As the ocean engineering structure serves in the critical underwater environment, a variety of defects, such as cracks and corrosions, always cause damage to the structure. It is still a key challenge to identify and evaluate these defects accurately in the underwater environment. In this article, the visual and intelligent identification methods are presented for the inspection of defects in underwater structures using the alternating current field measurement (ACFM) technique. The current perturbation theory is developed to analyze the disturbed current field and the distorted magnetic field caused by defects. The gradient imaging algorithm is presented as an image preprocessing method to highlight the visual morphology of defects. The underwater intelligent ACFM system is set up. The experiments are carried out to verify the gradient imaging algorithm. The convolutional neural network (CNN) deep learning algorithm is presented to identify the grey-scale map samples preprocessed by the gradient imaging algorithm. The results show that the current perturbation theory clarifies the relationship between the characteristic signal and the morphology of various defects. The Bz image reflects the surface morphology of defects. The gradient imaging algorithm can achieve visual detection of defects. The single crack, the irregular crack, and the corrosion can be identified intelligently by the CNN deep learning algorithm. These defects can be evaluated accurately after classification.
31 January 2023 in IEEE Transactions on Industrial Electronics
The decay of pulsed eddy currents (PEC) with depth in diffusion decreases resultant signal changes, thus bringing a challenge for reliable and accurate evaluation of deep subsurface defects. In this article, a novel cross-correlation inspired residual network (ResNet), termed CCResNet, is proposed to improve the capability for smart evaluation of subsurface defects. It consists of a cross-correlation layer, a ResNet, and a novel loss function, namely, focal-probability of detection (Focal-POD) loss. The customized Gaussian wavelet basis enables us to derive weak features from heavily noised PEC signals due to the similarity by cross-correlation operation, which is the origin of the constructed cross-correlation layer. Then, a Focal-POD loss is proposed to address class imbalance and endow CCResNet with powerful capability for detection of deep subsurface defects by increasing their loss values. Finally, a semisupervised framework is built to retrain CCResNet using pseudo and labeled dataset to obtain classified results as imaging features. The experimental results show that the developed CCResNet is featured as better imaging resolution, more accurate evaluation, and intelligence in detection of deeper subsurface defects.
28 October 2022 in IEEE Transactions on Industrial Informatics
The alternating current field measurement (ACFM) technique has been widely used in the defect detection of metal structures. However, the identification and reconstruction of defects depend on human experience or simple empirical formulas, which leads to misjudgment of defects and large quantization errors. In this article, we propose an end-to-end physics-informed neural network for defect identification and 3-D reconstruction. The high-precision automatic detection system with a specially designed probe is established to detect defects in any direction. The faster RCNN network is used to identify and classify defects. The physics-informed Pix2Pix network is constructed to realize 3-D reconstruction of defects with different types. The results show that the established end-to-end physics-informed neural network can realize the identification and 3-D reconstruction of defects in which the mean average precision is 0.9982, the average length error of cracks is 0.9249 mm, the average depth error of cracks is 0.3402 mm, the average volume error of corrosion is 0.0667, and the average maximum depth error of corrosion is 0.3464 mm.
As an important connection method between drill pipes, the thread is prone to fatigue cracks and stress corrosion cracks due to long-term exposure to complex forces and corrosive environments. The thread crack detection system of drill pipe based on altering current field measurement (ACFM) is proposed. The crack detection model of the drill pipe thread using ACFM is developed by the finite element software COMSOL. The distribution of the induced magnetic field above the thread is extracted. The relationships between the characteristic signal and the depth and angle of the crack are analyzed. The drill pipe thread detection probes and the detection system are established, and different types of cracks on two different specifications of threads are detected. The experimental results show that the detection system can achieve full coverage detection of the thread crest and root, and cracks at different angle can be identified effectively on the thread.
The circumferential 0-mode shear horizontal (CSH 0 ) guided waves have great advantages for the detection of cracks in pipelines due to the small attenuation of acoustic signals. However, the CSH 0 guided wave detection methods are confused by the signal mixing, low signal-to-noise ratio and poor quantitative accuracy of cracks. In this paper, the novel detection method which includes three circumferential uniformly distributed probes are proposed for quantitative detection of pipeline cracks. The signals received by two receiving probes constitute a couple of differential signals. A full physical field coupled finite element model is developed to study the effect of CSH guided wave-defect interactions. The differential signal enhancement algorithm can separate the crack signal from the overlapped signal, which greatly improves the accuracy of crack location. The characteristic coefficient method is introduced to improve the quantitative accuracy of the crack sizes. The localization and quantification methods are verified through experiments. The results show that the defect localization method proposed in this paper can improve the signal-to-noise ratio of the crack signal from -15 dB to 2.4 dB. The localization error of crack is less than 6.75%. The quantitative error of the crack sizes is less than 8.75% based on the characteristic coefficient method.
Capacitive Imaging (CI) sensors are capable of non-destructively detecting both surface and hidden defects in dielectric materials and characterizing conducting surfaces through a relatively thick insulation layer. However, the complex Measurement Sensitivity Distribution (MSD) of CI sensors render the sensor capacitance variation with lift-off highly non-linear, which may lead to misinterpretation of defect indications. This work systematically studied the lift-off effect using both Finite Element (FE) analysis and experimental approaches. Sensor MSD was used as a tool to predict the imaging performance. Normalized Variation Ratio (NVR) was introduced and used to characterise sensor responses due to defects for a CI sensor. Both the FE analysis and experiments suggest that the lift-off effect for a CI sensor is specimen type and condition dependent. For a given defect, the NVR may vary non-monotonically with increased lift-offs. A case study on a glass-fibre composite/aluminium hybrid structure with multiple artificial defects demonstrated the feasibility of defects discrimination using multiple CI scans with increased lift-offs.