Dynamic evolution investigation on crystal size and morphology of sodium sulfate based on a combination of ultrasonic and image on-line testing

Near-Surface Defects Identification of Polyethylene Pipes Based on Synchro-Squeezing Transform and Deep Learning

To conduct the ultrasonic weld inspection of polyethylene pipes, it is necessary to use low-frequency transducers due to the high sound energy attenuation of polyethylene. However, one of the challenges in this process is that the blind zone of the ultrasonic transducer may cover a part of the workpiece being tested. This leads to a situation where if a defect appears near the surface of the workpiece, its signal will be buried by the blind zone signal. This hinders the early identification of defects, which is not favorable in such a scenario. To address this issue, we propose a new approach to detect and locate the near-surface defects. We begin by performing a synchro-squeezing transform on the original A-scan signal to obtain an accurate time-frequency distribution. While successful in detecting and localizing near-surface defects, the method alone fails to identify the specific type of defect directly: a limitation shared with other signal processing methods. Thus, an effective and lightweight defect identification model was established that combines depth-wise separable convolution and an attention mechanism. Finally, the performance of the proposed model was compared and visually analyzed with other models. This paper successfully achieves the detection, localization, and identification of near-surface defects through the synchro-squeezing transform and the defect identification model. The results show that our model can identify both general and near-surface defects with an accuracy of 99.50% while having a model size of only 1.14 MB.

Ultrasonic resonance-based inspection of ultra-thin nickel sheets bonded to silicone

Abstract In the field of non-destructive testing (NDT), The detection of bonding defects between ultra-thin metal and silica gel is a difficult problem. In this study, In this study, ultrasonic resonance method was used to evaluate the bonding strength of ultra-thin metal to silica gel bonding structure. The composite parts of ultra-thin nickel sheet and silicon sheet with three different bonding states were studied. The bonding state of nickel sheet and silica gel is different, and the absorption of ultrasound is different. Using the resonance generated by high-frequency ultrasound in ultra-thin nickel sheet, the acoustic attenuation of the combination of ultra-thin nickel sheet and silicon rubber sheet was analyzed by resonance signal, and the bonding state between ultra-thin nickel sheet and silicon rubber sheet was characterized by bonding coefficient. Through experimental comparison, the results showed that the attenuation of ultrasonic signal in the nickel sheet and silicon film with different adhesive states characterize the adhesive state of ultra-thin nickel sheet and silicon film by the bonding coefficient, the bonding coefficient of good parts, weak adhesive parts and debonded parts is reduced successively. By setting an appropriate determination threshold value, the bonding state between the ultra-thin nickel sheet and the silicon film can be accurately determined according to the bonding coefficient obtained by detection.

Research on Classification and Recognition of Industrial Stainless Steel Welded Pipe Defects Based on Convolution Neural Network

Abstract For the classification and identification of industrial stainless steel welded pipe defects, a combined method of STFT and CNN based on vortex test is proposed. First, the collected original eddy current signal is STFT transformed to obtain a two-dimensional time-frequency map. Then, the two-dimensional time-frequency map was input into the two neural networks VGG-16 and GoogLeNet for model training, selecting a more accurate network model under the condition of the same learning rate. The trained network model is then classified using different learning rates. The results show that with the learning rate of 0.0001, the VGG-16 training model is better than the GoogLeNet training model, which has a certain reference significance for the classification and identification of defects in industrial stainless steel welded pipes.

Semantic segmentation of surface defects of smooth parts based on deep convolutional neural networks

Machine vision plays an increasingly important role in industrial product quality detection. During processing, scratches, dents and other defects are inevitable on the surface of a smooth part. Although surface defects do not affect the overall performance of the product, their existence is unacceptable when a perfect product is required. The surface defect detection method based on machine vision and deep convolutional neural networks overcomes, to a certain extent, the problem of low detection efficiency, high false detection and missing detection rates in the traditional detection method. In this paper, a multistream semantic segmentation neural network is proposed to identify defects on smooth parts. Taking a seatbelt buckle as an example, the scratch and crush defects on the surface are classified. The network takes DeepLabV3+ as the framework and three types of image stream as the input of the network. In the backbone feature extraction network, the Xception structure is improved to MobilenetV2 and the convolutional block attention module (CBAM) is introduced into the decoding network, which improves the operational efficiency and accuracy. Compared with other classical networks, this network demonstrates good performance in the image dataset of the seatbelt buckle and realises fast and accurate semantic segmentation and classification of surface defects. The evaluation results of the network model have been significantly improved.

Study on matching layer of ultrasonic transducer for defect detection of stainless steel

Abstract To solve the impedance mismatch problem between the piezoelectric wafer and the tested material when ultrasonic testing is used to detect the internal defects of materials, alumina powder and epoxy resin are compounded to prepare the matching layer material. Then the matching layer of the ultrasonic transducer is designed according to the quarter wavelength theory. The pulse-echo signal of the developed matching layer is tested and analyzed. The results show that the sound velocity of the matching layer is consistent with the changing trend of the acoustic impedance. That is, the acoustic impedance increases with the increase of the alumina mass fraction, and the density is 1080.73kg/m³. The ultrasonic transducer was designed by matching layer with a sound velocity of 1059.49m/s, with an acoustic impedance of 1.14mryal and acoustic attenuation coefficient of 0.1279dB/mm has good bandwidth and sensitivity.

Eddy current detection of hollowing defect in building exterior wall insulation layer based on RBF neural network

A Lightweight Network for Real-Time Target Semantic Segmentation Based on Dual Paths

Feasibility of Nonlinear Ultrasonic Method to Characterize the Aging Degree of Polyethylene Pipes

Detection and Analysis of Metal Induction Hardened Layer Based on Ultrasonic Technology

Abstract The quenched SAE 1552 steel has excellent mechanical properties. The depth of hardened layer is closely related to its mechanical properties, so it is of great significance to detect the depth of hardened layer. By using ultrasonic testing technology, the relationship between each ultrasonic parameter and quenchable material with different thickness is established. Through short-time Fourier transform, the ultrasonic signal is analyzed. It is observed that the time-domain waveform of steel with different hardened layer thickness is very different, and the ultrasonic frequency and signal amplitude are also different. This feature can be used as an important theoretical method of ultrasonic testing, which proves that the ultrasonic testing method can accurately detect the hardened metal layer.

Research on on-line ultrasonic testing of small diameter thin wall stainless steel straight welded pipe

Abstract In the production process of thin-walled stainless steel longitudinal welded pipe, there are many defects, such as cracks, porosity, partial welding, incomplete penetration, incomplete fusion. In this study, a local coupling water tank and six ultrasonic linear focusing probes with a center frequency of 5MHz were used to realize the defect identification of the straight welded pipe in the twisted state through the combined detection of the ultrasonic shear wave, longitudinal wave and creeping wave, which can meet the on-line inspection requirements of small-diameter thin-walled stainless steel straight welded pipe.

Measurement of hydraulic oil pressure in pipeline based on Short-time Fourier method