Publications

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Aria is a Galerkin finite element based program for solving coupled-physics problems described by systems of PDEs and is capable of solving nonlinear, implicit, transient and direct-to-steady state problems in two and three dimensions on parallel architectures. The suite of physics currently supported by Aria includes the incompressible Navier-Stokes equations, energy transport equation, species transport equations, nonlinear elastic solid mechanics, and electrostatics as well as generalized scalar, vector and tensor transport equations. Additionally, Aria includes support for arbitrary Lagrangian-Eulerian (ALE) and level set based free and moving boundary tracking. Coupled physics problems are solved in several ways including fully-coupled Newton's method with analytic or numerical sensitivities, fully-coupled Newton-Krylov methods, fully-coupled Picard's method, and a loosely-coupled nonlinear iteration about subsets of the system that are solved using combinations of the aforementioned methods. Error estimation, uniform and dynamic h-adaptivity and dynamic load balancing are some of Aria's more advanced capabilities. Aria is based on the Sierra Framework.

In recent years, considerable effort has been placed on developing efficient and robust solution algorithms for the incompressible Navier-Stokes equations based on preconditioned Krylov methods. These include physics-based methods, such as SIMPLE, and purely algebraic preconditioners based on the approximation of the Schur complement. All these techniques can be represented as approximate block factorization (ABF) type preconditioners. The goal is to decompose the application of the preconditioner into simplified sub-systems in which scalable multi-level type solvers can be applied. In this paper we develop a taxonomy of these ideas based on an adaptation of a generalized approximate factorization of the Navier-Stokes system first presented in [25]. This taxonomy illuminates the similarities and differences among these preconditioners and the central role played by efficient approximation of certain Schur complement operators. We then present a parallel computational study that examines the performance of these methods and compares them to an additive Schwarz domain decomposition (DD) algorithm. Results are presented for two and three-dimensional steady state problems for enclosed domains and inflow/outflow systems on both structured and unstructured meshes. The numerical experiments are performed using MPSalsa, a stabilized finite element code.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

The goal of our paper is to demonstrate the cost-effective use of the Lanczos method for estimating the critical time step in an explicit, transient dynamics code. The Lanczos method can provide a significantly larger estimate for the critical time-step than an element-based method (the typical scheme). However, the Lanczos method represents a more expensive method for calculating a critical time-step than element-based methods. Our paper shows how the additional cost of the Lanczos method can be amortized over a number of time steps and lead to an overall decrease in run-time for an explicit, transient dynamics code. We present an adaptive hybrid scheme that synthesizes the Lanczos-based and element-based estimates and allows us to run near the critical time-step estimate provided by the Lanczos method. Copyright © 2006 John Wiley & Sons, Ltd.

We have developed an accelerator architecture that can serve as the basis of the design of petawatt-class z-pinch drivers. The architecture has been applied to the design of two z-pinch accelerators, each of which can be contained within a 104-m-diameter cylindrical tank. One accelerator is driven by slow (∼1μs) Marx generators, which are a mature technology but which necessitate significant pulse compression to achieve the short pulses (1μs) required to drive z pinches. The other is powered by linear transformer drivers (LTDs), which are less mature but produce much shorter pulses than conventional Marxes. Consequently, an LTD-driven accelerator promises to be (at a given pinch current and implosion time) more efficient and reliable. The Marx-driven accelerator produces a peak electrical power of 500 TW and includes the following components: (i) 300 Marx generators that comprise a total of 1.8×104 capacitors, store 98 MJ, and erect to 5 MV; (ii) 600 water-dielectric triplate intermediate-store transmission lines, which also serve as pulse-forming lines; (iii) 600 5-MV laser-triggered gas switches; (iv) three monolithic radial-transmission-line impedance transformers, with triplate geometries and exponential impedance profiles; (v) a 6-level 5.5-m-diameter 15-MV vacuum insulator stack; (vi) six magnetically insulated vacuum transmission lines (MITLs); and (vii) a triple-post-hole vacuum convolute that adds the output currents of the six MITLs, and delivers the combined current to a z-pinch load. The accelerator delivers an effective peak current of 52 MA to a 10-mm-length z pinch that implodes in 95 ns, and 57 MA to a pinch that implodes in 120 ns. The LTD-driven accelerator includes monolithic radial transformers and a MITL system similar to those described above, but does not include intermediate-store transmission lines, multimegavolt gas switches, or a laser trigger system. Instead, this accelerator is driven by 210 LTD modules that include a total of 1×106 capacitors and 5×105 200-kV electrically triggered gas switches. The LTD accelerator stores 182 MJ and produces a peak electrical power of 1000 TW. The accelerator delivers an effective peak current of 68 MA to a pinch that implodes in 95 ns, and 75 MA to a pinch that implodes in 120 ns. Conceptually straightforward upgrades to these designs would deliver even higher pinch currents and faster implosions. © 2007 The American Physical Society.

In this paper, a system model is developed to investigate independent and coupled effects of resolution, noise and data processing algorithms on the accuracy of the scalar gradient and dissipation measurements in turbulent flows. Finite resolution effects are simulated by spectral filtering, noise is modelled as an additive source in the model spectrum and differencing stencils are analysed as digital filters. In the current study, the effective resolution is proposed to be a proper criterion for quantifying the resolution requirement for scalar gradient and dissipation measurement. Both effective resolution and noise-induced apparent dissipation are mainly determined by the system transfer function. The finite resolution results, based upon a model scalar energy spectrum, are shown to agree with non-reacting experimental data. The coupled resolution-noise results show three regions in the mean scalar dissipation rate measurement: noise-dominated region, noise-resolution correlated region and resolution-dominated region. Different noise levels lead to different resolution error curves for the measured mean scalar dissipation rate. Experimental procedures and guidelines to improve the scalar gradient and dissipation experiments are proposed, based on these model study results. Finally, the proposed system approach can also be applied to other derived quantities involving complex transfer functions.

Six computational methods for solution of the radiative transfer equation in an absorbing-emitting, nonscattering gray medium were compared for a 2-m JP-8 pool fire. The emission temperature and absorption coefficient fields were taken from a synthetic fire due to the lack of a complete set of experimental data for computing radiation for large and fully turbulent fires. These quantities were generated by a code that has been shown to agree well with the limited quantity of relevant data in the literature. Reference solutions to the governing equation were determined using the Monte Carlo method and a ray-tracing scheme with high angular resolution. Solutions using the discrete transfer method (DTM), the discrete ordinates method (DOM) with both S4 and LC11 quadratures, and a moment model using the M1 closure were compared to the reference solutions in both isotropic and anisotropic regions of the computational domain. Inside the fire, where radiation is isotropic, all methods gave comparable results with good accuracy. Predictions of DTM agreed well with the reference solutions, which is expected for a technique based on ray tracing. DOM LC11 was shown to be more accurate than the commonly used S4 quadrature scheme, especially in anisotropic regions of the fire domain. On the other hand, DOM S4 gives an accurate source term and, in isotropic regions, correct fluxes. The M1 results agreed well with other solution techniques and were comparable to DOM S4. This represents the first study where the M1 method was applied to a combustion problem occurring in a complex three-dimensional geometry. Future applications of M1 to fires and similar problems are recommended, considering its similar accuracy and the fact that it has significantly lower computational cost than DOM S4. © 2006 The Combustion Institute.

Robots for high-force interaction with humans face particular challenges to achieve performance and coupled stability. Because available actuators are unable to provide sufficiently high force density and low impedance, controllers for such machines often attempt to mask the robots physical dynamics, though this threatens stability. Controlling for passivity, the state-of-the-art means of ensuring coupled stability, inherently limits performance to levels that are often unacceptable. A controller that imposes passivity is compared to a controller designed by a new method that uses limited knowledge of human dynamics to improve performance. Both controllers were implemented on a testbed, and coupled stability and performance were tested. Results show that the new controller can improve both stability and performance. The different structures of the controllers yield key differences in physical behavior, and guidelines are provided to assist in choosing the appropriate approach for specific applications.