Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Relative accuracy”

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 775 records · Page 43
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

A viscoelastic viscoplastic damage model (VPVD)

This report documents the development, implementation, and parameter estimation procedures for aviscoelastic-viscoplastic-viscodamage model. This model is being developed to improve the predictive accuracy of simulations involving plastic bonded explosives (PBX). A representative PBX is PBX9501, which is 95% by weight HMX, 2.5% estane binder and 2.5% nitroplasticizer. Mechanical deformation of this material is primarily accommodated by straining of the polymer binder and rearranging the relatively rigid HMX crystals.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2022

Intelligent Process Visualization through Nuclear Operation Process Modeling, Reasoning, and Object Detection from Field Videos (Final Report)

This report is a deliverable for the “Final Report” task of DOE NEET Project 19-16790, "Context-Aware Safety Information Display for Nuclear Field Workers." This project's overall goal is to test the hypothesis that integrating computer vision and process reasoning methods will enable proactive visualization of the safe operation and maintenance processes of Nuclear Power Plants (NPP) for field workers. Augmented Reality (AR) glasses adopting such proactive safety information visualization techniques can significantly increase personnel safety and reduce the NPP’s operating costs. The current practice of monitoring NPPs requires workers to switch between digital models, data, and physical workspaces in identifying relevant but potentially occluded objects and in assessing the risks of operation and maintenance processes. On the other hand, frequently changed field conditions require field workers to report to supervisors for real-time guidance. Such guidance is essential to ensure that changing conditions will not invalidate or endanger the work order and other ongoing processes that may jeopardize NPP operations. Additionally, incorrect recognition of equipment objects can result in communication errors and safety problems. AR techniques can assist engineers in viewing the physical workspaces with objects labeled with detailed operation procedures and safety reminders during field operations. The project team developed an “Intelligent Context-Aware Safety Information Display” (ICAD) for supporting Nuclear Power Plant (NPP) field workers in achieving safe and efficient execution of a series of operational tasks in uncertain and changing workspaces of an NPP. Before designing the ICAD-AR prototype, the project team synthesized NPP operational knowledge models through literature review studies, surveys, interviews with domain experts, and knowledge modeling. The project team conducted an extensive study of the operational procedures of various NPPs, and digital technologies that can support the safe and efficient execution of those procedures in different NPP operational contexts. This literature review helped the project team conduct surveys and interviews with nuclear engineers and field workers to identify three categories of information. The NPP knowledge modeling efforts reveal that the three categories of information identified have different levels of importance in a typical procedure of carrying out a series of tasks to achieve a specific NPP operation goal (e.g., shutdown, mode changes). These three categories of information include 1) Workspace dynamics – the changing spatial arrangements of workspaces, tools, protection equipment, and supporting materials, 2) Workflow prognostics – the dynamic dependencies between different parts of an NPP that functionally support and influence each other in terms of safety and efficiency, and 3) Hazards – objects and spaces that contain hazardous materials or physical conditions that can pose risks to workers or mechanical systems. The project team has profiled the importance levels of these categories of information into a knowledge model. This knowledge model specifies what types of information are more critical for a given task in a given workspace so that computers can automatically identify critical objects and sensors in a scene for delivering context-ware safety information to field workers through AR devices. Significant research development of this project results in technical research outcomes and a prototyping system that illustrates the technical feasibility of establishing an ICAD-AR system supporting the proactive safety information display for nuclear field workers. This final report summarizes the project team’s technological achievements in the past three years. Overall, the project team completed the development and integration of five techniques into a prototype ICAD Augmented Reality (ICAD-AR) system and demonstrated the developed system’s real-time execution in a mechanical room. The project team completed the analysis of using this prototype in other types of workspaces based on 3D image data and digital design models collected from two additional workspaces (a water treatment plant and a flow loop training facility). The integrated techniques include 1) Natural Language Processing (NLP) algorithms supporting the generation and updates of nuclear fieldwork process models based on text analysis of work packages and operation manuals; 2) sensor log analysis for predicting control actions in given sensor reading contexts; 3) computer vision algorithms for automatic localization and navigation of workers; 4) object detection algorithms for identifying task-related objects and correlated sensors for safety checking; 5) AR technique as a platform for supporting the integration. The testing results of these five techniques have shown that 1) the sensor log analysis model can predict the next control action with an accuracy of 0.883; 2) the trained natural language processing model can extract more than 80% of the critical information from paper-based procedures (PBPs); 3) the navigation algorithm with the integration of Visual Inertial Odometry (VIO) and Non-Recursive Bayesian Filter methods make operator’s trajectory estimation resilient to drift error; 4) the computer vision algorithm can detect task-specific and safety-critical objects with an average accuracy of 95.3%. The project team used work procedures collected from a flow loop training facility and two datasets collected from two mechanical rooms simulating the workspaces of NPPs to demonstrate the technical capabilities of the developed ICAD-AR prototype. The demonstration validated the technical feasibility of establishing the ICAD-AR system for nuclear field workers and identified the challenges in 1) automatic text analysis of work packages; 2) use of limited samples of sensor logs for predicting the proper timings of control actions; 3) reliably tracking workers and their task progress in mechanical rooms with many similar objects.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Genome-scale model development and genomic sequencing of the oleaginous clade Lipomyces

The Lipomyces clade contains oleaginous yeast species with advantageous metabolic features for biochemical and biofuel production. Limited knowledge about the metabolic networks of the species and limited tools for genetic engineering have led to a relatively small amount of research on the microbes. Here, a genome-scale metabolic model (GSM) of Lipomyces starkeyi NRRL Y-11557 was built using orthologous protein mappings to model yeast species. Phenotypic growth assays were used to validate the GSM (66% accuracy) and indicated that NRRL Y-11557 utilized diverse carbohydrates but had more limited catabolism of organic acids. The final GSM contained 2,193 reactions, 1,909 metabolites, and 996 genes and was thus named iLst996. The model contained 96 of the annotated carbohydrate-active enzymes. iLst996 predicted a flux distribution in line with oleaginous yeast measurements and was utilized to predict theoretical lipid yields. Twenty-five other yeasts in the Lipomyces clade were then genome sequenced and annotated. Sixteen of the Lipomyces species had orthologs for more than 97% of the iLst996 genes, demonstrating the usefulness of iLst996 as a broad GSM for Lipomyces metabolism. Pathways that diverged from iLst996 mainly revolved around alternate carbon metabolism, with ortholog groups excluding NRRL Y-11557 annotated to be involved in transport, glycerolipid, and starch metabolism, among others. Overall, this study provides a useful modeling tool and data for analyzing and understanding Lipomyces species metabolism and will assist further engineering efforts in Lipomyces .

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Multifidelity Monte Carlo estimation for efficient uncertainty quantification in climate-related modeling

Abstract. Uncertainties in an output of interest that depends on the solution of a complex system (e.g., of partial differential equations with random inputs) are often, if not nearly ubiquitously, determined in practice using Monte Carlo (MC) estimation. While simple to implement, MC estimation fails to provide reliable information about statistical quantities (such as the expected value of the output of interest) in application settings such as climate modeling, for which obtaining a single realization of the output of interest is a costly endeavor. Specifically, the dilemma encountered is that many samples of the output of interest have to be collected in order to obtain an MC estimator that has sufficient accuracy – so many, in fact, that the available computational budget is not large enough to effect the number of samples needed. To circumvent this dilemma, we consider using multifidelity Monte Carlo (MFMC) estimation which leverages the use of less costly and less accurate surrogate models (such as coarser grids, reduced-order models, simplified physics, and/or interpolants) to achieve, for the same computational budget, higher accuracy compared to that obtained by an MC estimator – or, looking at it another way, an MFMC estimator obtains the same accuracy as the MC estimator at lower computational cost. The key to the efficacy of MFMC estimation is the fact that most of the required computational budget is loaded onto the less costly surrogate models so that very few samples are taken of the more expensive model of interest. We first provide a more detailed discussion about the need to consider an alternative to MC estimation for uncertainty quantification. Subsequently, we present a review, in an abstract setting, of the MFMC approach along with its application to three climate-related benchmark problems as a proof-of-concept exercise.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Physics-Informed Neural Network Method for Forward and Backward Advection-Dispersion Equations

Advection-dispersion equations (ADEs) are commonly used to describe transport phenomena in porous media. Even though mature discretization-based numerical methods for ADEs exist, some challenges still remain, especially when it comes to solving advection-dominated forward ADEs and diffusion-dominated backward ADEs. The latter problem usually arises in the source identification context and leads to numerically unstable grid-based solutions that require a form of regularization or should be treated as an inverse problem that is computationally more expensive because it requires solving the forward problem multiple times. In this study, we propose a discretization-free approach based on the physics-informed neural network (PINN) method for solving coupled ADE and Darcy flow equations with space-dependent hydraulic conductivity. In this approach, the hydraulic conductivity, hydraulic head, and concentration fields are approximated with deep neural networks (DNNs). We assume that the conductivity field is given by its values on a grid, and we use these values to train the conductivity DNN. The head and concentration DNNs are trained by minimizing the residuals of the flow equation and ADE and using the initial and boundary conditions as additional constraints. The PINN method is applied to one- and two-dimensional forward ADE problems, where its performance for various P\'{e}clet numbers ($Pe$) is compared with the analytical and numerical solutions. We find that the PINN method is accurate with errors of less than 1\% and outperforms some conventional discretization-based methods for $Pe$ larger that 100. Next, we demonstrate that the PINN method remains accurate for the backward ADEs, with the relative errors in most cases staying under 5\% compared to the reference concentration field. Finally, we show that when available, the concentration measurements can be easily incorporated in the PINN method and significantly improve (by more than 50\% in the considered cases) the accuracy of the PINN solution of the backward ADE.

He, Qizhi↗
DOE OSTI2024

Modeling neutrino-induced scale-dependent galaxy clustering for photometric galaxy surveys

Abstract The increasing statistical precision of photometric redshift surveys requires improved accuracy of theoretical predictions for large-scale structure observables to obtain unbiased cosmological constraints. In ΛCDM cosmologies, massive neutrinos stream freely at small cosmological scales, suppressing the small-scale power spectrum. In massive neutrino cosmologies, galaxy bias modeling needs to accurately relate the scale-dependent growth of the underlying matter field to observed galaxy clustering statistics. In this work, we implement a computationally efficient approximation of the neutrino-induced scale-dependent bias (NISDB). Through simulated likelihood analyses of Dark Energy Survey Year 3 (DESY3) and Legacy Survey of Space and Time Year 1 (LSSTY1) synthetic data that contain an appreciable NISDB, we examine the impact of linear galaxy bias and neutrino mass modeling choices on cosmological parameter inference. We find model misspecification of the NISDB approximation and neutrino mass models to decrease the constraining power of photometric galaxy surveys and cause parameter biases in the cosmological interpretation of future surveys. We quantify these biases and devise mitigation strategies.

Astronomy & Astrophysics↗
DOE OSTI2022

Evaluating the performance of a high resolution SLIM IM module integrated with an Agilent 6560 IM-QTOF MS - CRADA 413 (Final Report)

Developing new technologies that enhance our abilities to identify molecular signatures is important for the health of the public, the environment, and national security. Deploying these novel technologies requires multiple stages of validation and evaluation. This project focused on evaluating a technology developed at PNNL named Structures for Lossless Ion Manipulations (SLIM) in the context of a commercially available ion mobility mass spectrometer instrument from Agilent Technologies. There is a benefit to integrating the recently developed SLIM module into a real-world instrument platform. Data from the combined instrument will provide new standards for characterizing samples using ion mobility. These standards, related to both the drift time of ions and their resulting collisional cross sections (CCS) will be of value in characterizing samples from both humans and environmental samples, such as soil microbiomes. Because no previous instrument has been designed and implemented to simultaneously analyze the same sample in both constant and oscillating drift fields, the project will provide a new research capability that will support research in a wide variety of applications. The evaluation approach included measuring the resolving power, mass range, sensitivity, and accuracy of two SLIM variations (constant and oscillating fields). Although SLIM performed similarly to the commercial instrument in terms of sensitivity, mass range, and accuracy the resolving power of SLIM far exceeded that of the commercially-available instrument. This validation will provide critically useful data to our commercial partners which they can use to reach informed decisions on the advantages of the SLIM technology such as the outstanding performance and the low cost of manufacturing.

42 ENGINEERING↗
DOE OSTI2022

A physical method for downscaling land surface temperatures using surface energy balance theory

Fine-resolution land surface temperature (LST) derived from thermal infrared remote sensing images is a good indicator of surface water status and plays an essential role in the exchange of energy and water between land and atmosphere. A physical surface energy balance (SEB)-based LST downscaling method (DTsEB) is developed to downscale coarse remotely sensed thermal infrared LST products with fine-resolution visible and near-infrared data. Here, the DTsEB method is advantageous for its ability to mechanically interrelate surface variables contributing to the spatial variation of LST, to quantitatively weigh the contributions of each related variable within a physical framework, and to efficaciously avoid the subjective selection of scaling factors and the establishment of statistical regression relationships. The applicability of the DTsEB method was tested by downscaling 12 scenes of 990 m Moderate Resolution Imaging Spectroradiometer (MODIS) and aggregated Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) LST products to 90 m resolution at six overpass times between 2005 and 2015 over three 9.9 km by 9.9 km cropland (mixed by grass, tree, and built-up land) study areas. Three typical LST downscaling methods, namely the widely applied TsHARP, the later developed least median square regression downscaling (LMS) and the geographically weighted regression (GWR), were introduced for intercomparison. The results showed that the DTsEB method could more effectively reconstruct the subpixel spatial variations in LST within the coarse-resolution pixels and achieve a better downscaling accuracy than the TsHARP, LMS and GWR methods. The DTsEB method yielded, on average, root mean square errors (RMSEs) of 2.01 K and 1.42 K when applied to the MODIS datasets and aggregated ASTER datasets, respectively, which were lower than those obtained with the TsHARP method, with average RMSEs of 2.41 K and 1.71 K, the LMS method, with average RMSEs of 2.35 K and 1.63 K, and the GWR method, with average RMSEs of 2.38 K and 1.64 K, respectively. The contributions of the related surface variables to the subpixel spatial variation in the LST varied both spatially and temporally and were different from each other. In summary, the DTsEB method was demonstrated to outperform the TsHARP, LMS, and GWR methods and could be used as a good alternative for downscaling LST products from coarse to fine resolution with high robustness and accuracy.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Real-time semantic segmentation on FPGAs for autonomous vehicles with hls4ml

In this paper, we investigate how field programmable gate arrays can serve as hardware accelerators for real-time semantic segmentation tasks relevant for autonomous driving. Considering compressed versions of the ENet convolutional neural network architecture, we demonstrate a fully-on-chip deployment with a latency of 4.9 ms per image, using less than 30% of the available resources on a Xilinx ZCU102 evaluation board. The latency is reduced to 3 ms per image when increasing the batch size to ten, corresponding to the use case where the autonomous vehicle receives inputs from multiple cameras simultaneously. We show, through aggressive filter reduction and heterogeneous quantization-aware training, and an optimized implementation of convolutional layers, that the power consumption and resource utilization can be significantly reduced while maintaining accuracy on the Cityscapes dataset.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Generalized moving least squares vs. radial basis function finite difference methods for approximating surface derivatives

Approximating differential operators defined on two-dimensional surfaces is an important problem that arises in many areas of science and engineering. Over the past ten years, localized meshfree methods based on generalized moving least squares (GMLS) and radial basis function finite differences (RBF-FD) have been shown to be effective for this task as they can give high orders of accuracy at low computational cost, and they can be applied to surfaces defined only by point clouds. However, there have yet to be any studies that perform a direct comparison of these methods for approximating surface differential operators (SDOs). The first purpose of this work is to fill that gap. For this comparison, we focus on an RBF-FD method based on polyharmonic spline kernels and polynomials (PHS+Poly) since they are most closely related to the GMLS method. Additionally, we use a relatively new technique for approximating SDOs with RBF-FD called the tangent plane method since it is simpler than previous techniques and natural to use with PHS+Poly RBF-FD. Further, the second purpose of this work is to relate the tangent plane formulation of SDOs to the local coordinate formulation used in GMLS and to show that they are equivalent when the tangent space to the surface is known exactly. The final purpose is to use ideas from the GMLS SDO formulation to derive a new RBF-FD method for approximating the tangent space for a point cloud surface when it is unknown. For the numerical comparisons of the methods, we examine their convergence rates for approximating the surface gradient, divergence, and Laplacian as the point clouds are refined for various parameter choices. We also compare their efficiency in terms of accuracy per computational cost, both when including and excluding setup costs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Closing the Accuracy Gap in Tandem Photovoltaic Testing: An Accessible and Efficient Spectral Tuning Method Using LED-Based Simulators for Research Laboratories and Industry

Accurate performance calibration of multijunction (MJ) solar cells is critical for advancing this technology toward large-scale terrestrial application, yet existing testing methods developed by photovoltaics calibration laboratories remain prohibitively complex and resource intensive for most research laboratories. Current approaches rely on expensive multisource simulators and/or intricate spectral tuning algorithms, limiting accessibility and hindering standardized comparisons of emerging MJ technologies such as perovskite-based tandems. This paper introduces a streamlined spectral irradiance adjustment method for light-emitting diode (LED)-based solar simulators, which are increasingly adopted in the photovoltaics community due to their versatility and cost-effectiveness. The method we present bins LED channels into groups based on the number of junctions of the MJ photovoltaic device under test (DUT) and their corresponding band gaps and incorporates an automated tuning algorithm that eliminates the need to adjust each channel’s irradiance individually. This tuning algorithm requires the relative spectral responsivities of both the DUT and a broadband reference cell, as well as a calibrated spectroradiometer. Measurement validation across various MJ devices, including III-V and perovskite/silicon tandems, demonstrates excellent agreement within 1% with well-established xenon-tungsten multisource simulators and associated tuning algorithms. By enabling precise spectral tuning with readily available equipment and a simple tuning algorithm, our approach bridges the measurement accuracy gap between research laboratories and accredited testing facilities, fostering more reliable device comparisons and accelerating the translation of MJ technologies into real-world energy systems.

14 SOLAR ENERGY↗
DOE OSTI2024

Multifidelity methods for uncertainty quantification of a nonlocal model for phase changes in materials

This study is devoted to the construction of a multifidelity Monte Carlo (MFMC) method for the uncertainty quantification of a nonlocal, non-mass-conserving Cahn-Hilliard model for phase transitions with an obstacle potential. Here, we are interested in estimating the expected value of an output of interest (OoI) that depends on the solution of the nonlocal Cahn-Hilliard model. As opposed to its local counterpart, the nonlocal model captures sharp interfaces without the need for significant mesh refinement. However, the computational cost of the nonlocal Cahn-Hilliard model is higher than that of its local counterpart with similar mesh refinement, inhibiting its use for outer-loop applications such as uncertainty quantification. The MFMC method augments the desired high-fidelity, high-cost OoI with a set of lower-fidelity, lower-cost OoIs to alleviate the computational burden associated with nonlocality. Most of the computational budget is allocated to sampling the cheap surrogate models to achieve speedup, whereas the high-fidelity model is sparsely sampled to maintain accuracy. For the non-mass-conserving nonlocal Cahn-Hilliard model, the use of the MFMC method results in, for a given computational budget, about an order of magnitude reduction in the mean-squared error of the expected value of the OoI relative to that of the Monte Carlo method.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Simulation of electron Bernstein waves using FullWave with a 2D non-local hot plasma model

Hot plasma wave simulation capability is expanded in the FullWave code by updating the hybrid iterative solver in the code with a semi-implicit time stepping method. The new approach is used to simulate Electron Bernstein Wave (EBW) heating in over-dense spherical tokamak plasmas. The code’s hybrid iterative solver circumvents the prohibitive memory cost of direct methods by combining a time evolution of Maxwell’s equations with frequency-domain relaxation, while the conductivity kernel, calculated via 3D particle tracking, captures the essential non-local wave–particle interactions. One-dimensional EBW simulations verify the algorithm’s accuracy by demonstrating mode conversion from X-mode wave to EBW at the upper hybrid resonance and a strong cyclotron damping near the plasma core. Two-dimensional simulation reproduces the predicted short EBW wavelength and quantitatively matches the hot-plasma dispersion relation. This study demonstrates the fidelity of the hybrid solver for the electron cyclotron frequency range.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Blinding scheme for the scale-dependence bias signature of local primordial non-Gaussianity for DESI 2024

The next generation of spectroscopic surveys is expected to achieve an unprecedented level of accuracy in the measurement of cosmological parameters. To avoid confirmation bias and thereby improve the reliability of these results, blinding procedures become a standard practice in the cosmological analyses of such surveys. Blinding is especially crucial when the impact of observational systematics is important relative to the cosmological signal, and a detection of that signal would have significant implications. This is the case for local primordial non-gaussianity, as probed by the scale-dependent bias of the galaxy power spectrum at large scales that are heavily sensitive to the dependence of the target selection on the imaging quality, known as imaging systematics. We propose a blinding method for the scale-dependent bias signature of local primordial non-gaussianity at the density field level which consists in generating a set of weights for the data that replicate the scale-dependent bias. The applied blinding is predictable, and can be straightforwardly combined with other catalog-level blinding procedures that have been designed for the baryon acoustic oscillation and redshift space distortion signals. The procedure is validated through simulations that replicate data from the first year of observation of the Dark Energy Spectroscopic Instrument, but may find applications to other upcoming spectroscopic surveys.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Learning to Unknot

We introduce natural language processing into the study of knot theory, as made natural by the braid word representation of knots. We study the UNKNOT problem of determining whether or not a given knot is the unknot. After describing an algorithm to randomly generate N-crossing braids and their knot closures and discussing the induced prior on the distribution of knots, we apply binary classification to the UNKNOT decision problem. We find that the Reformer and shared-QK Transformer network architectures outperform fully-connected networks, though all perform at 95% accuracy. Perhaps surprisingly, we find that accuracy increases with the length of the braid word, and that the networks learn a direct correlation between the confidence of their predictions and the degree of the Jones polynomial. Finally, we utilize reinforcement learning (RL) to find sequences of Markov moves and braid relations that simplify knots and can identify unknots by explicitly giving the sequence of unknotting actions. Trust region policy optimization (TRPO) performs consistently well, reducing 80% of the unknots with up to 96 crossings we tested to the empty braid word, and thoroughly outperformed other RL algorithms and random walkers. Studying these actions, we find that braid relations are more useful in simplifying to the unknot than one of the Markov moves.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

2-D seismic wave propagation using the distributional finite-difference method: further developments and potential for global seismology

SUMMARY We present a time-domain distributional finite-difference scheme based on the Lebedev staggered grid for the numerical simulation of wave propagation in acoustic and elastic media. The central aspect of the proposed method is the representation of the stresses and displacements with different sets of B-splines functions organized according to the staggered grid. The distributional finite-difference approach allows domain-decomposition, heterogeneity of the medium, curvilinear mesh, anisotropy, non-conformal interfaces, discontinuous grid and fluid–solid interfaces. Numerical examples show that the proposed scheme is suitable to model wave propagation through the Earth, where sharp interfaces separate large, relatively homogeneous layers. A few domains or elements are sufficient to represent the Earth’s internal structure without relying on advanced meshing techniques. We compare seismograms obtained with the proposed scheme and the spectral element method, and we show that our approach offers superior accuracy, reduced memory usage, and comparable efficiency.

Geochemistry & Geophysics↗
DOE OSTI2023

Improving Deep Neural Networks’ Training for Image Classification With Nonlinear Conjugate Gradient-Style Adaptive Momentum

Momentum is crucial in stochastic gradient-based optimization algorithms for accelerating or improving training deep neural networks (DNNs). In deep learning practice, the momentum is usually weighted by a well-calibrated constant. However, tuning the hyperparameter for momentum can be a significant computational burden. In this article, we propose a novel adaptive momentum for improving DNNs training; this adaptive momentum, with no momentum-related hyperparame- ter required, is motivated by the nonlinear conjugate gradient (NCG) method. Stochastic gradient descent (SGD) with this new adaptive momentum eliminates the need for the momentum hyperparameter calibration, allows using a significantly larger learning rate, accelerates DNN training, and improves the final accuracy and robustness of the trained DNNs. For example, SGD with this adaptive momentum reduces classification errors for training ResNet110 for CIFAR10 and CIFAR100 from 5.25% to 4.64% and 23.75% to 20.03%, respectively. Furthermore, SGD, with the new adaptive momentum, also benefits adversarial training and, hence, improves the adversarial robustness of the trained DNNs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Wavelet-based local mesh refinement for rainfall–runoff simulations

A wavelet-based local mesh refinement (wLMR) strategy is designed to generate multiresolution and unstructured triangular meshes from real digital elevation model (DEM) data for efficient hydrological simulations at the catchment scale. The wLMR strategy is studied considering slope- and curvature-based refinement criteria to analyze DEM inputs: the slope-based criterion uses bed elevation data as input to the wLMR strategy, whereas the curvature-based criterion feeds the bed slope data into it. The performance of the wLMR meshes generated by these two criteria is compared for hydrological simulations; first, using three analytical tests with the systematic variation in topography types and then by reproducing laboratory- and real-scale case studies. The bed elevation on the wLMR meshes and their simulation results are compared relative to those achieved on the finest uniform mesh. Repeated analytical tests show that the slope- and curvature-based criteria are equally effective with the wLMR strategy, and that it is easier to decide which criterion to take in relation to the (regular) shape of the topography. For the realistic case studies: (i) slope analysis provides a better metric to assess the correlation of a wLMR mesh to the fine uniform mesh and (ii) both criteria predict outlet hydrographs with a close predictive accuracy to that on the uniform mesh, but the curvature-based criterion is found to slightly better capture the channeling patterns of real DEM data.

54 ENVIRONMENTAL SCIENCES↗