Data-Based System Reliability Design for Wind

目前的设计标准中使用的风荷载没有考虑各种不确定因素，如风洞记录、气候数据和表面粗糙度长度。这种简化的表征可能会产生不准确的风效应，结果是不能令人满意地估计建筑物在风荷载下的性能。该奖项将基于气候学数据、风洞记录和严格的概率/统计概念，为风设计开发一个实用的计算框架，从而克服当前标准的局限性。该框架将通过概率模型量化风荷载中的不确定性，通过有效的计算方法将这些不确定性传播到风效应中，并制定将建筑物视为系统而不是独立组件集合的设计条件。这项研究将得到一个教育项目的补充，该项目涉及课程开发、研究生和本科生研究人员，以及在国内和国际会议上的演讲。该计算框架将作为风设计的一个模块与NSF支持的自然灾害工程研究基础设施、计算建模和模拟中心共享。这项研究将有助于国家科学基金会在国家减少风暴影响计划中发挥作用。该项目将开发一个计算框架，构成风能工程的一个新范例。该框架将通过考虑建筑物的机械和空气动力学特性，将气候和风洞数据映射到建筑物的系统可靠性估计中。风设计的系统可靠性方法将基于：(1)风荷载的新的概率模型，其样本和实现，而不仅仅是它们的平均值和相关函数，将与风洞时间历史在统计上一致；(2)在平稳/非平稳、高斯/非高斯风荷载作用下，根据线性/非线性结构的风荷载样本计算风效应样本的新的有效方法；以及(3)结构性能的系统可靠性估计，不同于目前逐个成员检查极限/正常使用极限状态的方法，该方法将同时考虑所有结构成员的极限状态。最终/可维护性标准的格式将与当前标准类似。该裁决反映了NSF的法定使命，并已通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Wind loads used in current design standards do not account for various sources of uncertainties, such as wind tunnel records, climatological data, and surface roughness length. This simplified characterization may yield inaccurate wind effects and, as a result, unsatisfactory estimates of the performance of buildings subjected to wind loads. This award will overcome limitations of current standards by developing a practical computational framework for wind design, based on climatological data, wind tunnel records, and rigorous probabilistic/statistical concepts. The framework will quantify the uncertainties in wind loads by probabilistic models, propagate these uncertainties into wind effects by efficient computational methods, and develop design conditions which view buildings as systems rather than collections of independent components. The research will be complemented by an educational program involving course development, graduate and undergraduate researchers, and presentations at national and international conferences. The computational framework will be shared with the NSF-supported Natural Hazards Engineering Research Infrastructure, Computational Modeling and Simulation Center, as a module for wind design. This research will contribute to NSF's role in the National Windstorm Impact Reduction Program. This project will develop a computational framework that constitutes a new paradigm in wind engineering. The framework will map climatological and wind tunnel data into system reliability estimates for buildings subjected to wind loads by accounting for the mechanical and aerodynamic characteristics of these buildings. The system reliability methodology for wind design will be based on: (1) novel probabilistic models for wind loads whose samples and realizations, rather than just their mean and correlation functions, will be statistically consistent with wind tunnel time histories, (2) new efficient methods for calculating wind effect samples from wind load samples in linear/nonlinear structures subjected to stationary/nonstationary, Gaussian/non-Gaussian wind loads, and (3) system reliability estimates of structural performance that, unlike current approaches that check ultimate/serviceability limit states member-by-member, will consider these limit states simultaneously at all structural members. The format of the ultimate/serviceability criteria will be similar to that of current standards.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.