Underwater pipelines are the channels for oil transportation in the sea. In the course of pipeline operation, leakage accidents occur from time to time for natural and man-made reasons which result in economic losses and environmental pollution. To avoid economic losses and environmental pollution, damage detection of underwater pipelines must be carried out. In this paper, based on the histogram of oriented gradient (HOG) and support vector machine (SVM), a non-contact ultrasonic imaging method is proposed to detect the shedding damage of the metal underwater pipeline external anti-corrosion layer. Firstly, the principle of acoustic scattering characteristics for detecting the metal underwater pipelines is introduced. Following this, a HOG+SVM image-extracting algorithm is used to extract the pipeline area from the underwater ultrasonic image. According to the difference of mean gray value in the horizontal direction of the pipeline project area, the shedding damage parts are identified. Subsequently, taking the metal underwater pipelines with three layers of polyethylene outer anti-corrosive coatings as the detection object, an Autonomous Surface Vehicle (ASV) for underwater pipelines defect detection is developed to verify the detection effect of the method. Finally, the underwater ultrasonic image which used to detect the metal underwater pipeline shedding damage is obtained by acoustic sensor. The results show that the shedding damage can be detected by the proposed method. With the increase of shedding damage width, the effect of pipeline defect location detection is better.

Optimal sensor placement is a significant task for structural health monitoring (SHM). In this paper, an SHM system is designed which can recognize the different impact location and impact degree in the composite plate. Firstly, the finite element method is used to simulate the impact, extracting numerical signals of the structure, and the wavelet decomposition is used to extract the band energy. Meanwhile, principal component analysis (PCA) is used to reduce the dimensions of the vibration signal. Following this, the non-dominated sorting genetic algorithm (NSGA-II) is used to optimize the placement of sensors. Finally, the experimental system is established, and the Product-based Neural Network is used to recognize different impact categories. Three sets of experiments are carried out to verify the optimal results. When three sensors are applied, the average accuracy of the impact recognition is 59.14%; when the number of sensors is four, the average accuracy of impact recognition is 76.95%.

External vibration is the main disturbance condition in the practical monitoring of outdoor stranded structure using laser ultrasonic guided wave. It is difficult to extract and identify the real damage state under different vibration conditions due to the variation of the guided wave feature distribution. At present, there is no effective solution to this practical problem. In this paper, a new deep cross-domain adaptive semi-supervised damage identification method is proposed by using transfer learning method and combining with the actual demand of stranded guided wave monitoring. Firstly, a novel laser excitation-piezoelectric receiving sensing method is realized by taking full advantage of the non-contact characteristics, wide frequency band and high stability of the laser and piezoelectric sensors. Secondly, a multi-layer convolutional neural network is constructed to extract the damage features of the guided wave signals in the source domain and map them to the high-level hidden space. Then, a multi-core MMD method is designed to reduce the distribution difference of damage features between the target and source domain by using the optimal multi-core selection method, and the essential damage features of ultrasonic guided waves were learned. Finally, the different damage states of the target domain are effectively identified by feature identification. The experimental results illustrate that the proposed method can realize automatic extraction of inherent damage features and adaptive matching of multi-layer features, connect the source and target domain in the high-level feature space, and learn the invariant features of guided wave signals under different vibrations. Moreover, the proposed method has good performance both in the mean between-class average distance and the mean within-class average distance damage degree of feature under various vibration conditions, reaches 100% accuracy in damage degree identification under different vibration conditions, and shows better performance than the comparison methods.

Due to its good thermal conductivity and corrosion resistance, copper has become a common material for transmission pipelines. It is necessary to detect the early signs of damage in copper pipelines effectively and quickly. Laser ultrasound scanning is a non-contact and non-destructive damage identification method, which can realize high-precision, non-contact detection. At the same time, with the progress of internet technology, traditional damage testing has begun to use advanced technologies such as the internet of things and cloud computing to promote an upgrade of the testing industry from an offline industry to an online industry. However, obtaining a large amount of wavefield vibration data is time consuming. In this paper, we present a laser ultrasonic scanning cloud platform damage detection method for copper pipelines based on alternating learning blind compressive sensing (BCS) and the adjacent area difference coefficient (AADC); this approach can improve real-time performance and detection accuracy. First, the damage detection method is introduced in detail. BCS is used to compress the laser scanning signal at the data acquisition terminal, and then transmitted to a data processing cloud platform for reconstruction. Copper pipeline damage imaging is realized by taking the AADC value of each detection point as the pixel value. The simulated detection data of the copper pipeline are then obtained through a finite element model, and the weighted vectors of the AADC are determined by a genetic algorithm. Finally, experimental data are used to verify the effectiveness of this method, and the experimental results are analyzed and discussed. The AADC and other distance damage imaging methods are compared. The results demonstrate that this method can compress the wavefield data to 13% of the original data, and the detection of crack damage is realized.

Whole structure fast detection system suitable for multiple classes of materials is an engineering goal in structural health monitoring community. But mode selection caused by the difference of damage modes among different structures seriously affects the intelligence and automation of guided wave detection. In this paper, a scatter feature adaption intelligent detection method based on ultrasonic guided wave is proposed to monitor different structures according to deep feature generalization. Firstly, the deep convolutional neural network is constructed, and the multi-layer feature mapping relationship is constructed for both source domain and target domain signals. Secondly, the maximum mean difference(MMD) method was utilized to construct a quantitative difference evaluation mechanism for the guided wave scatter signals with different structures. At the same time, both the feature distance loss and the source-domain label loss are used to optimize the deep network model. Finally, the visual damage localization is realized by improved possibility imaging method according to the multi-sensor damage index. Experimental results show that the proposed method can realize the damage monitoring in another target domain by adapting the scatter signals in a single source domain and a single target domain, and has great advantages compared with the existing methods in such problems.