Energy Harvesting for Detecting Hidden Damages

Abstract

The paper presents how energy harvesting devices can be deployed to assess and estimate damages that are hidden to difficult to reach in terms of being able to provide instrumentation. To demonstrate this, the paper uses piezoelectric harvesters and analyses its open circuit dynamic response from different location of a damaged beam. This provides deflection and mode shape information. These responses are subsequently analysed via wavelets to identify damage location and extent. Subsequently, using prior modal and dynamic parameter information and knowledge, the paper uses a Bayesian statistical Identification Framework to determine the probabilities of crack location and extent within an obscured part of the structure or where the spatial modal shape or spatial wavelet is unavailable. The use of Recursive Bayesian Estimation can implement this framework as well. The paper demonstrates how novel sensors and materials can, with appropriate probabilistic techniques, provide better inspection and damage detection of our built infrastructure.