AI-Based Bolt Loosening Diagnosis with Deep Learning from Laser Ultrasonic Wave Data

Deep learning (DL) algorithms for non-destructive evaluation (NDE) are becoming more popular, especially for the use of laser ultrasonic methods to identify and estimate looseness in bolted joints. By examining the connection between the energy lost by the passage of ultrasound waves and the bolt head-plate's genuine contact area, this work advances the field. The study used ultrasonic wave travel imaging to produce a 3D full-field sonar dataset utilizing an acoustic emission camera and a Q-switched Nd: YAG pulsating laser. The research also highlighted how crucial signal processing methods are to improving DL performance for automated looseness estimate.