Data Driven Model Updating using Wireless Sensor Networks

In recent years, the structural engineering community has been actively exploring the use of wireless sensors in structural monitoring systems. A direct result of low-cost wireless sensors is wireless monitoring systems defined by high nodal densities offering extensive sets of structural response data. Such data sets can serve as a powerful tool for validation of structural design assumptions and formulation of performance-based design concepts. However, these benefits can only be gained if this data is analyzed. This paper explores the automated analysis of large sets of response data created by wireless sensor networks. Using the software modules and information technologies made available on the Network for Earthquake Engineering Simulation (NEES), structural response data collected with a wireless sensor network can be used to update analytical models commonly associated with the design and analysis of civil structures. For validation, a simple test structure is assembled in which wireless sensors are installed. Dynamic excitation and subsequent monitoring of the structure are performed using the NEES cyber-environment. Simulated annealing methods are employed to automate the updating of model parameters within OpenSees. This internet-based model updating framework can be further extended to serve as a scalable and fully automated structural health monitoring system.