An artificial intelligent bearing fault and hierarchical severity diagnosis framework is proposed in this study. The framework utilizes a combined deep belief networks (DBNs) and Dempster–Shafer (D-S) theory fault diagnosis scheme and adopts a two-stage approach in classifying (1) bearing fault conditions and (2) fault severities. The combined fault diagnostic scheme first employs two parameter-optimized DBNs to process the horizontal and vertical vibration data acquired from the bearing house of a test rig, where the parameters of the DBNs are optimized using a hybrid genetic algorithm and particle swarm optimization algorithm proposed in this study. The classification results from the two DBNs are fused further using the D-S theory to improve the diagnostic accuracy. The fault diagnosis scheme is used first to classify the bearing fault conditions in Stage 1 from a bulk dataset containing all bearing operation conditions under study. The same diagnosis scheme is applied once more to classify the hierarchical fault severities for each fault condition in Stage 2 using the pre-classified data from Stage 1. The effectiveness of the framework is then evaluated on a set of bearing condition monitoring data. A comparison study between the results obtained using the current method and those from existing published work is also presented in the article. It is shown that the accuracy for bearing fault and severity diagnosis can be substantially improved by using the current framework.
RaiAUpadhyaySH. A review on signal processing techniques utilized in the fault diagnosis of rolling element bearing. Tribol Int2016; 96: 289–306.
2.
YuKLinTRTanJ, et al. An adaptive sensitive frequency band selection method for empirical wavelet transform and its application in bearing fault diagnosis. Measurement2018; 134: 375–384.
3.
YanRQGaoRXChenXF. Wavelets for fault diagnosis of rotary machines: a review with applications. Signal Process2014; 96: 1–15.
4.
HuangNEShenZLongSR, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proc Royal Soc London A1998; 454(1971): 903–995.
5.
DaubechiesILuJWuH-T. Synchrosqueezed wavelet transform: an empirical mode decomposition like tool. Appl Comput Harmon Anal2011; 30: 243–261.
6.
YuGWangZZhaoP. Multisynchrosqueezing transform. IEEE T Ind Electron2019; 66: 5441–5455.
7.
FengZPZhouYZuoMJ, et al. Atomic decomposition and sparse representation for complex signal analysis in machinery fault diagnosis: a review with examples. Measurement2017; 103: 106–132.
8.
ZhouHChenJDongG, et al. Detection and diagnosis of bearing faults using shift-invariant dictionary learning and hidden Markov model. Mech Syst Signal Pr2016; 72–73: 65–79.
9.
CuiHZhangLKangR, et al. Research on fault diagnosis for reciprocating compressor valve using information entropy and SVM method. J Loss Prevent Proc2009; 22: 864–867.
10.
JegadeeshwaranRSugumaranV. Fault diagnosis of automobile hydraulic brake system using statistical features and support vector machines. Mech Syst Signal Pr2015; 52–53: 436–446.
11.
HoangD-TKangH-J. A survey on deep learning based bearing fault diagnosis. Neurocomputing2019; 335: 327–335.
12.
HintonGESalakhutdinovRR. Reducing the dimensionality of data with neural networks. Science2006; 313: 504–507.
13.
YuDDengL. Deep learning and its applications to signal and information processing [exploratory DSP]. IEEE Sig Process Magazine2011; 28: 145–154.
14.
BengioYLamblinPPopoviciD, et al. Greedy layer-wise training of deep networks. Adv Neur In 2007; 19: 153–160.
15.
LeCunYBoserBDenkerJS, et al. Handwritten digit recognition with a back-propagation network. Adv Neur In 1990; 88: 396–404.
16.
GilesCLKuhnGMWilliamsRJ. Dynamic recurrent neural networks: theory and applications. IEEE T Neural Networ1994; 5(2): 153–156.
17.
RafieiMHAdeliH. A novel unsupervised deep learning model for global and local health condition assessment of structures. Eng Struct2018; 156: 598–607.
18.
TamilselvanPWangP. Failure diagnosis using deep belief learning based health state classification. Reliab Eng Syst Safe2013; 115: 124–135.
19.
TranVTAlThobianiFBallA. An approach to fault diagnosis of reciprocating compressor valves using Teager-Kaiser energy operator and deep belief networks. Expert Syst Appl2014; 41: 4113–4122.
20.
ChenZLiCSanchezR. Gearbox fault identification and classification with convolutional neural networks. Shock Vib2015; 2015: 390134.
21.
GanMWangCZhuCA. Construction of hierarchical diagnosis network based on deep learning and its application in the fault pattern recognition of rolling element bearings. Mech Syst Signal Pr2016; 72–73: 92–104.
22.
ShaoHJiangHZhangX, et al. Rolling bearing fault diagnosis using an optimization deep belief network. Meas Sci Technol2015; 26: 115002.
23.
ThangarajRPantMAbrahamA, et al. Particle swarm optimization: hybridization perspectives and experimental illustrations. Appl Math Comput2011; 217: 5208–5226.
24.
BasirOYuanX. Engine fault diagnosis based on multi-sensor information fusion using Dempster-Shafer evidence theory. Inform Fusion2007; 8: 379–386.
25.
MoraBWulderMAWhiteJC. An approach using Dempster-Shafer theory to fuse spatial data and satellite image derived crown metrics for estimation of forest stand leading species. Inform Fusion2013; 14: 384–395.
26.
ZervasEMpimpoudisAAnagnostopoulosC, et al. Multisensor data fusion for fire detection. Inform Fusion2011; 12: 150–159.
27.
BoudraaAOBentabetASalzensteinF, et al. Dempster-Shafer’s basic probability assignment based on fuzzy membership functions. Electr Lett Comp Vis Image Anal2004; 4(1): 1–9.
28.
JiangWYangYLuoY, et al. Determining basic probability assignment based on the improved similarity measures of generalized fuzzy numbers. Int J Comput Commun2015; 10(3): 333–347.
29.
ShaferG. A mathematical theory of evidence. Princeton, NJ: Princeton University Press, 1976.
30.
HintonGEOsinderSTehYW. A fast learning algorithm for deep belief nets. Neural Comput2006; 18(7): 1527–1554.
31.
Al RahhalMMBaziYAlhichriH, et al. Deep learning approach for active classification of electrocardiogram signals. Inform Sci2016; 345: 340–354.
32.
GoodfellowIBengioYCourvilleA. Deep learning. Cambridge, MA: The MIT Press, 2016.
33.
LiuZCaoHChenX, et al. Multi-fault classification based on wavelet SVM with PSO algorithm to analyze vibration signals from rolling element bearings. Neurocomputing2013; 99: 399–410.
34.
ShankarTShanmugavelSRajeshA. Hybrid HSA and PSO algorithm for energy efficient cluster head selection in wireless sensor networks. Swarm Evol Comput2016; 30: 1–10.
35.
MitchellM. An introduction to genetic algorithms. Cambridge, MA: The MIT Press, 1996.
36.
ZhouCYinKCaoY, et al. Application of time series analysis and PSO-SVM model in predicting the Bazimen landslide in the Three Gorges Reservoir, China. Eng Geol2016; 204: 108–120.
37.
KennedyJEberhartR. Particle swarm optimization. In: Proceedings of 1995 IEEE international conference on neural networks, Perth, WA, Australia, 27 November–1 December 1995, pp. 1942–1948. New York: IEEE.
38.
AlirezaA. PSO with adaptive mutation and inertia weight and its application in parameter estimation of dynamic systems. Acta Automat Sinica2011; 37: 541–549.
39.
AhmadlouMAdeliH. Enhanced probabilistic neural network with local decision circles: a robust classifier. Integr Comput-Aid E2010; 17: 197–210.
40.
ShenWChenYWuAA. Low-complexity sinusoidal-assisted EMD (SAEMD) algorithms for solving mode-mixing problems in HHT. Digit Signal Process2014; 24: 170–186.
41.
HintonGE. A practical guide to training restricted Boltzmann machines. In: MontavonGOrrGBMüllerKR (eds) Neural networks: tricks of the trade. Berlin: Springer, 2012, pp. 599–619.
42.
HopfieldJJ. Neural networks and physical systems with emergent collective computational abilities. Proc Nat Acad Sci U S A1982; 79: 2554–2558.
43.
BottouL. Large-scale machine learning with stochastic gradient descent. In: LechevallierYSaportaG (eds) Proceedings of COMPSTAT’2010. London: Springer, pp. 177–186.
44.
HintonGE. Training products of experts by minimizing contrastive divergence. Neural Comput2002; 14(8): 1771–1800.
