Computational Methods for Multiscale Turbulent Reacting Flows

提出的研究目标是在湍流反应流动的多尺度建模框架内开发计算方法。该框架基于基于大涡模拟（LES）和随机低维模型一维湍流（ODT）模型的混合粗粒度和细粒度（resp.）模拟。ODT模型旨在捕捉湍流和分子运输以及化学/热释放之间的耦合。PI开发了燃烧框架，直接模拟与湍流-化学相互作用相关的亚网格尺度物理；但是，它可以很容易地在涉及低维随机解的三维确定性解的多尺度驱动流中有更广泛的应用。本文提出的工作扩展了模型公式，以解决多尺度计算工具的发展，以耦合两种解决方案。这些方法包括基于物理的滤波后的子网格尺度应力在LES和ODT之间的匹配，以及基于小波变换的多分辨率方法的实现。将制定和实施数据同化策略，以管理统计数据从随机解到高阶确定性LES解的转移。最后，将研究加速化学动力学和分子运输整合的策略，以提高混合LES-ODT配方的计算效率。各种公式将通过直接数值模拟（DNS）进行验证。提出了一种新的模拟框架来研究燃烧流动。化石燃料和新型燃料的燃烧构成了狮子？美国和其他国家的能源消费份额；随着计算资源和技术的扩展，模拟燃烧流的能力最终可以取代建造昂贵的发动机原型，加快从设计到生产的过程，并有助于减少污染物和温室气体的产生。燃烧模拟是复杂的，因为需要考虑从分子尺度到器件尺度的复杂物理；研究人员和他的同事将开发多尺度策略，通过混合模拟来模拟燃烧流动，旨在解决不同尺度范围的物理问题。综合结果是这些模拟的自举过程，结果是计算效率高的策略，以表示燃烧流的大范围尺度。很多工作需要深入了解燃烧流动的物理原理以及能源和环境影响；但是，其他方面的努力利用了信息技术的进步（包括访问超级计算资源，软件的进步）和计算数学。

The objective of the proposed research is to develop computational methods within a multiscale modeling framework for turbulent reacting flows. The framework is based on hybrid coarse-grained and fine-grained(resp.) simulations based on large-eddy simulations (LES) and a stochastic low-dimensional model, the one-dimensional turbulence (ODT) model. The ODT model is designed to capture the coupling between turbulent and molecular transport and chemistry/heat release. The framework has been developed by the PI for combustion to directly model subgrid scale physics associated with turbulence-chemistry interactions; but, it can easily have broader applications for multiscale driven flows involving 3D deterministic solutions with low-dimensional stochastic solutions. The work proposed here extends the model formulation to address the development of multiscale computational tools to couple the two solution schemes. The approaches involve physics-based matching of filtered subgrid scale stresses between LES and ODT as well as the implementation multiresolution methods based on the wavelet transform. Data assimilation strategies will be formulated and implemented for managing the transfer of statistics from the stochastic solution to the high-order deterministic LES solution. Finally, strategies to accelerate the integration of chemical kinetics and molecular transport will be investigated to improve the computational efficiency of the hybrid LES-ODT formulation. The various formulations will be validated with direct numerical simulations (DNS).The proposed effort addresses a novel simulation framework to study combustion flows. The combustion of fossil and novel fuels make up the lion?s share of energy consumption in the US and abroad; and the ability to simulate combustion flows, with expanding computational resources and know-how, can eventually replace building expensive prototypes of engines, and speed up the design-to-production process, and help mitigate the generation of pollutants and greenhouse gases. Combustion simulation is complex, because an account of complex physics from the molecular scale to the device scale is required; the investigator and his colleagues will develop multiscale strategies to simulate combustion flows through hybrid simulations that are designed to addresses physics at different ranges of scales. The combined outcome is a bootstrapping process for these simulations that results in computationally efficient strategies to represent the broad ranges of scales in combustion flows. Much of the effort requires insight to the physics of combustion flows and energy and environmental implications; but, other aspects of the effort draw upon advances in information technology (including access to supercomputing resources, advances in software) and computational mathematics.