Rail crack recognition based on Adaptive Weighting Multi-classifier Fusion Decision

Abstract In order to make the full use of three-dimensional information of Magnetic Flux Leakage (MFL) signals, an Adaptive Weighting Multi-classifier Fusion Decision Algorithm is adopted for rail crack recognition. Support Vector Machine (SVM) is used to classify MFL signals from single-channel and single-direction, and then adaptive weightings of different SVMs are assigned according to entropy calculated by posterior probabilities of different SVMs. Finally, weighted majority vote strategy is used to make a comprehensive decision by fusing classification results of different channels and different directions. Effectiveness of the proposed method is testified by experiments based on measured MFL signals.

[1]  Stuart L. Grassie,et al.  Studs and squats: The evolving story , 2016 .

[2]  Ping Wang,et al.  Experimental Studies and New Feature Extractions of MBN for Stress Measurement on Rail Tracks , 2013, IEEE Transactions on Magnetics.

[3]  Victoria J. Hodge,et al.  Wireless Sensor Networks for Condition Monitoring in the Railway Industry: A Survey , 2015, IEEE Transactions on Intelligent Transportation Systems.

[4]  Gui Yun Tian,et al.  PEC thermography for imaging multiple cracks from rolling contact fatigue , 2011 .

[5]  Hans-Martin Thomas,et al.  NDT techniques for railroad wheel and gauge corner inspection , 2004 .

[6]  E P Tomasini,et al.  Experimental investigation by laser ultrasonics for high speed train axle diagnostics. , 2015, Ultrasonics.

[7]  Rachel S. Edwards,et al.  Inspection of rail track head surfaces using electromagnetic acoustic transducers (EMATs) , 2004 .

[8]  Robin Clark,et al.  Rail flaw detection: overview and needs for future developments , 2004 .

[9]  Jin Sung Kim,et al.  Fatigue life prediction of a railway hollow axle with a tapered bore surface , 2015 .

[10]  Wang Hai.  3D Magnetic Field Analysis for Rail Track Surface Defect Using Magnetic Flux Leakage Testing , 2010 .

[11]  Gui Yun Tian,et al.  Feature extraction and selection for defect classification of pulsed eddy current NDT , 2008 .

[12]  Zhang Zhen.  Study on SVM-based Fusion Algorithm Identifying Pipeline Crack Flaw MFL Signal Characteristics , 2011 .

[13]  陈佩华,et al.  Application of eddy current testing in the quantitative evaluation of the rail cracks , 2011 .

[14]  Chao Lu,et al.  Application of Portable Ultrasonic Phased Array Instrument for Rail Welds Ultrasonic Inspection , 2013 .

[15]  Saurabh Maheshwari,et al.  Railway Security System based on Wireless Sensor Networks: State of the Art , 2014 .

[16]  Joseph L. Rose,et al.  Guided wave inspection potential of defects in rail , 2004 .

[17]  Ashutosh Tiwari,et al.  A review of key planning and scheduling in the rail industry in Europe and UK , 2016 .

[18]  Gui Yun Tian,et al.  Pulsed electromagnetic methods for defect detection and characterisation , 2007 .

[19]  Uwe Zerbst,et al.  Introduction to the damage tolerance behaviour of railway rails – a review , 2009 .

[20]  Gui Yun Tian,et al.  Features extraction of sensor array based PMFL technology for detection of rail cracks , 2014 .