Flexible and Optimal Dependency Management via Max-SMT
Explore the source record for details and available documents.
Engineering topics
Publications and source records attributed to Bell, J.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Abstract not provided
Abstract not provided
We describe the development of a flux-limited gray radiation solver for the compressible astrophysics code, CASTRO. CASTRO uses an Eulerian grid with block-structured adaptive mesh refinement based on a nested hierarchy of logically rectangular variable-sized grids with simultaneous refinement in both space and time. The gray radiation solver is based on a mixed-frame formulation of radiation hydrodynamics. In our approach, the system is split into two parts, one part that couples the radiation and fluid in a hyperbolic subsystem, and another parabolic part that evolves radiation diffusion and source-sink terms. The hyperbolic subsystem is solved explicitly with a high-order Godunov scheme, whereas the parabolic part is solved implicitly with a first-order backward Euler method.
We explore the dependence on spatial dimension of the viability of the neutrino heating mechanism of core-collapse supernova explosions. We find that the tendency to explode is a monotonically increasing function of dimension, with three dimensions (3D) requiring {approx}40%-50% lower driving neutrino luminosity than one dimension and {approx}15%-25% lower driving neutrino luminosity than two dimensions (2D). Moreover, we find that the delay to explosion for a given neutrino luminosity is always shorter in 3D than 2D, sometimes by many hundreds of milliseconds. The magnitude of this dimensional effect is much larger than the purported magnitude of a variety of other effects, such as nuclear burning, inelastic scattering, or general relativity, which are sometimes invoked to bridge the gap between the current ambiguous and uncertain theoretical situation and the fact of robust supernova explosions. Since real supernovae occur in three dimensions, our finding may be an important step toward unraveling one of the most problematic puzzles in stellar astrophysics. In addition, even though in 3D, we do see pre-explosion instabilities and blast asymmetries, unlike the situation in 2D, we do not see an obvious axially symmetric dipolar shock oscillation. Rather, the free energy available to power instabilities seems to be shared by more and more degrees of freedom as the dimension increases. Hence, the strong dipolar axisymmetry seen in 2D and previously identified as a fundamental characteristic of the shock hydrodynamics may not survive in 3D as a prominent feature.
The inclusion of rotationally induced mixing in stellar evolution can alter the structure and composition of pre-supernova stars. We survey the effects of progenitor rotation on nucleosynthetic yields in Population III and II supernovae (SNe) using the new adaptive mesh refinement code CASTRO. We examine piston-driven spherical explosions in 15, 25, and 40 M{sub sun} stars at Z = 0 and 10{sup -4} Z{sub sun} with three explosion energies and two rotation rates. Rotation in the Z = 0 models resulted in primary nitrogen production and a stronger hydrogen burning shell which led all models to die as red supergiants (in contrast to the blue supergiant progenitors made without rotation). On the other hand, the Z = 10{sup -4} Z{sub sun} models that included rotation ended their lives as compact blue stars. Because of their extended structure, the hydrodynamics favors more mixing and less fallback in the metal-free stars than the Z = 10{sup -4} models. As expected, higher energy explosions produce more enrichment and less fallback than do lower energy explosions, and at constant explosion energy, less massive stars produce more enrichment and leave behind smaller remnants than do more massive stars. We compare our nucleosynthetic yields to the chemical abundances in the three most iron-poor stars yet found and reproduce the abundance pattern of one, HE 0557-4840, with a zero metallicity, 15 M{sub sun}, 2.4 x 10{sup 51} erg SN. A Salpeter IMF-averaged integration of our yields for Z = 0 models with explosion energies of 2.4 x 10{sup 51} erg or less is in good agreement with the abundances observed in larger samples of extremely metal-poor (EMP) stars, provided 15 M{sub sun} stars are included. Since the abundance patterns of EMP stars likely arise from a representative sample of progenitors, our yields suggest that 15-40 M{sub sun} core-collapse SNe with moderate explosion energies contributed the bulk of the metals to the early universe.
The advent of highly accurate, large scale volumetric simulations has made data analysis and visualization techniques an integral part of the modern scientific process. To develop new insights from raw data, scientists need the ability to define features of interest in a flexible manner and to understand how changes in the feature definition impact the subsequent analysis of the data. Therefore, simply exploring the raw data is not sufficient. This paper presents a new topological framework for the analysis of large scale, time-varying, turbulent combustion simulations. It allows the scientists to explore interactively the complete parameter space of fuel consumption thresholds for an entire time-dependent combustion simulation. By computing augmented merge trees and their corresponding data segmentations, the system allows the user complete flexibility to segment, select, and track burning cells through time thanks to a linked view interface. We developed this technique in the context of low-swirl turbulent pre-mixed flame simulation analysis, where the topological abstractions enable an efficient tracking through time of the burning cells and provide new qualitative and quantitative insights into the dynamics of the combustion process.
Abstract not provided
Over the past half-century, the Applied Mathematics program in the U.S. Department of Energy's Office of Advanced Scientific Computing Research has made significant, enduring advances in applied mathematics that have been essential enablers of modern computational science. Motivated by the scientific needs of the Department of Energy and its predecessors, advances have been made in mathematical modeling, numerical analysis of differential equations, optimization theory, mesh generation for complex geometries, adaptive algorithms and other important mathematical areas. High-performance mathematical software libraries developed through this program have contributed as much or more to the performance of modern scientific computer codes as the high-performance computers on which these codes run. The combination of these mathematical advances and the resulting software has enabled high-performance computers to be used for scientific discovery in ways that could only be imagined at the program's inception. Our nation, and indeed our world, face great challenges that must be addressed in coming years, and many of these will be addressed through the development of scientific understanding and engineering advances yet to be discovered. The U.S. Department of Energy (DOE) will play an essential role in providing science-based solutions to many of these problems, particularly those that involve the energy, environmental and national security needs of the country. As the capability of high-performance computers continues to increase, the types of questions that can be answered by applying this huge computational power become more varied and more complex. It will be essential that we find new ways to develop and apply the mathematics necessary to enable the new scientific and engineering discoveries that are needed. In August 2007, a panel of experts in applied, computational and statistical mathematics met for a day and a half in Berkeley, California to understand the mathematical developments required to meet the future science and engineering needs of the DOE. It is important to emphasize that the panelists were not asked to speculate only on advances that might be made in their own research specialties. Instead, the guidance this panel was given was to consider the broad science and engineering challenges that the DOE faces and identify the corresponding advances that must occur across the field of mathematics for these challenges to be successfully addressed. As preparation for the meeting, each panelist was asked to review strategic planning and other informational documents available for one or more of the DOE Program Offices, including the Offices of Science, Nuclear Energy, Fossil Energy, Environmental Management, Legacy Management, Energy Efficiency & Renewable Energy, Electricity Delivery & Energy Reliability and Civilian Radioactive Waste Management as well as the National Nuclear Security Administration. The panelists reported on science and engineering needs for each of these offices, and then discussed and identified mathematical advances that will be required if these challenges are to be met. A review of DOE challenges in energy, the environment and national security brings to light a broad and varied array of questions that the DOE must answer in the coming years. A representative subset of such questions includes: (1) Can we predict the operating characteristics of a clean coal power plant? (2) How stable is the plasma containment in a tokamak? (3) How quickly is climate change occurring and what are the uncertainties in the predicted time scales? (4) How quickly can an introduced bio-weapon contaminate the agricultural environment in the US? (5) How do we modify models of the atmosphere and clouds to incorporate newly collected data of possibly of new types? (6) How quickly can the United States recover if part of the power grid became inoperable? (7) What are optimal locations and communication protocols for sensing devices in a remote-sensing network? (8) How can new materials be designed with a specified desirable set of properties? In comparing and contrasting these and other questions of importance to DOE, the panel found that while the scientific breadth of the requirements is enormous, a central theme emerges: Scientists are being asked to identify or provide technology, or to give expert analysis to inform policy-makers that requires the scientific understanding of increasingly complex physical and engineered systems. In addition, as the complexity of the systems of interest increases, neither experimental observation nor mathematical and computational modeling alone can access all components of the system over the entire range of scales or conditions needed to provide the required scientific understanding.
The atomic pair distribution function (PDF) method was used to study the structure of cesium aluminosilicate geopolymer. The geopolymer was prepared by reacting metakaolin with cesium silicate solution followed by curing at 50C for 24 h in a sealed container. Heating of Cs-geopolymer above 1000C resulted in formation of crystalline pollucite (CsAlSi{sub 2}O{sub 6}). PDF refinement of the pollucite phase formed displayed an excellent fit over the 10-30 {angstrom} range when compared with a cubic pollucite model. A poorer fit was attained from 1-10 {angstrom} due to an additional amorphous phase present in the heated geopolymer. On the basis of PDF analysis, unheated Cs-geopolymer displayed structural ordering similar to pollucite up to a length scale of 9 {angstrom}, despite some differences. Our results suggest that hydrated Cs{sup +} ions were an integral part of the Cs-geopolymer structure and that most of the water present was not associated with Al-OH or Si-OH bonds.
The atomic structure of geopolymers is often described as amorphous with a local structure that is equivalent to that of crystalline zeolites. However, this structural relationship has never been quantified beyond a first-nearest-neighbor bonding environment. In this study, the short to medium range (1 nm) structural order of metakaolin-based KAlSi{sub 2}O{sub 6}{center_dot}5.5H{sub 2}O geopolymer was quantified and compared to zeolitic tetragonal leucite (KAlSi2O6) using the X-ray atomic pair distribution function technique. Unheated KAlSi{sub 2}O{sub 6}{center_dot}5.5H{sub 2}O was found to be structurally similar to leucite out to a length of 8 {angstrom}, but had increased medium range disorder over the 4.5 {angstrom} < r < 8 {angstrom} range. On heating to >300 C, changes in the short to medium range structure were observed due to dehydration and removal of chemically bound water. Crystallization of leucite occurred in samples heated beyond 1050 C. Refinements of a leucite model against the PDF data for geopolymer heated to 1100 C for 24 h yielded a good fit.