No abstract

In this paper, a two-stage deep offline-to-online transfer learning framework (DOTLF) is proposed for longdistance pipeline leakage detection (PLD). At the offline training stage, a feature transfer-based long short-term memory network with regularization information (TL-LSTM-Ri) is developed where a maximum mean discrepancy regularization term is employed to extract domain-invariant features and an adjacentbias-corrected regularization term is introduced to extract early fault features from pipeline samples under different scenarios. At the online detection stage, the trained TL-LSTM-Ri is employed for motion prediction so as to monitor the operating condition of the pipeline in real time. To demonstrate its application potential, the DOTLF is successfully applied to handle the PLD problem on the long-distance oil-gas pipeline data. Experimental results demonstrate the effectiveness of the proposed DOTLF for realtime PLD under real-world scenarios.

No abstract

Data augmentation (DA) has the potential to address the issue of imbalanced and insufficient datasets (I&ID) in pipeline fault diagnosis. However, the majority of existing DA methods for time series are inspired by computer vision techniques, ignoring the temporal dynamic properties and finegrained fault features, which leads to limited performance of the augmentation. To tackle this problem, we introduce a novel DA approach called the subdomain-alignment adversarial selfattention network (SA-ASN), which takes into account both temporal association and semantic correlation. Our approach features a novel temporal association learning (TAL) mechanism, which transfers temporal information from the discriminator to the generator via a customized knowledge-sharing structure, improving the reliability of synthetic long-range associations. Additionally, we introduce a prototype-assisted subdomain alignment (PASA) strategy that forms a hierarchical structure in the synthetic dataset by incorporating local semantic correlation into the model training. With the support of TAL and PASA, our SA-ASN algorithm enhances the authenticity of temporal structure at the instance level and improves the discriminability of fault features at the category level. Our experimental results show that the SA-ASN algorithm provides a more diverse and accurate augmentation of pipeline data. The effectiveness of our SA-ASN algorithm encourages the use of data-driven diagnostic models in complex real-world oilfield pipeline networks.