Composite Panel Damage Detection using Ultrasonic Testing and Neural Networks

For modern aircraft, wind turbines, and other aerospace structures composites are replacing metal components. These aerospace structures will be designed and expected to perform well for many years ahead. Therefore, structural health monitoring (SHM) systems are being designed to monitor these structures to improve maintenance, repairing or replacing composites components when the need arises. The focus of this research is on creating an SHM system, using ultrasonic testing methods to monitor damage in a composite panel. The ultrasonic testing consists of either actively scanning the panel by introducing controlled strain waves or passively listening to noises created by damage, which is a technique called acoustic emission. Through acoustic emission, composite coupons with damage introduced in the form of fiber breakage, delamination, and matrix cracking are examined by introducing tensile loading, since these forms of damage are present in aircraft and wind turbines through normal use of the systems. Artificial neural networks are used to analyze the signals received from the ultrasonic testing method. These networks allow for fast processing of the strain waves even with some extra noise present. Through this study, a system, using ultrasonic testing techniques, was shown to be a plausible SHM system for aerospace systems of the future.