In Situ Monitoring and Defect Diagnosis Method Based on Synchronous Compression Short-Time Fourier Transform and K-Singular Value Decomposition for Al-Carbon Fiber-Reinforced Thermoplastic Friction Stir Lap Welding

Lightweight is one of the key technologies and research hotspots in the fields of aerospace and railway, which makes the demand for the joining of dissimilar materials such as Al-based and carbon fiber-based (Al-CFRTP) materials. In friction stir lap welding (FSLW), weld defects are easily generated due to differences in physical and chemical properties. In this paper, an in situ monitoring and defect diagnosis method for FSLW of Al-CFRTP is proposed. In this method, time–frequency features are extracted by synchronous compression short-time Fourier transform (SSTFT). The welding status is classified by the K-singular value decomposition (KSVD), including two welding states and three welding defects. From the SSTFT results, the frequency band shifts from 23 to 10 kHz accompanied by the nugget collapse defects and shifts from 25 to 17 kHz accompanied by the flash and surface galling defects. From the KSVD results, the recognition accuracy of defects reaches 90% based on KSVD prediction model. The computation speed is 100 times faster than neural networks with nearly same recognition precision. The superiority of this method is the precise characterization description of FSLW defects from time–frequency domain features based on SSTFT and the fast processing speed in the classification by KSVD.