CAREER: High Fidelity Numerical Modeling and Simulation of Fire Suppression

摘要：本研究的技术目标是开发一种先进的建模和仿真框架，用于预测使用水雾和喷雾扑灭大规模火灾。该模型基于大涡模拟（LES）技术，使用基于概率的亚网格尺度（SGS）模型来考虑浮力驱动的湍流、燃烧、抑制热化学、液滴输送和热辐射传热的多相耦合。具体来说，这项工作主要集中在：(1)改进的浮力驱动流动的SGS湍流模型，(2)新的SGS随机过程描述，以解释湍流-化学-辐射-粒子相互作用，以及(3)湍流多相流中的辐射传热。这些模型的实现还包括开发先进的数值算法，以便在大规模并行计算机上有效地实现蒙特卡罗求解程序，以及开发用于参与各向异性散射介质的无网格辐射方法。这组新模型和算法的组合将导致灭火现场建模的重大改进，并推进多相燃烧和辐射过程重要的其他应用的定量处理。这项工作的意义在于为扑灭最近引起全国关注的大规模火灾提供科学见解，从纽约双子塔的毁坏到肆虐西南地区的野火。控制这些火灾的挑战之一是预测供水系统的性能，这取决于对高湍流、强辐射、多相、反应流中火焰抑制过程动力学的理解。本研究的目标之一是提供一个高保真度的预测工具来模拟这些过程。这种工具将使消防工程师能够设计出更好的灭火系统，以确保我们国家关键基础设施的消防安全。

Abstract - DesJardin, 0348110The technical objective of this research is to develop an advanced modeling and simulation framework for predicting the suppression of large-scale fires using water mists and sprays. The modeling is based on Large Eddy Simulation (LES) techniques using probabilistic based sub-grid scale (SGS) models to account for multiphase coupling of buoyancy-driven turbulence, combustion, suppression thermochemistry, droplet transport, and thermal radiation heat transfer. Specifically, this effort is focused on: (1) improved SGS turbulence models for buoyancy-driven flows, (2) new SGS stochastic process descriptions to account for turbulence-chemistry-radiation-particle interactions, and (3) radiation heat transfer in turbulent multiphase flows. The implementation of these models also includes development of advanced numerical algorithms for efficient implementation of Monte Carlo solution procedures implemented on massively parallel computers and development of gridless radiation methodologies for participating anisotropic scattering media. This combined set of new models and algorithms should result in a significant improvements in fire-suppression field modeling as well as advance the quantitative treatment of other applications in which multiphase combustion and radiation processes are important.The significance of this effort is to offer scientific insight on suppressing large-scale fires that have recently caught the attention of our nation, ranging from the destruction of the twin towers in New York City to the wild fires that have raged through the southwest. One of the challenges in controlling these fires is predicting the performance of the water-delivery system, which depends on the understanding of dynamics of flame suppression processes in highly turbulent, strongly radiating, multiphase, reacting flows. One of the goals of this research is to provide a high fidelity predictive tool to simulate these processes. Such a tool will allow fire-protection engineers to design better fire-suppression systems to insure fire safety for our nations' critical infrastructures.