Defect identification by an ultrasonic cylindrical phased array

In this paper, a new method is presented for defects classification by ultrasonic cylindrical phased array. Firstly, a finite element model is conducted to simulate the defects identification by the cylindrical phased array transducer. A series of simulation are done for 4 types of defects with different sizes by a 64-element cylindrical phased transducer with the center frequency of 500 kHz. Then, the Wavelet-packet transform decompose algorithm is used to four-levers decompose, reconstruct and extract the feature of these echo signals. Finally, the reconstructed signals are used to the deep neural network to the defect classification. The accuracy of the known defects classification is 100%, which means the method is feasible for classification by ultrasonic cylindrical phased array.