227. STATISTICALLY-BASED DAMAGE DETECTION IN GEOMETRICALLY-COMPLEX STRUCTURES USING ULTRASONIC INTERROGATION

Department: Structural Engineering
Faculty Advisor(s): Michael Todd

Primary Student
Name: Colin M Haynes
Email: cmhaynes@ucsd.edu
Phone: 858-534-6560
Grad Year: 2014

Abstract
This poster summarizes a novel approach for the detection of damage in geometrically-complex structures. Two sparse networks of PZT sensor-actuators are bonded to a section of an aircraft fuselage rib and a three-story bolted frame structure, respectively. Two different forms of damage are implemented on each testbed, and ultrasonic guided wave scattering is used to inspect the structure for changes from the baseline condition. In order to determine the detector that provides the best performance, a receiver operating characteristic (ROC) data-driven approach and a theory-based approach are implemented. Several sensor fusion strategies are implemented, and the resulting detection performance gains are quantified.

Related Links:

  1. http://shm.ucsd.edu/Site/Research_Activities/Projects/Truss_UGWSHM.html
  2. http://shm.ucsd.edu/Site/Profiles/Entries/2010/12/26_Colin_Haynes.html

« Back to Posters or Search Results