摘要
为了进一步提升钢轨裂纹的识别精度,从新特征的角度出发,提出一种基于路图特征和支持向量机(Support Vector Machine,SVM)的钢轨裂纹识别方法。该方法基于图信号处理与图谱理论,计算由钢轨裂纹时域漏磁(Magnetic Flux Leakage,MFL)信号转换得到的路图信号的“时域”和“频域”统计量作为钢轨裂纹MFL信号的特征训练SVM分类器,有效实现了不同缺陷参数的钢轨裂纹识别。基于钢轨裂纹漏磁检测平台实测数据验证所提方法的有效性。实验结果表明,相比于传统漏磁信号特征,采用路图特征在钢轨裂纹识别中的精度更高、稳定性更好。
In order to further improve the identification accuracy of rail cracks,a rail crack identification method based on path graph features and support vector machine(SVM)is proposed from the perspective of new features.Based on the theory of graph signal processing and spectrum,the method calculates the statistics of“time domain”and“frequency domain”of the graph signal converted from the time domain magnetic flux leakage(MFL)signal as the features of MFL signal to train the SVM classifier,which effectivcly realizes the rail crack identification of different defect parameters.Effectiveness of the proposed method is verified by the measured data of magnetic leakage inspection platform for rail cracks.Experimental results show that compared with the traditional features of MFL signal,the graph features adopted have higher accuracy and better stability in rail crack identification.
作者
冷强
刘文波
赵旭东
杜晨琛
王平
LENG Qiang;LIU Wen-bo;ZHAO Xu-dong;DU Chen-chen;WANG Ping(College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)
出处
《测控技术》
2019年第10期5-9,39,共6页
Measurement & Control Technology
基金
国家质量基础的共性技术研究与应用重点专项基金(2017YFF0209700,2016YFF0103702,2016YFB1100205)
南京航空航天大学校开放基金(kfjj20180317)
关键词
漏磁检测
路图
SVM
钢轨裂纹识别
magnetic flux leakage testing
path graph
SVM
rail crack identification
