Bolt Stress Detection Algorithm Based on Bayesian Compressive Sensing

Ultrasonic stress testing serves as an essential tool in the realm of non-destructive evaluation, facilitating the stress examination of precision instruments to ensure their optimal functionality. However, the online monitoring of bolts is a long-term process, resulting in the accumulation of extensive data sets that pose significant challenges for data storage and processing. This work introduces a novel algorithm for ultrasonic time-of-flight (ToF) extraction based on Bayesian compressed sensing, named the Bc-T algorithm, which reduces amount of ultrasonic data. The proposed method begins by segmenting the first echoes within the ultrasonic signal through a timed window function. Data are compressed using the Bayesian compressive sensing algorithm and restored as required for processing. It then extracts the corresponding ToF for the first echo using the cross-correlation algorithm. In the tensile test of an M36 bolt with a length of 1100 mm, the practical effectiveness of the algorithm proposed in this paper was validated by comparing it to the mainstream wavelet thresholding algorithm.