Towards smart monitoring of systems: an integrated non-parametric Bayesian KDE and LSTM approach for anomaly detection of rotating machinery under uncertainties
Restricted accessResearch articleFirst published online May, 2023
Towards smart monitoring of systems: an integrated non-parametric Bayesian KDE and LSTM approach for anomaly detection of rotating machinery under uncertainties
The Bayesian inference metric is derived from the normality assumption of data, and is widely used in engineering applications. Owing to data scarcity and uncertainty, the assumption is usually violated, leading to inaccurate results in anomaly detection of large rotating machinery. This paper proposes a novel non-parametric Bayesian evaluation metric based on kernel density estimation of data to avoid the normality assumption. Based on kernel density estimation, a non-parametric Bayesian evaluation metric is mathematically derived from both null and alternative hypothesis tests to improve its accuracy and generality without any distribution assumption. It provides a quantitative indicator for assessing the status of a rotating machines under data uncertainties. A comparison with the traditional method is conducted to demonstrate the effectiveness and feasibility of the proposed methodology in terms of normal and non-normal distributions of data. A general Bayesian evaluation procedure is proposed for the anomaly detection of rotating machinery. Sensor data collected from a real-world steam turbine were used to illustrate the effectiveness of the proposed method.
TranVTYangBS. An intelligent condition-based maintenance platform for rotating machinery. Expert Syst Appl2012; 39(3): 2977–2988.
2.
KumarKPRaoKKrishnaKR, et al. Neural network based vibration analysis with novelty in data detection for a large steam turbine. Shock Vibr2012; 19(1): 25–35.
3.
HiranoTGuoZKirkR G. Application of computational fluid dynamics analysis for rotating machinery—part II: labyrinth seal analysis. J Eng Gas Turbines Power2005; 127(4): 820–826.
4.
WangYXuGZhangQ, et al. Rotating speed isolation and its application to rolling element bearing fault diagnosis under large speed variation conditions. J Sound Vibr2015; 348: 381–396.
5.
HeWMiaoQAzarianM, et al. Health monitoring of cooling fan bearings based on wavelet filter. Mech Syst Signal Process2015; 64: 149–161.
6.
SelakLButalaPSlugaA. Condition monitoring and fault diagnostics for hydropower plants. Comput Industry2014; 65(6): 924–936.
7.
LiuRYangBZioE, et al. Artificial intelligence for fault diagnosis of rotating machinery: a review. Mech Syst Signal Process2018; 108: 33–47.
8.
ZuberNBajrićRŠostakovR. Gearbox faults identification using vibration signal analysis and artificial intelligence methods. Eksploatacja i Niezawodność2014; 16(1): 61–65.
9.
QiYShenCWangD, et al. Stacked sparse autoencoder-based deep network for fault diagnosis of rotating machinery. IEEE Access2017; 5: 15066–15079.
10.
ZaherAMcArthurSInfieldD, et al. Online wind turbine fault detection through automated SCADA data analysis. Wind Energy2009; 12: 574–593.
11.
SchlechtingenMFerreira SantosI. Comparative analysis of neural network and regression based condition monitoring approaches for wind turbine fault detection. Mech Syst Signal Process2011; 25: 1849–1875.
12.
KoizumiYKawaguchiYImotoK, et al. Description and discussion on DCASE2020 challenge task2: unsupervised anomalous sound detection for machine condition monitoring. arXiv preprint arXiv:2006.05822, 2020.
13.
MohanrajTShankarSRajasekarR, et al. Tool condition monitoring techniques in milling process—a review. J Mater Res Technol2020; 9(1): 1032–1042.
14.
ZhouYXueW. Review of tool condition monitoring methods in milling processes. Int J Adv Manuf Technol2018; 96(5): 2509–2523.
15.
WelzZCobleJUpadhyayaB, et al. Maintenance-based prognostics of nuclear plant equipment for long-term operation. Nucl Eng Technol2017; 49(5): 914–919.
16.
ChoudharyAGoyalDShimiSL, et al. Condition monitoring and fault diagnosis of induction motors: a review. Arch Comput Methods Eng2019; 26(4): 1221–1238.
17.
SalamehJPCauetSEtienE, et al. Gearbox condition monitoring in wind turbines: a review. Mech Syst Signal Process2018, 111: 251–264.
18.
StetcoADinmohammadiFZhaoX, et al. Machine learning methods for wind turbine condition monitoring: a review. Renewable Energy2019; 133: 620–635.
19.
LiuZZhangL. A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings. Measurement2020; 149: 107002.
AlmqvistO. A comparative study between algorithms for time series forecasting on customer prediction: an investigation into the performance of ARIMA, RNN, LSTM, TCN and HMM, Thesis, University of Skövde, 2019.
22.
HochreiterSSchmidhuberJ. Long short-term memory. Neural Comput1997; 9(8): 1735–1780.
XueZYXiahouKSLiMS, et al. Diagnosis of multiple open-circuit switch faults based on long short-term memory network for DFIG-based wind turbine systems. IEEE J Emerg Sel Top Power Electron2019; 8(3): 2600–2610.
25.
HungS-LHuangC-SWenC-M, et al. Nonparametric identification of a building structure from experimental data using wavelet neural network. Computer-Aided Civ Infrastruct Eng2003; 18(5): 358–370.
26.
RederMDGonzalezEMeleroJJ. Wind turbine failures – Tackling current problems in failure data analysis. J Phys Conf Ser2016; 753: 072027.
27.
LingJLiuGJLiJL, et al. Fault prediction method for nuclear power machinery based on Bayesian PPCA recurrent neural network model. Nucl Sci Tech2020; 31(8): 1–11.
28.
JiangXMahadevanS. Bayesian wavelet methodology for structural damage detection. Struct Control Health Monit2008; 15(7): 974–991.
29.
JiangXMahadevanS. Bayesian probabilistic inference for nonparametric damage detection of structures. J Eng Mech2008; 134(10): 820–831.
30.
MahadevanSRebbaR. Validation of reliability computational models using Bayes networks. Reliab Eng Syst Saf2005; 87(2): 223–232.
31.
LiuGGuHShenX, et al. Bayesian long short-term memory model for fault early warning of nuclear power turbine. IEEE Access2020; 99: 1–1.
32.
MurakamiYMizuguchiK. Applying the Naïve Bayes classifier with kernel density estimation to the prediction of protein–protein interaction sites. Bioinformatics2010; 26(15): 1841–1848.
33.
PérezALarrañagaPInzaI. Bayesian classifiers based on kernel density estimation: Flexible classifiers. Int J Approximate Reasoning2009; 50(2): 341–362.
34.
LiXYangXYangY, et al. An intelligent diagnostic and prognostic framework for large-scale rotating machinery in the presence of scarce failure data. Struct Health Monit2020; 19(5): 1375–1390.
35.
ZhangFLiuYChenC, et al. Fault diagnosis of rotating machinery based on kernel density estimation and Kullback-Leibler divergence. J Mech Sci Technol2014; 28(11): 4441–4454.
36.
SheatherSJJonesMC. A reliable data-based bandwidth selection method for kernel density estimation. J Royal Stat Soc Series B (Methodological)1991; 53(3): 683–690.
37.
RosenblattM. Remarks on a multivariate transformation. Ann Math Stat1952; 23(3): 470–472.
38.
NatafA. Determination des distribution don t les marges sont donnees. Comptes Rendus de l Academie des Sci1962; 225: 42–43.
39.
BoxGEPCoxDR. An analysis of transformations. J R Stat Soc B1964; 26(2): 211–243.
40.
PericchiLR. A Bayesian approach to transformations to normality. Biometrika1981; 68(1): 35–43.
41.
KingmaDPBaJ. Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
42.
RebbaRMahadevanS. Validation of models with multivariate output. Reliab Eng Syst Saf2006; 91(8): 861–871.
43.
XuSJiangXHuangJ, et al. Bayesian wavelet PCA methodology for turbomachinery damage diagnosis under uncertainty. Mech Syst Signal Process2016; 80: 1–18.
44.
YoungIT. Proof without prejudice: use of the Kolmogorov-Smirnov test for the analysis of histograms from flow systems and other sources. J Histochem Cytochem1977; 25(7): 935–941.
