Search

Computer aided Interpretation of Ultrasonic signals depicting flaws in weld pieces is depicted in this work. In this work, feasibility of Long Short Term Memory (LSTM) for flaw characterization is studied. Owing to the advantage of LSTM, the first technique involves training LSTM directly with the signals as inputs and testing its ability to characterize the flaws from the input signals. Due to wide variation in the length of input sequences, which introduced sparseness in other sequences, overall accuracy is affected. Hence in the second technique, LSTM are trained with features of the signals and it is found that the overall accuracy for test data is 67.64%. These features are statistical parameters obtained from the approximation co-efficient of the input signals. The input signals are decomposed with a novel wavelet template.

Ultrasonic Testing is the highly reliable Non Destructive Testing (NDT) technique for flaw detection in steel weldments. Ultrasonic test signals are analyzes to identify the nature of the defect. In this paper, Hilbert transform is used for decomposing the ultrasonic test signals into high and low frequency components (namely Intrinsic Mode Function). These components are characterised in terms of Power Spectral Density (PSD). An attempt has been successfully made to classify the flaws based on PSD. It is found that power spectral density of planar defects is higher than volumetric defects for the fourth Intrinsic Mode Function.

Non-destructive testing (NDT) techniques are most commonly employed for detection and characterization of flaws in the component. Apart from flaw characteristics, another parameter which is equally important to assess the structural integrity of engineering components is the material property. Nondestructive testing techniques offer several advantages over the conventional coupon-based techniques. Ultrasonic Testing is one of the widely used NDT techniques for material characterization. In this paper flaw detection such as lack of fusion and slag of welding carbonized steel and stainless steel material is characterized using discrete wavelet transform (DWT).In order to categorize the type of flaws, the given ultrasonic signal is subjected into DWT decomposition. Then, energy is calculated from DWT approximation coefficient. Finally corresponding welding flaw is determined using computed energy.

Ultrasonic Testing is a highly reliable Non-Destructive Testing Technique for weld defect characterization. Defects occur either high frequency components (Porosity, Sidewall crack) or as low frequency components (Root, Lack of Fusion, Lack of penetration, slag) in the UT signal. Manual interpretation of these signals is subjective in nature and is dependent on the expertise of the individual. Hence it is necessary to develop automated signal analysis system that classifies the defect. As defect classification is non-linear in nature, neural network based classification techniques are cited in literature. However neural network based techniques are computationally complex and has prediction error. Hence in this paper, an effective range based classification system using statistical moments is proposed. Performance of the proposed technique is measured in terms sensitivity and specificity.