Author(s)

XiChun Luo, HaoRan Hu^*

Affiliation(s)

Yunnan Jingjian Rail Transit Investment Construction Co., Ltd., Kunming 650000, Yunnan, China.

Corresponding Author

HaoRan Hu

ABSTRACT

Wheel out-of-roundness (OOR) is a prevalent issue in rail transit vehicles, posing potential safety hazards to electric multiple units (EMUs) and significantly affecting passenger ride comfort. However, current research predominantly focuses on dynamic simulation analyses, with relatively few studies targeting the vibration characteristics associated with wheel OOR. To address this gap, this paper proposes a novel diagnostic method that utilizes Variational Mode Decomposition (VMD) to extract salient signal features and employs the Grey Wolf Optimizer (GWO) to determine the optimal parameters for Maximum Correlated Kurtosis Deconvolution (MCKD) based on minimum sample entropy. Finally, the fault characteristic frequencies are extracted through envelope spectrum analysis. The method was validated on real-world wheel OOR data collected from operational trains. The results demonstrate that the proposed approach effectively isolates the fault characteristic information of wheel OOR, providing a robust basis for further research and practical application in this domain.

KEYWORDS

Wheel out-of-roundness; Variational Mode Decomposition; Maximum Correlated Kurtosis Deconvolution; Greywolf optimizer

CITE THIS PAPER

XiChun Luo, HaoRan Hu. Diagnosis for wheelset out-of-roundness of metro vehicle using VMD combined with optimized MCKD. World Journal of Engineering Research. 2025, 3(2): 32-40. DOI: https://doi.org/10.61784/wjer3027.

REFERENCES

[1] Chen X, Li J H, Liu W S, et al. Research on countermeasures for wheel out-of-roundness of domestic Type A metro vehicles. Locomotive & Rolling Stock Technology, 2025, 61(1): 50-54.

[2] Hou S J. Initial Research on the Automatic Computer Inspection System on Wheel Treads. Railway Vehicles, 2001, 39(12): 3.

[3] Ji J C. Research on online dynamic measurement method of wheelset geometric parameters: Beijing Jiaotong University. 2015.

[4] Zhang J J, Ma Z Q, Wang M Q, et al. Rolling bearing fault feature extraction based on VMD and autocorrelation analysis. Journal of Electronic Measurement and Instrumentation, 2017, 31(9): 7.

[5] Sun K, Zhang L, Wang F Z. Partial discharge signal denoising method based on variational mode decomposition and singular value decomposition. Journal of Henan Polytechnic University (Natural Science), 2020, 39(6): 8.

[6] Fei H B, Zhang C, Wu L, et al. Denoising and impact feature enhancement method for weak fault signals based on VMD-MCKD. Mechanical & Electrical Engineering, 2025, 42(2): 237-246.

[7] Tang J, Zhao Q. Motor rolling bearing fault diagnosis based on MVMD energy entropy and GWO-SVM. Journal of Vibroengineering, 2023, 25(6): 1096-1107.

[8] Saari J, Strombergsson D, Lundberg J, et al. Detection and identification of windmill bearing faults using a one-class support vector machine (SVM). measurement, 2019, 137: 287-301.

[9] Dragomiretskiy K, Zosso D. Variational mode decomposition. IEEE transactions on signal processing, 2013, 62(3): 531-544.

[10] Wang X, You C, Li X, et al. Chatter feature extraction for milling thin-walled parts based on GWO-VMD and CMSE. The International Journal of Advanced Manufacturing Technology, 2025, 1-14.

[11] Yang J, Liu W, Li S, et al. editors. Fault Diagnosis Method of Bearings Based on RandWPSO-VMD-MCKD. Journal of Physics Conference Series, 2025, 2999(1): 012052. DOI: 10.1088/1742-6596/2999/1/012052.

[12] Zhao L, Chi X, Li P, et al. Incipient fault feature enhancement of rolling bearings based on CEEMDAN and MCKD. Applied Sciences, 2023, 13(9): 5688.

[13] Stylianou O, Susi G, Hoffmann M, et al. Multiscale detrended cross-correlation coefficient: estimating coupling in non-stationary neurophysiological signals. Frontiers in Neuroscience, 2024, 18: 1422085.