AQUADA PLUS: Automated damage inspection of cyclic-loaded large-scale composite structures using thermal imagery and computer vision

Abstract

This study develops and demonstrates a new method to locate and track multiple fatigue damages in large-scale composite structures subject to cyclic loads using thermography and computer vision. Progressive damages generate thermal features due to material friction, allowing efficient detection using passive thermography. Automated damage localization and evaluation are done by thermal image processing incorporated with thermodynamics principles. The proposed new method, AQUADA PLUS, is demonstrated on two composite wind turbine blades in which artificial defects are introduced to trigger damage growth when subject to cyclic loading. Multiple damage sites are located, tracked, and evaluated automatically despite the complex thermal background that changes considerably when the thermal videos are taken. A pioneering drone-based field test has been conducted to demonstrate the possibility of field application where complex environmental conditions and varying thermal backgrounds are present. The associated challenges are identified and the possible solutions are discussed for the further development of the proposed method toward real-world application. All original thermal videos presented in this study are shared with the public for future study.