EECS Summer 2017 Seminar

Joel Harley
Sparse Signal Processing and Machine Learning for Smart Structures and Materials
University of Utah
Glennan 313
11:30a - 12:30p
July 7, 2017

Structural health monitoring systems are designed to achieve a real-time assessment of the presence, location, extent, and characteristics of damage in structures and materials. Over the last decade, guided wave structural health monitoring systems have become a dominant tool for these purposes. Guided waves can travel throughout an entire structure from a single location and are sensitive to many types of damage. However, current guided wave structural health monitoring tools have several disadvantages. First, most methods are only applicable for simple geometric structures, such as large, rectangular plates. Second, many guided wave characteristics (such as their velocity and frequency domain properties) are often not accurately known and are essential for analysis. Third, most structural health monitoring methods are highly sensitive to environmental variations, such as temperature. These variations distort data and hide changes due to progressive damage.

This presentation discusses signal processing and machine learning methods to address these three challenges of complexity, uncertainty, and variability. We utilize sparse coding and dictionary learning methods to address complexity and uncertainty as well as dynamic mapping methods to address variability. These methods integrate accurate data-driven physical models of guided waves into our analysis to improve damage detection and localization. We demonstrate these methods on several experimental data sets of varying complexity and compare our approaches to more traditional guided wave analysis techniques and show significant improvements in performance.


Joel B. Harley is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of Utah, Salt Lake City, UT. His interests include the integration of complex wave propagation models with novel signal processing, machine learning, and data science methods for applications in smart infrastructure monitoring.

Dr. Harley is a recipient of a 2017 Air Force Young Investigator Award, a 2014 Carnegie Mellon A. G. Jordan Award (for academic excellence and exceptional service to the community), the 2009 National Defense Science and Engineering Graduate (NDSEG) Fellowship, the 2009 National Science Foundation (NSF) Graduate Research Fellowship, the 2009 Department of Homeland Security Graduate Fellowship (declined), and the 2008 Lamme/Westinghouse Electrical and Computer Engineering Graduate Fellowship. He has published more than 40 technical journal and conference papers, including four best student papers. He is a student representative advisor for the IEEE Ultrasonics, Ferroelectrics, and Frequency Control Society, the president of the Utah chapter of the IEEE Signal Processing Society, a member of the IEEE Signal Processing Society, and a member of the Acoustical Society of America.