journal article·2025-09-01Open access

A crack depth prediction method using transient geometric features of three-dimensional circular array MFL signals aided LSTM

Guoliang Zhou, Hongqiang Zhou, Zezhen Jiang, Wendong Gong

AIP Advances· Volume 15· Issue 9

Abstract

It is of great significance to improve the detection accuracy of initial fatigue cracks on the rail top surface for early warning of rail damage. The crack depth is especially difficult to measure directly in common methods. This study proposes an approach using transient characteristics of three-dimensional array MFL signals aided by Long Short-Term Memory (LSTM) algorithm to predict the crack depth. First, the characteristics of MFL signals are analyzed in the finite element simulation model when crack depth changes. Then, transient geometric features of the circular array MFL signals are extracted as a set of input feature vectors. Finally, the LSTM neural network algorithm is used to predict the crack depth. The experiment results show that the average prediction accuracy of crack depth is about 93.6%. It indicates that the method can explore more signal features and improve the crack depth prediction accuracy.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Non-Destructive Testing Techniques · Structural Health Monitoring Techniques · Ultrasonics and Acoustic Wave Propagation

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

School of Intelligent Manufacturing of Lishui Vocational and Technical College 1 , Lishui City, Zhejiang Province 323000,

Duke Kunshan University 2 , Kunshan 215316,

School of Urban Rail, Shandong Polytechnic 3 , Jinan 250104,

Funding

Lishui Science and Technology Bureau

References (47)

Show source-provided references
1. Failure analysis of cyclic damage to a railway rail—A case study10.1016/j.engfailanal.2020.104732 ↗
2. Detection and maintenance for railway track defects: A review10.1088/1755-1315/1140/1/012011 ↗
3. Nondestructive testing technologies for rail inspection: A review10.3390/coatings12111790 ↗
4. A literature review of artificial intelligence applications in railway systems10.1016/j.trc.2022.103679 ↗
5. Development of phased array ultrasonic system for detecting rail cracks10.1016/j.sna.2020.112086 ↗
6. Acoustic emission wave classification for rail crack monitoring based on synchrosqueezed wavelet transform and multi-branch convolutional neural network10.1177/1475921720922797 ↗
7. Investigation of DC electromagnetic-based motion induced eddy current on NDT for crack detection10.1109/jsen.2021.3049551 ↗
8. An MFL probe using shiftable magnetization angle for front and back side crack evaluation
9. Application of improved least-square generative adversarial networks for rail crack detection by AE technique10.1016/j.neucom.2018.12.057 ↗
10. An investigation on acoustic emission detection of rail crack in actual application by chaos theory with improved feature detection method10.1016/j.jsv.2018.09.014 ↗
11. Railway ballast fouling, inspection, and solutions—A review10.1177/09544097221148057 ↗
12. Inspection of RCF rail defects—Review of NDT methods10.1016/j.ymssp.2022.109568 ↗
13. Vision based rail track extraction and monitoring through drone imagery10.1016/j.icte.2017.11.010 ↗
14. Automated vision inspection of rail surface cracks: A double-layer data-driven framework10.1016/j.trc.2018.05.007 ↗
15. Developments in 3D visualisation of the rail tunnel subsurface for inspection and monitoring10.3390/app122211310 ↗
16. A visual detection system for rail surface defects10.1109/tsmcc.2012.2198814 ↗
17. On the use of ultrasonic guided waves for the health monitoring of rails10.1177/09544097211025898 ↗
18. Rapid detection of cracks in the rail foot by ultrasonic B-scan imaging using a shear horizontal guided wave electromagnetic acoustic transducer10.1016/j.ndteint.2021.102437 ↗
19. Rail diagnostics based on ultrasonic guided waves: An overview10.3390/app11031071 ↗
20. Ultrasonic guided wave monitoring of an operational rail track10.1177/1475921719893887 ↗
21. An ultrasonic guided wave mode selection and excitation method in rail defect detection10.3390/app9061170 ↗
22. AC-MFL testing method for surface crack on reground billets using flexible parallel-cable-based probe10.1016/j.ndteint.2023.102928 ↗
23. A review of magnetic flux leakage nondestructive testing10.3390/ma15207362 ↗
24. A magnetic shielding strategy for magnetic sensor in magnetic flux leakage testing10.1016/j.jmmm.2022.169888 ↗
25. MAR and UAR: Solutions for depth profile reconstruction of MFL inspection with high degree of freedoms10.1109/TASE.2023.3326496 ↗
26. Magnetic flux leakage course of inner defects and its detectable depth10.1186/s10033-021-00579-y ↗
27. An application of BRANN and MFL methods: Determining crack type and physical properties on M5 steel sheets10.1016/j.measurement.2019.02.064 ↗
28. Evaluation of transverse cracks detection depth in MFL rail NDT10.1134/S1061830914080026 ↗
29. Multiple cracks detection and visualization using magnetic flux leakage and eddy current pulsed thermography10.1016/j.sna.2015.09.011 ↗
30. A fast globally convergent particle swarm optimization for defect profile inversion using MFL detector10.3390/machines10111091 ↗
31. Developments, challenges, and perspectives of railway inspection robots10.1016/j.autcon.2022.104242 ↗
32. THMS-Net: A two-stage heterogeneous signals mutual supervision network for MFL weak defect detection10.1109/tim.2022.3198762 ↗
33. Evaluation of defect depth in ferromagnetic materials via magnetic flux leakage method with a double Hall sensor10.1016/j.jmmm.2022.169341 ↗
34. Rail crack detection using acoustic emission technique by joint optimization noise clustering and time window feature detection10.1016/j.apacoust.2019.107141 ↗
35. Adaptive filtering method of MFL signal on rail top surface defect detection10.1109/access.2021.3065044 ↗
36. Magnetic field frequency optimisation for MFL imaging using QWHE sensors10.1784/insi.2020.62.7.396 ↗
37. Corrosion detection in PSC bridge tendons using kernel PCA denoising of measured MFL signals10.3390/s20215984 ↗
38. MFL image recognition method of pipeline corrosion defects based on multilayer feature fusion multiscale GhostNet10.1109/tim.2022.3199247 ↗
39. An estimation method of defect size from MFL image using visual transformation convolutional neural network10.1109/tii.2018.2828811 ↗
40. Enhancement method of magnetic flux leakage signals for rail track surface defect detection10.1049/iet-smt.2018.5651 ↗
41. Short-circuit fault detection in laminated long stators of high-speed maglev track based on fractal dimension10.1016/j.measurement.2021.109177 ↗
42. Magnetic flux leakage detection method by the arrays consisted of three-dimensional hall sensor for rail top surface cracks10.1080/00150193.2022.2091999 ↗
43. A new rail crack detection method using LSTM network for actual application based on AE technology10.1016/j.apacoust.2018.08.020 ↗
44. CNN-LSTM network-based damage detection approach for copper pipeline using laser ultrasonic scanning10.1016/j.ultras.2022.106685 ↗
45. Characterisation of clustered cracks using an ACFM sensor and application of an artificial neural network10.1016/j.ndteint.2018.04.007 ↗
46. A survey on anomaly detection for technical systems using LSTM networks10.1016/j.compind.2021.103498 ↗
47. Detection of rail surface defects based on CNN image recognition and classification