Ordinal pattern (OP) mode decomposition (OPMD) is a novel signal decomposition method that can effectively reveal pseudo-periodic impulse components obscured by noise. However, OPMD inherently relies on the precision of OP anchor points, which diminish its performance in situations with a low signal-to-noise ratio. Moreover, the accuracy of OPMD can be also significantly compromised in the presence of periodic impulse interference. To enhance the applicability of OPMD in mechanical fault diagnosis, this article proposes an ensemble OPMD method, incorporating blind filter based on indicators of second-order cyclostationary (ICS2). First, the ICS2 blind filtering technique is proposed to mitigate excessive noise spectral line and periodic interference caused by the transmission path and the inherent vibrations of the equipment and a filtering signal based on the ICS2 blind filtering technique is utilized to identify anchors of potential periodic OPs. Subsequently, the phase-rectified signal averaging (PRSA) filter is designed using PRSA technique based on these anchors. Finally, a filter bank is constructed based on the ICS2 filter and the PRSA filter to decompose it into fault-related impulse patterns and nonimpulse residual patterns. To validate the effectiveness of the ensemble OPMD method for fault diagnosis, a simulation experiment and three case studies involving bearing and gear faults under typical operating conditions were conducted.
LiXWangYXingJ, et al. Causal graph inference with adaptive dynamic structure learning for mechanism-oriented fault diagnosis in dynamic industrial systems. Reliab Eng Syst Saf2025; 266: 111865.
2.
WangJShaoHHeJ, et al. A novel interpretable fault diagnosis method using multi-image feature extraction and attention fusion. Pattern Recognit Lett2025; 189: 38–47.
3.
ShaoHXiaoYLengJ, et al. Collaborative human-computer fault diagnosis via calibrated confidence estimation. Adv Eng Inform2025; 65: 103349.
4.
WangSXiangJTangH, et al. Minimum entropy deconvolution based on simulation-determined band pass filter to detect faults in axial piston pump bearings. ISA Trans2019; 88: 186–198.
5.
ChengYZhouNZhangW, et al. Application of an improved minimum entropy deconvolution method for railway rolling element bearing fault diagnosis. Journal of Sound and Vibration2018; 425: 53–69.
6.
McDonaldGLZhaoQZuoMJ.Maximum correlated kurtosis deconvolution and application on gear tooth chip fault detection. Mech Syst Signal Process2012; 33: 237–255.
7.
McDonaldGLZhaoQ.Multipoint optimal minimum entropy deconvolution and convolution fix: application to vibration fault detection. Mechanical Systems and Signal Processing2017; 82: 461–477.
8.
PeetersCAntoniJHelsenJ.Blind filters based on envelope spectrum sparsity indicators for bearing and gear vibration-based condition monitoring. Mech Syst Signal Process2020; 138: 106556.
9.
AntoniJ.Cyclostationarity by examples. Mech Syst Signal Process2009; 23(4): 987–1036.
10.
BuzzoniMAntoniJd’EliaG.Blind deconvolution based on cyclostationarity maximization and its application to fault identification. J Sound Vib2018; 432: 569–601.
11.
ZhangBMiaoYLinJ, et al. Adaptive maximum second-order cyclostationarity blind deconvolution and its application for locomotive bearing fault diagnosis. Mech Syst Signal Process2021; 158: 107736.
12.
HuangNEShenZLongSR, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proc R Soc Lond A1998; 454(1971): 903–995.
13.
DragomiretskiyKZossoD.Variational mode decomposition. IEEE Trans Signal Process2013; 62(3): 531–544.
14.
HouBWangDXiaT, et al. Difference mode decomposition for adaptive signal decomposition. Mech Syst Signal Process2023; 191: 110203.
15.
MiaoYZhangBLiC, et al. Feature mode decomposition: new decomposition theory for rotating machinery fault diagnosis. IEEE Trans Ind Electron2022; 70(2): 1949–1960.
16.
FlandrinPGonçalvesPRillingG.EMD equivalent filter banks, from interpretation to applications. World Sci2005; 16: 57–74.
17.
FlandrinPRillingGGonçalvesP.Empirical mode decomposition as a filter bank. IEEE Signal Process Lett2004; 11(2): 112–114.
18.
WuZHuangNE.Ensemble empirical mode decomposition: a noise-assisted data analysis method. Adv Adapt Data Anal2009; 1(01): 1–41.
19.
WangYMarkertR.Filter bank property of variational mode decomposition and its applications. Signal Process2016; 120: 509–521.
20.
LianJLiuZWangH, et al. Adaptive variational mode decomposition method for signal processing based on mode characteristic. Mech Syst Signal Process2018; 107: 53–77.
21.
GuoJLiuYYangR, et al. Ensemble difference mode decomposition based on transmission path elimination technology for rotating machinery fault diagnosis. Mech Syst Signal Process2024; 212: 111330.
22.
ThakurGWuHT.Synchrosqueezing-based recovery of instantaneous frequency from nonuniform samples. SIAM J Math Anal2011; 43(5): 2078–2095.
23.
YuGYuMXuC.Synchroextracting transform. IEEE Trans Ind Electron2017; 64(10): 8042–8054.
24.
ShaoHMingYLiuY, et al. Small sample gearbox fault diagnosis based on improved deep forest in noisy environments. Nondestruct Test Eval2025; 40(8): 3935–3956.
25.
QianQWenQTangR, et al. DG-Softmax: a new domain generalization intelligent fault diagnosis method for planetary gearboxes. Reliab Eng Syst Saf2025; 260: 111057.
26.
LiXGuJLiM, et al. A physically inspired and bounded metric learning framework for fault detection and multi-class classification in mechanical systems. Expert Syst Appl2025; 302: 130393.
27.
LiuDCuiLWangG, et al. Interpretable domain adaptation transformer: a transfer learning method for fault diagnosis of rotating machinery. Struct Health Monit2025; 24(2): 1187–1200.
28.
YooSKimMParkG, et al. Hybrid explainable anomaly detection framework of gas turbines for feature selection and fault localization. Struct Health Monit2025; 24(5): 14759217251333312.
29.
ZhangHHongJYangH, et al. A semisupervised fault frequency analysis method for rotating machinery based on restricted self-attention network. Struct Health Monit2025; 24(3): 1522–1537.
30.
JablounM.Ordinal pattern-based mode decomposition: a new approach to time series analysis. Signal Process2025; 230: 109802.
31.
BandtCShihaF.Order patterns in time series. J Time Ser Anal2007; 28(5): 646–665.
32.
BauerAKantelhardtJWBundeA, et al. Phase-rectified signal averaging detects quasi-periodicities in non-stationary data. Physica A2006; 364: 423–434.
33.
LemayMPrudatYJacquemetV, et al. Phase-rectified signal averaging used to estimate the dominant frequencies in ECG signals during atrial fibrillation. IEEE Trans Biomed Eng2008; 55(11): 2538–2547.
34.
JablounMVan ZaenJVesinJM. Time-frequency analysis based on the phase-rectified signal averaging method. In 2009 17th European signal processing conference, IEEE, Glasgow, Scotland, UK, 2009, pp. 2303–2307.
35.
HouBWangYWangD.Cycle-embedded sparsity measures as a generalized objective function of impulsive mode decomposition for impulsive fault component extraction. Mech Syst Signal Process2025; 231: 112566.
36.
WangBLeiYLiN, et al. A hybrid prognostics approach for estimating remaining useful life of rolling element bearings. IEEE Trans Reliab2018; 69(1): 401–412.
37.
ZamanianAHOhadiA.Gear fault diagnosis based on Gaussian correlation of vibration signals and wavelet coefficients. Appl Soft Comput2011; 11(8): 4807–4819.
