Phased array ultrasonic test signal enhancement and classification using Empirical Wavelet Transform and Deep Convolution Neural Network

In the recent past, Non-Destructive Testing (NDT) has become the most popular technique due to its efficiency and accuracy without destroying the object and maintaining its original structure and gathering while examining external and internal welding defects. Generally, the NDT environment is harmful which is distinguished by huge volatile fields of electromagnetic, elevated radiation emission instability, and elevated heat. Therefore, a suitable NDT approach could be recognized and practiced. In this paper, a novel algorithm is proposed based on a Phased array ultrasonic test (PAUT) for NDT to attain the proper test attributes. In the proposed methodology, the carbon steel welding section is synthetically produced with various defects and tested using the PAUT method. The signals which are acquired from the PAUT device are having noise. The Adaptive Least Mean Square (ALMS) filter is proposed to filter PAUT signal to eliminate random noise and Gaussian noise. The ALMS filter is the combination of low pass filter (LPF), high pass filter (HPF), and bandpass filter (BPF). The time-domain PAUT signal is converted into a frequency-domain signal to extract more features by applying the Empirical Wavelet Transform (EWT) algorithm. In the frequency domain signal, first order and second order features extraction techniques are applied to extract various features for further classification. The Deep Learning methodology is proposed for the classification of PAUT signals. Based on the PAUT signal features, the Deep Convolution Neural Network (DCNN) is applied for further classification. The DCNN will classify the welding signal as to whether it is defective or non-defective. The Confusion Matrix (CM) is used for the estimation of measurement of performance of classification as calculating accuracy, sensitivity, and specificity. The experiments prove that the proposed methodology for PAUT testing for welding defect classification is obtained more accurately and efficiently across existing methodologies by providing numerical and graphical results.