Skip to main navigationSkip to main content
The University of Southampton
Engineering

Research project: Prediction of ultrasonic wave propagation in aircraft structures for crack monitoring - Dormant

Currently Active: 
No

Ultrasonic waves have long since been used to detect flaws in structures. One such technique applicable to cracks uses ultrasonic sensors to ‘listen’ for high frequency (~100kHz) acoustic emissions which are generated when micro-cracks grow. We are developing computer models to simulate the signals that would arrive at sensors positioned remotely from a crack as a result of wave propagation in the structure. Our work is focussing on critical loadbearing substructures of aircraft, such as wing spars, in which many complicated waves can co-exist. Our intention is to use these models to help quantify the accuracy and robustness of a commercially available monitoring system developed by Ultra Electronics.

During growth a crack will emit numerous pulses that cause waves to travel througout the structure. These can be picked up by an array of ultrasonic sensors at various distances from the crack. The electrical signals generated by the sensors encapsulate different arrival times from which the crack location can be identified. However, many different types of waves can travel in 'thick' structures such as wing spars at ultrasonic frequencies, and they travel at varying speeds. This complicates the inference of crack location from time of arrival information.

Courtesy of Ultra Electronics Control
Acoustic emission monitoring system
Image of sketch
Sketch of wing spar
Exaggerated motion of the spar when a particular wave travels along it
Spar deformation at high frequency

We have developed Semi-Analytical Finite Element (SAFE) computer models to predict how (and which) waves propagate in critical loadbearing substructures of aircraft, such as wing spars, at ultrasonic frequencies. The model is able to simulate the response at any position on the structure due to an assumed excitation elsewhere. We are able to use the model to quantify the variability in inferred existence and position of a crack position due to many factors such as sensor type and placement, frequency, form of excitation, and scattering of waves at reflecting boundaries.

Related research groups

Dynamics Group
Signal Processing, Audio and Hearing Group
Share this research project Share this on Facebook Share this on Twitter Share this on Weibo
Privacy Settings