Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “gradient methods”

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 793 records · Page 44
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Co-Active Subspace Methods for the Joint Analysis of Adjacent Computer Models

Active subspace (AS) methods are a valuable tool for understanding the relationship between the inputs and outputs of a Physics simulation. In this article, an elegant generalization of the traditional ASM is developed to assess the co-activity of two computer models. This generalization, which we refer to as a Co-Active Subspace (Co-AS) Method, allows for the joint analysis of two or more computer models allowing for thorough exploration of the alignment (or non-alignment) of the respective gradient spaces. We define co-active directions, co-sensitivity indices, and a scalar “concordance” metric (and complementary “discordance” pseudo-metric) and we demonstrate that these are powerful tools for understanding the behavior of a class of computer models, especially when used to supplement traditional AS analysis. Details for efficient estimation of the Co-AS and an accompanying R package (concordance) are provided. Practical application is demonstrated through analyzing a set of simulated rate stick experiments for PBX 9501, a high explosive, offering insights into complex model dynamics.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Moment-Fourier approach to ion parallel fluid closures and transport for a toroidally confined plasma

A general method of solving the drift kinetic equation is developed for an axisymmetric magnetic field. Expanding a distribution function in general moments, a set of ordinary differential equations is obtained. Successively expanding the moments and magnetic-field involved quantities in Fourier series, a set of linear algebraic equations is obtained. The set of full (Maxwellian and non-Maxwellian) moment equations is solved to express the first-order density, temperature, and flow velocity in terms of radial gradients of the zeroth-order pressure and temperature. Closure relations that connect parallel heat flux density and viscosity to the radial gradients and parallel gradients of temperature and flow velocity are also obtained by solving the non-Maxwellian moment equations. The closure relations combined with the linearized fluid equations reproduce the same solution obtained directly from the full moment equations. Furthermore, the method can be generalized to derive closures and transport for an electron-ion plasma and a multi-ion plasma in a general magnetic field.

neoclassical transport↗
DOE OSTI2024

Generalized Korn’s inequalities for piecewise $H¹$ and $H²$ vector fields

The purpose of this paper is to construct a new class of discrete generalized Korn’s inequalities for piecewise $H^1$ vector fields and piecewise $H^2$ vector fields in three-dimensional space. The resulting Korn’s inequalities are different from the standard Korn’s inequalities, as they involve the trace-free symmetric gradient operator, in place of the usual symmetric gradient operator. Furthermore, it is anticipated that the new generalized Korn’s inequalities will be useful for the analysis of a broad range of finite element methods, including mixed finite element methods and discontinuous Galerkin methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

On Transfer Learning Of Neural Networks Using Bi-Fidelity Data For Uncertainty Propagation

Due to their high degree of expressiveness, neural networks have recently been used as surrogate models for mapping inputs of an engineering system to outputs of interest. Once trained, neural networks are computationally inexpensive to evaluate and remove the need for repeated evaluations of computationally expensive models in uncertainty quantification applications. However, given the highly parameterized construction of neural networks, especially deep neural networks, accurate training often requires large amounts of simulation data that may not be available in the case of computationally expensive systems. In this paper, to alleviate this issue for uncertainty propagation, we explore the application of transfer learning techniques using training data generated from both high- and low-fidelity models. We explore two strategies for coupling these two datasets during the training procedure, namely, the standard transfer learning and the bi-fidelity-weighted learning. In the former approach, a neural network model mapping the inputs to the outputs of interest is trained based on the low-fidelity data. The high-fidelity data are then used to adapt the parameters of the upper layer(s) of the low-fidelity network, or train a simpler neural network to map the output of the low-fidelity network to that of the high-fidelity model. In the latter approach, the entire low-fidelity network parameters are updated using data generated via a Gaussian process model trained with a small high-fidelity dataset. The parameter updates are performed via a variant of stochastic gradient descent with learning rates given by the Gaussian process model. Using three numerical examples, we illustrate the utility of these bi-fidelity transfer learning methods where we focus on accuracy improvement achieved by transfer learning over standard training approaches.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

An improved stochastic weighted particle method for boundary driven flows

Here, the stochastic weighted particle method (SWPM) is a generalization of the Direct Simulation Monte Carlo (DSMC) method where particle weights are variable and dynamic. SWPM is backed by a strong theoretical foundation but has not been critically evaluated for problems of practical interest. A thorough assessment of SWPM for boundary-driven flows reveals significant numerical artifacts near the boundary, notably a diverging heat flux. To correct the boundary heat flux, two modifications to SWPM are proposed: separated grouping and a spatially-dependent weight transfer function. To gauge the relative efficiency of SWPM in comparison to DSMC, a high-Mach-number wheel flow which forms a strong density gradient is also simulated.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Study of self-interaction errors in density functional predictions of dipole polarizabilities and ionization energies of water clusters using Perdew-Zunger and locally scaled self-interaction corrected methods

Here, we studied the effects of self-interaction error (SIE) on the static dipole polarizabilities of water clusters modelled with three increasingly sophisticated, non-empirical density functional approximations (DFAs), viz. the local spin density approximation (LDA), the Perdew-Burke-Ernzherof (PBE) generalized-gradient approximation (GGA), and the strongly constrained and appropriately normed (SCAN) meta-GGA, using the Perdew-Zunger self-interaction-correction (PZ-SIC) energy functional in the Fermi-Lowdin orbital SIC (FLO-SIC) framework. Our results show that while all three DFAs overestimate the cluster polarizabilities, the description systematically improves from LDA to PBE to SCAN. The self-correlation free SCAN predicts polarizabilities quite accurately with a mean absolute error (MAE) of 0.58 Bohr 3 with respect to coupled cluster singles and doubles (CCSD) values. Removing SIE using PZ-SIC correctly reduces the DFA polarizabilities, but over-corrects, resulting in underestimated polarizabilities in SIC-LDA, -PBE, and -SCAN. Finally, we applied a recently proposed local-scaling SIC (LSIC) method using a quasi self-consistent scheme and using the kinetic energy density ratio as an iso-orbital indicator. The results show that the LSIC polarizabilities are in excellent agreement with mean absolute error of 0.08 Bohr 3 for LSIC-LDA and 0.06 Bohr 3 for LSIC-PBE with most recent CCSD polarizabilities. Likewise, the ionization energy estimates as an absolute of highest occupied energy eigenvalue predicted by LSIC are also in excellent agreement with CCSD(T) ionization energies with MAE of 0.4 eV for LSIC-LDA and 0.04 eV for LSIC-PBE. The LSIC-LDA predictions of ionization energies are comparable to the reported GW ionization energies while the LSIC-PBE ionization energies are more accurate than reported GW results.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Accurate parameterization of the kinetic energy functional

The absence of a reliable formulation of kinetic energy density functional has hindered the development of orbital free density functional theory. Using the data-aided learning paradigm, we propose a simple prescription to accurately model the kinetic energy density of any system. Our method relies on a dictionary of functional forms for local and nonlocal contributions which have been proposed in the literature and the appropriate coefficients are calculated via a linear regression framework. To model the nonlocal contributions, we explore two new nonlocal functionals - a functional that captures fluctuations in electronic density and a functional that incorporates gradient information. Since, the analytical functional forms of the kernels present in these nonlocal terms are not known from theory, we propose a basis function expansion to model these seemingly difficult nonlocal quantities. This allows us to easily reconstruct kernels for any system using only a few structures. The proposed method is able to learn kinetic energy densities and total kinetic energies of molecular and periodic systems, such as H 2 , LiH, LiF and a one-dimensional chain of 8 hydrogens using data from Kohn-Sham density functional theory calculations for only a few structures. For the ease of reproduction, codes used to generate the models are provided in the supporting materials.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2022

CODE-TO-CODE BENCHMARK STUDY FOR THERMAL STRESS MODELING AND PRELIMINARY ANALYSIS OF THE HIGH-TEMPERATURE SINGLE HEAT PIPE EXPERIMENT

Microreactors are very small nuclear reactors with typical thermal-energy output of up to 20 MWth. The microreactor concept is gaining more and more attention for safe, robust, and reliable supply of electricity for remote locations and to meet industrial process heat needs. Heat-pipe-cooled microreactor is one of the microreactors being investigated at Idaho National Laboratory. In the heat-pipe-cooled microreactor, heat pipes remove heat from the reactor core as a passive heat-transfer device, so the fluid circulation is not required for cooling, which can substantially simplify the overall reactor design. However, given the extremely high temperatures in the core region and potentially large temperature gradients across the structure materials, thermal stresses need to be well-analyzed to ensure structural integrity during normal operations and accident scenarios for the heat-pipe-cooled microreactor designs. This paper discusses finite element method-based thermal-stress analysis for the high-temperature single heat-pipe test article in the Single Primary Heat Extraction and Removal Emulator (SPHERE) facility at Idaho National Laboratory, using two commercial software packages, Abaqus and STAR-CCM+. A code-to-code benchmark study was performed to crosscheck the model setup and capability of each code and to gain insights into the potential thermal stress concerns from the current experimental setup. It is observed the significant thermal stresses happen at the locations where the largest temperature gradients appeared between heat pipe and electric heater. The temperature fields have good agreements between Abaqus and STAR-CCM+, while the induced thermal stresses show modest deviations probably due to differences of meshing engines used in these two codes.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2023

Chemical Upcycling of Polyethylene, Polypropylene, and Mixtures to High-Value Surfactants

Conversion of plastic wastes to fatty acids is an attractive means to supplement the sourcing of these high-value, high-volume chemicals. We report a method for transforming polyethylene (PE) and polypropylene (PP) at ~80% conversion to fatty acids with number-average molar masses of up to ~700 and 670 daltons, respectively. The process is applicable to municipal PE and PP wastes and their mixtures. Temperature-gradient thermolysis is the key to controllably degrading PE and PP into waxes and inhibiting the production of small molecules. The waxes are upcycled to fatty acids by oxidation over manganese stearate and subsequent processing. PP ..beta..-scission produces more olefin wax and yields higher acid-number fatty acids than does PE ..beta..-scission. We further convert the fatty acids to high-value, large-market-volume surfactants. Industrial-scale technoeconomic analysis suggests economic viability without the need for subsidies.

deconstruction↗
DOE OSTI2025

Bias-Variance Trade-Off in Physics-Informed Neural Networks with Randomized Smoothing for High-Dimensional PDEs

Physics-Informed Neural Networks (PINNs) have triggered a paradigm shift in scientific computing, leveraging mesh-free properties and robust approximation capabilities. While proving effective for low-dimensional partial differential equations (PDEs), the computational cost of PINNs remains a hurdle in high-dimensional scenarios. This is particularly pronounced when computing high-order and high-dimensional derivatives in the physics-informed loss. Randomized Smoothing PINN (RS-PINN) introduces Gaussian noise for stochastic smoothing of the original neural net model, enabling the use of Monte Carlo methods for derivative approximation, which eliminates the need for costly automatic differentiation. Despite its computational efficiency, especially in the approximation of high-dimensional derivatives, RS-PINN introduces biases in both loss and gradients, negatively impacting convergence, especially when coupled with stochastic gradient descent (SGD) algorithms. We present a comprehensive analysis of biases in RS-PINN, attributing them to the nonlinearity of the Mean Squared Error (MSE) loss as well as the intrinsic nonlinearity of the PDE itself. We propose tailored bias correction techniques, delineating their application based on the order of PDE nonlinearity. The derivation of an unbiased RS-PINN allows for a detailed examination of its advantages and disadvantages compared to the biased version. Specifically, the biased version has a lower variance and runs faster than the unbiased version, but it is less accurate due to the bias. To optimize the bias-variance trade-off, we combine the two approaches in a hybrid method that balances the rapid convergence of the biased version with the high accuracy of the unbiased version. In addition to methodological contributions, we present an enhanced implementation of RS-PINN. Extensive experiments on diverse high-dimensional PDEs, including Fokker-Planck, Hamilton-Jacobi-Bellman (HJB), viscous Burgers’, Allen-Cahn, and Sine-Gordon equations, illustrate the bias-variance trade-off and highlight the effectiveness of the hybrid RS-PINN. Empirical guidelines are provided for selecting biased, unbiased, or hybrid versions, depending on the dimensionality and nonlinearity of the specific PDE problem.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

A method for convex black-box integer global optimization

Here we study the problem of minimizing a convex function on a nonempty, finite subset of the integer lattice when the function cannot be evaluated at noninteger points. We propose a new underestimator that does not require access to (sub)gradients of the objective; such information is unavailable when the objective is a blackbox function. Rather, our underestimator uses secant linear functions that interpolate the objective function at previously evaluated points. These linear mappings are shown to underestimate the objective in disconnected portions of the domain. Therefore, the union of these conditional cuts provides a nonconvex underestimator of the objective. We propose an algorithm that alternates between updating the underestimator and evaluating the objective function. We prove that the algorithm converges to a global minimum of the objective function on the feasible set. We present two approaches for representing the underestimator and compare their computational effectiveness. We also compare implementations of our algorithm with existing methods for minimizing functions on a subset of the integer lattice. We discuss the difficulty of this problem class and provide insights into why a computational proof of optimality is challenging even for moderate problem sizes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Numerical Analysis of Regular Material Point Method and its Application to Multiphase Flows

The material point method (MPM) is gaining wide popularity in engineering research to model and simulate complex multiphase flow dynamics. The method relies on solving the governing equations of motion and transport in a Lagrangian framework using particles also known as material points. The fluid and kinematic properties are stored on the material points while the spatial gradient calculation and temporal integration are performed on a background grid. This Lagrangian framework allows for large deformations, easy integration of constitutive models, and direct import of complex geometries as particles. However, despite their increasing popularity, very few studies have addressed the issues of numerical resolution and stability of MPM techniques. The presence of additional factors such as the number of material points-per-cell, the location of the material points, the CFL-like condition used in time update, and the grid shape functions also increase the complexity of the error analysis when compared to other finite element methods. In this presentation, we analyze the various forms of error incurred in the application of MPM to continuum mechanics and multiphase flows. The effect of the previously mentioned factors on the error dynamics is studied. The application of these principles to canonical and industrial problems is also presented.

high pressure reverse osmosis↗
DOE OSTI2020

Reinforcement learning based automated history matching for improved hydrocarbon production forecast

History matching aims to find a numerical reservoir model that can be used to predict the reservoir performance. An engineer and model calibration (data inversion) method are required to adjust various parameters/properties of the numerical model in order to match the reservoir production history. In this study, we develop deep neural networks within the reinforcement learning framework to achieve automated history matching that will reduce engineers’ efforts, human bias, automatically and intelligently explore the parameter space, and remove the need of large set of labeled training data. To that end, a fast-marching-based reservoir simulator is encapsulated as an environment for the proposed reinforcement learning. The deep neural-network-based learning agent interacts with the reservoir simulator within reinforcement learning framework to achieve the automated history matching. Reinforcement learning techniques, such as discrete Deep Q Network and continuous Deep Deterministic Policy Gradients, are used toth, used to train the learning agents. The continuous actions enable the Deep Deterministic Policy Gradients to explore more states at each iteration in a a learning episode; consequently, a better history matching is achieved using this algorithm as compared to Deep Q Network. For simplified dual-target composite reservoir models, the best history-matching performances of the discrete and continuous learning methods in terms of normalized root mean square errors are 0.0447 and 0.0038, respectively. Furthermore, our study shows that continuous action space achieved by the deep deterministic policy gradient drastically outperforms deep Q network.

42 ENGINEERING↗
DOE OSTI2020

Fast Evaluation of Aircraft Icing Severity Using Machine Learning Based on XGBoost

Aircraft icing represents a serious hazard in aviation which has caused a number of fatal accidents over the years. In addition, it can lead to substantial increase in drag and weight, thus reducing the aerodynamics performance of the airplane. The process of ice accretion on a solid surface is a complex interaction of aerodynamic and environmental variables. The complex relationship makes machine learning-based methods an attractive alternative to traditional numerical simulation-based approaches. In this study, we introduce a purely data-driven approach to find the complex pattern between different flight conditions and aircraft icing severity prediction. The supervised learning algorithm Extreme Gradient Boosting (XGBoost) is applied to establish the prediction framework which makes prediction based on any set of observations. The input flight conditions for the proposed prediction framework are liquid water content, droplet diameter and exposure time. The proposed approach is demonstrated in three cases: maximum ice thickness prediction, icing area prediction and icing severity level evaluation. Performance comparison studies and error analysis are also conducted to verify the effectiveness and performance of the proposed method. Results show that the proposed method has reasonable capability in evaluating aircraft icing severity.

Li, Sibo (ORCID:000000021705844X)↗
DOE OSTI2024

Proximal Galerkin: A Structure-Preserving Finite Element Method for Pointwise Bound Constraints

The proximal Galerkin finite element method is a high-order, low iteration complexity, nonlinear numerical method that preserves the geometric and algebraic structure of pointwise bound constraints in infinite-dimensional function spaces. This paper introduces the proximal Galerkin method and applies it to solve free boundary problems, enforce discrete maximum principles, and develop a scalable, mesh-independent algorithm for optimal design with pointwise bound constraints. This paper also introduces the latent variable proximal point (LVPP) algorithm, from which the proximal Galerkin method derives. When analyzing the classical obstacle problem, we discover that the underlying variational inequality can be replaced by a sequence of second-order partial differential equations (PDEs) that are readily discretized and solved with, e.g., the proximal Galerkin method. Throughout this work, we arrive at several contributions that may be of independent interest. These include (1) a semilinear PDE we refer to as the entropic Poisson equation; (2) an algebraic/geometric connection between high-order positivity-preserving discretizations and certain infinite-dimensional Lie groups; and (3) a gradient-based, bound-preserving algorithm for two-field, density-based topology optimization. The complete proximal Galerkin methodology combines ideas from nonlinear programming, functional analysis, tropical algebra, and differential geometry and can potentially lead to new synergies among these areas as well as within variational and numerical analysis. Open-source implementations of our methods accompany this work to facilitate reproduction and broader adoption.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

A More La Niña–Like Response to Radiative Forcing after Flux Adjustment in CESM2

Abstract In response to greenhouse gas forcing, most coupled global climate models project the tropical Pacific SST trend toward an “El Niño–like” state, with a reduced zonal SST gradient and a weakened Walker circulation. However, observations over the last five decades reveal a trend toward a more “La Niña–like” state with a strengthening zonal SST gradient. Recent research indicates that the identified trend differences are unlikely to be entirely due to internal variability and probably result, at least in part, from systematic model biases. In this study, Community Earth System Model, version 2 (CESM2), is used to explore how mean-state biases within the model may influence its forced response to radiative forcing in the tropical Pacific. The results show that using flux adjustment to reduce the mean-state bias in CESM2 over the tropical regions results in a more La Niña–like trend pattern in the tropical Pacific, with a strengthening of the tropical Pacific zonal SST gradient and a relatively enhanced Walker circulation, as hypothesized to occur if the ocean thermostat mechanism is stronger than the atmospheric mechanisms which by themselves would weaken the Walker circulation. We also find that the historical strengthening of the tropical Pacific zonal gradient is transient but persists into the near term in a high-emissions future warming scenario. These results suggest the potential of flux adjustment as a method for developing alternative projections that represent a wider range of possible future tropical Pacific warming scenarios, especially for a better understanding of regional patterns of climate risk in the near term.

Zhuo, Jing-Yi [Lamont-Doherty Earth Observatory, C↗
DOE OSTI2023

Reinforcement Learning-based Output Structured Feedback for Distributed Multi-Area Power System Frequency Control

Load frequency control (LFC) is a key factor to maintain the stable frequency in multi-area power systems. As the modern power systems evolve from centralized to decentralized paradigm, LFC needs to consider the decentralized scheme that considers limited information from the information-exchange graph for the generator control of each interconnected area. This paper aims to solve a data-driven constrained LQR problem with mean-variance risk constraints and output structured feedback, and applies this framework to solve the LFC problem in multi-area power systems. By reformulating the constrained optimization problem into a minimax problem, the stochastic gradient descent max-oracle (SGDmax) algorithm with zero-order policy gradient (ZOPG) is adopted to find the optimal feedback gain from the learning, while guaranteeing the convergence. In addition, to improve the adaptation of the proposed learning method to new or varying models, we construct an emulator grid that approximates the dynamics of a physical grid and performs training based on this model. Once the feedback gain is obtained from the emulator grid, it is applied to the physical grid with a robustness test to check whether the controller from the approximated emulator applies to the actual system. Numerical tests show that the obtained feedback controller can successfully control the frequency of each area, while mitigating the uncertainty from the loads, with reliable robustness that ensures the adaptability of the obtained feedback gain to the actual physical grid.

Kwon, Kyung-bin↗
DOE OSTI2023

Highly cascaded first-order sapphire optical fiber Bragg gratings fabricated by a femtosecond laser

This Letter reports an innovative technique for fabricating large-scale, highly cascaded first-order sapphire optical fiber Bragg gratings (FBGs) using a femtosecond laser-assisted point-by-point inscription method. For the first time, to the best of our knowledge, this study successfully demonstrates a distributed array of 10 FBGs within highly multimode sapphire crystal fiber, made possible by employing a high-power laser technique to generate larger reflectors with a Gaussian intensity profile. These first-order FBGs offer advantages such as enhanced reflectivity, shorter fabrication time, and simplified spectral characteristics, making them easier to interpret compared with high-order FBGs. The FBGs’ resilience and effectiveness are analyzed by subjecting them to temperature tests, proving their capacity for accurate temperature monitoring up to 1500°C—a testament to their suitability for harsh environments. This novel approach broadens the scope for sensing and communication applications in sapphire fibers, particularly under challenging conditions. The novelty of our work lies in successfully overcoming the limitations of previous designs by integrating a cascade of 10 FBGs in sapphire fibers, thereby enhancing multiplexing capabilities, minimizing overlapping of FBG peaks, and ensuring reliable temperature monitoring in industries and applications with thermal gradients.

47 OTHER INSTRUMENTATION↗