Sponsor: ONR

PI: Salvatore Salamone

The increase in loads applied to the aging and deteriorating structures, such as aircrafts and vessels, their usage beyond the designed life, and the desire to reduce downtime associated with regular maintenance operations have all sparked interests and researches into SHM methods. Structural health monitoring methods based on low profile piezoelectric transducers, which have the capability of generating and receiving ultrasonic stress waves, and thus can triangulate to detect defects are among the most promising candidates. Although these novel SHM systems have seen significant developments, very few if any have been implemented in real structures. A main reason for this lack of acceptance is the potential for these systems to emit false positives: that is alarming operators and inspectors that a structural fault has occurred when it has not. One of the main culprits in false positives are the multiple reflections generated by structural features (e.g., joints, stiffeners, fasteners, etc.) and geometric boundaries, that if not properly removed often show up as damage. In addition, the inherent uncertainty in sensor measurements, caused not only by temperature variations and noise, but also from their multimode nature and dispersive behavior, may hamper their reliability in terms of automatic damage detection.

The objective of the proposed program is to investigate the hypothesis that multiple reflections generated by structural features (e.g., edges, joints, welds, fasteners, etc.) can be leveraged for the structural health monitoring (SHM) of waveguide-like structures which are ubiquitous in Navy structures. It is hypothesized that through the analysis of multiple echoes and reverberations present in recorded waveforms, damage diagnostics and characterization can be achieved while using fewer transducers than conventional techniques. This program will set the foundation for future multidisciplinary endeavors bridging SHM, probability theory and signal processing. It is envisioned that, the algorithms created in this project will have many opportunities for applications that include not only the structural diagnostic by ultrasonic stress waves measurements, but also other areas of acoustic research such as analysis of low level acoustic noise in large bodies of water, that could be relevant for other Naval applications.