报告题目：New Directions in Structural Health Monitoring
报告人：Khalid M. Mosalam （PE Taisei Professor of Civil Engineering Director, Pacific Earthquake Engineering Research (PEER) Center University of California, Berkeley）
报告摘要：This presentation covers several ongoing efforts of PEER research in several areas with emphasis on Structural Health Monitoring (SHM). First, a data-driven damage assessment is presented focused on using data from instrumented buildings to compute the values of damage features. Using machine learning algorithms, these damage features are used for rapid identification of the level and location of damage after earthquakes. Second, a vision-based damage assessment is discussed in relation to automated damage identification from images. Deep learning techniques are used to automatically conduct several identification tasks from images, examples of which are the structural component type, level of damage, and type of damage. The objective is to use crowdsourcing, allowing the general public (i.e., any person with a smartphone) to take photographs of damage and upload them to a server where damage can be automatically identified as a result of deep learning algorithms. Third, a sensor development for SHM with the objective of developing affordable sensor systems using microcontrollers. Examples of such developments include monitoring bridges and developing a laser-based sensor for settlement. The presentation concludes with PEER’s efforts in engaging the earthquake engineering community in these developments, an example of which is the PEER Hub Image Net (PHI-Net).