Lamb Wave Flaw Classification in Al Plates Using Time Reversal and Deep Neural Networks

The interpretation of Lamb wave signals in real transducers needs careful excitation of Lamb wave mode and complex signal processing technique. In this study, the low mode antisymmetric Lamb wave is generated at one side and recorded at the other side, in a solid plate with different length and thickness. The time reversal technique is applied to Lamb wave signal processing and deep neural network is used to classify various Lamb wave FE signals. In industrial applications, the ultrasonic signals are not noise free. The white Gaussian noise is added to the FE simulation signals to augment the database. Without extracting the features of ultrasonic signal, the applied deep neural network to the ultrasonic noisy signals shows a high accuracy performance to classify the defect one and non-defect ones.