Date of Award


Document Type

Open Access Master's Report

Degree Name

Master of Science in Civil Engineering (MS)

Administrative Home Department

Department of Civil and Environmental Engineering

Advisor 1

Raymond A. Swartz

Committee Member 1

Bo Chen

Committee Member 2

William Bulleit


Fatigue damage is an important concern in structural steel connections where frequent load reversals are expected. Detection of fatigue damage is typically done with infrequent visual inspections that are subjective and limited to surface features. A permanent embedded structural health monitoring (SHM) system could be helpful in detecting damage as it occurs. Current methods for detecting fatigue damage include inference of fatigue life expended through cycle counting from long-term strain measurement campaigns, and through short-term impedance measurements. The first approach has the advantage that it is relatively simple from an algorithmic point of view, but it is an indirect measure of damage, and it does require that strain gauges be present and operational to measure the entire strain history of the component in question. Impedance measurements from piezoelectric transducers (PZTs) bonded to the surface of the specimen can detect damage directly and do not require the use of historical data (though baseline health impedance signatures are needed for reference), but the processing of impedance data in this application can be difficult and subjective. In addition, many studies focused on impedance measurements for damage detection using overly simplified coupon geometries for experimental validation that do not capture the full complexity of a structural steel connection. In this study, an acoustic wave propagation method is proposed to detect fatigue damage in a bolted seated connection. A PZT located on the connected column measures the energy that propagates through the connection. Only the top angle for the connection (typically used for stability) is damaged in the study to allow multiple tests to be made with minimal specimen preparation. Signal processing methods including vii matched filter to separate the input signal from the signal distortions are used to improve the sensitivity of the approach. Features examined include energy transmitted through the connection and pole information associated with the signal residuals (error information) with only the latter being sensitive to fatigue damage.