Pipeline leakage aperture identification method based on pseudolabel learning

Abstract

Aiming at the problem of insufficient label data in the pipeline leak detection field, this paper proposes a pseudolabel adaptive learning method (PLALM) based on multiscale convolutional neural network (MSCNN) with the idea of transfer learning for pipeline leak aperture identification. First, the convolutional and pooling layers for transfer learning feature extraction are improved by using a dual-channel MSCNN. Second, the KL divergence function after dimensionality reduction is used to calculate the distribution distance between the source domain and the target domain to improve the robustness of distribution alignment in high-noise environments. In addition, considering the interference of pseudo-label noise, this paper develops a pseudolabel(PL) dynamic threshold to achieve the purpose of PL adaptive updating. Compared with the fixed threshold, the improved pseudolabel learning(PLL) can effectively improve the prediction accuracy of the model. The effectiveness of the method proposed in this paper is verified by predicting pipeline leakage conditions at different distances and under different pressures. The comparative analysis results show that the method in this paper is superior to other transfer learning methods in terms of prediction accuracy, stability, and convergence speed.

Acoustic-based approach for micro-leakage detection and localization in water supply pipelines

“…Acoustic leak detection techniques, which utilize the propagation of sound waves in pipe walls, offer numerous benefits, such as rapid response time, cost-effectiveness in maintenance, and long-term applicability. 13 However, there are still some issues that require attention and resolution to further enhance their effectiveness. Although continuous and automatic monitoring of pipelines is achievable, accurately pinpointing leaks remains a present challenge.…”

AI-enabled industrial equipment monitoring, diagnosis and health management

“…Chen et al [10] conducted a comparison of 26 classical deep learning models to detect eight types of leather surface defects, with the aim of laying the groundwork for the development of innovative inspection strategies. In Yuan et al [11], a pseudo-label adaptive learning approach utilizing a multiscale convolutional neural network (CNN) was introduced for identifying pipeline leak apertures, addressing the challenge posed by limited labeled data. Li et al [12] introduced the transfer multiscale adaptive CNN (TMACNN) for diagnosing bearing faults with limited labeled training data and diverse operational conditions.…”