I-Corps: Turbulent Stress Model in Support of Computational Fluid Dynamics

I-Corps项目对湍流计算流体动力学（CFD）软件应用的广泛影响/商业潜力可能具有变革性。湍流无处不在；此外，CFD方法通常用于评估与工业规模流动相关的流动保障和安全问题。尽管计算机硬件和软件技术已经取得了很大的进步，但目前紊流应力闭合模型严重限制了CFD方法的商业价值。最近有关湍流应力闭合问题的研究为改进CFD软件的性能提供了机会，该软件用于设计下一代化学和核反应堆、飞机发动机、汽车发动机和风力发电涡轮机。拟议的创新还将支持水力旋流分离器的原型设计以及下一代工业混合器的原型设计。I-Corps项目为长期存在的湍流问题提供了一个软件解决方案。尽管湍流应力闭合问题是一个多世纪以来研究的核心问题，但目前还没有任何商业软件包支持完全可实现的闭合模型。该项目将通过提供补充软件来解决这一需求，该软件实现了一类新的完全可实现的代数应力模型，这些模型与在旋转和非旋转湍流中观察到的低阶统计特性一致。最重要的是，不管选择哪个基准流来校准模型参数，闭包模型都是完全可实现的。这一发现将对当前紊流计算流体动力学软件的使用产生重大影响。

The broader impact/commercial potential of this I-Corps project on the use of computational fluid dynamic (CFD) software for turbulent flows may be transformational. Turbulent flows occur ubiquitously; and, CFD methods are routinely used to assess flow assurance and safety issues associated with industrial scale flows. Although computer hardware and software technologies have advanced significantly, current closure models for the turbulent stress seriously limit the commercial value of CFD methods. Recent research related to the turbulent stress closure problem provides an opportunity to improve the performance of CFD software used to design next generation chemical and nuclear reactors, aircraft engines, automotive engines, and wind turbines for power generation. The proposed innovation will also support prototype designs of hydrocyclone separators as well as prototype designs for next generation industrial mixers.This I-Corps project offers a software solution to a longstanding problem related to turbulence. Although the turbulent stress closure problem has been a core issue of research for more than a century, a fully-realizable closure model is not presently supported by any commercial software package. This project will address this need by providing complementary software that implements a new class of fully-realizable algebraic stress models consistent with the low-order statistical properties observed in rotating and non-rotating turbulent flows. Most significantly, the closure models are fully-realizable regardless of the benchmark flows selected to calibrate model parameters. This discovery will have a significant impact on the current use of computational fluid dynamic software for turbulent flows.