An Acoustic Emission and Artificial Intelligence Approach to Structural Health Monitoring for Aerospace Application

In the area of aerospace and other applications, structural health monitoring (SHM) has been a significant and growing area of research in recent years. Throughout the operational life of aerospace structures, various damage scenarios may manifest, and it is of great concern to the aerospace community to develop methodologies for detecting and assessing these damage scenarios. In this paper, fundamental research on the use of the acoustic emission (AE) approach to SHM for fatigue crack growth is presented. In general, the AE approach to SHM and non-destructive evaluation (NDE) involves the sensing of ultrasonic Lamb waves propagating through a structure. Piezoelectric wafer active sensors (PWAS) have proven to be an effective tool in sensing these ultrasonic Lamb waves. The goal of this research was to conduct fundamental investigations into the use of PWAS for AE sensing of fatigued aerospace-grade aluminum 2024-T3 and the use of artificial intelligence approaches for AE signal classification efforts. The signal classification efforts presented in this thesis involve: (i) locating the source of the acoustic emission (source localization); (ii) determining whether an AE signal sensed is crack-related or noise; (iii) determining the crack length from which an AE originates. Ultimately, it is hypothesized and desired that the techniques developed in this paper and similar literature may be applied to production efforts of aerospace structures to identify and locate damage, optimize aircraft maintenance efforts, and prevent disastrous failure