基于随机森林的管道漏磁缺陷检测与量化
作者:
作者单位:

1.天津大学;2.北京华航无线电测量研究所

基金项目:

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [34]
  • | | | |
  • 文章评论
    摘要:

    油气管道的缺陷尺寸量化是管道检测的关键问题和最终目标。传统的缺陷检测方法往往停留在缺陷分类的阶段,数据处理不具体给后续结果分析增加了难度;智能识别方法又对漏磁数据的质量有更高要求。因此提出一种粒子群优化和随机森林相结合的PSO-RF算法,实现管道缺陷长、宽、深的自动量化。首先对一组缺陷漏磁数据进行多维度的特征提取,然后利用随机森林算法进行回归预测;针对随机森林算法最佳参数不宜获得的难点,使用粒子群优化算法进行超参数调优,最终获得比较准确的缺陷长、宽、深预测数据。两种算法相结合得到PSO-RF算法,并与经典的卷积神经网络和PSO-SVR训练算法进行对比,对长、宽、深的量化精度分别提高了28%、32%、68%,验证了PSO-RF算法的有效性与优越性。最后使用一组带标签的管道缺陷数据对算法进行验证,长、宽、深量化误差在20%以内的数据分别达到80.3%、88.5%和95.9%。

    Abstract:

    The quantification of defect size in oil and gas pipelines is a key issue and ultimate goal of pipeline inspection. Traditional defect detection methods often remain in the stage of defect classification, and the lack of detailed data increases the difficulty of subsequent processing; Intelligent recognition methods have higher requirements for the quality of magnetic leakage data however. Therefore, a PSO-RF algorithm combining particle swarm optimization and random forest is proposed to quantify the length, width, and depth of pipeline defects. Firstly, multi-dimensional feature extraction is performed on a set of defect magnetic leakage data, and then the random forest algorithm is used for regression prediction; In view of the difficulty of obtaining the best parameters of random forest algorithm, particle swarm optimization algorithm is used to optimize the hyperparameters, and finally more accurate prediction data of defect length, width and depth are obtained. The PSO-RF algorithm was obtained by combining two algorithms, and compared with classical CNN and PSO-SVR training algorithms. The quantization accuracy of length, width and depth was improved by 28%, 32% and 68% respectively, verifying the effectiveness and superiority of the PSO-RF algorithm. Finally, a set of labeled pipeline defect data was used to validate the algorithm, and the data with quantization errors of length, width and depth within 20% achieved 80.3%, 88.5% and 95.9% respectively.

    参考文献
    [1] 王国涛, 郭天昊. 油气管道特殊缺陷的漏磁信号识别方法[J]. 沈阳工业大学学报, 2019, 41(04): 401-405.
    WANG G T, GUO T H. Recognition method of magnetic flux leakage signal for special defect of oil and gas pipeline. Journal of Shenyang University of Technology. 2019, 41(4): 401-405.
    [3] [2] 焦敬品, 常予, 李光海, 石福涛, 何存富, 吴斌. 金属包覆层管道外壁损伤低频漏磁检测技术[J]. 北京工业大学学报, 2018, 44(12): 1471-1477.
    JIAO J P, CHANG Y, LI G H, SHI F T, HE C F, WU B. Outer Wall Damage of Cladding Tube Based on Low Frequency Magnetic Flux Leakage Technology[J]. Journal of Beijing University of Technology, 2018, 44(12): 1471-1477.
    [5] [3] 刘金海, 付明芮, 唐建华. 基于漏磁内检测的缺陷识别方法[J]. 仪器仪表学报, 2016, 37(11): 2572-2581.
    LIU J H, FU M R, TANG J H. MFL inner detection based defect recognition method[J]. Chinese Journal of Scientific Instrument, 2016, 37(11): 2572-2581.
    [7] [4] 杨理践, 曹辉. 基于深度学习的管道焊缝法兰组件识别方法[J]. 仪器仪表学报, 2018, 39(02): 193-202.
    YANG L J, CAO H. Deep learning based weld and flange identification in pipeline[J]. Chinese Journal of Scientific Instrument, 2018, 39(02): 193-202.
    [9] [5] CHEN J Z, KANG X W, ZHANG X W, HE R Y, MENG T. (2022). Research on Unsaturated Magnetization MFL Detection of Gouge in Oil and Gas Pipeline. In: Liu, G., Cen, F. (eds) Advances in Precision Instruments and Optical Engineering. Springer Proceedings in Physics, vol 270. Springer, Singapore.
    [10] [6] 赵春华, 罗顺, 谭金铃, 等. 基于PC-YOLOv7算法钢材表面缺陷检测[J]. 国外电子测量技术, 2023, 42(09): 137-145.
    ZHAO C H, LUO S, TAN J L, et al. Steel Surface defect detection based on PC-YOLOv7 algorithm [J]. Foreign Electronic Measurement Technology, 2023, 42(09): 137-145.
    [12] [7] RAVAN M, AMINEH R K, KOZIEL S, NIKOLOVA N K and REILLY J P, "Sizing of 3-D Arbitrary Defects Using Magnetic Flux Leakage Measurements," in IEEE Transactions on Magnetics, vol. 46, no. 4, pp. 1024-1033, April 2010.
    [13] [8] XU Z, LIU K, GU B, YAN L, PANG X and GAO K. (2023). Image recognition model of pipeline magnetic flux leakage detection based on deep learning. Corrosion Reviews, 41(6), 689-701.
    [14] [9] 秦浩东, 张颖, 赵鹏程. 改进Canny算子的小管径弯头漏磁缺陷图像量化方法[J]. 电子测量技术, 2024, 47(05): 150-157.
    QIN H D, ZHANG Y, ZHAO P C. Image quantization method of magnetic leakage defect of small diameter elbow with improved Canny operator [J]. Electronic Measurement Technique, 2024, 47(05): 150-157.
    [16] [10] SUN H et al., "Development of a Physics-Informed Doubly Fed Cross-Residual Deep Neural Network for High-Precision Magnetic Flux Leakage Defect Size Estimation," in IEEE Transactions on Industrial Informatics, vol. 18, no. 3, pp. 1629-1640, March 2022.
    [17] [11] 刘金海, 赵真, 付明芮, 等. 基于主动小样本学习的管道焊缝缺陷检测方法[J].仪器仪表学报, 2022, 43(11): 252-261.
    LIU J H, ZHAO Z, FU M R, et al. Pipeline weld defect detection method based on active small sample learning [J]. Chinese Journal of Scientific Instrument, 2022, 43(11): 252-261.
    [19] [12] 赵治. 基于机器学习的管道缺陷识别量化方法研究[D]. 沈阳工业大学, 2023.
    ZHAO Z. Research on Quantitative Method for Pipeline Defect Recognition Based on Machine Learning [D]. Shenyang University of Technology
    [21] [13] HUANG S, PENG L, SUN H, LI S. Deep Learning for Magnetic Flux Leakage Detection and Evaluation of Oil & Gas Pipelines: A Review. Energies 2023, 16, 1372.
    [22] [14] 刘杰. 速度可控的管内漏磁检测机器人研究[D]. 中国石油大学(北京), 2023.
    LIU J. Research on a Speed Controllable Magnetic Leakage Detection Robot inside a Pipe [D]. China University of Petroleum (Beijing), 2023.
    [24] [15] 常梦容, 王海瑞, 肖杨. mRMR特征筛选和随机森林的故障诊断方法研究[J]. 电子测量与仪器学报, 2022, 36(03): 175-183.
    CHANG MR, WANG HR, XIAO Y. Research on fault diagnosis method based on mRMR feature screening and random forest[J]. Journal of Electronic Measurement and Instrument, 2022, 36(3): 175-183.
    [26] [16] GEORGANOS S, GRIPPA T, NIANG GADIAGA A, LINARD C, LENNERT M, VANHUYSSE S, KALOGIROU S. (2019). Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling. Geocarto International, 36(2), 121–136.
    [27] [17] SPEISER J L, MILLER M E, TOOZE J, IP E. A Comparison of Random Forest Variable Selection Methods for Classification Prediction Modeling. Expert Syst Appl. 2019 Nov 15; 134: 93-101.
    [28] [18] SHI Y, CUI L, QI Z, MENG F and CHEN Z. "Automatic Road Crack Detection Using Random Structured Forests," in IEEE Transactions on Intelligent Transportation Systems, vol. 17, no. 12, pp. 3434-3445, Dec. 2016.
    [29] [19] GONG Y J et al., "Genetic Learning Particle Swarm Optimization," in IEEE Transactions on Cybernetics, vol. 46, no. 10, pp. 2277-2290, Oct. 2016.
    [30] [20] SHAMI T M, EL-SALEH A A, ALSWAITTI M, AL-TASHI Q, SUMMAKIEH M A and MIRJALILI S. "Particle Swarm Optimization: A Comprehensive Survey," in IEEE Access, vol. 10, pp. 10031-10061, 2022.
    [31] [21] BONYADI M R, MICHALEWICZ Z. Particle Swarm Optimization for Single Objective Continuous Space Problems: A Review. Evol Comput 2017; 25 (1): 1–54.
    [32] [22] HAN Q H, GUI C Q, XU J, LACIDOGNA G. A generalized method to predict the compressive strength of high-performance concrete by improved random forest algorithm, Construction and Building Materials, Volume 226, 2019, Pages 734-742, ISSN 0950-0618.
    [33] [23] 李郅琴, 杜建强, 聂斌, 熊旺平, 黄灿奕, 李欢. 特征选择方法综述[J]. 计算机工程与应用, 2019, 55(24): 10-19.
    LI Z Q, DU J Q, NIE B, XIONG W P, HUANG C Y, LI H. Summary of Feature Selection Methods[J]. Computer Engineering and Applications, 2019, 55(24): 10-19.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-07-08
  • 最后修改日期:2025-01-03
  • 录用日期:2025-01-08
文章二维码