Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Numerical benchmark”

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 973 records · Page 54
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Short-Term Solar Forecasting Performance of Popular Machine Learning Algorithms: Preprint

A framework for assessing the performance of short-term solar forecasting is presented in conjunction with a range of numerical results using global horizontal irradiation (GHI) from the open-source Surface Radiation Budget (SURFRAD) data network. A suite of popular machine learning algorithms is compared according to a set of statistically distinct metrics and benchmarked against the persistence-of-cloudiness forecast and a cloud motion forecast. Results show significant improvement compared to the benchmarks with trade-offs among the machine learning algorithms depending on the desired error metric. Training inputs include time series observations of GHI for a history of years, historical weather and atmospheric measurements, and corresponding date and time stamps such that training sensitivities might be inferred. Prediction outputs are GHI forecasts for 1, 2, 3, and 4 hours ahead of the issue time, and they are made for every month of the year for 7 locations. Photovoltaic power and energy outputs can then be made using the solar forecasts to better understand power system impacts.

14 SOLAR ENERGY↗
NASA NTRS2022

The TRAPPIST-1 Habitable Atmosphere Intercomparison (THAI). I. Dry Cases—The Fellowship of the GCMs

With the commissioning of powerful, new-generation telescopes such as the James Webb Space Telescope (JWST) and the ground-based Extremely Large Telescopes, the first characterization of a high molecular weight atmosphere around a temperate rocky exoplanet is imminent. Atmospheric simulations and synthetic observables of target exoplanets are essential to prepare and interpret these observations. Here we report the results of the first part of the TRAPPIST-1 Habitable Atmosphere Intercomparison (THAI) project, which compares 3D numerical simulations performed with four state-of-the-art global climate models (ExoCAM, LMD-Generic, ROCKE-3D, Unified Model) for the potentially habitable target TRAPPIST-1e. In this first part, we present the results of dry atmospheric simulations. These simulations serve as a benchmark to test how radiative transfer, subgrid-scale mixing (dry turbulence and convection), and large-scale dynamics impact the climate of TRAPPIST-1e and consequently the transit spectroscopy signature as seen by JWST. To first order, the four models give results in good agreement. The intermodel spread in the global mean surface temperature amounts to 7 K (6 K) for the N2-dominated (CO2-dominated) atmosphere. The radiative fluxes are also remarkably similar (intermodel variations less than 5%), from the surface (1 bar) up to atmospheric pressures ∼5 mbar. Moderate differences between the models appear in the atmospheric circulation pattern (winds) and the (stratospheric) thermal structure. These differences arise between the models from (1) large-scale dynamics, because TRAPPIST-1e lies at the tipping point between two different circulation regimes (fast and Rhines rotators) in which the models can be alternatively trapped, and (2) parameterizations used in the upper atmosphere such as numerical damping.

Martin Turbet↗
DOE OSTI2024

The battery failure databank: Insights from an open-access database of thermal runaway behaviors of Li-ion cells and a resource for benchmarking risks

The thermal response of Li-ion cells can greatly vary for identical cell designs tested under identical conditions, the distribution of which is costly to fully characterize experimentally. The open-source Battery Failure Databank presented here contains robust, high-quality data from hundreds of abuse tests spanning numerous commercial cell designs and testing conditions. Data was gathered using a fractional thermal runaway calorimeter and contains the fractional breakdown of heat and mass that was ejected, as well as high-speed synchrotron radiography of the internal dynamic response of cells during thermal runaway. The distribution of thermal output, mass ejection, and internal response of commercial cells are compared for different abuse-test conditions, which when normalized on a per amp-hour basis show a strong positive correlation between heat output from cells, the fraction of mass ejected from the cells, their energy- and power-density. Ejected mass was shown to contain 10x more heat per gram than non-ejected mass. The causes of 'outlier' thermal and ejection responses i.e., extreme cases, are elucidated by high-speed radiography which showed how occurrences such as vent clogging can create more hazardous conditions. High-speed radiography also demonstrated how the time-resolved interplay of thermal runaway propagation and mass ejection influences the total heat generated.

25 ENERGY STORAGE↗
DOE OSTI2018

A Monte Carlo importance-splitting analytic benchmark - 25539

This paper describes a procedure to create analytic benchmarks for Monte Carlo variance reduction techniques that can be used to predict and/or verify the technique's mean and variance. The process uses the solutions of the first and second history-score moment equations (HSMEs) to obtain the analytic solution for the Monte Carlo mean and variance. For this work, a simple demonstration of the procedure examines the behavior of uncollided particles subject to importance splitting. In addition, a forward approach is given for calculating the same parameters. Finally, the analytic solutions are compared to both Monte Carlo calculations that directly compute the mean and variance and deterministic calculations that solve the HSMEs. All analytic and numeric results agree within the associated Monte Carlo uncertainties. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2012

Hierarchical fractional-step approximations and parallel kinetic Monte Carlo algorithms

We present a mathematical framework for constructing and analyzing parallel algorithms for lattice kinetic Monte Carlo (KMC) simulations. The resulting algorithms have the capacity to simulate a wide range of spatio-temporal scales in spatially distributed, non-equilibrium physiochemical processes with complex chemistry and transport micro-mechanisms. Rather than focusing on constructing exactly the stochastic trajectories, our approach relies on approximating the evolution of observables, such as density, coverage, correlations and so on. More specifically, we develop a spatial domain decomposition of the Markov operator (generator) that describes the evolution of all observables according to the kinetic Monte Carlo algorithm. This domain decomposition corresponds to a decomposition of the Markov generator into a hierarchy of operators and can be tailored to specific hierarchical parallel architectures such as multi-core processors or clusters of Graphical Processing Units (GPUs). Based on this operator decomposition, we formulate parallel Fractional step kinetic Monte Carlo algorithms by employing the Trotter Theorem and its randomized variants; these schemes, (a) are partially asynchronous on each fractional step time-window, and (b) are characterized by their communication schedule between processors. The proposed mathematical framework allows us to rigorously justify the numerical and statistical consistency of the proposed algorithms, showing the convergence of our approximating schemes to the original serial KMC. The approach also provides a systematic evaluation of different processor communicating schedules. We carry out a detailed benchmarking of the parallel KMC schemes using available exact solutions, for example, in Ising-type systems and we demonstrate the capabilities of the method to simulate complex spatially distributed reactions at very large scales on GPUs. Finally, we discuss work load balancing between processors and propose a re-balancing scheme based on probabilistic mass transport methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Linearised Fokker–Planck collision model for gyrokinetic simulations

We introduce a gyrokinetic, linearised Fokker–Planck collision model that satisfies conservation laws and is accurate at arbitrary collisionalities. The differential test-particle component of the operator is exact; the integral field-particle component is approximated using a spherical harmonic and a modified Laguerre polynomial expansion developed by Hirshman and Sigmar (1976 Phys. Fluids 19 1532). The numerical methods of the implementation in the δf-gyrokinetic code stella (Barnes et al 2019 J. Comput. Phys. 391 365–80) are discussed, and conservation properties of the operator are demonstrated. The collision model is then benchmarked against the collision model of the gyrokinetic solver GS2 in the limiting cases of a reduced test-particle collision operator and energy- and momentum-conserving operator. The accuracy of the full collision model is investigated by solving the parallel Spitzer-Härm problem for the transport coefficients. It is shown that retaining collisional energy flux and higher-order terms in the field-particle operator reduces errors in the transport coefficients from 10%–25% for a simple momentum- and energy-conserving model to under 1%.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Wind Turbine Aeroelastic Stability in OpenFAST

Wind turbines are growing in size and increasingly suffer from aeroelastic instabilities. Unfortunately, numerical models often show inconsistent results during verification studies. We address this gap by first introducing novel linearization capabilities within the open-source aero-hydro-servo-elastic framework OpenFAST. Next, a code-to-code benchmark study is presented that compares modal parameters between OpenFAST and HAWCStab2 for a land-based version of the International Energy Agency 15-MW reference wind turbine modeled with quasi-steady aerodynamics. The two solvers are in strong agreement except for discrepancies in the second rotor flapwise modes. The differences are attributed to the torsional flexibility of the tower, which is assumed torsionally stiff in the OpenFAST model. Work is ongoing to close this modeling gap. The aeroelastic stability of a low-specific-power land-based wind turbine is also investigated. The impact of design choices is discussed, high-lighting how narrow the margins are between a stable design and an unstable design.

17 WIND ENERGY↗
DOE OSTI2010

Hydrodynamic study of plasma amplifiers for soft-x-ray lasers: A transition in hydrodynamic behavior for plasma columns with widths ranging from 20 {mu}m to 2 mm

Plasma-based seeded soft-x-ray lasers have the potential to generate high energy and highly coherent short pulse beams. Due to their high density, plasmas created by the interaction of an intense laser with a solid target should store the highest amount of energy density among all plasma amplifiers. Our previous numerical work with a two-dimensional (2D) adaptive mesh refinement hydrodynamic code demonstrated that careful tailoring of plasma shapes leads to a dramatic enhancement of both soft-x-ray laser output energy and pumping efficiency. Benchmarking of our 2D hydrodynamic code in previous experiments demonstrated a high level of confidence, allowing us to perform a full study with the aim of the way for 10-100 {mu}J seeded soft-x-ray lasers. In this paper, we describe in detail the mechanisms that drive the hydrodynamics of plasma columns. We observed transitions between narrow plasmas, where very strong bidimensional flow prevents them from storing energy, to large plasmas that store a high amount of energy. Millimeter-sized plasmas are outstanding amplifiers, but they have the limitation of transverse lasing. In this paper, we provide a preliminary solution to this problem.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2012

δf Monte Carlo calculation of neoclassical transport in perturbed tokamaks

Non-axisymmetric magnetic perturbations can fundamentally change neoclassical transport in tokamaks by distorting particle orbits on deformed or broken flux surfaces. This so-called non-ambipolar transport is highly complex, and eventually a numerical simulation is required to achieve its precise description and understanding. A new δf particle code (POCA) has been developed for this purpose using a modi ed pitch angle collision operator preserving momentum conservation. POCA was successfully benchmarked for neoclassical transport and momentum conservation in axisymmetric conguration. Non-ambipolar particle flux is calculated in the non-axisymmetric case, and results show a clear resonant nature of non-ambipolar transport and magnetic braking. Neoclassical toroidal viscosity (NTV) torque is calculated using anisotropic pressures and magnetic fi eld spectrum, and compared with the generalized NTV theory. Calculations indicate a clear δB 2 dependence of NTV, and good agreements with theory on NTV torque pro les and amplitudes depending on collisionality.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS2001

Multifunctional Collaborative Modeling and Analysis Methods in Engineering Science

Engineers are challenged to produce better designs in less time and for less cost. Hence, to investigate novel and revolutionary design concepts, accurate, high-fidelity results must be assimilated rapidly into the design, analysis, and simulation process. This assimilation should consider diverse mathematical modeling and multi-discipline interactions necessitated by concepts exploiting advanced materials and structures. Integrated high-fidelity methods with diverse engineering applications provide the enabling technologies to assimilate these high-fidelity, multi-disciplinary results rapidly at an early stage in the design. These integrated methods must be multifunctional, collaborative, and applicable to the general field of engineering science and mechanics. Multifunctional methodologies and analysis procedures are formulated for interfacing diverse subdomain idealizations including multi-fidelity modeling methods and multi-discipline analysis methods. These methods, based on the method of weighted residuals, ensure accurate compatibility of primary and secondary variables across the subdomain interfaces. Methods are developed using diverse mathematical modeling (i.e., finite difference and finite element methods) and multi-fidelity modeling among the subdomains. Several benchmark scalar-field and vector-field problems in engineering science are presented with extensions to multidisciplinary problems. Results for all problems presented are in overall good agreement with the exact analytical solution or the reference numerical solution. Based on the results, the integrated modeling approach using the finite element method for multi-fidelity discretization among the subdomains is identified as most robust. The multiple-method approach is advantageous when interfacing diverse disciplines in which each of the method's strengths are utilized. The multifunctional methodology presented provides an effective mechanism by which domains with diverse idealizations are interfaced. This capability rapidly provides the high-fidelity results needed in the early design phase. Moreover, the capability is applicable to the general field of engineering science and mechanics. Hence, it provides a collaborative capability that accounts for interactions among engineering analysis methods.

Ransom, Jonathan B.↗
NASA NTRS2001

On Multifunctional Collaborative Methods in Engineering Science

Multifunctional methodologies and analysis procedures are formulated for interfacing diverse subdomain idealizations including multi-fidelity modeling methods and multi-discipline analysis methods. These methods, based on the method of weighted residuals, ensure accurate compatibility of primary and secondary variables across the subdomain interfaces. Methods are developed using diverse mathematical modeling (i.e., finite difference and finite element methods) and multi-fidelity modeling among the subdomains. Several benchmark scalar-field and vector-field problems in engineering science are presented with extensions to multidisciplinary problems. Results for all problems presented are in overall good agreement with the exact analytical solution or the reference numerical solution. Based on the results, the integrated modeling approach using the finite element method for multi-fidelity discretization among the subdomains is identified as most robust. The multiple method approach is advantageous when interfacing diverse disciplines in which each of the method's strengths are utilized.

Ransom, Jonathan B.↗
NASA NTRS2000

Applications of the Space-Time Conservation Element and Solution Element (CE/SE) Method to Computational Aeroacoustic Benchmark Problems

The Internal Propagation problems, Fan Noise problem, and Turbomachinery Noise problems are solved using the space-time conservation element and solution element (CE/SE) method. The problems in internal propagation problems address the propagation of sound waves through a nozzle. Both the nonlinear and linear quasi 1D Euler equations are solved. Numerical solutions are presented and compared with the analytical solution. The fan noise problem concerns the effect of the sweep angle on the acoustic field generated by the interaction of a convected gust with a cascade of 3D flat plates. A parallel version of the 3D CE/SE Euler solver is developed and employed to obtain numerical solutions for a family of swept flat plates. Numerical solutions for sweep angles of 0, 5, 10, and 15 deg are presented. The turbomachinery problems describe the interaction of a 2D vortical gust with a cascade of flat-plate airfoils with/without a downstream moving grid. The 2D nonlinear Euler Equations are solved and the converged numerical solutions are presented and compared with the corresponding analytical solution. All the comparisons demonstrate that the CE/SE method is capable of solving aeroacoustic problems with/without shock waves in a simple and efficient manner. Furthermore, the simple non-reflecting boundary condition used in the CE/SE method which is not based on the characteristic theory works very well in 1D, 2D and 3D problems.

Wang, Xiao-Yen↗
NASA NTRS1983

Numerical simulations of fuel droplet flows using a Lagrangian triangular mesh

The incompressible, Lagrangian, triangular grid code, SPLISH, was converted for the study of flows in and around fuel droplets. This involved developing, testing and incorporating algorithms for surface tension and viscosity. The major features of the Lagrangian method and the algorithms are described. Benchmarks of the algorithms are given. Several calculations are presented for kerosene droplets in air. Finally, extensions which make the code compressible and three dimensional are discussed.

Fritts, M. J.↗
DOE OSTI2014

Modeling of transport phenomena in tokamak plasmas with neural networks

A new transport model that uses neural networks (NNs) to yield electron and ion heat ux pro les has been developed. Given a set of local dimensionless plasma parameters similar to the ones that the highest delity models use, the NN model is able to efficiently and accurately predict the ion and electron heat transport pro les. As a benchmark, a NN was built, trained, and tested on data from the 2012 and 2013 DIII-D experimental campaigns. It is found that NN can capture the experimental behavior over the majority of the plasma radius and across a broad range of plasma regimes. Although each radial location is calculated independently from the others, the heat ux pro les are smooth, suggesting that the solution found by the NN is a smooth function of the local input parameters. This result supports the evidence of a well-de ned, non-stochastic relationship between the input parameters and the experimentally measured transport uxes. Finally, the numerical efficiency of this method, requiring only a few CPU-μs per data point, makes it ideal for scenario development simulations and real-time plasma control.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2015

A study of overflow simulations using MPAS-Ocean: Vertical grids, resolution, and viscosity

MPAS-Ocean is used to simulate an idealized, density-driven overflow using the dynamics of overflow mixing and entrainment (DOME) setup. Additionally, numerical simulations are carried out using three of the vertical coordinate types available in MPAS-Ocean, including z-star with partial bottom cells, z-star with full cells, and sigma coordinates. The results are first benchmarked against other models, including the MITgcm’s z-coordinate model and HIM’s isopycnal coordinate model, which are used to set the base case used for this work. A full parameter study is presented that looks at how sensitive overflow simulations are to vertical grid type, resolution, and viscosity. Horizontal resolutions with 50 km grid cells are under-resolved and produce poor results, regardless of other parameter settings. Vertical grids ranging in thickness from 15 m to 120 m were tested. A horizontal resolution of 10 km and a vertical resolution of 60 m are sufficient to resolve the mesoscale dynamics of the DOME configuration, which mimics real-world overflow parameters. Mixing and final buoyancy are least sensitive to horizontal viscosity, but strongly sensitive to vertical viscosity. This suggests that vertical viscosity could be adjusted in overflow water formation regions to influence mixing and product water characteristics. In conclusion, the study shows that sigma coordinates produce much less mixing than z-type coordinates, resulting in heavier plumes that go further down slope. Sigma coordinates are less sensitive to changes in resolution but as sensitive to vertical viscosity compared to z-coordinates.

58 GEOSCIENCES↗
DOE OSTI2017

Universality and chaoticity in ultracold K+KRb chemical reactions

A fundamental question in the study of chemical reactions is how reactions proceed at a collision energy close to absolute zero. This question is no longer hypothetical: quantum degenerate gases of atoms and molecules can now be created at temperatures lower than a few tens of nanokelvin. Here we consider the benchmark ultracold reaction between, the most-celebrated ultracold molecule, KRb and K. We map out an accurate ab initio ground-state potential energy surface of the K 2 Rb complex in full dimensionality and report numerically-exact quantum-mechanical reaction dynamics. The distribution of rotationally resolved rates is shown to be Poissonian. An analysis of the hyperspherical adiabatic potential curves explains this statistical character revealing a chaotic distribution for the short-range collision complex that plays a key role in governing the reaction outcome.

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

Magnetic field mapping of inaccessible regions using physics-informed neural networks

A difficult problem concerns the determination of magnetic field components within an experimentally inaccessible region when direct field measurements are not feasible. In this paper, we propose a new method of accessing magnetic field components using non-disruptive magnetic field measurements on a surface enclosing the experimental region. Magnetic field components in the experimental region are predicted by solving a set of partial differential equations (Ampere’s law and Gauss’ law for magnetism) numerically with the aid of physics-informed neural networks (PINNs). Prediction errors due to noisy magnetic field measurements and small number of magnetic field measurements are regularized by the physics information term in the loss function. We benchmark our model by comparing it with an older method. The new method we present will be of broad interest to experiments requiring precise determination of magnetic field components, such as searches for the neutron electric dipole moment.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗