EAGER: Advanced Digital Twin Capability for Turbulent Wind Fields in the NHERI Boundary Layer Wind Tunnel at the University of Florida

这项早期概念探索性研究资助（EAGER）将为佛罗里达大学（UF）的国家科学基金会（NSF）支持的自然灾害工程研究基础设施（NHERI）边界层风洞（BLWT）建立，验证和传播先进的数字孪生能力。风洞试验仍然是评估结构风荷载和通知抗风设计以降低损坏成本的最常用方法。然而，风洞实验有其局限性，如测量分辨率和获得风速和压力场的同时记录的挑战。数值模拟，如大涡模拟（LES），提供了一个机会，以填补这些空白，但这种模拟能力目前没有最佳利用的研究界。一个重要的障碍是，目前的数值模拟能力主要是针对静止的，标准的中性风廓线;相比之下，风洞，如在佛罗里达州的BLWT越来越多地实施先进的功能，以再现更复杂的湍流风场，造成结构破坏。该研究项目将建立这些复杂风场的数值模拟能力。为了最大限度地发挥该项目的潜在影响，验证测试用例和相应的数字孪生工具集和模拟功能教程将通过与当前UF BLWT用户群的结构化访谈来定义。由此产生的数字孪生能力将使联合利用数值和实验模型成为可能，以提高对导致建筑物和民用基础设施损坏的湍流风荷载的理解，并推进风弹性设计。模拟数据和记录的源代码将在NHERI数据库（https：//www.example.com）中存档并公开提供。www.DesignSafe-ci.org EAGER将有助于NSF在国家减少风暴影响计划中发挥作用。研究的具体目标是建立和推广一种数值模拟策略，用于再现UF BLWT中产生的复杂湍流风场。对于标准中性对数律风场，入流边界条件通常采用人工湍流生成方法。由于人工湍流的速度统计在计算域中演变，因此需要某种形式的校准以确保正确再现目标风场。当目标是模拟更复杂的湍流风场时，这种校准挑战会加剧，例如在UF BLWT中可以产生具有明显粗糙度子层的边界层，这对于低层建筑物很重要，其中建筑物浸没在粗糙度子层中（边界层的粗糙度高度与建筑物高度相当）。这个项目的第一个目标是探索计算效率和精确的方法，数值再现这些粗糙子层。人工湍流入流发生器、上游粗糙度解析模拟、源项强迫方法和机器学习方法的不同组合将根据实验数据进行研究和验证。该项目的第二个目标是通过与BLWT用户群共同设计一个数字孪生工具集和一个教程来支持所产生的湍流生成方法的广泛传播。数字风洞还可以帮助确定物理测试的最佳测量位置，并在物理测量的空间分辨率有限的情况下支持数据填充。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This EArly-concept Grant for Exploratory Research (EAGER) will establish, validate, and disseminate an advanced digital twin capability for the National Science Foundation (NSF)-supported Natural Hazards Engineering Research Infrastructure (NHERI) boundary layer wind tunnel (BLWT) at the University of Florida (UF). Wind tunnel testing remains the most common approach for assessing wind loads on structures and informing wind resistant design to reduce the cost of damage. However, wind tunnel experiments have limitations, such as the measurement resolution and the challenge of obtaining simultaneous records of wind velocity and pressure fields. Numerical simulations, such as Large Eddy Simulation (LES), offer an opportunity to fill in these gaps, but such simulation capabilities are currently not optimally leveraged by the research community. An important barrier is that current numerical modeling capabilities are mostly tailored to stationary, standard neutral wind profiles; in contrast, wind tunnels such as the BLWT at UF are increasingly implementing advanced capabilities to reproduce more complex turbulent wind fields that cause structural damage. This research project will establish numerical simulation capabilities for these complex wind fields. To maximize the potential impact of the project, validation test cases and a corresponding digital twin tool set and tutorial for the simulation capabilities will be defined through structured interviews with the current UF BLWT user base. The resulting digital twin capability will make it possible to jointly leverage numerical and experimental models to improve understanding of the turbulent wind loads that drive damage to buildings and civil infrastructure and to advance wind resilient design. Simulation data and documented source codes will be archived and made publicly available in the NHERI Data Depot (https://www.DesignSafe-ci.org). This EAGER will contribute to the NSF role in the National Windstorm Impact Reduction Program. The specific goal of the research is to establish and disseminate a numerical modeling strategy for reproducing complex turbulent wind fields generated in the UF BLWT. For standard neutral log-law wind fields, inflow boundary conditions commonly employ artificial turbulence generation methods. Since the velocity statistics of artificial turbulence evolve within the computational domain, some form of calibration is required to ensure that the target wind field is correctly reproduced. This calibration challenge is exacerbated when the objective is to model more complex turbulent wind fields, such as the boundary layer with a pronounced roughness sublayer that can be produced in the UF BLWT, which is important for low-rise buildings, where the building is immersed in the roughness sublayer (the roughness height of the boundary layer is on the order of the building height). The first objective of this project is to explore computationally efficient and accurate methods for numerically reproducing these roughness sublayers. Different combinations of artificial turbulence inflow generators, upstream roughness resolving simulations, source term forcing methods, and machine learning approaches will be investigated and validated against experimental data. The second objective of this project is to support broad dissemination of the resulting turbulence generation method by co-designing a digital twin tool set and a tutorial with the BLWT user base. The digital wind tunnel can also help identify optimal measurement locations for physical testing and potentially support data infilling where there are limits on the spatial resolution of physical measurements.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.