Outlier analysis of nonlinear solitary waves for health monitoring applications

Abstract

The structural health monitoring/nondestructive evaluation method based on the generation and detection of highly nonlinear solitary waves is emerging as a cost-effective technique to monitor or inspect a variety of structures and materials. These waves possess unique characteristics not seen in conventional ultrasounds. Outlier analysis is a statistic tool able to identify anomalies in data that diverge from a set of baseline data. Although outlier analysis has received considerable attention for defect detection using modal data, guided ultrasonic waves, or other nondestructive approaches, its application for the analysis of solitary waves has never been explored. In the study presented in this article, the use of outlier analysis in terms of discordancy test and Mahalanobis squared distance was investigated to enhance the damage detection capability of a monitoring system based on highly nonlinear solitary waves. Two experiments were performed to demonstrate the procedure. In the first experiment, a thick steel plate was probed with a solitary wave transducer placed above the plate, and damage was simulated in terms of a foreign object magnetically attached to the bottom of the plate, at different distances from the transducer. In the second experiment, two aluminum plates were placed above each other in dry contact with the top plate subjected to localized, mostly hidden, defects. The transducer used in the first experiment was in this second test encased in a small cart with wheels to scan the sample at discrete positions. For both experiments, a few features were extracted from the time waveforms and fed to a univariate and a multivariate analysis that compared the testing data to a set of baseline data. The results show that the outlier analysis significantly improves the ability to detect damage using solitary waves.