Search NASA⌕ Search

SEARCH · Search NASA

Results for “Approximate Computing”

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 1,531 records · Page 85
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

SU-C-BRC-03: Development of a Novel Strategy for On-Demand Monte Carlo and Deterministic Dose Calculation Treatment Planning and Optimization for External Beam Photon and Particle Therapy

Purpose: Accurate and fast dose calculation is a prerequisite of precision radiation therapy in modern photon and particle therapy. While Monte Carlo (MC) dose calculation provides high dosimetric accuracy, the drastically increased computational time hinders its routine use. Deterministic dose calculation methods are fast, but problematic in the presence of tissue density inhomogeneity. We leverage the useful features of deterministic methods and MC to develop a hybrid dose calculation platform with autonomous utilization of MC and deterministic calculation depending on the local geometry, for optimal accuracy and speed. Methods: Our platform utilizes a Geant4 based “localized Monte Carlo” (LMC) method that isolates MC dose calculations only to volumes that have potential for dosimetric inaccuracy. In our approach, additional structures are created encompassing heterogeneous volumes. Deterministic methods calculate dose and energy fluence up to the volume surfaces, where the energy fluence distribution is sampled into discrete histories and transported using MC. Histories exiting the volume are converted back into energy fluence, and transported deterministically. By matching boundary conditions at both interfaces, deterministic dose calculation account for dose perturbations “downstream” of localized heterogeneities. Hybrid dose calculation was performed for water and anthropomorphic phantoms. Results: We achieved <1% agreement between deterministic and MC calculations in the water benchmark for photon and proton beams, and dose differences of 2%–15% could be observed in heterogeneous phantoms. The saving in computational time (a factor ∼4–7 compared to a full Monte Carlo dose calculation) was found to be approximately proportional to the volume of the heterogeneous region. Conclusion: Our hybrid dose calculation approach takes advantage of the computational efficiency of deterministic method and accuracy of MC, providing a practical tool for high performance dose calculation in modern RT. The approach is generalizable to all modalities where heterogeneities play a large role, notably particle therapy.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Relativistic resolution-of-the-identity with Cholesky integral decomposition

In this study, we present an efficient integral decomposition approach called the restricted-kinetic-balance resolution-of-the-identity (RKB-RI) algorithm, which utilizes a tunable RI method based on the Cholesky integral decomposition for in-core relativistic quantum chemistry calculations. The RKB-RI algorithm incorporates the restricted-kinetic-balance condition and offers a versatile framework for accurate computations. Notably, the Cholesky integral decomposition is employed not only to approximate symmetric large-component electron repulsion integrals but also those involving small-component basis functions. In addition to comprehensive error analysis, we investigate crucial conditions, such as the kinetic balance condition and variational stability, which underlie the applicability of Dirac relativistic electronic structure theory. Here we compare the computational cost of the RKB-RI approach with the full in-core method to assess its efficiency. To evaluate the accuracy and reliability of the RKB-RI method proposed in this work, we employ actinyl oxides as benchmark systems, leveraging their properties for validation purposes. This investigation provides valuable insights into the capabilities and performance of the RKB-RI algorithm and establishes its potential as a powerful tool in the field of relativistic quantum chemistry.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Exact spectral gaps of random one-dimensional quantum circuits

The spectral gap of local random quantum circuits is a fundamental property that determines how close the moments of the circuit's unitaries match those of a Haar random distribution. When studying spectral gaps, it is common to bound these quantities using tools from statistical mechanics or via quantum information-based inequalities. Here, by focusing on the second moment of one-dimensional unitary circuits where nearest-neighboring gates act on sets of qudits (with open and closed boundary conditions), we show that one can exactly compute the associated spectral gaps. Indeed, having access to their functional form allows us to prove several important results, such as the fact that the spectral gap for closed boundary condition is exactly the square of the gap for open boundaries, as well as improve on previously known bounds for approximate design convergence. Finally, we verify our theoretical results by numerically computing the spectral gap for systems of up to 70 qubits, as well as comparing them to gaps of random orthogonal and symplectic circuits.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Wigner phase space distribution via classical adiabatic switching

Evaluation of the Wigner phase space density for systems of many degrees of freedom presents an extremely demanding task because of the oscillatory nature of the Fourier-type integral. We propose a simple and efficient, approximate procedure for generating the Wigner distribution that avoids the computational difficulties associated with the Wigner transform. Starting from a suitable zeroth-order Hamiltonian, for which the Wigner density is available (either analytically or numerically), the phase space distribution is propagated in time via classical trajectories, while the perturbation is gradually switched on. According to the classical adiabatic theorem, each trajectory maintains a constant action if the perturbation is switched on infinitely slowly. We show that the adiabatic switching procedure produces the exact Wigner density for harmonic oscillator eigenstates and also for eigenstates of anharmonic Hamiltonians within the Wentzel-Kramers-Brillouin (WKB) approximation. We generalize the approach to finite temperature by introducing a density rescaling factor that depends on the energy of each trajectory. Time-dependent properties are obtained simply by continuing the integration of each trajectory under the full target Hamiltonian. Further, by construction, the generated approximate Wigner distribution is invariant under classical propagation, and thus, thermodynamic properties are strictly preserved. Numerical tests on one-dimensional and dissipative systems indicate that the method produces results in very good agreement with those obtained by full quantum mechanical methods over a wide temperature range. The method is simple and efficient, as it requires no input besides the force fields required for classical trajectory integration, and is ideal for use in quasiclassical trajectory calculations.

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

An approximate inertial manifold (AIM) based closure for turbulent flows

A closure model for turbulent flows is developed based on a dynamical system theory. An appropriately discretized formulation of the governing equations is considered for this process. The key ingredient is an approximation of the system’s attractor, where all the trajectories in phase space are confined. This approximate inertial manifold based approach provides a path to track trajectories of the system in a lower-dimensional subspace. Unlike conventional coarse-graining approaches, the turbulent field is decomposed into resolved and unresolved dynamics using the properties of the governing equations. The novelty of the approach relies on the reconstruction of the unresolved field constrained by the governing equations. A posteriori tests for homogeneous isotropic turbulence and the Kuramoto–Sivashinsky equation show promising results for considerable dimension reduction with strong convergence properties. The proposed model outperforms the dynamic Smagorinsky model, and the computational overhead is competitive with similar approaches.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Tensor Decompositions for Count Data that Leverage Stochastic and Deterministic Optimization

There is growing interest to extend low-rank matrix decompositions to multi-way arrays, or tensors. One fundamental low-rank tensor decomposition is the canonical polyadic decomposition (CPD). The challenge of fitting a low-rank, nonnegative CPD model to Poisson-distributed count data is of particular interest. Several popular algorithms use local search methods to approximate the global maximum likelihood estimator from local minima. Simultaneously, a recent trend in theoretical computer science and numerical linear algebra leverages randomization to solve very large, hard problems. The typical approach is to use randomization for a fast approximation and determinism for refinement to yield effective algorithms with theoretical guarantees. Two popular algorithms for Poisson CPD reflect that emergent dichotomy: CP Alternating Poisson Regression is a deterministic algorithm and Generalized Canonical Polyadic decomposition makes use of stochastic algorithms in several variants. This work extends recent work to develop two new methods that leverage randomized and deterministic algorithms for improved accuracy and performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Calculating the dissociation chemistry and equation of state properties of materials [Slides]

Many experimental techniques are accurate enough to reveal glaring weaknesses in the material models used by physics codes to design them. This is requiring ever more stringent equation of state (EOS) development. Formerly, EOS models were built using limited gas gun shock data and a few general physics assumptions. With the ever increasing computer power available, we began using density functional theory (DFT) to approximate other EOS regimes such as the principal isentrope, critical point, and sometimes melt. With Sequoia, and only because of Sequoia’s capacity, we are now calculating a limited range, general EOS table and then deriving the above EOS properties based on that table, often on a shorter time frame and more thoroughly than we used to compute just the EOS sections listed above. The first project was to calculate a limited range EOS for CO 2 and specifically track the dissociated species present at each density and temperature. We would not see all of the possible species due to our small sample size and limited time, however, we should see the dominant species and general trends. This work was done to support a new, chemistry aware EOS model that is still under development at LANL. The second project was done for two reasons. The first is LANL and SNL are developing a new amorphous SiO 2 EOS for use in hydrocodes. There is extensive principal Hugoniot and isentrope data, but very little lower density data. There is also no real estimate of the critical point so we found the approximate value of that as well. The second reason for lower density SiO 2 is to test a new method for calculating the Hugoniot of porous materials. Because aerogel (low density SiO 2 ) is used extensively in many shock experiments, there was a large data set to compare with. As can be seen in the figure, the new method appears to work well and is tracking the porous Hugoniot within the experimental error bars (not shown). Finally, SNL is developing a new EOS for LiF because we were seeing discrepancies with the SESAME table. DFT has some issues calculating EOS values for LiF, so we will make some corrections, however, our first attempt appears to have very good results which confirm several experimentally seen properties, as seen on the previous slide.

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

Phoebe: a high-performance framework for solving phonon and electron Boltzmann transport equations

Understanding the electrical and thermal transport properties of materials is critical to the design of electronics, sensors, and energy conversion devices. Computational modeling can accurately predict material properties but, in order to be reliable, requires accurate descriptions of electron and phonon states and their interactions. While first-principles methods are capable of describing the energy spectrum of each carrier, using them to compute transport properties is still a formidable task, both computationally demanding and memory intensive, requiring integration of fine microscopic scattering details for estimation of macroscopic transport properties. To address this challenge, we present Phoebe—a newly developed software package that includes the effects of electron–phonon, phonon–phonon, boundary, and isotope scattering in computations of electrical and thermal transport properties of materials with a variety of available methods and approximations. This open source C++ code combines MPI-OpenMP hybrid parallelization with GPU acceleration and distributed memory structures to manage computational cost, allowing Phoebe to effectively take advantage of contemporary computing infrastructures. We demonstrate that Phoebe accurately and efficiently predicts a wide range of transport properties, opening avenues for accelerated computational analysis of complex crystals.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Parallel Randomized Tucker Decomposition Algorithms

The Tucker tensor decomposition is a natural extension of the singular value decomposition (SVD) to multiway data. Here, we propose to accelerate Tucker tensor decomposition algorithms by using randomization and parallelization. We present two algorithms that scale to large data and many processors, significantly reduce both computation and communication cost compared to previous deterministic and randomized approaches, and obtain nearly the same approximation errors. The key idea in our algorithms is to perform randomized sketches with Kronecker-structured random matrices, which reduces computation compared to unstructured matrices and can be implemented using a fundamental tensor computational kernel. We provide probabilistic error analysis of our algorithms and implement a new parallel algorithm for the structured randomized sketch. Our experimental results demonstrate that our combination of randomization and parallelization achieves accurate Tucker decompositions much faster than alternative approaches. We observe up to a 16X speedup over the fastest deterministic parallel implementation on 3D simulation data.

Tucker decompositions↗
DOE OSTI2011

Variable order spherical harmonic expansion scheme for the radiative transport equation using finite elements

Highlights: {yields} We developed a variable order global basis scheme to solve light transport in 3D. {yields} Based on finite elements, the method can be applied to a wide class of geometries. {yields} It is computationally cheap when compared to the fixed order scheme. {yields} Comparisons with local basis method and other models demonstrate its accuracy. {yields} Addresses problems encountered n modeling of light transport in human brain. - Abstract: We propose the P{sub N} approximation based on a finite element framework for solving the radiative transport equation with optical tomography as the primary application area. The key idea is to employ a variable order spherical harmonic expansion for angular discretization based on the proximity to the source and the local scattering coefficient. The proposed scheme is shown to be computationally efficient compared to employing homogeneously high orders of expansion everywhere in the domain. In addition the numerical method is shown to accurately describe the void regions encountered in the forward modeling of real-life specimens such as infant brains. The accuracy of the method is demonstrated over three model problems where the P{sub N} approximation is compared against Monte Carlo simulations and other state-of-the-art methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2011

Simple algorithm for computing the geometric measure of entanglement

We present an easy implementable algorithm for approximating the geometric measure of entanglement from above. The algorithm can be applied to any multipartite mixed state. It involves only the solution of an eigenproblem and finding a singular value decomposition; no further numerical techniques are needed. To provide examples, the algorithm was applied to the isotropic states of three qubits and the three-qubit XX model with external magnetic field.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

A structure preserving Lanczos algorithm for computing the optical absorption spectrum

We present a new structure preserving Lanczos algorithm for approximating the optical absorption spectrum in the context of solving full Bethe-Salpeter equation without Tamm-Dancoff approximation. The new algorithm is based on a structure preserving Lanczos procedure, which exploits the special block structure of Bethe-Salpeter Hamiltonian matrices. A recently developed technique of generalized averaged Gauss quadrature is incorporated to accelerate the convergence. We also establish the connection between our structure preserving Lanczos procedure with several existing Lanczos procedures developed in different contexts. Numerical examples are presented to demonstrate the effectiveness of our Lanczos algorithm.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Solving inverse problems in stochastic models using deep neural networks and adversarial training

Inverse problems associated with stochastic models constitute a significant portion of scientific and engineering applications. In such cases the unknown quantities are distributions. The applicability of traditional methods is limited because of their demanding assumptions or prohibitive computational consumption; for example, maximum likelihood methods require closed-form density functions, and Markov Chain Monte Carlo needs a large number of simulations. We propose a new method that estimates the unknown distribution by matching the statistical properties between observed and simulated random processes. We leverage the expressive power of neural networks to approximate the unknown distribution and use a discriminative neural network for computing the statistical discrepancies between the observed and simulated random processes. Here we demonstrated numerically that the proposed methods can estimate both the model parameters and learn complicated unknown distributions.

42 ENGINEERING↗
DOE OSTI2017

Surrogate Models for TREAT Transient Calculations

The Transient Reactor Test Facility (TREAT) at Idaho National Laboratory (INL) is currently preparing for resumption of operation after more than 20 years in a long-term standby state. In parallel with preparation for operations and a new transient testing program, the DOENE Advanced Modeling and Simulation program (NEAMS) has been supporting the development of three-dimensional transient multi-physics modeling capabilities within the reactor physics application MAMMOTH. Recently, the MAMMOTH reactor physics team has begun development of uncertainty and sensitivity analysis for the TREAT core. This research will be used to support the analysis of reactor operation and experiment planning. Traditional approaches, i.e., Monte Carlo simulations, for such tasks require a large number of model evaluations under both steady state and transient conditions. However, exhaustive sets of sampling runs are simply not practical for TREAT since evaluations of each model can take a substantial amount of time. A solution to this problem is to build a surrogate model employing a smaller number of model evaluations. The surrogate model is designed to provide an accurate and fast approximation to the detailed TREAT model. In this manner, the surrogate model can be used in place of the TREAT transient model for uncertainty and sensitivity analysis at a fraction of the computational cost. The high dimensional model representation (HDMR) method is employed here for approximating the responses of interest in terms of functions of lower dimensions. This method has been implemented in RAVEN and all TREAT transient calculations are performed using the multiphysics code MAMMOTH. In addition, a code interface has also been developed in order to couple RAVEN with MAMMOTH. The surrogate model is then tested and compared for TREAT startup testing program Transient Test no. 15 core, and its accuracy is measured by its ability to estimate the mean, coefficient of variance, skewness, excess kurtosis and probability density function of the outputs of Transient Test no. 15 model, as well as root mean square error metric.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2014

Class II algorithm for charged-particle transport simulation - Paper 142

Simulation of the transport of high-energy charged particles is difficult because of the large number of interactions experienced by these particles before being brought to rest. Possible strategies to deal with this difficulty were analyzed by Berger. Most modern Monte Carlo codes use a mixture of complete-grouping (class I) and mixed (class II) schemes. In class I simulations, the global effect of all the interactions that occur along a trajectory segment of a given length s is described by means of precalculated distributions derived from multiple-scattering theories. In class II simulations, hard (or catastrophic) interactions (with energy loss W or polar scattering angle {theta} larger than given cutoff values) are sampled individually, while the effect of soft interactions (with energy losses and scattering angles smaller than the respective cutoffs) is described using a multiple-scattering approach. Although Berger effectively employed class II simulation only for energy-loss events, he did recognize the advantages of class II over class I also for multiple elastic scattering. On the one hand, because hard interactions are described exactly according to the adopted differential cross sections (DCSs), class II schemes are more accurate. On the other hand, multiple-scattering approximations, which are still required to describe the cumulative effect of soft interactions, are simpler and more accurate when individual interactions are limited to small energy transfers and angular deflections. For instance, for path lengths that are not too small, but still much less than the CSDA range, the energy-loss distribution due to multiple soft events is approximately Gaussian, by virtue of the central limit theorem. Similarly, the angular distribution caused by multiple small-angle deflections is also approximately Gaussian. However, class II schemes are more difficult to implement because of their dependence on specificities of the DCSs. In this article we briefly present the essentials of the class II algorithm for charged particles implemented in the last version of the penelope code, and provide evidence of its accuracy and stability with the simulation parameters. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2018

Geometries of edge and mixed dislocations in bcc Fe from first principles calculations

We use density functional theory (DFT) to compute the core structures of $a_0[100](010)$ edge, $a_0[100](011)$ edge, $a_0/2[\bar{1}\bar{1}1](1\bar{1}0)$ edge, and $a_0/2[111](1\bar{1}0)$ $71^{\circ}$ mixed dislocations in body-centered cubic (bcc) Fe. The calculations are performed using flexible boundary conditions (FBC), which effectively allow the dislocations to relax as isolated defects by coupling the DFT core to an infinite harmonic lattice through the lattice Green function (LGF). We use the LGFs of the dislocated geometries in contrast to most previous FBC-based dislocation calculations that use the LGF of the bulk crystal. The dislocation LGFs account for changes in the topology of the crystal in the core as well as local strain throughout the crystal lattice. A simple bulk-like approximation for the force constants in a dislocated geometry leads to dislocation LGFs that optimize the core structures of the $a_0[100](010)$ edge, $a_0[100](011)$ edge, and $a_0/2[111](1\bar{1}0)$ $71^{\circ}$ mixed dislocations. This approximation fails for the $a_0/2[\bar{1}\bar{1}1](1\bar{1}0)$ dislocation however, so in this case we derive the LGF from more accurate force constants computed using a Gaussian approximation potential. The standard deviations of the dislocation Nye tensor distributions quantify the widths of the dislocation cores. The relaxed cores are compact, and the local magnetic moments on the Fe atoms closely follow the volumetric strain distributions in the cores. We also compute the core structures of these dislocations using eight different classical interatomic potentials, and quantify symmetry differences between the cores using the Fourier coefficients of their Nye tensor distributions. Most of the core structures computed using the classical potentials agree well with the DFT results. Furthermore the DFT core geometries provide benchmarking for classical potential studies of work-hardening, as well as substitutional and interstitial sites for computing solute-dislocation interactions that serve as inputs for mesoscale models of solute strengthening and solute diffusion near dislocations.

36 MATERIALS SCIENCE↗
DOE OSTI2012

Coupled full core neutron transport/CFD simulations of pressurized water reactors

Recently as part of the CASL project, a capability to perform 3D whole-core coupled neutron transport and computational fluid dynamics (CFD) calculations was demonstrated. This work uses the 2D/1D transport code DeCART and the commercial CFD code STAR-CCM+. It builds on previous CASL work demonstrating coupling for smaller spatial domains. The coupling methodology is described along with the problem simulated and results are presented for fresh hot full power conditions. An additional comparison is made to an equivalent model that uses lower order T/H feedback to assess the importance and cost of high fidelity feedback to the neutronics problem. A simulation of a quarter core Combustion Engineering (CE) PWR core was performed with the coupled codes using a Fixed Point Gauss-Seidel iteration technique. The total approximate calculation requirements are nearly 10,000 CPU hours and 1 TB of memory. The problem took 6 coupled iterations to converge. The CFD coupled model and low order T/H feedback model compared well for global solution parameters, with a difference in the critical boron concentration and average outlet temperature of 14 ppm B and 0.94 deg. C, respectively. Differences in the power distribution were more significant with maximum relative differences in the core-wide pin peaking factor (Fq) of 5.37% and average relative differences in flat flux region power of 11.54%. Future work will focus on analyzing problems more relevant to CASL using models with less approximations. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2017

Recent Advancements in the Numerical Simulation of Surface Irradiance for Solar Energy Applications: Preprint

This paper briefly reviews the National Renewable Energy Laboratory's recent efforts on developing all-sky solar irradiance models for solar energy applications. The Fast All-sky Radiation Model for Solar applications (FARMS) utilizes the simulation of clear-sky transmittance and reflectance and a parameterization of cloud transmittance and reflectance to rapidly compute broadband irradiances on horizontal surfaces. FARMS delivers accuracy that is comparable to the two-stream approximation, but it is approximately 1,000 times faster. A FARMS-Narrowband Irradiance over Tilted surfaces (FARMS-NIT) has been developed to compute spectral irradiances on photovoltaic (PV) panels in 2002 wavelength bands. Further, FARMS-NIT has been extended for bifacial PV panels.

14 SOLAR ENERGY↗