Research on Oil and Gas Pipeline Leakage Detection Based on MSCNN-Transformer
Source abstract
Abstract
The leakage detection of oil and gas is very important for the safe operation of pipelines. The existing working condition recognition methods have limitations in processing and capturing complex multi-category leakage signal characteristics. In order to improve the accuracy of oil and gas pipeline leakage detection, a multi-scale convolutional neural network-Transformer (MSCNN-Transformer)-based oil and gas pipeline leakage condition recognition method is proposed. Firstly, in order to capture the global information and nonlinear characteristics of the time series signal, STFT is used to generate the time-frequency image. Furthermore, in order to enrich the feature information from different dimensions, the one-dimensional signal and the two-dimensional time-frequency image are sampled by multi-scale convolution, and the global relationship is established by combining the multi-head attention mechanism of the Transformer module. Finally, the leakage signal is accurately identified by fusing features and classifiers. The experimental results show that the proposed method shows high performance on the GPLA-12 data set, and the recognition accuracy is 96.02%. Compared with other leakage signal recognition methods, the proposed method has obvious advantages.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Water Systems and Optimization · Fire Detection and Safety Systems · Structural Integrity and Reliability Analysis
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
College of Mathematics and Statistics, Northeast Petroleum University, Daqing 163319, China
Artificial Intelligence Energy Research Institute, Northeast Petroleum University, Daqing 163319, China
School of Electrical Information Engineering, Northeast Petroleum University, Daqing 163319, China
Funding
the Heilongjiang Province Postdoctoral Science Foundation
References (34)
Show source-provided references
1. Review of China’s natural gas development in 2024 and outlook for 2025
2. Natural gas pipeline leak detection based on acoustic signal analysis and feature reconstruction10.1016/j.apenergy.2023.121975 ↗
3. Leakage aperture identification of natural gas pipeline based on compressed sensing combined with convolutional network
4. Identification of operating conditions of refined oil pipelines based on machine learning
5. Understanding and improving deep learning-based rolling bearing fault diagnosis with attention mechanism10.1016/j.sigpro.2019.03.019 ↗
6. Acoustic emission signal recognition of gas pipeline leakage based on one-dimensional convolutional neural network
7. Valve leakage ultrasonic signal recognition method based on improved CNN
8. Zhong, M.S., Li, Z.F., Xiong, X.Z., and Wang, X.G. (2015, January 28–30). A natural gas pipeline leak detection model based on one-dimensional multi-scale deep separable convolution. Proceedings of the 34th China Process Control Conference, Hangzhou, China.
9. One-dimensional convolutional and long short-term memory neural network method for pipeline leak detection
10. A novel method for bearing fault diagnosis based on BiLSTM neural networks10.1007/s00170-022-10792-1 ↗
11. Research on small-scale leakage detection of water supply pipelines based on VMD and deep learning
12. Research on small leakage detection technology of water supply pipe based on acoustic signal
13. Tang, Y.J., Liu, D., Xiao, Z.H., Hu, X., and Lai, X. (2021). Research on fault diagnosis method of hydropower units based on convolutional neural network and singular value decomposition. China Rural Water Conserv. Hydropower, 175–181.
14. A fault identification method for axle box bearings of high-speed EMUs based on improved VMD
15. Leakage signal identification method of DAS pipeline based on CNN-Transformer
16. Leakage fault diagnosis of heating pipe network based on convolutional neural network and Transformer
17. Research on DC microgrid fault diagnosis based on CNN-BiLSTM-Attention
18. Fault diagnosis of modular multilevel matrix converter based on CNN-GRU deep learning
19. Parallel multi-layer sensor fusion for pipe leak detection using multi-sensors and machine learning10.1016/j.engappai.2025.110923 ↗
20. Identification method of buried pipeline complex conditions based on TA-DBF model and information fusion technique10.1016/j.measurement.2025.118085 ↗
21. Fault diagnosis of wind turbine bearings based on BO-SDAE multi-source signals
22. Leak detection in pipelines based on a multilayer network of time–frequency domain cross-visibility graphs and graph convolutional networks10.1016/j.ymssp.2025.113250 ↗
23. Enhanced spectrum convolutional neural architecture: An intelligent leak detection method for gas pipeline10.1016/j.psep.2020.12.011 ↗
24. A Non-Metallic pipeline leak size recognition method based on CWT acoustic image transformation and CNN10.1016/j.apacoust.2024.110180 ↗
25. Fine-Grained Leakage Detection for Water Supply Pipelines Based on CNN and Selective State-Space Models10.3390/w17081115 ↗
26. Luo, Z.Y., Li, M.Z., and Dong, X. (2025). Effective diagnosis of rolling bearing fault diagnosis method of Vision Transformer network. J. Chongqing Univ. Posts Telecommun. (Nat. Sci. Ed.), 1–9.
27. A new method for natural gas pipeline leak detection based on gated attention network model
28. Leakage diagnosis of heating pipe network based on CABC-BP model
29. Classification and identification of leakage acoustic signals in water supply network
30. Valery, N., Fabien, M., and Pierre, T. (2022). Evaluation of the Leak Detection Performance of Distributed Kalman Filter Algorithms in WSN-Based Water Pipeline Monitoring of Plastic Pipes. Computation, 10.10.3390/computation10040055 ↗
31. Numerical Simulation of Non-Darcy Flow in Naturally Fractured Tight Gas Reservoirs for Enhanced Gas Recovery10.3390/gases4030015 ↗
32. Hamouda, A.A., Gupta, S., and Bahadori, A. (2017). Enhancing Oil Recovery from Chalk Reservoirs by a Low-Salinity Water Flooding Mechanism and Fluid/Rock Interactions. Energies, 10.10.3390/en10040576 ↗
33. Predicting the mechanical properties of polypropylene and recycled polypropylene by the application of sigmoid and ReLU functions in neural networks10.1016/j.rineng.2025.108459 ↗
34. Li, J., and Yao, L. (2021). GPLA-12: An Acoustic Signal Dataset of Gas Pipeline Leakage. arXiv.