CAREER: Fusing Meta-Learning Systems and Field Observations to Enhance the Simulation of Extreme Winds and their Impact on Civil Infrastructure

该学院早期职业发展（Career）奖将资助旨在促进对飓风期间发生的极端风场的地点依赖性及其对建筑环境影响的基本理解的研究。研究的动机是在实地研究活动中收集的地面风观测数据与传统边界层风洞模拟的理想风流条件之间存在明显差异。该研究将利用元学习系统、飓风登陆期间收集的全尺寸风速数据和高通量实验测试来模拟真实世界的风危害条件及其与沿海社区复杂地形特征的相互作用。该合同将扩大原本稀疏的地面风速测量的地理空间覆盖范围，以支持风暴后的现场侦察小组综合观测到的特定风力灾害条件下的结构破坏。该研究将与教育和公众参与活动相结合，包括开发学习模块，以加速将基于人工智能的工具引入土木工程课程，为塔拉哈西市的中学生举办学习型学习（L2L）研讨会，并促进佛罗里达狭长地带沿海社区的公众和利益相关者参与。本项目生成的数据将在自然灾害工程研究基础设施（NHERI）数据仓库（https:/www.DesignSafe-c.org）中存档并公开提供。该奖项将有助于国家科学基金会在国家减少风暴影响计划（NWIRP）中的作用。这些研究活动旨在推动更广泛的社会成果，直接解决与国家减轻和适应极端风灾害事件不断变化的威胁的能力相关的持久挑战。本研究的具体目标包括：(i)使用多扇流量控制仪器在多个几何尺度上重建真实世界的飓风记录；（ii）使用高分辨率数字表面模型（DSM）对沿海社区的真实世界地形形态进行物理模拟；（iii）整合元学习算法，提取影响地面飓风的远端和近端地形特征的显著特征。（iv）开发基于实际风流轨迹的现场特定风（组件和建筑级）负载推断工具。实验工作将利用佛罗里达大学的NHERI边界层风洞。该奖项反映了美国国家科学基金会的法定使命，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 Faculty Early Career Development (CAREER) award will fund research that intends to advance fundamental understanding of the site-dependent nature of extreme wind fields occurring during hurricanes and their impact on the built environment. The research is motivated by the palpable differences between ground-level wind observations collected during field research activities and the idealized wind flow conditions simulated in traditional boundary layer wind tunnels. This research will leverage meta-learning systems, full-scale anemometric data collected during landfalling hurricanes, and high-throughput experimental testing to simulate real-world wind hazard conditions and their interaction with complex terrain features of coastal communities. The award will expand the geospatial coverage of inherently sparse ground-based anemometric measurements to support post-storm field reconnaissance teams in synthesizing observed structural damage to site-specific wind hazard conditions. The research will be integrated with educational and public engagement activities, including the development of learning modules to accelerate the implementation of artificial intelligence-based tools into civil engineering curricula, hosting Learning-to-Learn (L2L) workshops for middle and high school students from the City of Tallahassee, and fostering public and stakeholder engagement from coastal communities in the Florida Panhandle. Data generated from this project will be archived and made publicly available in the Natural Hazards Engineering Research Infrastructure (NHERI) Data Depot (https:/www.DesignSafe-c.org). This award will contribute to the National Science Foundation's role in the National Windstorm Impact Reduction Program (NWIRP).The research activities are intended to drive broader societal outcomes that directly address enduring challenges related to the nation's ability to mitigate and adapt to the ever-changing threat of extreme wind hazard events. The specific goals of this research include (i) the recreation of real-world hurricane wind records at multiple geometric scales using a multi-fan flow-control instrument, (ii) the physical simulation of real-world terrain morphology of coastal communities using high-resolution digital surface models (DSM), (iii) the integration of meta-learning algorithms to extract salient features of distal and proximate terrain characteristics that influence ground-level hurricane winds, and (iv) the development of a site-specific wind (component- and building-level) load inference tool based on realistic wind flow traces. The experimental work will utilize the NHERI Boundary Layer Wind Tunnel at the University of Florida.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.