Project description: The goal of vibration based structural health monitoring (SHM) is to replace costly inspection campaigns by real-time instrumentation and automated data processing in order to estimate the current state of health and the remaining lifetime of structures. For damage detection, automated data-based methods using adequate statistical techniques have been developed recently. This project deals with the next level of SHM which is damage localization and focuses on the problem of cracking in prestressed concrete. The challenge is to design a data-based damage localization system based on output-only ambient vibrations. The main ingredients of the proposed method are: (i) the use of very large strain sensor networks, (ii) the development of automated and efficient data processing methods based on local spatial filters or several layers of spatio-temporal filters, and (iii) alarm triggering techniques based on unsupervised learning methods. The major innovation in this project is the use of advanced computational techniques for the creation of a large data set of virtual campaigns which are used for the design and optimization of the SHM system. In addition to these virtual campaigns, experimental developments are also foreseen, with a specific emphasis on the evaluation of different techniques for dynamic strain measurements.

We have an immediate opening in the Department of Engineering Science and Mechanics for a postdoctoral research associate in multiscale modeling and computational mechanics. The focus of this position is on the modeling, simulation, and analysis of the nonlinear evolution of coupled macro/microstructure in failure processes, with an emphasis toward structural health monitoring. Coupled-field models using continuum damage variables will be used to study the fundamental spatiotemporal features of damage evolution, particularly regarding the nature and origin of unpredictability in damage processes. The candidate must have a solid general background in theoretical mechanics and dynamics, and be familiar with nonlinear finite element analysis