高架桥梁声屏障随机车流扰动风荷载谱与应力概率模型

The vehicle-induced aerodynamic load (VIAL) has been studied by the applicant under support provided by National Science Youth Fund, “The mechanism study on moving vehicle induced aerodynamic load and the load effect characteristics on sound barriers on urban viaducts”. The influence factors and the variation rules of VIAL on vertical sound barriers has been revealed, and the expression of peak pressure is given based on potential flow theory. The existing results laid a solid foundation for the subsequent study on fatigue load on sound barriers caused by VIAL. Therefore, this research project focuses on exploring the way of combining the deterministic VIAL and the stochastic vehicle flow, and establishing the method to obtain VIAL and its effect on various types of sound barriers in a probabilistic context. Specifically, the numerical simulation tools should be conducted and validated by on-site tests in the first place, by which the spatial and temporal distribution of the VIAL, as well as the responses of sound barriers could be obtained. Afterwards, with the help of computer-aided visualization, machine learning and vehicle flow simulation, massive vehicles running on the viaduct can be identified or simulated, providing the statistic characteristics of vehicle type, speed and vehicle-barrier separation distance with respect to different recurrence period. Eventually, the load spectrum of VIAL, as well as the probabilistic model of stresses at key points, are established via data association. The results will provide basic data, numerical tools and practical methods for the structural design of the sound barrier on viaducts.

申请者主持完成的国家自然科学基金青年基金“城市高架道路声屏障受行驶汽车扰动的机理及其风荷载特性研究”揭示了车致直立型声屏障风压极值的影响因素和变化规律，并基于势流理论和直立边界假定给出了车致极值风压表达式。已有成果为后续声屏障随机车流扰动作用疲劳荷载研究奠定了重要基础。本课题探索将车辆扰动确定性和车流特征随机性相整合，建立声屏障随机车流扰动风荷载和风效应的概率表达方法，并能涵盖多种型式的声屏障。具体地，通过实测校验建立车致风荷载和风效应数值分析工具，揭示多参数影响下声屏障车致风压时空分布规律及其与结构响应的时空关系，给出相应的数学表达；结合计算机视觉、机器学习和车流模拟技术，实现高架桥汽车行驶特征识别和模拟外推，获得不同重现期车型、车速、车屏距离三参数特征向量；通过数据关联构建随机车流扰动风荷载谱和应力概率模型。成果将为随机车流作用的高架桥声屏障设计提供基础数据、数值工具和实用方法。

本项目通过计算流体动力学、结构动力学、计算机视觉和人工智能等领域的交叉与融合，采用理论分析、试验研究与数值模拟等手段，建立了车致声屏障风荷载模拟和风效应模拟的数值工具，揭示了汽车与声屏障空间关系变化过程对声屏障风压时空分布和响应特征的影响规律，提出了车致风荷载和风效应的时空分布模型，改进了车流特征识别和模拟方法，获得了车流特征向量的概率模型表达，进而建立了车流扰动风荷载谱和应力概率模型。综上，项目研究最终系统地建立了声屏障随机车流扰动风荷载与风效应研究的数值工具、基础理论和实用方法，对汽车行驶气流扰动及周边物体风荷载作用机制的科学问题揭示提供了新方法和新认识，具有明显的科学意义。相关成果发表SCI收录期刊论文7篇、EI收录期刊论文2篇、国际会议论文5篇，授权发明专利1项，编入国家标准《声屏障结构技术标准（GB/T51335-2018）》，获得上海市科技进步二等奖1项，培养毕业博士和硕士各1名；并进一步拓展应用到声屏障车致风荷载智慧监测、大场域车辆时空分布获取、工程结构模型三维重建、结构表观特征视觉智能识别、风致行车安全评估等方面，应用前景广阔。