|
| |||||||||||
![]()
|
The Center for the Simulation of Accidental Fires & Explosions, located at the University of Utah, is one of five Advanced Simulation and Computing (ASC) centers created through the Academic Strategic Alliances Program (ASAP).
Executive Summary The University of Utah has created an alliance with the DOE Advanced Simulation and Computing (ASC) program to form the Center for the Simulation of Accidental Fires and Explosions (C-SAFE). It focuses specifically on providing state-of-the-art, science-based tools for the numerical simulation of accidental fires and explosions, especially within the context of handling and storage of highly flammable materials. The objective of C-SAFE is to provide a system comprising a problem-solving environment in which fundamental chemistry and engineering physics are fully coupled with non-linear solvers, optimization, computational steering, visualization and experimental data verification. The availability of simulations using this system will help to better evaluate the risks and safety issues associated with fires and explosions. Our team will integrate and deliver a system that will be validated and documented for practical application to accidents involving both hydrocarbon and energetic materials. Although the ultimate C-SAFE goal is to simulate fires involving a diverse range of accident scenarios including multiple high-energy devices, complex building/surroundings geometries and many fuel sources, the initial efforts will focus on the computation of three scenarios:
These large-scale problems require consideration of fundamental gas and condensed phase chemistry, structural mechanics, turbulent reacting flows, convective and radiative heat transfer, and mass transfer, in a time-accurate, full-physics simulation of accidental fires. This simulation will be expansive enough to include the physical and chemical changes in containment vessels and structures, the mechanical stress and rupture of the container, and the chemistry and physics of organic, metallic and energetic material inside the vessel. We will include deflagration-to-detonation transitions (DDT) of any energetic material in the fire, but the simulation will end when/if detonation occurs. C-SAFE will provide coupling of the micro-scale and meso-scale contributions to the macroscopic application in order to provide full-physics across the breadth of supporting mechanistic disciplines, and to achieve efficient utilization of ASC program supercomputers. We will utilize a simulation development roadmap (SDRM) consisting of three distinct, sequential steps, which parallel the events in our physical problem: Ignition andFire Spread, Container Dynamics and High Energy Transformations. A fire or explosion is initiated by an ignition which depending upon the magnitude of heat generation and dissipative terms, a perturbation by an ignition source either decays or grows into a flame, followed by a spreading fire and possibly explosion. The fire or explosion can cause the container of HE material to undergo changes, perhaps rupture and, simultaneously or sequentially, the HE material itself can undergo transformations which lead to an explosion. The overall mission is to integrate these computational steps into a coupled fire and explosion system. To fulfill this mission we will draw on three core disciplines available at the University: molecular fundamentals, computational engineering, and computer science. The thrust of the molecular fundamentals team will be to perform micro-scale analyses of physical and chemical processes. To this end, they are concerned with aspects of molecular dynamics, electronic structure, and statistical mechanics in an integrated fashion to dynamically obtain properties for all materials (condensed phases, vaporized phases, and structures) in the fire and explosion. The thrust of the computational engineering team is to develop meso-scale models that bridge the ranges of length and time scales between microscopic and macroscopic properties. They will also develop large-scale Eulerian and Lagrangian models to describe structural and transport processes with geometric and mechanistic fidelity. The computer science effort will focus on a system development framework which combines target architecture performance analysis tools at the lowest level with an integrated, higher level scientific problem solving environment to provide interactive computational steering, visualization and large data set analysis capabilities. Twenty key U of U faculty have joined with strategically selected faculty from nearby BYU(2) and WPI(1) and experimental scientists from Thiokol Corporation to create C- SAFE. Twenty-five post doctoral/professionals and many graduate students are involved in the interdisciplinary academic program of the center. Because of the problem complexity, the personnel are organized in two types of interdisciplinary teams with a matrix structure: 1) three SDRM Step Teams that are each charged with the development of one subset of the overall application, and 2) four Discipline Teams that are responsible for the fundamental science/engineering of each discipline. Each C-SAFE participant (faculty or student) is a member of one of the discipline teams. The SDRM step teams are composed of the area leader from each of the disciplines and other key participants from within the disciplines. Thus, each center participant is a member of both types of teams. We believe this tightly integrated structure is essential to simultaneously ensure that (1) the common objective of developing a verified, fire and explosive simulation system is attained and (2) modern scientific/computational techniques are used throughout. Decisions regarding selection of key components to integrate into each step will be based on nonlinear sensitivity analysis and numerical optimization of our overall accidental fire simulation. Our delivered product will be the C-SAFE system that embodies the complete technology for performing integrated and validated simulations of full-physics fires. Our C-SAFE system requires a computational infrastructure that can support multi-physics modeling of large- scale, complex phenomena. C-SAFE models the physical complexities from the molecular level of HE materials, through millimeter-sized representations of the container, to the meter-sized representations of the fire spread. At each of these levels, the simulations will involve up to 10^9 discrete mesh points. Due to the multiple scales, the spatial requirements may exceed the Terabyte range for the full simulation. The computation will also require 10^10 time-steps to compute the physical time scales ranging from microseconds to minutes or hours. Thus the storage requirements far exceed the capacities of most computing facilities. Not only are the memory and storage requirements at the terascale, the computational demands are also on the order of tens to hundreds of teraflops. When these requisites are compounded with the visualization needs, successful realization of the C-SAFE system involves dataset management, model building, simulation, and visualization at the Terascale level. The C-SAFE system will be validated by rigorous comparison with experimental data for a variety of conditions at four different levels:
Initially, the primary focus will be on utilizing existing data, especially from the National Labs and Thiokol. Because it is inevitable that additional, detailed data will ultimately be required to validate the C-SAFE computations, a new integrated fire test facility will be created on-site using University cost sharing funds to directly support this program. In later years, limited large-scale supplemental testing with high energy materials will also be conducted at Thiokol. The University of Utah is ideally suited to create this center because it is just completing construction of a new, interdisciplinary scientific computations building, the planned home for C-SAFE and more than half of the proposed participants. We have recently been named the site of the first SGI Center for Visual Supercomputing which means that the C-SAFE teams will likely have direct, on-site access to both a 60 CPU Origin 2000 with eight Infinite Reality Graphics Engines and a 64 node IBM SP with 13 Gbytes of memory. The faculty have extensive, proven background in the microscopic modeling of static and dynamic condensed phase systems; visualization, parallel architectures, and object-oriented scientific programming and macroscopic modeling of combustion and reactive-flow processes. The NSF-funded Advanced Combustion Engineering Research Center is just completing 11 years of operation and the University is home to the world-renowned Henry Eyring Center for Theoretical Chemistry. | ||
|
Site Feedback | |||