Search NASA⌕ Search

SEARCH · Search NASA

Results for “Numerical Optimization”

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 1,657 records · Page 92

Robust block preconditioners for poroelasticity

In this paper we develop and analyze several preconditioners for the linear systems arising from discretized poroelasticity problems. The preconditioners include one block preconditioner for the two-field Biot model and several preconditioners for the classical three-field Biot model. We manage to analyze these different preconditioners under a same theoretical framework and show that all of them are uniformly optimal with respect to material and discretization parameters. Numerical tests demonstrate the robustness of these preconditioners.

42 ENGINEERING↗

Simultaneous shape and topology optimization of inflatable soft robots

Simultaneous shape and topology optimization is used to design pressure-activated inflatable soft robots. The pressure loaded boundary is meshed conformingly and shape optimized, while the morphology of the robot is topology optimized. The design objective is to exert maximum force on an object, i.e. to produce soft “grippers”. The robot’s motion is modeled using nearly incompressible finite deformation hyperelasticity. To ensure stability of the robot, the buckling load factors obtained via linearized buckling analyses are constrained. The finite element method is used to evaluate the optimization cost and constraint functions and the adjoint method is employed to compute their sensitivities. The numerical examples produce pressure-driven soft robots with varying complexity. We also compare our simultaneous optimization results to those obtained via sequential topology and then shape optimization.

42 ENGINEERING↗

Thermohydraulic experiments on a supercritical carbon dioxide–air microtube heat exchanger

Heat exchangers are critical components in supercritical CO 2 Brayton cycles and provide necessary heat input, recovery, and dissipation. Tubular heat exchangers with unconventionally small tube sizes (tube diameters less than 5 mm) are promising components for supercritical CO 2 cycles and potentially provide excellent structural stability with wide scope of application. This paper provides essential design and fabrication guidelines for a compact shell-and-tube heat exchanger with microtubes (with an inner diameter equal to 1.75 mm). A heat exchanger test rig is used to evaluate the thermohydraulic performance of this heat exchanger with supercritical CO 2 and air as working fluids. Thermohydraulic data are reported for more than forty sets of experiments with varying Reynolds numbers for shell and tube flows. Critical performance metrics are calculated from the data and compared with predictions from a previously described numerical model. The average deviations between the experimental and model results fall within 10% for all critical metrics. This excellent agreement validates the numerical model for supercritical CO 2 heat exchanger optimization and scale-up.

42 ENGINEERING↗

Cohesive finite element modeling of the delamination of HTPB binder and HMX crystals under tensile loading

Accurately modeling the mechanical behavior of the polymer binders and the degradation of interfaces between binder and crystal is important to science-based understanding of the macro-scale response of polymer bonded explosives. The paper presents a description of relatively a simple bi-crystal HMX-HTPB specimen and associated tensile loading experiment including computed tomography imaging, the pertinent constitutive theory, and details of numerical simulations used to infer the behavior of the material during the delamination process. Within this work, mechanical testing and direct numerical simulation of this relatively simple bi-crystal system enabled reasonable isolation of binder-crystal interface delamination, in which the effects of the complicated thermomechanical response of explosive crystals were minimized. Cohesive finite element modeling of the degradation and delamination of the interface between a modified HTPB binder and HMX crystals was used to reproduce observed results from tensile loading experiments on bi-crystal specimens. Several comparisons are made with experimental measurements in order to identify appropriate constitutive behavior of the binder and appropriate parameters for the cohesive traction-separation behavior of the crystal-binder interface. This research demonstrates the utility of directly modeling the delamination between binder and crystal within crystal-binder-crystal tensile specimen towards characterizing the behavior of these interfaces in a manner amenable to larger scale simulation of polycrystalline PBX materials. One critical aspect of this approach is micro computed tomography imaging conducted during the experiments, which enabled comparison of delamination patterns between the direct numerical simulation and actual specimen. In addition to optimizing the cohesive interface parameters, one important finding from this investigation is that understanding and representing the strain-hardening plasticity of HTPB binder is important within the context of using a cohesive traction-separation model for the delamination of a crystal-binder system.

42 ENGINEERING↗

An interrupted tensile testing at high strain rates for pure copper bars

A high-speed tensile facility (HSTF) invented by us was applied to interrupting the tests for pure copper specimen bars controlled locally at different levels of elongation. It was realized to isolate and identify the different stages of the dynamic fracture process of the pure copper specimen bar under impact tension. The results of scanning electron microscopical (SEM) investigation of the recovered pure copper specimens show that the void evolution near the surface of the minimum cross-section of the necking area is more severe than that at the middle of the necking area, which may be connected with the findings discussed by Alves and Jones [J. Mech. Phys. Solids 47, 643 (1999)]. The constitutive models in a certain range of strain determined from the tensile split Hopkinson bar optimized by us were employed and adjusted in numerically simulating the large deformation of the pure copper specimen in the interrupted tensile tests on HSTF. The dependence of the instability strain of thermoviscoplastic materials in simple tension on material parameters delineated by Batra and Wei [Int. J. Impact Eng. 34, 448 (2007)] was inspected in predicting the diffuse necking of the specimen bar. The axisymmetric necking rod model with a central void under static tension presented by Ragab [Eng. Fract. Mech. 71, 1515 (2004)] was extended to predicting the local necking and fracture of the specimen bar under impact tension.

42 ENGINEERING↗

Numerical matching of the sheath and presheath solutions for a spherical probe in radial-motion theory

The theory of positive-ion collection by a probe immersed in a low-pressure plasma was reviewed and extended by Allen, Boyd, and Reynolds [Proc. Phys. Soc. 70, 297 (1957)]. For a given value of the ion current, the boundary values of the matched “nonneutral” or “sheath” solution V{sup ~}{sub nn}{sup (m)}(r; r{sub m}) were obtained from the “quasineutral” or “presheath” solution V{sup ~}{sub qn}(r) by choosing the small potential and electric-field values corresponding to some large “matching radius” r{sub m}. Here, a straightforward but efficient numerical method is presented for systematically determining an optimal value of the matching radius at which the presheath and sheath solutions are joined to yield the “matched” potential profile. Some suitable initial matching radius r{sub m1} is chosen and the related potential and electric-field values of the quasineutral solution are calculated. Using these as boundary conditions, Poisson's equation is integrated to yield the matched nonneutral solution including the corresponding potential at the probe surface. This procedure is repeated for increasing values r{sub m2}, r{sub m3},…. until the resulting potential at the probe surface becomes practically constant. The corresponding value of r{sub m} is taken as the “optimal” matching radius r{sub mo} at which the presheath and sheath solutions are ultimately joined to yield the “optimal” matched potential profile in the entire plasma-probe transition region. It is also shown that the Bohm criterion is inapplicable in the present problem.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Innovative diffraction gratings for high-resolution resonant inelastic soft x-ray scattering spectroscopy

High-resolution Resonant Inelastic X-ray Scattering (RIXS) requires diffraction gratings with very exacting characteristics. The gratings should provide both very high dispersion and high efficiency which are conflicting requirements and extremely challenging to satisfy in the soft x-ray region for a traditional grazing incidence geometry. To achieve high dispersion one should increase the groove density of a grating; this however results in a diffraction angle beyond the critical angle range and results in drastic efficiency loss. The problem can be solved by use of multilayer coated blazed gratings (MBG). In this work we have investigated the diffraction characteristics of MBGs via numerical simulations and have developed a procedure for optimization of grating design for a multiplexed high resolution imaging spectrometer for RIXS spectroscopy to be built in sector 6 at the Advanced Light Source (ALS). We found that highest diffraction efficiency can be achieved for gratings optimized for 4{sup th} or 5{sup th} order operation. Fabrication of such gratings is an extremely challenging technological problem. We present a first experimental prototype of these gratings and report its performance. High order and high line density gratings have the potential to be a revolutionary new optical element that should have great impact in the area of soft x-ray RIXS.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

7.86 kV GaN-on-GaN PN power diode with BaTiO3 for electrical field management

Devices based on gallium nitride (GaN) have great potential for high power switching applications due to the high breakdown field and high electron mobility. In this work, we present a vertical GaN-on-GaN PN power diode using high dielectric constant material, BaTiO3, for electrical field management and high breakdown voltages, in together with an optimized guard-ring and field plate design. Numerical simulation shows that with high-k dielectrics implemented, the peak electrical field at the PN interface is mitigated from 3.5 to 3.1 MV/cm under a reverse bias of −9.05 kV. The device design with BaTiO3 shows a breakdown voltage of 9.65 kV or about 600 V improvement. The fabricated diodes with a 57 μm thick drift layer demonstrate a breakdown voltage of 7.86 kV on a bulk GaN substrate. The device has an on-resistance of 2.8 mΩ cm2 and a Baliga figure of merit of 22 GW/cm2.

Physics↗

Mathematical simulation of an experimental prototype of a high-speed nonreturn-to-zero differential phase-shift-keying fibre-optic communication system

The influence of chromatic dispersion, optical power, and nonlinear distortions in a fibre-optic communication system on the quality of data transmission based on nonreturn-to-zero differential phase-shift keying at a rate of 40 Gbit s{sup -1} in one spectral channel have been numerically simulated and experimental studied. The results of direct numerical calculations and estimates based on the quality factor (Q factor) are in qualitative agreement with the experimental data. It is found experimentally that the dependence of the error rate on the accumulated dispersion has a plateau in the range from -50 to 50 ps nm{sup -1}; a similar dependence is obtained in the numerical calculation based on the Q factor. The optimal calculated value of the power launched into each of 10 sections of a line with a total length of 1000 km is 2 - 4 dBm; it corresponds to the experimental value of 3 dBm.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Possibilities of laser amplification and measurement of the field structure of ultrashort pulses in the range of 2.7 – 3μm in tellurite glass fibres doped with erbium ions

The amplification of ultrashort signals in the range 2.7 – 3 μm in active tellurite glass fibres with high Kerr nonlinearity and high concentration of erbium ions in the core is theoretically investigated under diode pumping at a wavelength of 0.975 μm. The possibility of producing pulses with the energy ∼10 nJ and the duration smaller than 100 fs with ∼20 dB gain is shown. Moreover, using a specially developed iterative computer algorithm, the possibility of determining the intensity and phase of the generated pulses by measuring their fundamental spectra and two additional spectra obtained by means of nonlinear-optical conversion in short pieces of these optical fibres is demonstrated. Optical fibre samples are made of high-purity TeO{sub 2} – ZnO – La{sub 2}O{sub 3} – Na{sub 2}O glasses with 10{sup 21} cm{sup −3} concentration of erbium ions in the core and a low content of hydroxyl groups that causes small losses in the spectral range under consideration. The group velocity dispersions, effective mode fields, nonlinear coefficients and field overlap integrals of the LP{sub 01} fundamental modes with a doped core for optical fibres with different core diameters are calculated, and the emission and absorption cross sections are calculated basing on the experimental luminescence spectrum. The optimal parameters of the amplifier are determined by numerically modelling the dynamics of ultrashort pulses using the measured and calculated characteristics. (paper)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Including the vacuum energy in stellarator coil design

Being three-dimensional, stellarators have the advantage that plasma currents are not essential for creating rotational-transform; however, the external current-carrying coils in stellarators can have strong geometrical shaping, which can complicate the construction. Reducing the inter-coil electromagnetic forces acting on strongly shaped 3D coils and the stress on the support structure while preserving the favorable properties of the magnetic field is a design challenge. In this work, we recognize that the inter-coil forces are the gradient of the vacuum magnetic energy. We introduce an objective functional built on the usual quadratic flux on a prescribed target surface together with a weighed penalty on the vacuum energy. The Euler–Lagrange equation for stationary states is derived, and numerical illustrations are computed using a modern stellarator optimization framework. A study of the effect of the energy functional on the inter-coil forces is conducted and the energy is shown to be a promising quantity in producing coils with low forces.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Quantum optimal control of superconducting qubits based on machine-learning characterization

Implementing fast and high-fidelity quantum operations using open-loop quantum optimal control relies on having an accurate model of the quantum dynamics. Any deviations between this model and the complete dynamics of the device, such as the presence of spurious modes or pulse distortions, can degrade the performance of optimal controls in practice. Here, we propose an experimentally simple approach to realize optimal quantum controls tailored to the device parameters and environment while specifically characterizing this quantum system. Concretely, we use physics-inspired machine learning to infer an accurate model of the dynamics from experimentally available data and then optimize our experimental controls on this trained model. We show the power and feasibility of this approach by optimizing arbitrary single-qubit operations in detailed numerical simulations of a superconducting transmon qubit. Furthermore, we demonstrate that this framework produces an accurate description of the device dynamics under arbitrary controls, together with the precise pulses achieving arbitrary single-qubit gates with a high fidelity of ∼99.99%.

Artificial neural networks↗

Detector configuration of DECIGO/BBO and identification of cosmological neutron-star binaries

The primary target for the planned space-borne gravitational wave interferometers DECIGO/BBO (Decihertz Interferometer Gravitational Wave Observatory/Big Bang Observer) is a primordial gravitational wave background. However there exist astrophysical foregrounds, and among them, gravitational waves from neutron-star (NS) binaries are the solid and strong component that must be identified and subtracted. In this paper, we discuss the geometry of detector configurations preferable for identifying the NS/NS binary signals. As a first step, we analytically estimate the minimum signal-to-noise ratios of the binaries for several static detector configurations that are characterized by adjustable geometrical parameters, and determine the optimal values for these parameters. Next we perform numerical simulations to take into account the effect of detector motions, and find reasonable agreements with the analytical results. We show that, with the standard network formed by 4 units of triangle detectors, the proposed BBO sensitivity would be sufficient in receiving gravitational waves from all the NS/NS binaries at z{<=}5 with signal-to-noise ratios higher than 25. We also discuss the minimum sensitivity of DECIGO required for the foreground identification.

79 ASTRONOMY AND ASTROPHYSICS↗

Synchronous and Concurrent Multidomain Computing Method for Cloud Computing Platforms

We present a numerical method for synchronous and concurrent solution of transient elastodynamics problem where the computational domain is divided into subdomains that may reside on separate computational platforms. Here, this work employs the variational multiscale discontinuous Galerkin (VMDG) method to develop interdomain transmission conditions for transient problems. The fine-scale modeling concept leads to variationally consistent coupling terms at the common interfaces. The method admits a large class of time discretization schemes, and decoupling of the solution for each subdomain is achieved by selecting any explicit algorithm. Numerical tests with a manufactured solution problem show optimal convergence rates. The energy history in a free vibration problem is in agreement with that of the solution from a monolithic computational domain.

97 MATHEMATICS AND COMPUTING↗

Binary Optimal Control of Single-Flux-Quantum Pulse Sequences

We introduce a binary, relaxed gradient, trust-region method for optimizing pulse sequences for single flux quanta (SFQ) control of a quantum computer. The pulse sequences are optimized with the goal of realizing unitary gate transformations. Each pulse has a fixed amplitude and duration. Here we model this process as an binary optimal control problem, constrained by Schrödinger’s equation, where the binary variables indicate whether each pulse is on or off. We introduce a first-order trust-region method, which takes advantage of a relaxed gradient to determine an optimal pulse sequence that minimizes the gate infidelity, while also suppressing leakage to higher energy levels. The proposed algorithm has a computational complexity of O(p log(p)), where p is the number of pulses in the sequence. We present numerical results for the H and X gates, where the optimized pulse sequences give gate fidelity’s better than 99.9%, in ≈ 25 trust-region iterations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Enhancing Gaussian Process Surrogates for Optimization and Posterior Approximation via Random Exploration

This paper proposes novel noise-free Bayesian optimization strategies that rely on a random exploration step to enhance the accuracy of Gaussian process surrogate models. The new algorithms retain the ease of implementation of the classical GP-UCB algorithm, but the additional random exploration step accelerates their convergence, nearly achieving the optimal convergence rate. Furthermore, to facilitate Bayesian inference with intractable likelihoods, we propose to utilize optimization iterates for maximum a posteriori estimation to build a Gaussian process surrogate model for the unnormalized log-posterior density. We provide bounds for the Hellinger distance between the true and the approximate posterior distributions in terms of the number of design points. We demonstrate the effectiveness of our Bayesian optimization algorithms in nonconvex benchmark objective functions, in a machine learning hyperparameter tuning problem, and in a black-box engineering design problem. The effectiveness of our posterior approximation approach is demonstrated in two Bayesian inference problems for parameters of dynamical systems.

Bayesian inference↗

COMPASS: A Framework for Automated Performance Modeling and Prediction

Flexible, accurate performance predictions offer numerous benefits such as gaining insight into and optimizing applications and architectures. However, the development and evaluation of such performance predictions has been a major research challenge, due to the architectural complexities. To address this challenge, we have designed and implemented a prototype system, named COMPASS, for automated performance model generation and prediction. COMPASS generates a structured performance model from the target application's source code using automated static analysis, and then, it evaluates this model using various performance prediction techniques. As we demonstrate on several applications, the results of these predictions can be used for a variety of purposes, such as design space exploration, identifying performance tradeoffs for applications, and understanding sensitivities of important parameters. COMPASS can generate these predictions across several types of applications from traditional, sequential CPU applications to GPU-based, heterogeneous, parallel applications. Our empirical evaluation demonstrates a maximum overhead of 4%, flexibility to generate models for 9 applications, speed, ease of creation, and very low relative errors across a diverse set of architectures.

Lee, Seyong↗