Large Scale Siesmic Performance of Urban Regions

该奖项来自工程研究中心项目，旨在将密西西比州立大学计算场模拟工程研究中心在大规模计算模拟和可视化方面的专业知识，与加州大学伯克利分校太平洋地震工程研究中心的岩土和结构响应模拟，以及卡内基梅隆大学的先进计算地面运动和土壤-基础-结构-相互作用建模，以开发先进的计算能力，用于模拟和可视化城市地区地震对已建成基础设施的影响，并将此能力应用于模拟城市地区建筑物和其他结构的性能。最终目标是预测整个城市地区的损失数量和分布。提出的方法将整合“端到端”震源、路径、盆地和该地区地质结构对地面运动的地表影响，以及建筑物和桥梁的现实模型，包括土壤-基础-结构-相互作用效应，以开发分布式、高分辨率的模拟能力。所提出的分布式高性能计算仿真环境的主要特点是：结构类型、几何形状和性能的逼真表示；结构附近土体结构的精细建模；使用真实的表面拓扑结构和地下岩土力学特性；地面运动节点密度的高分辨率，以捕获确定结构响应所需的更高频率；明确考虑土-结构-基础相互作用效应；分析建筑物组合的同时地震反应，以检查通过土壤的结构对结构的相互作用效应和建筑环境对自由场地震地面运动的影响；自动数据存储、访问和传输；表示单个结构和集合体行为的复杂和大型数据集的可视化；以及这些组件在分布式计算环境中的集成。这个项目将使用模拟环境来调查地震对洛杉矶市区的影响。这个问题对于减轻灾害和减少地震风险非常重要，因为评估建筑物在其使用寿命期间将暴露的地面运动，并预测其对这种地面运动的反应，包括潜在的破坏，是适当设计和改造抗震建筑基础设施的必要步骤。基于性能的地震工程方法的动机是需要科学和透明的方法来将地震危害与结构性能和损失联系起来。预测损害和损失对应急计划和管理也有很大用处。可视化城市地区的损失可以帮助决策者和利益相关者就如何减少地震损失做出明智的决定。拟议的模拟环境将使用Globus工具包访问计算网格，NSF计划支持其部署，作为乔治·e·布朗地震工程模拟网络（NEES）的一部分。开发完成后，模拟环境将通过NEES向地震工程界开放。

This award from the Engineering Research Centers Program is to connect the expertise in large-scale computational simulation and visualization at the Engineering Research Center on Computational Field Simulation at Mississippi State University, in geotechnical and structural response simulation at the Pacific Earthquake Engineering Research Center at the University of California, Berkeley, and in advanced computational ground motion and soil-foundation-structure-interaction modeling at Carnegie Mellon University to develop an advanced computational capability for modeling and visualizing the effects of earthquakes in urban regions on the built infrastructure, and to apply this capability to simulate the performance of collections of buildings and other structures in an urban region. The ultimate goal is to forecast the amount and distribution of damage throughout an urban region. The proposed methodology will integrate "end-to-end" the earthquake source, path, basin, and surficial effects of the geological structure of the region on ground motion, with realistic models of buildings and bridges, including soil-foundation-structure-interaction effects, to develop a distributed, high resolution simulation capability. Key features of the proposed distributed high-performance computational simulation environment are the following: realistic representation of the structure type, geometry and properties; detailed modeling of the soil structure in the near region of the structure; use of real surface topology and sub-surface geotechnical properties; high resolution in the node density of the ground motion to capture the higher frequencies required for determining structural response; explicit consideration of soil-structure-foundation interaction effects; analysis of the simultaneous earthquake response of a portfolio of buildings to examine structure-to-structure interaction effects through the soil and the effects of the built environment on the free-field earthquake ground motion; automated data storage, access, and transport; visualization of complex and large datasets representing behavior of individual structures and aggregates; and integration of these components in a distributed computational environment. This project will use the simulation environment to investigate the effects of earthquakes in the Los Angeles urban region. This problem is of great importance to hazard mitigation and seismic risk reduction because assessing the ground motion to which structures will be exposed during their lifetimes, and predicting their response to this ground motion, including potential damage, is an essential step for the appropriate design and retrofit of earthquake-resistant built infrastructure. Performance-based earthquake engineering methodologies are motivated by the need for scientific and transparent methods to relate seismic hazard to structural performance and loss. Forecasting damage and loss can also be of great use for emergency planning and management purposes. Visualization of damage in an urban region can aid policy makers and stakeholders in making informed decisions on how to reduce earthquake losses. The proposed simulation environment will use the Globus toolkit for access to the computational grid, the deployment of which NSF plans to support as part of the George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES). When developed, the simulation environment will be accessible to the earthquake engineering community through NEES.