There may be several internal defects in railway track work that have different shapes and distribution rules, and these defects affect the safety of high-speed trains. Establishing reliable detection models and methods for these internal defects remains a challenging task. To address this challenge, in this study, an intelligent detection method based on a generalization feature cluster is proposed for internal defects of railway tracks. First, the defects are classified and counted according to their shape and location features. Then, generalized features of the internal defects are extracted and formulated based on the maximum difference between different types of defects and the maximum tolerance among same defects’ types. Finally, the extracted generalized features are expressed by function constraints, and formulated as generalization feature clusters to classify and identify internal defects in the railway track. Furthermore, to improve the detection reliability and speed, a reduced-dimension method of the generalization feature clusters is presented in this paper. Based on this reduced-dimension feature and strongly constrained generalized features, the K-means clustering algorithm is developed for defect clustering, and good clustering results are achieved. Regarding the defects in the rail head region, the clustering accuracy is over 95%, and the Davies-Bouldin index (DBI) index is negligible, which indicates the validation of the proposed generalization features with strong constraints. Experimental results prove that the accuracy of the proposed method based on generalization feature clusters is up to 97.55%, and the average detection time is 0.12 s/frame, which indicates that it performs well in adaptability, high accuracy, and detection speed under complex working environments. The proposed algorithm can effectively detect internal defects in railway tracks using an established generalization feature cluster model.
There may be several internal defects in railway track work that have different shapes and distribution rules, and these defects affect the safety of high-speed trains. Establishing reliable detection models and methods for these internal defects remains a challenging task. To address this challenge, in this study, an intelligent detection method based on a generalization feature cluster is proposed for internal defects of railway tracks. First, the defects are classified and counted according to their shape and location features. Then, generalized features of the internal defects are extracted and formulated based on the maximum difference between different types of defects and the maximum tolerance among same defects’ types. Finally, the extracted generalized features are expressed by function constraints, and formulated as generalization feature clusters to classify and identify internal defects in the railway track. Furthermore, to improve the detection reliability and speed, a reduced-dimension method of the generalization feature clusters is presented in this paper. Based on this reduced-dimension feature and strongly constrained generalized features, the K-means clustering algorithm is developed for defect clustering, and good clustering results are achieved. Regarding the defects in the rail head region, the clustering accuracy is over 95%, and the Davies-Bouldin index (DBI) index is negligible, which indicates the validation of the proposed generalization features with strong constraints. Experimental results prove that the accuracy of the proposed method based on generalization feature clusters is up to 97.55%, and the average detection time is 0.12 s/frame, which indicates that it performs well in adaptability, high accuracy, and detection speed under complex working environments. The proposed algorithm can effectively detect internal defects in railway tracks using an established generalization feature cluster model.
[1] M Leboeuf. High-speed rail: Opportunities and threats. Engineering, 2016, 2(4): 36-50.
[2] Y Xia, Y G Wei, Y H Lai, et al. Research on the line plan of passenger-like transport system of railway container based on transportation organization mode. Journal of the China Railway Society, 2019, 41(4): 10–15. (in Chinese)
[3] Q Y Peng, Y X Zhang, G Y Lu, et al. Arrival and departure tracks utilization adjustment method in railway passenger stations based on time-space resources. Journal of Tongji University (Natural Science), 2019, 47(7): 1011–1021. (in Chinese)
[4] A Poudel, B Lindeman, R Wilson. Current practices of rail inspection using ultrasonic methods: A review. Materials Evaluation, 2019, 77(7): 870-883.
[5] H Zhang, Y N Song, Y N Wang, et al. Review of rail defect non-destructive testing and evaluation. Chinese Journal of Scientific Instrument, 2019, 40(2): 11-25. (in Chinese)
[6] X C Bian, H G Jiang, Y M Chen. Accumulative deformation in railway track induced by high-speed traffic loading of the trains. Earthquake Engineering and Engineering Vibration, 2010, 9(3): 319-326.
[7] H M Shi, Z J Yu, L Q Zhu, et al. Study on the monitoring method of longitudinal displacement for the high-speed continuous welded rail. Chinese Journal of Scientific Instrument, 2016, 37(4): 811–817. (in Chinese)
[8] H M Yu, Q Y Li, Y Q Tan, et al. A coarse-to-fine model for rail surface defect detection. IEEE Transactions on Instrumentation and Measurement, 2018, 68(3): 656-666.
[9] A K Dubey, Z A Jaffery. Maximally stable extremal region marking-based railway track surface defect sensing. IEEE Sensors Journal, 2016, 16(24): 9047-9052.
[10] Q Y Li, S W Ren. A visual detection system for rail surface defects. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), 2012, 42(6): 1531-1542.
[11] H Zhang, X T Jin, Q M J Wu, et al. Automatic visual detection system of railway surface defects with curvature filter and improved gaussian mixture model. IEEE Transactions on Instrumentation and Measurement, 2018, 67(7): 1593-1608.
[12] G Y Tian, B Gao, Y L Gao, et al. Review of railway rail defect non-destructive testing and monitoring. Chinese Journal of Scientific Instrument, 2016, 37(8): 1763-1780. (in Chinese)
[13] S Mariani, T Nguyen, X Zhu, et al. Field test performance of noncontact ultrasonic rail inspection system. Journal of Transportation Engineering, Part A: Systems, 2017, 143(5): 04017007.
[14] H Cygan, L Girardi, Patrice Aknin, et al. B-scan ultrasonic image analysis for internal rail defect detection, WCRR 2003 Edinburgh Scotland, 2003.
[15] X Y Huang, Y S Shi, Y H Zhang, et al. BP neural network based on rail flaw classification of RFD car’s B-scan data. China Railway, 2018(3): 82-87. (in Chinese)
[16] C S Sun, J Liu, Y Qin, et al. Intelligent detection method for rail flaw based on deep learning. China Railway Science, 2018, 39(5): 51-57. (in Chinese)
[17] A Y Liang, S Sternini, L D S Francesco, et al. Rail flaw identification using ultrasonic imaging. Transportation Research Board 97th Annual Meeting, 2018.
[18] W J Wang, R Lewis, B Yang, et al. Wear and damage transitions of wheel and rail materials under various contact conditions. Wear, 2016, 362-363: 146-152.
[19] S Bahamon-Blanco, S Rapp, C Rupp, et al. Recognition of track defects through measured acceleration - part 1. IOP Conference Series Materials Science and Engineering, 2019, 615(1): 012121.
[20] Y Y Dong. Study on ultrasonic flaw detection method of rail base metal. Architectural Engineering Technology and Design, 2018(19): 110. (in Chinese)
[21] P W Loveday. Guided wave inspection and monitoring of railway track. Journal of Nondestructive Evaluation, 2012, 31(4): 303-309.
[22] D Mukherjee, Q M J Wu, G H Wang. A comparative experimental study of image feature detectors and descriptors. Machine Vision and Applications, 2015, 26(4): 443-466.
[23] S Kenderian, B B Djordjevic. Point and line source laser generation of ultrasound for inspection of internal and surface flaws in rail and structural materials. Research in Nondestructive Evaluation, 2001, 13(4): 189-200.
[24] Z D He, Y N Wang, J X Mao, et al. Research on inverse P-M diffusion-based rail surface defect detection. Acta Automatica Sinica, 2014, 40(8): 1667-1679. (in Chinese)
[25] A V Favorskaya, S V Kabisov, I B Petrov. Modeling of ultrasonic waves in fractured rails with an explicit approach. Doklady Mathematics, 2018, 98(1): 401-404.
[26] Q R Xu, Y S Shi, L H Xiong, et al. Research on detection technology of top-surface damage and magnetic leakage based on large flaw detection vehicle for rails. China Railway, 2017(10): 39-44. (in Chinese)
[27] R Stock, R Pippan. Rail grade dependent damage behaviour- characteristics and damage formation hypothesis. Wear, 2014, 314(1-2): 44-50.
[28] C H Yang, C Liu. Skeleton extraction algorithm of optimized gradient path. Journal of Xiamen University (Natural Science), 2014, 53(2): 201-205. (in Chinese)
[29] Z Y Wang, S Farhand, G Tsechpenakis. Fading affect bias: Improving the trade-off between accuracy and efficiency in feature clustering. Machine Vision and Applications, 2019, 30(2): 255-268.
[30] S X Lu, G Y Lin, H L Liu, et al. A weekly load data mining approach based on hidden markov model. IEEE Access, 2019, 7: 34609-34619.
