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

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