2023

Graph neural networks predict energetic and mechanical properties for models of solid solution metal alloy phases

Here, we developed a PyTorch-based architecture called HydraGNN that implements graph convolutional neural networks (GCNNs) to predict the formation energy and the bulk modulus for models of solid solution alloys for various atomic crystal structures and relaxed volumes. We trained the GCNN surrogate model on a dataset for nickel–niobium (NiNb) generated by the embedded atom model (EAM) empirical interatomic potential for demonstration purposes. The dataset was generated by calculating the formation energy and the bulk modulus as a prototypical elastic property for optimized geometries starting from initial body-centered cubic (BCC), face-centered cubic (FCC), and hexagonal compact packed (HCP) crystal structures, with configurations spanning the possible compositional range for each of the three types of initial crystal structures. Numerical results show that the GCNN model effectively predicts both the formation energy and the bulk modulus as function of the optimized crystal structure, relaxed volume, and configurational entropy of the model structures for solid solution alloys.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Neural-Network-Enhanced COTSIM: Advancing Predictive Capabilities for Fast DIII-D Simulations

Sustaining fusion reactions in tokamaks requires heating plasma to thermonuclear temperatures while maintaining confinement and stability. Neutral beam injection (NBI) provides heating, current drive, torque, and fueling, while electron cyclotron (EC) waves are widely used for heating and current drive; together, these actuators shape the plasma current, temperature, and density profiles. The control-oriented tokamak simulator (COTSIM), a predictive, control-oriented code, has been enhanced with neural-network surrogates for transport and sources. Turbulent transport is predicted by MMMnet—a neural-network version of the updated multimode model (MMM 9.0.10)—with significantly reduced computation time relative to MMM; neoclassical transport follows the Chang–Hinton model. NUBEAMnet, a surrogate of the Monte Carlo NUBEAM module, predicts beam-driven heating, current, and torque. EC heating and current drive use a control-oriented, empirically scaled source model; plasma resistivity follows the Spitzer formulation; bootstrap current uses the Sauter model. Equilibrium is computed using both prescribed and fixed-boundary solvers (FBSs), and the pedestal structure is modeled with an empirical pedestal model. For a representative DIII-D discharge, COTSIM predicts electron and ion temperature and safety-factor profiles in close agreement with TRANSP predictive and interpretive simulations while extending predictions through the pedestal region to the plasma edge (versus 80% of the minor radius in TRANSP). Furthermore, the equivalent COTSIM simulation runs in under 3 min compared to about 2 h for TRANSP, enabling rapid scenario planning, optimization of tokamak operation, and between-pulse control design.

Control-oriented tokamak simulator (COTSIM)↗
DOE OSTI2022

Predicting Solar Energetic Particles Using SDO/HMI Vector Magnetic Data Products and a Bidirectional LSTM Network

Solar energetic particles (SEPs) are an essential source of space radiation, and are hazardous for humans in space, spacecraft, and technology in general. In this paper, we propose a deep-learning method, specifically a bidirectional long short-term memory (biLSTM) network, to predict if an active region (AR) would produce an SEP event given that (i) the AR will produce an M- or X-class flare and a coronal mass ejection (CME) associated with the flare, or (ii) the AR will produce an M- or X-class flare regardless of whether or not the flare is associated with a CME. The data samples used in this study are collected from the Geostationary Operational Environmental Satellite's X-ray flare catalogs provided by the National Centers for Environmental Information. We select M- and X-class flares with identified ARs in the catalogs for the period between 2010 and 2021, and find the associations of flares, CMEs, and SEPs in the Space Weather Database of Notifications, Knowledge, Information during the same period. Each data sample contains physical parameters collected from the Helioseismic and Magnetic Imager on board the Solar Dynamics Observatory. Experimental results based on different performance metrics demonstrate that the proposed biLSTM network is better than related machine-learning algorithms for the two SEP prediction tasks studied here. We also discuss extensions of our approach for probabilistic forecasting and calibration with empirical evaluation

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Enhancing generative molecular design via uncertainty-guided fine-tuning of variational autoencoders

In recent years, deep generative models have been successfully applied to various molecular design tasks, particularly in the life and materials sciences. One critical challenge for pre-trained generative molecular design (GMD) models is to fine-tune them to be better suited for downstream design tasks that aim at optimizing specific molecular properties. However, redesigning and training an existing effective generative model from scratch for each new design task are impractical. Furthermore, the black-box nature of typical downstream tasks that involve property prediction makes it nontrivial to optimize the generative model in a task-specific manner. In this work, we propose an uncertainty-guided fine-tuning strategy that can effectively enhance a pre-trained variational autoencoder (VAE) for GMD through performance feedback in an active learning setting. The strategy begins by quantifying the model uncertainty of the generative model using an efficient active subspace-based UQ (uncertainty quantification) scheme. Next, the decoder diversity within the characterized model uncertainty class is explored to expand the viable space of molecular generation. The low-dimensionality of the active subspace makes this exploration tractable using a black-box optimization scheme, which in turn enables us to identify and leverage a diverse set of high-performing models to generate enhanced molecules. Empirical results across six target molecular properties using multiple VAE-based generative models demonstrate that our uncertainty-guided fine-tuning strategy consistently leads to improved models that outperform the original pre-trained models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations

In the last 50 years there has been a tremendous progress in solving numerically the Navier-Stokes equations using finite differences, finite elements, spectral, and even meshless methods. Yet, in many real cases, we still cannot incorporate seamlessly (multi-fidelity) data into existing algorithms, and for industrial-complexity applications the mesh generation is time consuming and still an art. Moreover, solving ill-posed problems (e.g., lacking boundary conditions) or inverse problems is often prohibitively expensive and requires different formulations and new computer codes. Here, we employ physics-informed neural networks (PINNs), encoding the governing equations directly into the deep neural network via automatic differentiation, to overcome some of the aforementioned limitations for simulating incompressible laminar and turbulent flows. We develop the Navier-Stokes flow nets (NSFnets) by considering two different mathematical formulations of the Navier-Stokes equations: the velocity-pressure (VP) formulation and the vorticity-velocity (VV) formulation. Since this is a new approach, we first select some standard benchmark problems to assess the accuracy, convergence rate, computational cost and flexibility of NSFnets; analytical solutions and direct numerical simulation (DNS) databases provide proper initial and boundary conditions for the NSFnet simulations. The spatial and temporal coordinates are the inputs of the NSFnets, while the instantaneous velocity and pressure fields are the outputs for the VP-NSFnet, and the instantaneous velocity and vorticity fields are the outputs for the VV-NSFnet. This is unsupervised learning and, hence, no labeled data are required beyond boundary and initial conditions and the fluid properties. The residuals of the VP or VV governing equations, together with the initial and boundary conditions, are embedded into the loss function of the NSFnets. No data is provided for the pressure to the VP-NSFnet, which is a hidden state and is obtained via the incompressibility constraint without extra computational cost. Unlike the traditional numerical methods, NSFnets inherit the properties of neural networks (NNs), hence the total error is composed of the approximation, the optimization, and the generalization errors. Here, we empirically attempt to quantify these errors by varying the sampling (“residual”) points, the iterative solvers, and the size of the NN architecture. For the laminar flow solutions, we show that both the VP and the VV formulations are comparable in accuracy but their best performance corresponds to different NN architectures. The initial convergence rate is fast but the error eventually saturates to a plateau due to the dominance of the optimization error. For the turbulent channel flow, we show that NSFnets can sustain turbulence at , but due to expensive training we only consider part of the channel domain and enforce velocity boundary conditions on the subdomain boundaries provided by the DNS data base. We also perform a systematic study on the weights used in the loss function for balancing the data and physics components, and investigate a new way of computing the weights dynamically to accelerate training and enhance accuracy. In the last part, we demonstrate how NSFnets should be used in practice, namely for ill-posed problems with incomplete or noisy boundary conditions as well as for inverse problems. We obtain reasonably accurate solutions for such cases as well without the need to change the NSFnets and at the same computational cost as in the forward well-posed problems. As a result, we also present a simple example of transfer learning that will aid in accelerating the training of NSFnets for different parameter settings.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Data-Driven Discovery of Linear Molecular Probes with Optimal Selective Affinity for PFAS in Water

Approaches to tackle the wide and growing variety of highly persistent per- and polyfluoroalkyl substances (PFAS) are of pressing global need because of their detrimental human health effects, such as cancer, birth defects, and hormone imbalance. Sensitive, selective, and easy-to-use real-time sensors to monitor and detect PFAS and sorbents to extract them are critical to meeting government-mandated environmental concentrations. In this work, we combine all-atom molecular dynamics simulations, enhanced sampling, deep representational learning, and Bayesian optimization to perform high-throughput virtual screening for highly sensitive and selective molecular probes. Our molecular design space consists of 3850 linear hydrocarbon chains with varying degrees of halogenation with and without amine- and phosphine-based headgroups. By employing a data-driven search process, we efficiently explore the molecular design space to optimize the sensitivity to perfluorooctanesulfonic acid (PFOS) as a prototypical PFAS analyte and selectivity relative to a sodium dodecyl sulfate (SDS) interferent. We calculate 504 Gibbs free energies of probe-analyte and probe-interferent interactions and identify probes with PFOS association free energies of up to (-ΔG PFOS ) = 9.8 ± 0.2 kJ/mol and selectivities relative to SDS of (-ΔΔG PFOS–SDS ) = 3.1 ± 1.5 kJ/mol. A C 11 Br 23 P(CH 3 ) 2 probe containing 11 backbone brominated carbons and a tertiary phosphine headgroup possesses the most sensitive binding constant to PFOS within the defined search space of K b PFOS = 177.4 ± 12.7, and a semibrominated probe C 5 H 11 C 7 Br 14 N(CH 3 ) 2 containing 12 backbone carbons and a tertiary amine headgroup possesses the highest selectivity relative to SDS of K b PFOS /K b SDS = 4.6 ± 1.7. A retrospective analysis of our data to extract interpretable design rules reveals that the sensitivity of linear hydrogenated probes increases by approximately 1 kJ/mol per C–C bond. The addition or removal of halogen atoms and amine or phosphine headgroups produces nonmonotonic changes in both sensitivity and selectivity with changes to the sensitivity of up to 2.5 kJ/mol. Finally, this work places empirical limitations on the performance of a wide range of linear probes for PFOS detection and offers a generic strategy for high-throughput computational screening to promote selective and sensitive binding.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Short-term Electricity Price Forecasting with Constrained Regressors

The volatility of electricity price presents a challenge to market participants as their decision-making process are highly depend on the accuracy of price forecasts. However, there is growing empirical evidence of increasing price volatility and price spikes in electricity markets as a result of variable renewable energy generation, extreme weather events, and other factors. The distribution shift caused by spikes in electricity price data differentiates the forecasting tasks from other renewable energy sources. Moreover, the observations may be compromised by cyberattacks and thus not available in the testing phase. To this end, we propose a Similarity-Enhanced Electricity Decomposition Forecasting model (SEED-Forecaster) to address the missing response problem and spikes capturing in short-term electricity price forecasting. The effectiveness of the proposed framework is tested on real-world electricity price data from California Independent System Operator (CAISO). Numerical results of case studies show that the proposed SEED-Forecsater can enhance forecasting performance, particularly in capturing electricity spikes, even under conditions without regressors during testing stage.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

Learning curves for drug response prediction in cancer cell lines

Motivated by the size and availability of cell line drug sensitivity data, researchers have been developing machine learning (ML) models for predicting drug response to advance cancer treatment. As drug sensitivity studies continue generating drug response data, a common question is whether the generalization performance of existing prediction models can be further improved with more training data. We utilize empirical learning curves for evaluating and comparing the data scaling properties of two neural networks (NNs) and two gradient boosting decision tree (GBDT) models trained on four cell line drug screening datasets. The learning curves are accurately fitted to a power law model, providing a framework for assessing the data scaling behavior of these models. The curves demonstrate that no single model dominates in terms of prediction performance across all datasets and training sizes, thus suggesting that the actual shape of these curves depends on the unique pair of an ML model and a dataset. The multi-input NN (mNN), in which gene expressions of cancer cells and molecular drug descriptors are input into separate subnetworks, outperforms a single-input NN (sNN), where the cell and drug features are concatenated for the input layer. In contrast, a GBDT with hyperparameter tuning exhibits superior performance as compared with both NNs at the lower range of training set sizes for two of the tested datasets, whereas the mNN consistently performs better at the higher range of training sizes. Moreover, the trajectory of the curves suggests that increasing the sample size is expected to further improve prediction scores of both NNs. These observations demonstrate the benefit of using learning curves to evaluate prediction models, providing a broader perspective on the overall data scaling characteristics. A fitted power law learning curve provides a forward-looking metric for analyzing prediction performance and can serve as a co-design tool to guide experimental biologists and computational scientists in the design of future experiments in prospective research studies.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

A data-driven method for modelling dissipation rates in stratified turbulence

We present a deep probabilistic convolutional neural network (PCNN) model for predicting local values of small-scale mixing properties in stratified turbulent flows, namely the dissipation rates of turbulent kinetic energy and density variance, $\varepsilon$ and $\chi$ . Inputs to the PCNN are vertical columns of velocity and density gradients, motivated by data typically available from microstructure profilers in the ocean. The architecture is designed to enable the model to capture several characteristic features of stratified turbulence, in particular the dependence of small-scale isotropy on the buoyancy Reynolds number $Re_b:=\varepsilon /(\nu N^2)$ , where $\nu$ is the kinematic viscosity and $N$ is the background buoyancy frequency, the correlation between suitably locally averaged density gradients and turbulence intensity and the importance of capturing the tails of the probability distribution functions of values of dissipation. Empirically modified versions of commonly used isotropic models for $\varepsilon$ and $\chi$ that depend only on vertical derivatives of density and velocity are proposed based on the asymptotic regimes $Re_b\ll 1$ and $Re_b\gg 1$ , and serve as an instructive benchmark for comparison with the data-driven approach. When trained and tested on a simulation of stratified decaying turbulence which accesses a range of turbulent regimes (associated with differing values of $Re_b$ ), the PCNN outperforms assumptions of isotropy significantly as $Re_b$ decreases, and additionally demonstrates improvements over the fitted empirical models. A differential sensitivity analysis of the PCNN facilitates a comparison with the theoretical models and provides a physical interpretation of the features enabling it to make improved predictions.

42 ENGINEERING↗
DOE OSTI2023

Improving the Estimation of the Atmospheric Water Vapor Pressure Using Interpretable Long Short-Term Memory Networks: Dataset, Python code, and trained models

Atmospheric water vapor pressure is an essential meteorological control on land surface and hydrologic processes. It is not as frequently observed as other meteorologic conditions, but often inferred through the August–Roche–Magnus formula by simply assuming dew point and daily minimum temperatures are equivalent or by empirically correlating the two temperatures using an aridity correction. The performance of both methods varies considerably across different regions and during different time periods; obtaining consistently accurate estimates across space and time remains a great challenge. We applied an interpretable Long Short-Term Memory (iLSTM) network conditioned on static, location specific attributes to estimate daily vapor pressure for 83 FLUXNET sites in the United States and Canada. This data package includes all raw data of the 83 FLUXNET sites, input data for model training/validation/test, trained models and results, and python codes for the manuscript "Improving the Estimation of the Atmospheric Water Vapor Pressure Using an Interpretable Long Short-term Memory Network". Specifically, it consists of five parts. - First, "1_Daymet_data_83sites.zip" includes raw data downloaded from Daymet for the 83 sites used in the paper according to their longitude and latitude, in which vapor pressure is used. It also includes a pre-processed CSV data file combining all data from the 83 sites which is specifically used for the paper. - Second, "2_Fluxnet2015_data_83sites.zip" includes raw half hourly data of the 83 sites downloaded from FLUXNET2015 data portal, pre-processed daily data of the 83 sites, a CSV file including combined pre-processed daily data of the 83 sites, and a CSV file including the information (site ID, site name, latitude, longitude, data available period) of the 83 sites. - Third, "3_MODIS_LAI_data_83sites_raw.zip" includes raw leaf area index (LAI) data downloaded from the AppEEARs data portal. - Fourth, "4_Scripts.zip" includes all scripts related to model training and post-processing of a trained model, and a jupyter notebook showing an example for model post-processing. Two typo errors in files titled "run2get_args.py" and "postprocess.py" were corrected on March 27, 2024 to avoid confusions. - Finally, "Trained_models_and_results.zip" includes three folders and three files with suffix ".npy", and each folder corresponds to one file with suffix ".npy" with the same title. Each of the three folders include all trained models associated with one iLSTM model configuration (35 models for each configuration, details are described in the paper). Each file with suffix ".npy" includes the post-processed results of the corresponding 35 models under one iLSTM model configuration.

54 ENVIRONMENTAL SCIENCES↗