Research on non-destructive testing technology of rock bolt grouting compactness based on ultrasonic guided wave and deep learning

Abstract

In this paper, a new Deep Trans-Fourier Transform (DT-FT) method is proposed, which combines deep Q network (DQN), fast Fourier transform (FFT) and Transformer. It is applied to the non-destructive testing technology of rock bolt grouting based on ultrasonic guided wave and deep learning. There are many limitations in the traditional method of testing the density of bolt grouting, such as low accuracy and complicated operation. DT-FT method uses FFT to convert ultrasonic guided wave signals in frequency domain, and extracts rich frequency features. DQN is used to optimize the decision-making process and select the best feature combination to improve the detection accuracy. At the same time, Transformer is introduced to enhance the model's ability to process long sequence data and improve the accuracy and real-time detection. In the experimental part, we compared DT-FT with traditional detection techniques. The experimental results show that the DT-FT method has the advantages of high detection accuracy and strong adaptability, which fully verifies the feasibility and effectiveness of DT-FT non-destructive testing technology in the density of bolt grouting.