[Diagnostic intelligent des défauts de roulements en conditions non stationnaires basé sur le rééchantillonnage angulaire et les machines à vecteurs de support]
Au cours des dernières décennies, les méthodes intelligentes de diagnostic des roulements sont devenues un sujet de recherche majeur. Ces méthodes nécessitent la construction d’un vecteur forme (VF), généralement composé d’indicateurs calculés à partir de signaux de vibration échantillonnés à pas de temps constant. Cependant, en conditions non stationnaires, ces signaux exigent l’application de méthodes complexes, dont le temps de calcul est particulièrement élevé. C’est pourquoi l’utilisation des techniques de rééchantillonnage angulaire est recommandée, car elles permettent de s’affranchir des fluctuations de vitesse et d’employer des méthodes de traitement plus simples.
Dans ce travail, nous proposons d’utiliser des signaux d’accélération rééchantillonnés angulairement pour le diagnostic intelligent des défauts de roulements en conditions non stationnaires. Il s’agit de comparer trois types de vecteurs formes : des indicateurs angulaires classiques sur les signaux angulaires x(θ), des indicateurs spectraux d’ordre originaux (amplitude de pic) ou la combinaison de ces deux familles d’indicateurs. Ensuite, la phase de sélection est réalisée à l’aide de l’algorithme Minimum Redundancy Maximum Relevance (MRMR) afin de choisir les indicateurs les plus pertinents. Enfin, la classification est effectuée par la méthode machine à vecteurs de support cubique (SVM Cubique) pour les étapes de détection et d’identification des différentes conditions de défaut des roulements. L’efficacité de la méthode proposée permet d’atteindre un taux de classification parfait de 100 %.
Over the past decades, intelligent bearing diagnostic methods have become a research hotspot. These methods require the construction of a Feature Vector (FV) generally composed of indicators calculated from time sampled vibration signals. However, in non-stationary conditions, these signals require the application of complex methods whose calculation time is really important. For this reason, the use of angular resampling techniques is recommended because they make it possible to get rid of speed fluctuations and to employ simple processing methods.
In this work, we propose to use angularly resampled acceleration signals for intelligent bearing fault diagnosis under non-stationary conditions. It is the question of comparing three types of FV: classic angular indicators on angular signals x(θ), original order spectrum indicators (peak amplitude), or the combination of the two previous families of indicators. Then, the selection phase is performed by the Minimum Redundancy Maximum Relevance (MRMR) algorithm to select the most relevant features. Finally, the classification is carried out by a cubic support vector machine (SVM) for the detection and identification stages of various bearings fault conditions. The effectiveness of the proposed method achieves a perfect classification rate of 100%.
Révisé le :
Accepté le :
Publié le :
Mots-clés : Roulements, Diagnostic des défauts, Machines à vecteurs de support, Vitesse de rotation variable, Rééchantillonnage angulaire
Fakhreddine Bouali 1 ; Semchedine Fedala 1 ; Hugo André 2 ; Ahmed Felkaoui 1
CC-BY 4.0
@article{CRMECA_2025__353_G1_499_0,
author = {Fakhreddine Bouali and Semchedine Fedala and Hugo Andr\'e and Ahmed Felkaoui},
title = {Intelligent bearing faults diagnosis in non-stationary conditions based on angular resampling and support vector machine},
journal = {Comptes Rendus. M\'ecanique},
pages = {499--518},
publisher = {Acad\'emie des sciences, Paris},
volume = {353},
year = {2025},
doi = {10.5802/crmeca.292},
language = {en},
}
TY - JOUR AU - Fakhreddine Bouali AU - Semchedine Fedala AU - Hugo André AU - Ahmed Felkaoui TI - Intelligent bearing faults diagnosis in non-stationary conditions based on angular resampling and support vector machine JO - Comptes Rendus. Mécanique PY - 2025 SP - 499 EP - 518 VL - 353 PB - Académie des sciences, Paris DO - 10.5802/crmeca.292 LA - en ID - CRMECA_2025__353_G1_499_0 ER -
%0 Journal Article %A Fakhreddine Bouali %A Semchedine Fedala %A Hugo André %A Ahmed Felkaoui %T Intelligent bearing faults diagnosis in non-stationary conditions based on angular resampling and support vector machine %J Comptes Rendus. Mécanique %D 2025 %P 499-518 %V 353 %I Académie des sciences, Paris %R 10.5802/crmeca.292 %G en %F CRMECA_2025__353_G1_499_0
Fakhreddine Bouali; Semchedine Fedala; Hugo André; Ahmed Felkaoui. Intelligent bearing faults diagnosis in non-stationary conditions based on angular resampling and support vector machine. Comptes Rendus. Mécanique, Volume 353 (2025), pp. 499-518. doi : 10.5802/crmeca.292. https://comptes-rendus.academie-sciences.fr/mecanique/articles/10.5802/crmeca.292/
[1] A survey on fault diagnosis of rotating machinery based on machine learning, Meas. Sci. Technol, Volume 35 (2024) no. 10, 102001 | DOI
[2] Fault diagnosis of various rotating equipment using machine learning approaches—a review, Proc. Inst. Mech. Eng. E: J. Process Mech. Eng., Volume 235 (2021) no. 2, pp. 629-642 | DOI
[3] Fault diagnosis of rolling element bearing using continuous wavelet transform and K-nearest neighbour, Mater. Today: Proc., Volume 92 (2023), pp. 56-60 | DOI
[4] Time to failure prediction of rotating machinery using dynamic feature extraction and gaussian process regression, Int. J. Adv. Manuf. Technol., Volume 130 (2024) no. 5, pp. 2939-2955 | DOI
[5] Prognostics of rotating machines through generalized Gaussian hidden Markov models, Mech. Syst. Signal Process., Volume 185 (2023), 109767 | DOI
[6] Convolutional neural network-based hidden markov models for rolling element bearing fault identification, Knowl.-Based Syst., Volume 144 (2018), pp. 65-76 | DOI
[7] Remaining useful life prediction of rolling bearings based on Bayesian neural network and uncertainty quantification, Qual. Reliab. Eng. Int., Volume 39 (2023) no. 5, pp. 1756-1774 | DOI
[8] Highly accurate gear fault diagnosis based on support vector machine, J. Vib. Eng. Technol., Volume 11 (2023) no. 7, pp. 3565-3577 | DOI
[9] Fault diagnosis of rotating machinery based on multi-class support vector machines, J. Mech. Sci. Technol., Volume 19 (2005) no. 3, pp. 846-859 | DOI
[10] Contribution of angular measurements to intelligent gear faults diagnosis, J. Intell. Manuf., Volume 29 (2018), pp. 1115-1131 | DOI
[11] Bearing fault diagnosis using multiclass support vector machines with binary particle swarm optimization and regularized Fisher’s criterion, J. Intell. Manuf., Volume 28 (2017), pp. 405-417 | DOI
[12] Rolling element bearing diagnostics—a tutorial, Mech. Syst. Signal Process., Volume 25 (2011) no. 2, pp. 485-520 | DOI
[13] The spectral kurtosis: a useful tool for characterising non-stationary signals, Mech. Syst. Signal Process., Volume 20 (2006) no. 2, pp. 282-307 | DOI
[14] Rotating machinery fault diagnosis based on typical resonance demodulation methods: a review, IEEE Sens. J., Volume 23 (2023) no. 7, pp. 6439-6459 | DOI
[15] Rotating machinery fault diagnosis under time-varying speeds: a review, IEEE Sens. J., Volume 23 (2023) no. 24, pp. 29969-29990 | DOI
[16] et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis, Proc. R. Soc. Lond. A, Volume 454 (1998) no. 1971, pp. 903-995 | DOI
[17] A study of the characteristics of white noise using the empirical mode decomposition method, Proc. R. Soc. Lond. A, Volume 460 (2004) no. 2046, pp. 1597-1611 | DOI
[18] The local mean decomposition and its application to EEG perception data, J. R. Soc. Interface R. Soc., Volume 2 (2005) no. 5, pp. 443-454 | DOI
[19] Gear fault diagnosis under non-stationary operating mode based on EMD, TKEO, and shock detector, C. R. Mec., Volume 347 (2019) no. 9, pp. 663-675 | DOI
[20] A rolling bearing fault diagnosis approach based on LCD and fuzzy entropy, Mech. Mach. Theory, Volume 70 (2013), pp. 441-453 | DOI
[21] A new automated classification framework for gear fault diagnosis using fourier–bessel domain-based empirical wavelet transform, Machines, Volume 11 (2023) no. 12, 1055 | DOI
[22] An IGSA-VMD method for fault frequency identification of cylindrical roller bearing, Proc. Inst. Mech. Eng. Part C: J. Mech. Eng. Sci., Volume 238 (2024) no. 18, pp. 9307-9320 | DOI
[23] Adaptive local iterative filtering for signal decomposition and instantaneous frequency analysis, Appl. Comput. Harmon. Analysis, Volume 41 (2016) no. 2, pp. 384-411 | DOI
[24] A tacho-less order tracking technology for large speed variations, Mech. Syst. Signal Process., Volume 40 (2013) no. 1, pp. 76-90 | DOI
[25] Practical performances of high-speed measurement of gear transmission error or torsional vibrations with optical encoders, Meas. Sci. Technol., Volume 9 (1998) no. 3, pp. 347-353 | DOI
[26] Comparison between angular sampling and angular resampling methods applied on the vibration monitoring of a gear meshing in non stationary conditions, Proceedings of the International Conference on Noise and Vibration Engineering: ISMA2010 including USD2010, Leuven, Belgium, Curran Associates, Inc., 2010, pp. 2727-2736
[27] Use of the acceleration of a gearbox in order to perform angular resampling (with limited speed fluctuation), Mech. Syst. Signal Process., Volume 19 (2005) no. 4, pp. 766-785 | DOI
[28] Analysis of computed order tracking, Mech. Syst. Signal Process., Volume 11 (1997) no. 2, pp. 187-205 | DOI
[29] A fault diagnosis method for nuclear power plant rotating machinery based on adaptive deep feature extraction and multiple support vector machines, Progr. Nucl. Energy, Volume 164 (2023), 104862 | DOI
[30] Fault detection and classification of the rotor unbalance based on dynamics features and support vector machine, Meas. Control, Volume 56 (2023) no. 5–6, pp. 1075-1086 | DOI
[31] A two-stage fault diagnosis methodology for rotating machinery combining optimized support vector data description and optimized support vector machine, Measurement, Volume 200 (2022), 111651 | DOI
[32] Application of SVM and ANN for image retrieval, Eur. J. Oper. Res., Volume 173 (2006) no. 3, pp. 938-950 | DOI
[33] A comparison of methods for multi-class support vector machines, IEEE Trans. Neural Netw., Volume 13 (2002) no. 2, pp. 415-425 | DOI
[34] Gear fault diagnosis based on angular measurements and support vector machines in normal and nonstationary conditions, Advances in Condition Monitoring of Machinery in Non-Stationary Operations. CMMNO 2014. Applied Condition Monitoring (F. Chaari; R. Zimroz; W. Bartelmus; M. Haddar, eds.), Volume 4, Springer International Publishing, Cham, 2015, pp. 291-308 | DOI
[35] Bearing vibration data collected under time-varying rotational speed conditions, Data in Brief, Volume 21 (2018), pp. 1745-1749 | DOI
[36] Bearing Vibration Data under Time-varying Rotational Speed Conditions, University of Ottawa, Canada, 2019 | DOI
[37] Diagnostic Vibratoire en Maintenance Préventive, Dunod, Paris, 1998
[38] Machinery Condition Monitoring, Principles and Practices, CRC Press/Taylor & Francis Group, Boca Raton, 2014 | DOI
[39] Crack detection in a rotating shaft using artificial neural networks and PSD characterisation, Meccanica, Volume 49 (2014), pp. 255-266 | DOI
[40] A comprehensive study of the bias and variance of frequency-response-function measurements : Optimal window selection and overlapping strategies, Automatica, Volume 43 (2007) no. 10, pp. 1723-1736 | DOI
[41] Noise and Vibration Analysis: Signal Analysis and Experimental Procedures, John Wiley & Sons, Ltd, Chichester, 2011 | DOI
[42] Rolling element bearing diagnostics in run-to-faillure lifetime testing, Mech. Syst. Signal Process., Volume 15 (2001) no. 5, pp. 979-993 | DOI
[43] Blind deconvolution based on cyclostationarity maximization and its application to fault identification, J. Sound Vib., Volume 432 (2018), pp. 569-601 | DOI
[44] Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics, J. Sound Vib., Volume 289 (2006) no. 4–5, pp. 1066-1090 | DOI
[45] A two-stage blind deconvolution strategy for bearing fault vibration signals, Mech. Syst. Signal Process., Volume 134 (2019), 106307 | DOI
[46] Bearing condition diagnostics via vibration and acoustic emission measurments, Mech. Syst. Signal Process., Volume 11 (1997) no. 5, pp. 693-705 | DOI
[47] Order-based identification of bearing defects under variable speed condition, Appl. Sci., Volume 11 (2021) no. 9, 3962 | DOI
[48] Natural roller bearing fault detection by angular measurement of true instantaneous angular speed, Mech. Syst. Signal Process., Volume 24 (2010) no. 7, pp. 1998-2011 | DOI
[49] Bearing fault feature extraction and fault diagnosis method based on feature fusion, Sensors, Volume 21 (2021) no. 7, 2524 | DOI
[50] Pattern recognition based fault diagnosis in industrial processes: review and application, Pattern Recognition Recent Advances (A. Herout, ed.), IntechOpen, Rijeka, 2010, pp. 483-508 (ISBN: 978-953-7619-90-9) | DOI
[51] Wrapper for feature subset selection, Artif. Intell., Volume 97 (1997) no. 1–2, pp. 273-324 | DOI
[52] Estimation of the information by an adaptive partitioning of the observation space, IEEE Trans. Inform. Theory, Volume 45 (1999) no. 4, pp. 1315-1321 | DOI
[53] Minimum redundancy feature selection from microarray gene expression data, J. Bioinform. Comput. Biol., Volume 3 (2005) no. 2, pp. 185-205 | DOI
Cité par Sources :
Commentaires - Politique