Rail Crack Monitoring Using Acoustic Emission Technique (eBook)

Supervisor’s Foreword 6
Parts of this thesis have been published in the following journal articles:Li, D. *, Kuang, K. S. C., Koh, C. G. (2017). Rail crack monitoring based on Tsallis synchrosqueezed wavelet entropy of acoustic emission signals: a field study. Structural Health Monitoring. Prepublished online December 4, 2017, DOI: 10.1177/1475921717742339.Li, D., Kuang, K. S. C. *, Koh, C. G. (2017). Fatigue crack sizing in rail steel using crack closure-induced acoustic emission waves. Measurement Science and Technology. 28(6): 065601.Kuang, K. S. C., Li, D.*, Koh, C. G. (2016). Acoustic emission source location and noise cancellation for crack detection in rail head. Smart Structures and Systems. 18(5): 1063–1085. 8
Acknowledgements 9
Contents 10
Abbreviations 13
Nomenclature 14
List of Figures 16
List of Tables 21
Summary 22
1 Introduction 24
1.1 Background 24
1.2 Objectives and Scope of Research 26
1.3 Research Significance 26
1.4 Thesis Outline 27
References 29
2 Literature Review 30
2.1 Common Defects of Rail Track 30
2.1.1 Surface Cracks 30
2.1.2 Internal Cracks 34
2.2 Current Rail Monitoring Techniques 34
2.2.1 Acceleration-Based Technique 35
2.2.2 Automated Visual Technique 35
2.2.3 Ultrasonic Techniques 36
2.2.4 Electromagnetic Techniques 38
2.2.5 Magnetic Induction Technique 38
2.3 AE Technique and Its Applications 39
2.3.1 Introduction to AE Technique 39
2.3.2 Characterization of AE Waves 41
2.3.3 Relevant Applications of AE Technique 43
2.4 State-of-Art of Rail Condition Monitoring Using AE 45
References 47
3 Propagation Features and Source Location 52
3.1 Introduction 52
3.2 Experimental Procedure 54
3.2.1 Pencil Lead Break (PLB) 55
3.2.2 Field PLB Test 55
3.2.3 Field Train Pass-by Test 56
3.2.4 AE Data Acquisition 57
3.3 Time-Frequency Representation of AE Waves 58
3.3.1 Continuous Wavelet Transform (CWT) 58
3.3.2 Optimal Mother Wavelet Selection 59
3.3.3 Time-Frequency Characteristics of AE Waves 61
3.4 Propagation Features of AE Waves 62
3.4.1 Theory of Ultrasonic Propagation 63
3.4.2 Attenuation of AE Waves in Rail Head 66
3.4.3 Dispersion of AE Waves in Rail Head 67
3.5 Source Location Methods 68
3.5.1 Time-of-Arrival (TOA) Method 68
3.5.2 Wavelet Transform-Based Modal Analysis Location (WTMAL) Method 69
3.6 Hilbert Transform-Based Noise Cancellation Method 71
3.7 Results and Discussion 73
3.7.1 Influence of Operational Noise on Crack Detection 73
3.7.2 Source Location Without Noise Using TOA Method 76
3.7.3 Source Location Without Noise Using WTMAL Method 76
3.7.4 Source Location with Noise Using WTMAL Method 79
3.8 Concluding Remarks 82
References 85
4 Sizing of Fatigue Cracks 87
4.1 Introduction 87
4.2 Experimental Procedure 88
4.2.1 Rail Steel Specimens 88
4.2.2 Fatigue Tests 89
4.2.3 AE Data Acquisition 91
4.2.4 Crack Length Calculation 92
4.2.5 Crack Surface Observation 93
4.3 AE Wave Classification 93
4.3.1 Wavelet Power (WP)-Based Classification Index 94
4.3.2 Threshold Determination for the Classification Index 99
4.3.3 Frequency Bands Selection for the Classification Index 101
4.4 Fatigue Crack Sizing Methods 101
4.4.1 Traditional Method Based on CP-Induced AE Waves 101
4.4.2 Novel Method Based on CC-Induced AE Waves 102
4.4.3 Comparison of Crack Sizing Methods 103
4.5 Results and Discussion 103
4.5.1 AE Waves Classification 103
4.5.2 Crack Sizing Using the Traditional Method 105
4.5.3 Crack Sizing Using the Novel Method 108
4.6 Concluding Remarks 111
References 112
5 Field Monitoring of Rail Cracks 114
5.1 Introduction 114
5.2 Experimental Procedure 117
5.2.1 Field Tests 117
5.2.2 AE Data Acquisition 119
5.3 Time-Frequency Representation of AE Waves 120
5.3.1 Synchrosqueezed Wavelet Transform (SWT) 120
5.3.2 Analysis of Example AE Waves 123
5.4 Crack Identification and Location Based on Time-Tsallis Synchrosqueezed Wavelet Entropy (TSWE) 125
5.4.1 Tsallis Entropy 126
5.4.2 TSWE with Time 130
5.4.3 Determination of Parameters in TSWE 133
5.4.4 Results and Discussion 138
5.5 Classification of Crack-Related AE Waves Based on Enhanced SWT Scalogram 145
5.5.1 SWT Enhanced by Gamma Correction 146
5.5.2 Results and Discussion 146
5.6 Concluding Remarks 148
References 151
6 Conclusions and Future Work 154
6.1 Conclusions 154
6.2 Future Work 156