Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Numerical differentiation”

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 1,603 records · Page 89

Differentiable Multiphysics Codes: A Breakthrough Technology for Simulation and Computing

This document summarizes the findings of a strategic planning exercise commissioned by the Weapons Simulation and Computing, Computational Physics (WSC/CP) program at the Lawrence Livermore National Laboratory (LLNL) in FY24. During the year, the committee met with multiple stakeholder communities to gather input, opinions, suggestions and concerns which have been incorporated throughout this document. The key findings from this exercise are summarized: • The development of multiphysics modelling and simulation (mod/sim) codes and software technologies, their deployment on exascale compute platforms, and their broad adoption across the NNSA is a major success of the Advanced Simulation and Computing (ASC) program and the Exascale Computing Project (ECP). Sustained investment in these core technologies is essential. • Today’s state of the art involves running ensembles of O(100K) simulations to perform uncertainty quantification (UQ) and design studies using multiple statistical methods such as Bayesian optimization to understand sensitivities of our models and explore parameterized design spaces. Even with exascale computing, we are practically limited to O(10) parameters in these studies since the number of simulations required to sample the space scales exponentially with the number of design parameters. • The data from these simulation ensembles is increasingly being used to train machine learned (ML) surrogates (or reduced order models, ROMs) which can then be used for optimization or real time design exploration. However, the trained surrogates are still limited in the number of parameters they can represent due to the sampling limitations previously noted. • Augmenting our suite of integrated multiphysics simulation codes, both current and emerging, with the ability to compute gradients (solution derivatives) of arbitrary simulation outputs with respect to (some or all) simulation inputs would be a breakthrough technology, opening the door to a new era of efficient and automated inverse design based on verified and validated mod/sim capabilities. • This capability, which we refer to as differentiable multiphysics codes (DMCs), would revolutionize both UQ and optimization studies by breaking the curse of dimensionality that presently limits our “gradient-free” ensemble based computing approach. A similar breakthrough occurred in the AI/ML community once the ability to compute gradients of arbitrary loss functions using back-propagation became commonplace. Gradient information from the multiphysics codes can also be used to dramatically improve the efficiency and scale of training of ML/ROM surrogates for rapid assessments. • Achieving this in our suite of codes will be a grand challenge, similar to the amount of effort that was required to transition from CPU to GPU computing. It will require buy-in from the entire WSC/CP program and beyond, including all integrated codes, physics and engineering models, third-party library dependencies and performance portability abstractions. It will also require investment in research and development of numerical methods for computing adjoints of coupled physics across multiple adaptively refined moving meshes and of stochastic (Monte Carlo) and mesh free (SPH) methods. • New software and numerical techniques, largely pioneered by the AI/ML community, make this feasible. Chief among these is automatic differentiation (AD), the ability to employ AD at point-wise locations in a physics calculation (instead of traditional black-box approaches) and the ability to perform “back-propagation in time” (or reverse mode AD) for non-linear partial differential equations (PDEs). Fundamentally, the conclusion of this strategic planning exercise is that the time is right to undertake a large scale effort in WSC, centered on the existing integrated codes, to continue the natural evolution of mod/sim in the age of AI/ML. Instead of attempting to replace mod/sim with purely data driven AI/ML models, we believe the key to success is to integrate AI/ML by building on top of the decades of hard-won knowledge and the verified/validated multiphysics modelling capability that is the hallmark of the ASC program.

97 MATHEMATICS AND COMPUTING↗

Testing the Delayed Gamma Capability in MCNP6 - Paper 86

The mission of the Domestic Nuclear Detection Office (DNDO) is to quickly and reliably detect unauthorized attempts to import or transport special nuclear material (SNM) for use against the United States. Developing detection equipment to meet this objective requires accurate simulation of both the detectable signature and detection mechanism. MCNP6 is capable of transporting the various types of radiation encountered in these scenarios; however, the accuracy of MCNP6 must be quantified such that we can be confident in the results of the simulations. Delayed gamma transport is one important component of these detection scenarios. Delayed gamma transport is defined as the combination of gamma emission from radioactive and subsequent transport through matter. MCNP always possessed the ability to emit delayed neutrons from fission; however, these delayed neutron spectra and intensity were functional fits and not directly sampled from the created fission products. Furthermore, there existed no method to emit the decay photons from radioactive fission products (or any radioactive nucleus) A delayed particle capability was initially added to MCNPX 2.6.A in 2005 to sample the radioactive fission product parents and emit decay particles resulting from the decay chain. At the point of capture, a residual nucleus is created and decayed using the CINDER90 depletion code. MCNP only sends CINDER90 a residual radioactive nucleus and not a flux (i.e. capture rate). As a result, the Bateman equations for the solution are coupled first order differential equations with constant coefficients. Theoretically, these equations could be solved analytically; however, CINDER90 employs a numerical Markov linear chain scheme to solve these equations. This paper addresses a more simplified approach to benchmarking in order to best assess: (1) whether or not deficiencies exist in the current methodology for certain types of decay chains; and (2) paths forward for improving the delayed particle emission capability. We utilize closed form analytic solutions to the granddaughter equations for particular sets of isotope systems to compare the accuracy of the delayed gamma capability in MCNP6. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗

Comparison of the Chebyshev Method and the Generalized Crank-Nicholson Method for time Propagation in Quantum Mechanics

We compare efficiency of two methods for numerical solution of the time-dependent Schroedinger equation, namely the Chebyshev method and the recently introduced generalized Crank-Nicholson method. As a testing system the free propagation of a particle in one dimension is used. The space discretization is based on the high-order finite diferences to approximate accurately the kinetic energy operator in the Hamiltonian. We show that the choice of the more effective method depends on how many wave functions must be calculated during the given time interval to obtain relevant and reasonably accurate information about the system, i.e. on the choice of the time step.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Role of Combined NNBI and ICRH Heating in FAST H-mode plasmas

The combination of ICRH+NNBI in FAST allows the generation of fast ion populations with different velocity space anisotropy and radial profiles. These energetic ion populations can excite meso-scale fluctuations with the same characteristics of those expected in reactor conditions and, for this reason, FAST can address a number of important burning plasma physics issues. Numerical simulation and modeling of energetic particle physics are based on the use of transport codes that are iteratively coupled with a bi-dimensional full wave-quasi-linear solver for ICRH, in order to determinate the normalized supra-thermal population pressure ss{sub hot}. The value of ss{sub hot}. as well the energetic particle distribution functions can be used as initial condition for numerical simulation studies, investigating the destabilization and saturation of fast ion driven Alfvenic modes.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Numerical study of two-photon ionization of helium using an ab initio numerical framework

Few-photon-induced breakup of helium is studied using a newly developed ab initio numerical framework for solving the six-dimensional time-dependent Schroedinger equation. We present details of the method and calculate (generalized) cross sections for the process of two-photon nonsequential (direct) double ionization at photon energies ranging from 39.4 to 54.4 eV, a process that has been very much debated in recent years and is not yet fully understood. In particular, we have studied the convergence property of the total cross section in the vicinity of the upper threshold ({approx} 54.4 eV) versus the pulse duration of the applied laser field. We find that the cross section exhibits an increasing trend near the threshold, as has also been observed by others, and show that this rise cannot solely be attributed to an unintended inclusion of the sequential two-photon double ionization process caused by the bandwidth of the applied field.

74 ATOMIC AND MOLECULAR PHYSICS↗

The soaring kite: a tale of two punctured tori

We consider the 5-mass kite family of self-energy Feynman integrals and present a systematic approach for constructing an ε-form basis, along with its differential equation pulled back onto the moduli space of two tori. Each torus is associated with one of the two distinct elliptic curves this family depends on. We demonstrate how the locations of relevant punctures, which are required to parametrize the full image of the kinematic space onto this moduli space, can be extracted from integrals over maximal cuts. A boundary value is provided such that the differential equation is systematically solved in terms of iterated integrals over g-kernels and modular forms. Then, the numerical evaluation of the master integrals is discussed, and important challenges in that regard are emphasized. In an appendix, we introduce new relations between g-kernels.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

An analytical technique for approximating unsteady aerodynamics in the time domain

An analytical technique is presented for approximating unsteady aerodynamic forces in the time domain. The order of elements of a matrix Pade approximation was postulated, and the resulting polynomial coefficients were determined through a combination of least squares estimates for the numerator coefficients and a constrained gradient search for the denominator coefficients which insures stable approximating functions. The number of differential equations required to represent the aerodynamic forces to a given accuracy tends to be smaller than that employed in certain existing techniques where the denominator coefficients are chosen a priori. Results are shown for an aeroelastic, cantilevered, semispan wing which indicate a good fit to the aerodynamic forces for oscillatory motion can be achieved with a matrix Pade approximation having fourth order numerator and second order denominator polynomials.

Dunn, H. J.↗

Asynchronous and corrected-asynchronous numerical solutions of parabolic PDES on MIMD multiprocessors

A major problem in achieving significant speed-up on parallel machines is the overhead involved with synchronizing the concurrent process. Removing the synchronization constraint has the potential of speeding up the computation. The authors present asynchronous (AS) and corrected-asynchronous (CA) finite difference schemes for the multi-dimensional heat equation. Although the discussion concentrates on the Euler scheme for the solution of the heat equation, it has the potential for being extended to other schemes and other parabolic partial differential equations (PDEs). These schemes are analyzed and implemented on the shared memory multi-user Sequent Balance machine. Numerical results for one and two dimensional problems are presented. It is shown experimentally that the synchronization penalty can be about 50 percent of run time: in most cases, the asynchronous scheme runs twice as fast as the parallel synchronous scheme. In general, the efficiency of the parallel schemes increases with processor load, with the time level, and with the problem dimension. The efficiency of the AS may reach 90 percent and over, but it provides accurate results only for steady-state values. The CA, on the other hand, is less efficient, but provides more accurate results for intermediate (non steady-state) values.

Amitai, Dganit↗

Differential detection of Gaussian MSK in a mobile radio environment

Minimum shift keying with Gaussian shaped transmit pulses is a strong candidate for a modulation technique that satisfies the stringent out-of-band radiated power requirements of the mobil radio application. Numerous studies and field experiments have been conducted by the Japanese on urban and suburban mobile radio channels with systems employing Gaussian minimum-shift keying (GMSK) transmission and differentially coherent reception. A comprehensive analytical treatment is presented of the performance of such systems emphasizing the important trade-offs among the various system design parameters such as transmit and receiver filter bandwidths and detection threshold level. It is shown that two-bit differential detection of GMSK is capable of offering far superior performance to the more conventional one-bit detection method both in the presence of an additive Gaussian noise background and Rician fading.

Simon, M. K.↗

End-to-End QoS for Differentiated Services and ATM Internetworking

The Internet was initially design for non real-time data communications and hence does not provide any Quality of Service (QoS). The next generation Internet will be characterized by high speed and QoS guarantee. The aim of this paper is to develop a prioritized early packet discard (PEPD) scheme for ATM switches to provide service differentiation and QoS guarantee to end applications running over next generation Internet. The proposed PEPD scheme differs from previous schemes by taking into account the priority of packets generated from different application. We develop a Markov chain model for the proposed scheme and verify the model with simulation. Numerical results show that the results from the model and computer simulation are in close agreement. Our PEPD scheme provides service differentiation to the end-to-end applications.

Su, Hongjun↗

Source-free integration method for black hole perturbations and self-force computation: Radial fall

Perturbations of Schwarzschild-Droste black holes in the Regge-Wheeler gauge benefit from the availability of a wave equation and from the gauge invariance of the wave function, but lack smoothness. Nevertheless, the even perturbations belong to the C{sup 0} continuity class, if the wave function and its derivatives satisfy specific conditions on the discontinuities, known as jump conditions, at the particle position. These conditions suggest a new way for dealing with finite element integration in the time domain. The forward time value in the upper node of the (t,r*) grid cell is obtained by the linear combination of the three preceding node values and of analytic expressions based on the jump conditions. The numerical integration does not deal directly with the source term, the associated singularities and the potential. This amounts to an indirect integration of the wave equation. The known wave forms at infinity are recovered and the wave function at the particle position is shown. In this series of papers, the radial trajectory is dealt with first, being this method of integration applicable to generic orbits of EMRI (Extreme Mass Ratio Inspiral).

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Modeling jets in cross flow

Various approaches to the modeling of jets in cross flow are reviewed. These are grouped into four classes, namely: empirical models, integral models, perturbation models, and numerical models. Empirical models depend largely on the correlation of experimental data and are mostly useful for first-order estimates of global properties such as jet trajectory and velocity and temperature decay rates. Integral models are based on some ordinary-differential form of the conservation laws, but require substantial empirical calibration. They allow more details of the flow field to be obtained; simpler versions have to assume similarity of velocity and temperature profiles, but more sophisticated ones can actually calculate these profiles. Perturbation models require little empirical input, but the need for small parameters to ensure convergent expansions limits their application to either the near-field or the far-field. Therefore, they are mostly useful for the study of flow physics. Numerical models are based on conservation laws in partial-differential form. They require little empirical input and have the widest range of applicability. They also require the most computational resources. Although many qualitative and quantitative features of jets in cross flow have been predicted with numerical models, many issues affecting accuracy such as grid resolution and turbulence model are not completely resolved.

Demuren, A. O.↗

CuSn(OH){sub 6} submicrospheres: Room-temperature synthesis, growth mechanism, and weak antiferromagnetic behavior

Highlights: Black-Right-Pointing-Pointer CuSn(OH){sub 6} spheres have been synthesized via an aqueous solution method at room temperature. Black-Right-Pointing-Pointer The diameters of the CuSn(OH){sub 6} spheres can be tuned by adjusting the molar ratio of SnO{sub 3}{sup 2-} to Cu{sup 2+}. Black-Right-Pointing-Pointer The as-obtained CuSn(OH){sub 6} spheres are antiferromagnetic and have a weak spin-Peierls transition at about 78 K -- Abstract: CuSn(OH){sub 6} submicrospheres with diameters of 400-900 nm have been successfully fabricated using a simple aqueous solution method at room temperature. Influencing factors such as the dosage of reactants and reaction time on the preparation were systematically investigated. The products were characterized with X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), transmission electron microscopy (TEM), thermogravimetric analysis (TG) and differential thermal analysis (DTA). Results reveal that the CuSn(OH){sub 6} spheres are built from numerous nanoparticles. It is found that the diameter of CuSn(OH){sub 6} spheres can be readily tuned by adjusting the molar ratio of SnO{sub 3}{sup 2-} to Cu{sup 2+}. A possible growth mechanism for the CuSn(OH){sub 6} submicrospheres has been proposed. Amorphous CuSnO{sub 3} submicrospheres were obtained after thermal treatment of the CuSn(OH){sub 6} submicrospheres at 300 Degree-Sign C for 4 h. Standard magnetization measurements demonstrate that the CuSn(OH){sub 6} submicrospheres are antiferromagnetic and have a weak spin-Peierls transition at about 78 K.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA-UCLA Workshop on Computational Techniques in Identification and Control of Flexible Flight Structures, Lake Arrowhead, CA, Nov. 2-4, 1989, Proceedings

The present conference on the identification and control of flexible flight structures with computational techniques encompasses existing and planned testbeds, modeling techniques, adaptive control, and numerical computation and simulation. Specific issues addressed include the NASA/MFSC ground test facility, an active control test on the NASA Minimast, the Astrex testbed for large precision space structures, the dynamic analysis of truss-beam modeling, combined structure-controls-integrated optimization, distributed parameter modeling applicable to the control of flexible flight structures, and approaches to nonclassically damped dynamic systems. Also addressed are adaptive control concepts, techniques for vibration and decentralized slew maneuver control, the discretization of second-order-in-time differential equations which are applicable to nonlinear wave problems, the finite-dimensional approximations of Riccati equations, and numerical simulations of actively controlled space structures.

Balakrishnan, A. V.↗

Evolution equations for the joint probability of several compositions in turbulent combustion

One-point statistical simulations of turbulent combustion require models to represent the molecular mixing of species mass fractions, which then determine the reaction rates. For multi-species mixing the Dirichlet distribution has been used to characterize the assumed joint probability density function (PDF) of several scalars, parametrized by solving modeled evolution equations for their means and the sum of their variances. The PDF is then used to represent the mixing state and to obtain the chemical reactions source terms in moment closures or large eddy simulation. We extend the Dirichlet PDF approach to transported PDF methods by developing its governing stochastic differential equation (SDE). The transport equation, as opposed to parametrizing the assumed PDF, enables (1) the direct numerical computation of the joint PDF (and therefore the mixing model to directly account for the flow dynamics (e.g. reaction) on the shape of the evolving PDF), and (2) the individual specification of the mixing timescales of each species. From the SDE, systems of equations are derived that govern the first two moments, based on which constraints are established that provide consistency conditions for material mixing. A SDE whose solution is the generalized Dirichlet PDF is also developed and some of its properties from the viewpoint of material mixing are investigated. The generalized Dirichlet distribution has the following advantages over the standard Dirichlet distribution due to its more general covariance structure: (1) its ability to represent differential diffusion (i.e. skewness) without affecting the scalar means, and (2) it can represent both negatively and positively correlated scalars. The resulting development is a useful representation of the joint PDF of inert or reactive scalars in turbulent flows: (1) In moment closures, the mixing physics can be consistently represented by one underlying modeling principle, the Dirichlet or the generalized Dirichlet PDF, and (2) based on the SDEs transported PDF mixing models for multi-species diffusion can be constructed by specifying the SDE coefficients.

97 MATHEMATICS AND COMPUTING↗

An improved numerical process for solution of solid mechanics problems

This paper gives an overview of the development and status of an improved numerical process for the solution of solid mechanics problems. The proposed process uses a mixed formulation with the fundamental unknowns consisting of both stress and displacement parameters. The problem is formulated either by means of first-order partial differential equations or in a variational form by using a Hellinger-Reissner-type mixed variational principle. For presentation purposes, the components of a numerical process are characterized and the criteria for an ideal process are outlined. Commonly used finite-difference and finite-element procedures are examined in the light of these criteria and it is shown that they fall short in a number of ways. The proposed numerical process, on the other hand, satisfies most of the optimality criteria and appears to be particularly suited for use with the forthcoming generation computers.

Noor, A. K.↗

Efficiency enhancement of a harmonic lasing free-electron laser

The harmonic lasing free-electron laser amplifier, in which two wigglers is employed in order for the fundamental resonance of the second wiggler to coincide with the third harmonic of the first wiggler to generate ultraviolet radiation, is studied. A set of coupled nonlinear first-order differential equations describing the nonlinear evolution of the system, for a long electron bunch, is solved numerically by CYRUS code. Solutions for the non-averaged and averaged equations are compared. Remarkable agreement is found between the averaged and non-averaged simulations for the evolution of the third harmonic. Thermal effects in the form of longitudinal velocity spread are also investigated. For efficiency enhancement, the second wiggler field is set to decrease linearly and nonlinearly at the point where the radiation of the third harmonic saturates. The optimum starting point and the slope of the tapering of the amplitude of the wiggler are found by a successive run of the code. It is found that tapering can increase the saturated power of the third harmonic considerably. In order to reduce the length of the wiggler, the prebunched electron beam is considered.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

GPU-enabled extreme-scale turbulence simulations: Fourier pseudo-spectral algorithms at the exascale using OpenMP offloading

Fourier pseudo-spectral methods for nonlinear partial differential equations are of wide interest in many areas of advanced computational science, including direct numerical simulation of three-dimensional (3-D) turbulence governed by the Navier-Stokes equations in fluid dynamics. This paper presents a new capability for simulating turbulence at a new record resolution up to 35 trillion grid points, on the world's first exascale computer, Frontier, comprising AMD MI250x GPUs with HPE's Slingshot interconnect and operated by the US Department of Energy's Oak Ridge Leadership Computing Facility (OLCF). Key programming strategies designed to take maximum advantage of the machine architecture involve performing almost all computations on the GPU which has the same memory capacity as the CPU, performing all-to-all communication among sets of parallel processes directly on the GPU, and targeting GPUs efficiently using OpenMP offloading for intensive number-crunching including 1-D Fast Fourier Transforms (FFT) performed using AMD ROCm library calls. With 99% of computing power on Frontier being on the GPU, leaving the CPU idle leads to a net performance gain via avoiding the overhead of data movement between host and device except when needed for some I/O purposes. Memory footprint including the size of communication buffers for MPI_ALLTOALL is managed carefully to maximize the largest problem size possible for a given node count. Detailed performance data including separate contributions from different categories of operations to the elapsed wall time per step are reported for five grid resolutions, from 2048 3 on a single node to 32768 3 on 4096 or 8192 nodes out of 9408 on the system. Both 1D and 2D domain decompositions which divide a 3D periodic domain into slabs and pencils respectively are implemented. The present code suite (labeled by the acronym GESTS, GPUs for Extreme Scale Turbulence Simulations) achieves a figure of merit (in grid points per second) exceeding goals set in the Center for Accelerated Application Readiness (CAAR) program for Frontier. The performance attained is highly favorable in both weak scaling and strong scaling, with notable departures only for 2048 3 where communication is entirely intra-node, and for 32768 3 , where a challenge due to small message sizes does arise. Communication performance is addressed further using a lightweight test code that performs all-to-all communication in a manner matching the full turbulence simulation code. Performance at large problem sizes is affected by both small message size due to high node counts as well as dragonfly network topology features on the machine, but is consistent with official expectations of sustained performance on Frontier. Overall, although not perfect, the scalability achieved at the extreme problem size of 32768 3 (and up to 8192 nodes — which corresponds to hardware rated at just under 1 exaflop/sec of theoretical peak computational performance) is arguably better than the scalability observed using prior state-of-the-art algorithms on Frontier's predecessor machine (Summit) at OLCF. New science results for the study of intermittency in turbulence enabled by this code and its extensions are to be reported separately in the near future.

3D fast Fourier transform↗