SEARCH · Engineering Papers

Results for “mixed finite element method”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Droplet formation simulation using mixed finite elements

Droplet formation happens in finite time due to the surface tension force. The linear stability analysis is useful to estimate the size of a droplet but fails to approximate the shape of the droplet. This is due to a highly nonlinear flow description near the point where the first pinch-off happens. A one-dimensional axisymmetric mathematical model was first developed by Eggers and Dupont [“Drop formation in a one-dimensional approximation of the Navier–Stokes equation,” J. Fluid Mech. 262, 205–221 (1994)] using asymptotic analysis. This asymptotic approach to the Navier–Stokes equations leads to a universal scaling explaining the self-similar nature of the solution. Numerical models for the one-dimensional model were developed using the finite difference [Eggers and Dupont, “Drop formation in a one-dimensional approximation of the Navier–Stokes equation,” J. Fluid Mech. 262, 205–221 (1994)] and finite element method [Ambravaneswaran et al., “Drop formation from a capillary tube: Comparison of one-dimensional and two-dimensional analyses and occurrence of satellite drops,” Phys. Fluids 14, 2606–2621 (2002)]. The focus of this study is to provide a robust computational model for one-dimensional axisymmetric droplet formation using the Portable, Extensible Toolkit for Scientific Computation. The code is verified using the Method of Manufactured Solutions and validated using previous experimental studies done by Zhang and Basaran [“An experimental study of dynamics of drop formation,” Phys. Fluids 7, 1184–1203 (1995)]. The present model is used for simulating pendant drops of water, glycerol, and paraffin wax, with an aspiration of extending the application to simulate more complex pinch-off phenomena.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRS2016

In Search of a Time Efficient Approach to Crack and Delamination Growth Predictions in Composites

Analysis benchmarking was used to assess the accuracy and time efficiency of algorithms suitable for automated delamination growth analysis. First, the Floating Node Method (FNM) was introduced and its combination with a simple exponential growth law (Paris Law) and Virtual Crack Closure technique (VCCT) was discussed. Implementation of the method into a user element (UEL) in Abaqus/Standard(Registered TradeMark) was also presented. For the assessment of growth prediction capabilities, an existing benchmark case based on the Double Cantilever Beam (DCB) specimen was briefly summarized. Additionally, the development of new benchmark cases based on the Mixed-Mode Bending (MMB) specimen to assess the growth prediction capabilities under mixed-mode I/II conditions was discussed in detail. A comparison was presented, in which the benchmark cases were used to assess the existing low-cycle fatigue analysis tool in Abaqus/Standard(Registered TradeMark) in comparison to the FNM-VCCT fatigue growth analysis implementation. The low-cycle fatigue analysis tool in Abaqus/Standard(Registered TradeMark) was able to yield results that were in good agreement with the DCB benchmark example. Results for the MMB benchmark cases, however, only captured the trend correctly. The user element (FNM-VCCT) always yielded results that were in excellent agreement with all benchmark cases, at a fraction of the analysis time. The ability to assess the implementation of two methods in one finite element code illustrated the value of establishing benchmark solutions.

Krueger, Ronald↗

NASA NTRS1986

Elimination sequence optimization for SPAR

SPAR is a large-scale computer program for finite element structural analysis. The program allows user specification of the order in which the joints of a structure are to be eliminated since this order can have significant influence over solution performance, in terms of both storage requirements and computer time. An efficient elimination sequence can improve performance by over 50% for some problems. Obtaining such sequences, however, requires the expertise of an experienced user and can take hours of tedious effort to affect. Thus, an automatic elimination sequence optimizer would enhance productivity by reducing the analysts' problem definition time and by lowering computer costs. Two possible methods for automating the elimination sequence specifications were examined. Several algorithms based on the graph theory representations of sparse matrices were studied with mixed results. Significant improvement in the program performance was achieved, but sequencing by an experienced user still yields substantially better results. The initial results provide encouraging evidence that the potential benefits of such an automatic sequencer would be well worth the effort.

Hogan, Harry A.↗

NASA NTRS1987

Surface flaw reliability analysis of ceramic components with the SCARE finite element postprocessor program

The SCARE (Structural Ceramics Analysis and Reliability Evaluation) computer program on statistical fast fracture reliability analysis with quadratic elements for volume distributed imperfections is enhanced to include the use of linear finite elements and the capability of designing against concurrent surface flaw induced ceramic component failure. The SCARE code is presently coupled as a postprocessor to the MSC/NASTRAN general purpose, finite element analysis program. The improved version now includes the Weibull and Batdorf statistical failure theories for both surface and volume flaw based reliability analysis. The program uses the two-parameter Weibull fracture strength cumulative failure probability distribution model with the principle of independent action for poly-axial stress states, and Batdorf's shear-sensitive as well as shear-insensitive statistical theories. The shear-sensitive surface crack configurations include the Griffith crack and Griffith notch geometries, using the total critical coplanar strain energy release rate criterion to predict mixed-mode fracture. Weibull material parameters based on both surface and volume flaw induced fracture can also be calculated from modulus of rupture bar tests, using the least squares method with known specimen geometry and grouped fracture data. The statistical fast fracture theories for surface flaw induced failure, along with selected input and output formats and options, are summarized. An example problem to demonstrate various features of the program is included.

Gyekenyesi, John P.↗

NASA NTRS1987

Surface flaw reliability analysis of ceramic components with the SCARE finite element postprocessor program

Gyekenyesi, J. P.↗

NASA NTRS1982

Studies of implicit and explicit solution techniques in transient thermal analysis of structures

Studies aimed at an increase in the efficiency of calculating transient temperature fields in complex aerospace vehicle structures are reported. The advantages and disadvantages of explicit and implicit algorithms are discussed and a promising set of implicit algorithms with variable time steps, known as GEARIB, is described. Test problems, used for evaluating and comparing various algorithms, are discussed and finite element models of the configurations are described. These problems include a coarse model of the Space Shuttle wing, an insulated frame tst article, a metallic panel for a thermal protection system, and detailed models of sections of the Space Shuttle wing. Results generally indicate a preference for implicit over explicit algorithms for transient structural heat transfer problems when the governing equations are stiff (typical of many practical problems such as insulated metal structures). The effects on algorithm performance of different models of an insulated cylinder are demonstrated. The stiffness of the problem is highly sensitive to modeling details and careful modeling can reduce the stiffness of the equations to the extent that explicit methods may become the best choice. Preliminary applications of a mixed implicit-explicit algorithm and operator splitting techniques for speeding up the solution of the algebraic equations are also described.

Adelman, H. M.↗

NASA NTRSDate not supplied

Edge Based Viscous Method for Node-Centered Formulations

This paper presents a novel, efficient, conservative, edge-based method for evaluation of mean flow viscous fluxes and turbulence-model diffusion terms of the Reynolds-averaged Navier-Stokes equations on tetrahedral grids. The new method is implemented in a practical, node-centered, finite-volume computational fluid dynamics solver. The baseline finite-volume scheme that is equivalent to a second-order accurate finite-element Galerkin approximation of viscous stresses is reformulated. The order of operations to compute the cell-based Green-Gauss gradients is changed to combine the operations by edge, which leads to an equivalent formulation on tetrahedral grids, improves efficiency, and preserves the compact discretization stencil based on the nearest neighbors. The computational results presented in this paper verify the implementation of this edge-based method by comparing its accuracy and iterative convergence with those of the well verified and validated baseline formulation. Efficiency gains for residual and Jacobian evaluations result in significant reduction of time to solution. This novel edge-based formulation on tetrahedra can be seamlessly combined with the baseline formulation on cells of other types for computing solutions on mixed-element grids.

Edge Based↗

NASA NTRSDate not supplied

NASA Langley FUN3D Analyses in Support of the 1st AIAA Stability and Control Prediction Workshop

This work summarizes the results of FUN3D analyses conducted for the 1st AIAA Stability and Control Workshop on behalf of participants from the NASA Langley Research Center. The workshop was created to establish best practices for the prediction of stability and control derivatives using computational fluid dynamics and assess the limitations of these methods when those best practices are applied. The inaugural workshop considered the ONERA version of the NASA/Boeing Common Research Model, which includes the wing, body, horizontal tail, and a vertical tail designed by ONERA. Wind tunnel data at small sideslip angles remain unpublished and served as ‘blind’ data for computational comparisons. The present research generated workshop test case data using the NASA FUN3D code, which is a parallelized, unstructured, node-based, finite-volume discretization, Reynolds-averaged Navier-Stokes flow solver. Steady- state numerical simulations were conducted for workshop test cases investigating the following: grid convergence, Mach number effect on static stability, wind tunnel sting increments, static stability-derivative calculations, and a sideslip angle sweep. Results were generated for two series of unstructured, mixed-element grids, one set provided by the workshop and another set created using the HeldenMesh grid generation software. The results provided include total- and component-level breakdowns of the force and moment coefficients, in addition to sectional pressure distributions for the wing and tail components for comparisons to wind tunnel data.

CFD↗

NASA NTRS1997

Flowfield-Dependent Mixed Explicit-Implicit (FDMEL) Algorithm for Computational Fluid Dynamics

Despite significant achievements in computational fluid dynamics, there still remain many fluid flow phenomena not well understood. For example, the prediction of temperature distributions is inaccurate when temperature gradients are high, particularly in shock wave turbulent boundary layer interactions close to the wall. Complexities of fluid flow phenomena include transition to turbulence, relaminarization separated flows, transition between viscous and inviscid incompressible and compressible flows, among others, in all speed regimes. The purpose of this paper is to introduce a new approach, called the Flowfield-Dependent Mixed Explicit-Implicit (FDMEI) method, in an attempt to resolve these difficult issues in Computational Fluid Dynamics (CFD). In this process, a total of six implicitness parameters characteristic of the current flowfield are introduced. They are calculated from the current flowfield or changes of Mach numbers, Reynolds numbers, Peclet numbers, and Damkoehler numbers (if reacting) at each nodal point and time step. This implies that every nodal point or element is provided with different or unique numerical scheme according to their current flowfield situations, whether compressible, incompressible, viscous, inviscid, laminar, turbulent, reacting, or nonreacting. In this procedure, discontinuities or fluctuations of an variables between adjacent nodal points are determined accurately. If these implicitness parameters are fixed to certain numbers instead of being calculated from the flowfield information, then practically all currently available schemes of finite differences or finite elements arise as special cases. Some benchmark problems to be presented in this paper will show the validity, accuracy, and efficiency of the proposed methodology.

Garcia, S. M.↗

DOE OSTI2020

Mesoscopic structure of mixed type domain walls in multiaxial ferroelectrics

The structure of a 180° uncharged rotational domain wall in a multiaxial ferroelectric film is studied in the framework of an analytical Landau-Ginzburg-Devonshire (LGD) approach. Finite element modeling (FEM) is used to solve numerically the system of the coupled nonlinear Euler-Lagrange (EL) second-order differential equations for two components of polarization. We show that the structure of the domain wall and corresponding metastable or stable phase of the film are controlled by a single parameter—the dimensionless ferroelectric anisotropy factor μ. We fitted the static profile of a solitary domain wall, calculated by FEM, with kinklike functions for polarization components, and extracted the five μ-dependent parameters from the fitting to FEM curves. The surprisingly high accuracy of the fitting results for two polarization components in the entire μ range allows us to conclude that the analytical functions, which are trial functions in the direct variational method, can be treated as a high-accuracy variational solution of the static EL equations. We further derive exact two-component analytical solutions of the static EL equations for a polydomain 180° domain structure in a multiaxial ferroelectric film. Using these, we derive analytical expressions for the system free energy and analyze its dependence on the film thickness and boundary conditions at the film surfaces. The single-domain state is ground for zero polarization derivative at the surfaces, while the polydomain states minimize the system energy for zero polarization at the surfaces. Counterintuitively, the energy of the polydomain states split into two levels, “0” and “1,” for zero polarization at the surfaces, and each of the levels contains a large number of close-energy sublevels, whose structure is characterized by a different number and type of domain walls. The analytical solutions can become a useful tool for Bayesian analysis of high-resolution scanning transmission electron microscopy images in ferroelectric films.

36 MATERIALS SCIENCE↗

NASA NTRS2008

Towards Verification of Unstructured-Grid Solvers

New methodology for verification of finite-volume computational methods using unstructured grids is presented. The discretization order properties are studied in computational windows, easily constructed within a collection of grids or a single grid. Tests are performed within each window and address a combination of problem-, solution-, and discretization/grid-related features affecting discretization error convergence. The windows can be adjusted to isolate particular elements of the computational scheme, such as the interior discretization, the boundary discretization, or singularities. Studies can use traditional grid-refinement computations within a fixed window or downscaling, a recently-introduced technique in which computations are made within windows contracting toward a focal point of interest. Grids within the windows are constrained to be consistently refined, allowing a meaningful assessment of asymptotic error convergence on unstructured grids. Demonstrations of the method are shown, including a comparative accuracy assessment of commonly-used schemes on general mixed grids and the identification of local accuracy deterioration at boundary intersections. Recommendations to enable attainment of design-order discretization errors for large-scale computational simulations are given.

Thomas, James L.↗

DOE OSTI2021

Thermomechanical Microstructural Predictions of Fracture Nucleation of Zircaloy-4 Alloys With δ and ε Hydride Distributions

Abstract A crystalline dislocation-density formulation that was incorporated with a nonlinear finite-element (FE) method was utilized to understand and to predict the thermomechanical behavior of an hexagonal closest packed (h.c.p.) zircaloy system with hydrides with either face-centered cubic (f.c.c.) or body-centered cubic (b.c.c.) hydrides. This formulation was then used with a recently developed fracture methodology that is adapted for finite inelastic strains and multiphase crystalline systems to understand how different microstructurally based fracture modes nucleate and propagate. The interrelated microstructural characteristics of the different crystalline hydride and matrix phases with the necessary orientation relationships (ORs) have been represented, such that a detailed physical understanding of fracture nucleation and propagation can be predicted for the simultaneous thermomechanical failure modes of hydride populations and the matrix. The effects of volume fraction, morphology, crystalline structure, and orientation and distribution of the hydrides on simultaneous and multiple fracture modes were investigated for radial, circumferential, and mixed distributions. Another key aspect was accounting for temperatures changes due to the effects of thermal conduction and dissipated plastic work and their collective effects on fracture. For hydrided aggregates subjected to high temperatures, thermal softening resulted in higher ductility due to increased dislocation-density activity, which led to higher shear strain accumulation and inhibited crack nucleation and growth. The predictions provide validated insights into why circumferential hydrides are more fracture-resistant than radial hydrides for different volume fractions and thermomechanical loading conditions.

Engineering↗

DOE OSTI2025

Design of a Structure for Assembly and Cooling the Magnet of the Next-Generation 45 GHz ECR Ion Source MARS-D

Here, the current Electron Cyclotron Resonance Ion Sources (ECRISs), constructed with Nb-Ti wires and the conventional racetrack-and-solenoid structure, have achieved operating frequencies up to 28 GHz and utilized about 90% of the critical current of the Nb-Ti wire. A Mixed Axial and Radial field System Demonstrator (MARS-D) is being developed at Lawrence Berkeley National Laboratory (LBNL). This system, which consists of an innovative hexagonal Closed-Loop Coil (CLC) and a set of solenoids, can generate higher magnetic fields (up to 150% ) while requiring only about 50% of the superconducting wire, enabling Nb-Ti wires to be used in the next-generation 45 GHz ECRIS. However, the assembly and cooling of such an efficient and compact magnet are particularly challenging due to the small radial gap between the CLC and solenoids, as well as the tight operating temperature margin. To address these challenges, a structure was developed that combines a three-section radially split solenoid mandrel with a series of shrink-fit reinforcement rings and cooling channels. This paper presents the detailed structure, manufacturing method, assembly procedure, impregnation method, mechanical Finite Element Analysis (FEA) comparison, and thermal FEA comparison.

ECR ion sources↗

NASA NTRS2004

Penalty-Based Interface Technology for Prediction of Delamination Growth in Laminated Structures

An effective interface element technology has been developed for connecting and simulating crack growth between independently modeled finite element subdomains (e.g., composite plies). This method has been developed using penalty constraints and allows coupling of finite element models whose nodes do not necessarily coincide along their common interface. Additionally, the present formulation leads to a computational approach that is very efficient and completely compatible with existing commercial software. The present interface element has been implemented in the commercial finite element code ABAQUS as a User Element Subroutine (UEL), making it easy to test the approach for a wide range of problems. The interface element technology has been formulated to simulate delamination growth in composite laminates. Thanks to its special features, the interface element approach makes it possible to release portions of the interface surface whose length is smaller than that of the finite elements. In addition, the penalty parameter can vary within the interface element, allowing the damage model to be applied to a desired fraction of the interface between the two meshes. Results for double cantilever beam DCB, end-loaded split (ELS) and fixed-ratio mixed mode (FRMM) specimens are presented. These results are compared to measured data to assess the ability of the present damage model to simulate crack growth.

Averill, Ronald C.↗

NASA NTRS2016

Multiscale Static Analysis of Notched and Unnotched Laminates Using the Generalized Method of Cells

The generalized method of cells (GMC) is demonstrated to be a viable micromechanics tool for predicting the deformation and failure response of laminated composites, with and without notches, subjected to tensile and compressive static loading. Given the axial [0], transverse [90], and shear [+45/-45] response of a carbon/epoxy (IM7/977-3) system, the unnotched and notched behavior of three multidirectional layups (Layup 1: [0,45,90,-45](sub 2S), Layup 2: [0,60,0](sub 3S), and Layup 3: [30,60,90,-30, -60](sub 2S)) are predicted under both tensile and compressive static loading. Matrix nonlinearity is modeled in two ways. The first assumes all nonlinearity is due to anisotropic progressive damage of the matrix only, which is modeled, using the multiaxial mixed-mode continuum damage model (MMCDM) within GMC. The second utilizes matrix plasticity coupled with brittle final failure based on the maximum principle strain criteria to account for matrix nonlinearity and failure within the Finite Element Analysis--Micromechanics Analysis Code (FEAMAC) software multiscale framework. Both MMCDM and plasticity models incorporate brittle strain- and stress-based failure criteria for the fiber. Upon satisfaction of these criteria, the fiber properties are immediately reduced to a nominal value. The constitutive response for each constituent (fiber and matrix) is characterized using a combination of vendor data and the axial, transverse, and shear responses of unnotched laminates. Then, the capability of the multiscale methodology is assessed by performing blind predictions of the mentioned notched and unnotched composite laminates response under tensile and compressive loading. Tabulated data along with the detailed results (i.e., stress-strain curves as well as damage evolution states at various ratios of strain to failure) for all laminates are presented.

GMC↗

DOE OSTI2021

High-Fidelity Simulations of Gravity Currents Using a High-Order Finite-Difference Spectral Vanishing Viscosity Approach

This numerical work investigates the potential of a high-order finite-difference spectral vanishing viscosity approach to simulate gravity currents at high Reynolds numbers. The method introduces targeted numerical dissipation at small scales through altering the discretisation of the second derivatives of the viscous terms in the incompressible Navier-Stokes equations to mimic the spectral vanishing viscosity (SVV) operator, originally designed for the regularisation of spectral element method (SEM) solutions of pure advection problems. Using a sixth-order accurate finite-difference scheme, the adoption of the SVV method is straightforward and comes with a negligible additional computational cost. In order to assess the ability of this high-order finite-difference spectral vanishing viscosity approach, we performed large-eddy simulations (LES) of a gravity current in a channelised lock-exchange set-up with our SVV model and with the well-known explicit static and dynamic Smagorinsky sub-grid scale (SGS) models. The obtained data are compared with a direct numerical simulation (DNS) based on more than 800 million mesh nodes, and with experimental measurements. A framework for the energy budget is introduced to investigate the behaviour of the gravity current. First, it is found that the DNS is in good agreement with the experimental data for the evolution of the front location and velocity field as well as for the stirring and mixing inside the gravity current. Secondly, the LES performed with less than 0.4% of the total number of mesh nodes compared to the DNS, can reproduce the main features of the gravity currents, with the SVV model yielding slightly more accurate results. It is also found that the dynamic Smagorinsky model performs better than its static version. For the present study, the static and dynamic Smagorinsky models are 1.8 and 2.5 times more expensive than the SVV model, because the latter does not require the calculation of explicit SGS terms in the Navier-Stokes equations nor spatial filtering operations.

direct numerical simulation↗

DOE OSTI2020

Multi-scale characterization and simulation of impact welding between immiscible Mg/steel alloys

Vaporizing foil actuator spot welding method is used in this paper to join magnesium alloy AZ31 and uncoated high-strength steel DP590, which are typically considered as un-weldable due to their high physical property disparities, low mutual solubility, and the lack of any intermetallic phases. Characterization results from scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM) of the weld interface indicate that the impact creates an Mg nanocrystalline interlayer with abundant Fe particles. The interlayer exhibits intact bonding with both DP590 and AZ31 substrates. To investigate the fundamental bond formation mechanisms at the interface, a finite element (FE)-based process simulation is first performed to calculate the local temperature and deformation at the interface under the given macroscopic experimental condition. Finally, taking the FE results at the interface as inputs, molecular dynamics (MD) simulations are conducted to study the interlayer formation at the Mg/Fe interface during the impact and cooling. The results found a high velocity shearing-induced mechanical mixing mechanism that mixes Mg/Fe atoms at the interface and creates the interlayer, leading to the metallurgical bond between Mg/steel alloys.

36 MATERIALS SCIENCE↗

DOE OSTI2026

JIMWLK on a quantum computer

We propose a method for solving the Jalilian-Marian-Iancu-McLerran-Weigert-Leonidov-Kovner (JIMWLK) evolution equation on quantum computers. Our approach exploits the reformulation of the JIMWLK equation as a Lindblad master equation governing the rapidity evolution of the hadronic density matrix, as established in prior work. To render the problem tractable for quantum simulation, we introduce several approximations: the two-dimensional transverse plane is reduced to a one-dimensional radial lattice by assuming azimuthal symmetry of the jump operators; the gauge group is restricted to SU(2); and the infinite Wilson lines of the JIMWLK equation are replaced by finite Wilson links along the light-cone direction. The resulting bosonic Hilbert space is truncated using the electric field basis familiar from Hamiltonian lattice gauge theory, with states restricted to angular momenta 𝑗 ≤ 𝑗 max . We derive the matrix elements of the JIMWLK Lindblad jump operators in this basis. As a benchmark, we demonstrate rapid convergence of the fundamental dipole expectation value with 𝑗 max for both pure and mixed Gaussian initial density matrices. For the simplest truncation, 𝑗 max =1/2, we implement the Lindblad evolution using a quantum simulation algorithm verified with the Qiskit statevector simulator by decomposing the non-unitary evolution operator into a linear combination of unitaries. This work establishes a concrete pathway toward quantum simulation of high-energy quantum chromodynamics evolution equations, with direct relevance to the physics program of the Electron-Ion Collider.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗