Laser-ultrasonic technology based on adaptive VMD and spatio-temporal wave-field domains to detect defects in titanium alloys under high-temperature conditions

Abstract

Titanium alloys are widely used in key structural components in aerospace and other industries, owing to their excellent performance. However, detecting damages in high-temperature environments is difficult, and the signal-to-noise ratio of ultrasonic signals is also significantly low. This poses a serious challenge for traditional non-destructive testing methods to accurately identify and characterize internal defects. In this work, a novel high-temperature laser-ultrasonic defect detection scheme is proposed. The adaptive variational mode decomposition (VMD) method introduces permutation entropy as the fitness function in the particle-swarm-optimization algorithm to identify the optimal penalty factor and number of modal decompositions, thereby achieving effective optimal decomposition and noise reduction of ultrasonic signals in high-temperature environments. Furthermore, by establishing a mapping relationship between the surface wave amplitude and surface defect depth, as well as the internal defect diameter, quantitative identification and detection of defects in titanium alloy materials is achieved. The results show that damage characterization of the titanium alloy is realized at 800 °C. The adaptive-VMD processed signal exhibits an average signal-to-noise ratio enhancement of 2.453 dB, maximum magnitude reduction of 0.177 %, average waveform correlation coefficient of 0.907, and average alignment entropy of 0.693. The method proposed in this paper has significant implications for time-domain signal processing and high-temperature laser ultrasonic defect detection.