基于支持向量机的油管内外表面缺陷识别方法

蹇清平; 艾志久; 张勇; 杨赟达

doi:10.13433/j.cnki.1003-8728.2015.0125

基于支持向量机的油管内外表面缺陷识别方法

doi: 10.13433/j.cnki.1003-8728.2015.0125

1.
西南石油大学机电工程学院, 成都 610500

详细信息

作者简介:
蹇清平（1984-），博士研究生，研究方向为无损检测，jianqingping2004@163.com

通讯作者:
艾志久，教授，aizhijiu123@vip.sina.com

计量
- 文章访问数: 232
- HTML全文浏览量: 30
- PDF下载量: 12
- 被引次数: 0
出版历程
- 收稿日期: 2013-07-02
- 刊出日期: 2015-01-05

Identification of Oil Tube Defects on Internal and External Surfaces Based on SVM

1.
School of Mechatronic Engineering, Southwest Petroleum University, Chengdu 610500

摘要

摘要: 由于根据漏磁信号难以准确识别出油管内、外表面缺陷，为此提出了基于支持向量机（SVM）的油管内外表面缺陷识别方法。采用时频分析技术提取了用于区分油管内外表面缺陷的漏磁信号时域和频域特征量，然后将其作为油管内外表面缺陷识别SVM模型的样本数据，采用改进的云自适应粒子群（MACPSO）优化算法对SVM识别模型的参数进行优选，结合优选的模型参数和样本数据训练构建油管内外表面缺陷识别SVM模型。实验结果表明：该智能识别方法能够有效区分油管的内外表面缺陷，识别准确率高于90%。
- 油管 /
- 支持向量机 /
- 内外表面缺陷 /
- 粒子群优化 /
- 漏磁信号
Abstract: There are some difficulties to accurately identify the defects on the internal and external surfaces of the oil tube by the magnetic flux leakage(MFL) signal. Hence, this paper presents an identification method to achieve it based on support vector machines (SVM). First, the time-frequency analysis approach is employed to acquire the feature vectors of the time domain and frequency domain from MFL signal. Then, the feature vectors are used as the sample data of the proposed model of defect identification based on SVM. The parameters of this model are optimized by the Modified Adaptive Particle Swarm Optimization(MCAPSO)algorithm based on cloud theory. Finally, with the optimized model parameters and training sample data, the identification model of defect on internal and external surfaces based on SVM was established. The experimental results show that the proposed approach is effective on the identification of the oil tube defects between the internal and external surfaces, and its accuracy is higher than 90%.
- feature extraction /
- frequency domain analysis /
- internal and external surface defects /
- magnetic flux leakage /
- oil tube

HTML全文

参考文献(15)

[1]	Bubenik T, Nestlroth J, Eiber R, et al. Magnetic flux leakage (MFL) technology for natural gas pipeline inspection[J]. NDT & E International,1997,30(1):36-36
[2]	McJunkin T R, Miller K S, Tolle C R. Observations and characterization of defects in coiled tubing from magnetic flux leakage data[C]// SPE/ICoTA Coiled Tubing and Well Intervention Conference and Exhibition, USA: Society of Petroleum Engineers,2006:1-23
[3]	Mikkola C, Case C, Garrity K. Inline corrosion inspection verifies integrity of nonpiggable, noninterruptible gas lines[J]. Oil and Gas Journal,2005,103(16):64-69
[4]	李久政,钢管漏磁探伤中的内外伤区分方法[D].武汉:华中科技大学,2009 Li J Z. The discrimination methods of the inside and outside defects in the ferromagnetic tubular products during magnetic flux leakage testing[D]. Wuhan:Huazhong University of Science and Tcchnology,2009 (in Chinese)
[5]	Clapham L, Babbar V, Byrne J. Detection of mechanical damage using the magnetic flux leakage technique[C]// 16th World Conference Non-destructive Testing, Montreal: ASME,2004:983-990
[6]	Carvalho A. MFL signals and artificial neural networks applied to detection and classification of pipe weld defects[J]. NDT & E International,2006,39:661-667
[7]	Kopp G, Willems H. Sizing limits of metal loss anomalies using tri-axial MFL measurements: a model study[J]. NDT & E International,2013,55:75-81
[8]	王太勇,刘兴荣,秦旭达.熵谱分析方法在漏磁信号特征提取中的应用[J].天津大学学报,2004,37(3):216-220 Wang T Y, Liu X R, Qin X D. Spectrum entropy and its application in characteristics abstraction of magnetic flux leakage signals[J]. Journal of Tianjin University,2004,37(3):216-220 (in Chinese)
[9]	杨志军.铁磁性平板腐蚀缺陷多通道漏磁信号的反演与重构[D].大庆:东北石油大学,2011 Yang Z J. Inversion and reconstruction of multi-channel magnetic flux leakage signals of ferromagnetic floor corrosion defects[D]. Daqing:Northeast Petroleum University,2011 (in Chinese)
[10]	Cortes C, Vapnik V. Support-vector network[J]. Machine Learning,1995,20(3):273-297
[11]	Kennedy J, Eberhart R. Particle swarm optimization[C]// Proceedings of the 1995 IEEE International Conference on Neural Networks, Perth: the University of Western Australia,1995,3:1942-1948
[12]	李爱,陈果,侯民利.航空发动机油样光谱分析的PSO-LSSVM组合预测方法[J].机械科学与技术,2013,32(1):120-125 Li A, Chen G, Hou M L. Combinational forecast method based on PSO-LSSVM in spectrometric oil analysis of the aircraft engine[J]. Mechanical Science and Technology for Aerospace Engineering,2013,32(1):120-125 (in Chinese)
[13]	张锦华.改进的云自适应粒子群算法[J].计算机工程与应用,2012,48(5):29-31 Zhang J H. Modified adaptive PSO algorithm based on cloud theory[J]. Computer Engineering and Applications,2012,48(5):29-31 (in Chinese)
[14]	Ganyun L V, Cheng H Z, Zhai H B, et al. Fault diagnosis of power transformer based on multi-layer SVM classifier[J]. Electric Power Systems Research,2005,74(1):1-7
[15]	Yuan S F, Chu F L. Support vector machines-based fault diagnosis for turbo-pump rotor[J]. Mechanical Systems and Signal Processing,2006,20(4):939-952