Application of GPR System With Convolutional Neural Network Algorithm Based on Attention Mechanism to Oil Pipeline Leakage Detection

被引:10
作者
Li, Jiadai [1 ]
Yang, Ding [1 ]
Guo, Cheng [1 ]
Ji, Chenggao [2 ]
Jin, Yangchao [1 ]
Sun, Haijiao [3 ]
Zhao, Qing [1 ]
机构
[1] Univ Elect Sci & Technol China, Sch Resources & Environm, Chengdu, Peoples R China
[2] CNPCLogging Co Ltd Tianjin Branch, Tianjin, Peoples R China
[3] SINOPEC NorthwestCompany, Inst Petr Engn Technol, Urumqi, Peoples R China
关键词
unconventional detection; convolutional neural network; ground-penetrating radar; oil pipeline leakage; attention mechanism;
D O I
10.3389/feart.2022.863730
中图分类号
P [天文学、地球科学];
学科分类号
07 ;
摘要
High-efficiency and high-quality detection of oil pipeline will significantly reduce environmental pollution and economic loss, so an unconventional oil pipeline anomaly detection convolutional neural network (CNN) algorithm based on attention mechanism is proposed in this article. By taking the simulated ground-penetrating radar (GPR) data as prior knowledge, the structure of the convolutional neural network based on the attention mechanism is constructed, and finally, the location and working condition of the underground oil pipeline are recognized in the simulation data and measured data. The simulation results show that after using the new optimized convolutional neural network, the accuracy rates of the leakage discrimination from horizontal data acquired along the oil pipeline and the classification of the target from longitudinal data acquired perpendicular to the oil pipeline are 94.5% and 84.6%, respectively. Compared with the original convolutional neural network without an attention mechanism, the accuracy rates of the leakage discrimination and the classification of the target are improved by 6.2% and 7.8%, respectively. We further train measured data with an optimized convolutional neural network, results show that compared with a conventional network, the new network can increase the corresponding accuracy rates of the leakage discrimination and the targets classification by 5.4% and 6.9%, reaching 92.3% and 84.4%, respectively. According to our study, the ground-penetrating radar oil pipeline recognition algorithm based on an attention mechanism can well accomplish the identification of underground oil pipelines.
引用
收藏
页数:14
相关论文
共 27 条
  • [1] Cao Z, 1996, GEOL EXPLOR, V32, P34
  • [2] Network-based optimization modeling of manhole setting for pipeline transportation
    Chen, Jingxu
    Wang, Shuaian
    Liu, Zhiyuan
    Guo, Yanyong
    [J]. TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW, 2018, 113 : 38 - 55
  • [3] NEOCOGNITRON - A NEW ALGORITHM FOR PATTERN-RECOGNITION TOLERANT OF DEFORMATIONS AND SHIFTS IN POSITION
    FUKUSHIMA, K
    MIYAKE, S
    [J]. PATTERN RECOGNITION, 1982, 15 (06) : 455 - 469
  • [4] Reducing the dimensionality of data with neural networks
    Hinton, G. E.
    Salakhutdinov, R. R.
    [J]. SCIENCE, 2006, 313 (5786) : 504 - 507
  • [5] Calculation of Average Reservoir Pore Pressure Based on Surface Displacement Using Image-To-Image Convolutional Neural Network Model
    Hu, Chaoyang
    Wang, Fengjiao
    Ai, Chi
    [J]. FRONTIERS IN EARTH SCIENCE, 2021, 9
  • [6] Deep learning-based automated underground cavity detection using three-dimensional ground penetrating radar
    Kang, Man-Sung
    Kim, Namgyu
    Lee, Jong Jae
    An, Yun-Kyu
    [J]. STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2020, 19 (01): : 173 - 185
  • [7] Underground Cylindrical Objects Detection and Diameter Identification in GPR B-Scans via the CNN-LSTM Framework
    Lei, Wentai
    Luo, Jiabin
    Hou, Feifei
    Xu, Long
    Wang, Ruiqing
    Jiang, Xinyue
    [J]. ELECTRONICS, 2020, 9 (11) : 1 - 16
  • [8] Automatic hyperbola detection and fitting in GPR B-scan image
    Lei, Wentai
    Hou, Feifei
    Xi, Jingchun
    Tan, Qianying
    Xu, Mengdi
    Jiang, Xinyue
    Liu, Gengye
    Gu, Qingyuan
    [J]. AUTOMATION IN CONSTRUCTION, 2019, 106
  • [9] [李嘉 LI Jia], 2007, [地球物理学进展, Progress in Geophysiscs], V22, P629
  • [10] Tree Root Automatic Recognition in Ground Penetrating Radar Profiles Based on Randomized Hough Transform
    Li, Wentao
    Cui, Xihong
    Guo, Li
    Chen, Jin
    Chen, Xuehong
    Cao, Xin
    [J]. REMOTE SENSING, 2016, 8 (05)