Most current generative adversarial network (GAN) cannot simultaneously consider the quality and diversity of generated samples due to limited data and variable working condition. To solve the problem, a Transformer-based conditional GAN transfer learning network is proposed. Firstly, a transformer-based conditional GAN (TCGAN) generative network is constructed with sample label information, enhancing the quality of generated data while retaining the diversity of generated signals. Secondly, a transfer learning network based on TCGAN is established, and a "generation-transfer" collaborative training strategy based on the expectation maximization is introduced to realize parallel updating of the parameters of the generative network and the transfer network. Finally, the effectiveness of the proposed method is verified using bearing datasets from CWRU and the self-made KUST-SY. The results show that the proposed method can generate higher quality data than comparative methods such as TTS-GAN and CorGAN, which provides a new solution for improving the cross-domain fault diagnosis performance.

Titanium (Ti) is an ideal structural material whose use is gradually emerging in civil engineering. Regular defect evaluation is indispensable during the long-term use of Ti sheets, which can be performed effectively using eddy current (EC) imaging, a method of visualizing defects that is convenient for inspectors. However, as EC scan images contain abundant information and have discrepancies in terms of their quality, it is difficult to extract effective features, thus affecting the evaluation results. In this article, we propose a method that combines the EC imaging technology with a quantitative evaluation method for Ti sheet defects based on the deep belief network (DBN) and least squares support vector machine (LSSVM). A multilayer DBN is constructed to extract the effective features from EC scan images for Ti sheet defects. Based on the extracted feature vectors, a multi-objective regression model of defect dimensions is established using the LSSVM algorithm. Then, the dimensions of Ti sheet defects such as length, diameter, and depth are quantitatively evaluated by the effective features and the efficient regression model. The experimental results show that the evaluation errors for the defect lengths and depths tested are less than 3 and 5%, respectively, confirming the validity of the proposed method.

Siphons have been widely used in water supply systems and sewage networks. However, it is difficult to implement non-destructive testing due to structural complexity and limited accessibility. In this paper, a novel condition classification method for water-filled underground siphons is proposed, which uses the acoustic signals received from acoustic sensors installed in the siphon. The proposed method has the advantages of simpler operation, lower cost, and higher detection efficiency. The acoustic wave forms in the siphons reflect on the system characteristics. Seven typical conditions of a water-filled underground siphon were investigated, and a series of experiments were conducted. Acoustic signals were recorded and transformed into acoustic pressure responses for further analysis. The variational mode decomposition (VMD) and the acoustic energy flow density were used for signal processing and feature extraction. The acoustic energy flux density eigenvectors were input to three different classifiers to classify the siphon conditions. The results demonstrate that the proposed acoustic-based approach can effectively classify the blockage and damage conditions of siphons, and the recognition accuracy of the proposed method is higher than 94.4%. Therefore, this research has value for engineering applications.

Targeting the challenge of determining the degree of blockage in buried pipelines and the difficulty of effectively extracting blockage features, a blockage detection method integrating variational mode decomposition (VMD) and information gain is proposed. Acoustic impulse response signals were obtained by deconvolving the output signals of the system, which were then subjected to VMD to obtain 12 components in different frequency ranges. Next, information gain (IG) was introduced to characterize the 12 components quantitatively, through which the components containing rich information about the pipe conditions were selected out. Meanwhile, sound pressure level conversion was performed on the selected components to amplify any changes in the sound field. Finally, the root mean square entropy (RMSE) was calculated to constitute the feature eigenvectors, which were input into Random Forests (RF) classifier for defect identification of pipeline. As the experimental results demonstrate, the proposed method is capable of determining the degree of blockage effectively in the running state. Meanwhile, it can also eliminate the interference of functional parts such as lateral connections during the identification process, thereby improving the identification accuracy. The present study has shown both theoretical significance and application value in the field of defect detection and recognition.