Strain Response Measurement-Based Optimal Damage Identification of Civil Structures

本项目的目标是为建筑和桥梁结构中使用的复杂和大型结构系统制定一种有效的损伤识别方法。迫切需要一种方法，能够可靠和有效地检测结构损坏，而无需进行广泛的人工检查，以监测民用基础设施的健康状况。现有的方法侧重于利用容易测量但损伤敏感性较低的线性结构加速度响应，但它们对于损伤识别还不够可靠。该项目将侧重于利用对损伤更敏感和可直接测量的动态应变响应来开发一种可靠的损伤识别方法。这项研究涉及在一个新的系统识别概念内综合应变和加速度传感器数据，以解决国家和国际关注的一个重要问题。该项目还将制定确定传感器最佳位置的方法，为实现最有效的损害鉴定目标提供最佳数据。该方法将通过利用选定位置的高灵敏度应变测量来最大化Fisher信息矩阵的适当范数，以达到信息论中经常使用的最低Cramer-Rao界限。为了验证新的损伤识别方案，将使用针对真实建筑物和桥梁模型获得的模拟和实验室响应数据。这种基于应变测量的健康监测方法的主要特点是对测量误差的稳健性、分析效率和准确性，可直接应用于民用基础设施的关键部件。本研究将为现代建筑和桥梁中使用的复杂和大型结构系统的有效损伤识别提供一个具有挑战性的问题的解决方案。该项目将通过对学生的高级培训来加强国家的教育和研究基础设施。它将为高效、经济的土木结构审问和健康监测提供急需的技术。项目成果将通过在专业会议和会议上的技术陈述、专门的互联网站、在同行评议的技术期刊上发表以及课堂教学等方式向从业人员和研究人员提供。

The objective of this project is to formulate an effective damage identification approach for complex and large structural systems that are used in buildings and bridge structures. The need for an approach that can reliably and efficiently detect structural damage without extensive manual inspections to monitor the health of civil infrastructure is urgent. The existing methods have focused on the utilization of easily measured but less damage sensitive linear structural acceleration responses but they are not yet adequately reliable for damage identification. This project will focus on the utilization of more damage sensitive and directly measurable dynamic strain responses to develop an approach for reliable damage identification. The study involves a synthesis of strain and acceleration sensor data within a new system identification concept to solve an important problem of national and international interest. The project will also develop methodologies to identify optimal locations of sensors that provide the best data to achieve the most effective damage identification objectives. The methodology will maximize a suitable norm of the Fisher information matrix through the utilization of highly sensitive strain measurements at selected locations to achieve the lowest Cramer-Rao bound often used in the information theory. For the validation of new damage identification schemes, the simulated and laboratory response data obtained for realistic models of buildings and bridges will be used. The main feature of this strain measurement-based health monitoring approach will be its robustness to the measurement errors, analytical efficiency and accuracy for direct implementation on the critical components of civil infrastructure.This study will provide a solution of a challenging problem for effective damage identification of complex and large structural systems used in modern buildings and bridges. The project will enhance the nation's educational and research infrastructure by advanced training of students. It will provide a much needed technology for efficient and economical interrogation and health monitoring of civil structures. The project results will be made available to practitioners and researchers through technical presentations in professional meetings and conferences, dedicated Internet sites, publication in peer reviewed technical journals, and classroom teaching.