Endüstriyel sistemlerde sağlam rulman arıza teşhisi için çok ölçekli entropi özellik çıkarımının hafif sınıflandırıcılarla bütünleştirilmesi

Öz

Arka Plan—Rulmanlar, endüstriyel makinelerin en kritik bileşenleri arasında yer almakta olup, arızaları beklenmedik duruşlara, ekonomik kayıplara ve güvenlik risklerine neden olabilmektedir. Endüstri 4.0 uygulamalarının yaygınlaşmasıyla birlikte, değişen endüstriyel koşullarda çalışabilen, güvenilir, hesaplama açısından verimli ve gerçek zamanlı arıza teşhis sistemlerine duyulan ihtiyaç artmaktadır.

Amaç—Bu çalışmanın amacı, çok ölçekli entropi tabanlı özellik çıkarım yöntemlerini geleneksel makine öğrenmesi sınıflandırıcılarıyla bütünleştirerek sağlam ve hafif bir rulman arıza teşhis çerçevesi geliştirmektir. Çalışmada, farklı entropi ölçülerinin değişen çalışma koşullarında doğru rulman arıza sınıflandırmasındaki etkinliği, sağlamlığı ve parametre duyarlılığının sistematik olarak değerlendirilmesi ve aynı zamanda uç bilişim ve gerçek zamanlı endüstriyel izleme uygulamalarına uygunluğunun ortaya konulması amaçlanmıştır.

Yöntem—Önerilen çerçevede, Çok Ölçekli Örnek Entropisi (MSE), Çok Ölçekli Bulanık Entropi (MFE) ve Çok Ölçekli Permütasyon Entropisi (MPE) olmak üzere üç çok ölçekli entropi yöntemi; k-En Yakın Komşular, Destek Vektör Makineleri, Lojistik Regresyon, Geri Yayılımlı Sinir Ağı, Aşırı Öğrenme Makinesi ve Softmax Regresyon olmak üzere altı hafif sınıflandırma algoritmasıyla birleştirilmiştir. Deneysel değerlendirmeler, Case Western Reserve University (CWRU) tahrik tarafı rulman veri seti ile gerçekçi çalışma koşullarında toplanan CISAR üç eksenli endüstriyel titreşim veri seti kullanılarak gerçekleştirilmiştir. Hiperparametre seçimi, adil model karşılaştırması sağlamak ve aşırı öğrenmeyi önlemek amacıyla tabakalı 10 katlı çapraz doğrulama ile gerçekleştirilmiştir. Ayrıca, entropi ölçek faktörü 6 ile 18 arasında değiştirilerek bu parametrenin sınıflandırma performansı ve özellik kararlılığı üzerindeki etkisini incelemek amacıyla duyarlılık analizi gerçekleştirilmiştir.

Bulgular—Deneysel sonuçlar, farklı arıza türleri, arıza çapları ve çalışma koşulları arasında genel olarak yüksek sınıflandırma performansı elde edildiğini göstermiştir. İncelenen özellik çıkarım yöntemleri arasında MFE, deneysel koşullar genelinde karşılaştırmalı olarak yüksek sınıflandırma doğruluğu ve kararlı bir performans sergilemiş ve birden fazla sınıflandırıcı kombinasyonunda sınıflandırma doğruluğu sıklıkla %99'un üzerine çıkmıştır. CWRU veri setinde bazı özellik-sınıflandırıcı kombinasyonlarında kusursuz sınıflandırma performansı elde edilmiştir. Duyarlılık analizi sonucunda, en uygun ölçek faktörü aralığının 12 ile 14 arasında olduğu belirlenmiştir. MSE ve MPE ile karşılaştırıldığında MFE'nin ölçek faktöründeki değişimlere karşı daha düşük duyarlılık gösterdiği ve bu durumun çalışma koşullarındaki değişkenlik ve sinyal gürültüsüne karşı daha güçlü bir sağlamlığa işaret ettiği görülmüştür.

Sonuç— Önerilen çerçeve, kaynakları kısıtlı izleme ortamlarında uygulanma potansiyeline sahip, doğru ve hafif bir rulman arıza teşhis yaklaşımı sunmaktadır. Çok ölçekli entropi özelliklerinin hafif sınıflandırıcılarla bütünleştirilmesi, değişen endüstriyel koşullarda güvenilir kestirimci bakım uygulamalarını desteklemektedir. Elde edilen sonuçlar, MFE tabanlı özellik çıkarımının deneysel olarak daha yüksek performans sergilediğini ortaya koymakta ve sağlam Endüstri 4.0 arıza izleme sistemlerinin geliştirilmesi için pratik bir ölçek faktörü aralığı sunmaktadır.

Anahtar Kelimeler

• Rulman arıza teşhisi
• Aşırı öğrenme makinesi
• Özellik çıkarımı
• Çok ölçekli entropi
• Ölçek faktörü duyarlılığı
• Titreşim analizi

Integrating multiscale entropy feature extraction with lightweight classifiers for robust bearing fault diagnosis in industrial systems

Öz

Context—Rolling-element bearings are among the most critical components in industrial machinery, and their failures may result in unexpected downtime, economic losses, and safety risks. As Industry 4.0 applications continue to expand, there is an increasing demand for reliable, computationally efficient, and real-time fault diagnosis systems capable of operating under varying industrial conditions.

Objective—This study aims to develop a robust and lightweight bearing fault diagnosis framework by integrating multiscale entropy-based feature extraction methods with conventional machine learning classifiers. The objective is to systematically evaluate the effectiveness, robustness, and parameter sensitivity of different entropy measures for accurate bearing fault classification under diverse operating conditions while maintaining suitability for edge-computing and real-time industrial monitoring applications.

Method—The proposed framework combines three multiscale entropy methods—Multiscale Sample Entropy, Multiscale Fuzzy Entropy, and Multiscale Permutation Entropy—with six lightweight classification algorithms: k-Nearest Neighbors, Support Vector Machine, Logistic Regression, Backpropagation Neural Network, Extreme Learning Machine, and Softmax Regression. Experimental evaluations were conducted using two benchmark datasets: the Case Western Reserve University (CWRU) drive-end bearing dataset and the CISAR three-axis industrial vibration dataset collected under realistic conditions. Hyperparameter selection was performed through stratified 10-fold cross-validation to ensure fair model comparison and prevent overfitting. In addition, sensitivity analysis was conducted by varying the entropy scale factor between 6 and 18 to investigate its effect on classification performance and feature stability.

Results—Experimental findings demonstrate consistently high classification performance across different fault types, fault diameters, and operating conditions. Among the investigated feature extraction methods, MFE showed comparatively high classification accuracy and stable performance across the experimental settings, frequently exceeding 99% classification accuracy across multiple classifier combinations. Several classifier-feature pairs achieved perfect classification performance on the CWRU dataset. The sensitivity analysis revealed that the optimal scale-factor range lies between τ = 12 and 14. Compared with MSE and MPE, MFE showed descriptively lower sensitivity to scale-factor variation, indicating stronger robustness against operational variability and signal noise.

Conclusion—The proposed framework provides an accurate and lightweight approach for bearing fault diagnosis, with potential applicability to resource-constrained monitoring environments. The integration of multiscale entropy features with lightweight classifiers enables reliable predictive maintenance under varying industrial conditions. The results highlight the empirically higher performance of MFE-based feature extraction and provide practical guidance for developing robust Industry 4.0 fault monitoring systems.

Anahtar Kelimeler

• Bearing fault diagnosis
• Extreme learning machine
• Feature extraction
• Multiscale entropy
• Scale–factor sensitivity
• Vibration analysis

Kaynakça

1. K. C. Rath, A. Khang, D. Roy, “The role of internet of things (IoT) technology in industry 4.0 economy”, Advanced IoT Technologies and Applications in the Industry 4.0 Digital Economy, 1st Edition, A. Khang, V. Abdullayev, V. Hahanov, V. Shah, Eds. Boca Raton, Florida, USA, CRC Press, 2024, Ch. 1. https://doi.org/10.1201/9781003434269.
2. N. D. Thuan, H. S. Hong, “HUST bearing: a practical dataset for ball bearing fault diagnosis”, BMC Research Notes, 16(1), 138, 2023. https://doi.org/10.1186/s13104-023-06400-4.
3. C. Wang, J. Yang, H. Jie, Z. Zhao, W. Wang, “An energy-efficient mechanical fault diagnosis method based on neural dynamics-inspired metric SpikingFormer for insufficient samples in industrial internet of things”, IEEE Internet of Things Journal, 12(1), 1081-1097, 2025. https://doi.org/10.1109/JIOT.2024.3476034.
4. Y. Shang, X. Tang, G. Zhao, P. Jiang, T. R. Lin, “A remaining life prediction of rolling element bearings based on a bidirectional gate recurrent unit and convolution neural network”, Measurement, 202, 111893, 2022. https://doi.org/10.1016/j.measurement.2022.111893.
5. Q. Ni, J. C. Ji, K. Feng, “Data-driven prognostic scheme for bearings based on a novel health indicator and gated recurrent unit network”, IEEE Transactions on Industrial Informatics, 19(2), 1301-1311, 2023. https://doi.org/10.1109/TII.2022.3169465.
6. J. Guo, Z. Li, M. Li, “A review on prognostics methods for engineering systems”, IEEE Transactions on Reliability, 69(3), 1110–1129, 2020. https://doi.org/10.1109/TR.2019.2957965.
7. M. A. Abbasi, S. P. Huang, A. S. Khan, “Fault detection and classification of motor bearings under multiple operating conditions”, ISA Transactions, 156, 61–69, 2025. https://doi.org/10.1016/j.isatra.2024.11.008.
8. J. Jiao, M. Zhao, J. Lin, K. Liang, “A comprehensive review on convolutional neural network in machine fault diagnosis”, Neurocomputing, 417, 36–63, 2020. https://doi.org/10.1016/j.neucom.2020.07.088.

9. Y. Wang, J. Wu, Y. Cheng, J. Wang, K. Hu, “Memory-enhanced hybrid deep learning networks for remaining useful life prognostics of mechanical equipment”, Measurement, 187, 110354, 2022. https://doi.org/10.1016/j.measurement.2021.110354.
10. M. A. Abbasi, S. Huang, “Development of a fault diagnostics and tolerance system: An application to continuous stirred tank reactor”, Measurement Science and Technology, 35(6), 066203, 2024. https://doi.org/10.1088/1361-6501/ad2da7.
11. Q. Ni, J. C. Ji, K. Feng, Y. Zhang, D. Lin, J. Zheng, “Data-driven bearing health management using a novel multi-scale fused feature and gated recurrent unit”, Reliability Engineering & System Safety, 242, 109753, 2024. https://doi.org/10.1016/j.ress.2023.109753.
12. Y. Lei, B. Yang, X. Jiang, F. Jia, N. Li, A. K. Nandi, “Applications of machine learning to machine fault diagnosis: A review and roadmap”, Mechanical Systems and Signal Processing, 138, 106587, 2020. https://doi.org/10.1016/j.ymssp.2019.106587.
13. S. Chikkam, S. Singh, “Condition monitoring and fault diagnosis of induction motor using DWT and ANN”, Arabian Journal for Science and Engineering, 48(5), 6237–6252, 2023. https://doi.org/10.1007/s13369-022-07294-3.
14. Y. Li, X. Wang, S. Si, S. Huang, “Entropy based fault classification using the Case Western Reserve University data: A benchmark study”, IEEE Transactions on Reliability, 69(2), 754–767, 2020. https://doi.org/10.1109/TR.2019.2896240.
15. P. Wang, H. Xiong, H. He, “Bearing fault diagnosis under various conditions using an incremental learning-based multi-task shared classifier”, Knowledge-Based Systems, 266, 110395, 2023. https://doi.org/10.1016/j.knosys.2023.110395.
16. P. K. Kankar, S. C. Sharma, S. P. Harsha, “Fault diagnosis of ball bearings using machine learning methods”, Expert Systems with Applications, 38(3), 1876–1886, 2011. https://doi.org/10.1016/j.eswa.2010.07.119.
17. L. Frosini, C. Harlisca, L. Szabo, “Induction machine bearing fault detection by means of statistical processing of the stray flux measurement”, IEEE Transactions on Industrial Electronics, 62(3), 1846–1854, 2015. https://doi.org/10.1109/TIE.2014.2361115.
18. R. Zhao, R. Yan, Z. Chen, K. Mao, P. Wang, R. X. Gao, “Deep learning and its applications to machine health monitoring”, Mechanical Systems and Signal Processing, 115, 213–237, 2019. https://doi.org/10.1016/j.ymssp.2018.05.050.
19. Y. Yang, M. M. M. Haque, D. Bai, W. Tang, “Fault diagnosis of electric motors using deep learning algorithms and its application: A review”, Energies, 14(21), 7017, 2021. https://doi.org/10.3390/en14217017.
20. J. Yin, G. Cen, “Intelligent motor bearing fault diagnosis using channel attention-based CNN”, World Electric Vehicle Journal, 13(11), 208, 2022. https://doi.org/10.3390/wevj13110208.
21. G. Li, Y. Chen, W. Wang, Y. Wu, R. Liu, “Automatic Transmission Bearing Fault Diagnosis Based on Comprehensive Index Method and Convolutional Neural Network”, World Electric Vehicle Journal, 13(10), 184, 2022. https://doi.org/10.3390/wevj13100184.
22. S. Zhang, S. B. Zhang, B. Wang, T. G. Habetler, “Deep learning algorithms for bearing fault diagnostics—A comprehensive review”, IEEE Access, 8, 29857–29881, 2020. https://doi.org/10.1109/ACCESS.2020.2972859.
23. Z. Liu, P. Zhang, Y. Yu, M. Li, Z. Zeng, “A novel fault diagnosis model of rolling bearing under variable working conditions based on attention mechanism and domain adversarial neural network”, Journal of Mechanical Science and Technology, 38(3), 1101–1111, 2024. https://doi.org/10.1007/s12206-024-1208-8
24. L. H. Wang, X. P. Zhao, J. X. Wu, Y. Y. Xie, Y. H. Zhang, “Motor fault diagnosis based on short-time Fourier transform and convolutional neural network”, Chinese Journal of Mechanical Engineering, 30(6), 1357–1368, 2017. https://doi.org/10.1007/s10033-017-0190-5.
25. C. Deng, Z. Deng, S. Lu, M. He, J. Miao, Y. Peng, “Fault diagnosis method for imbalanced data based on multi-signal fusion and improved deep convolution generative adversarial network”, Sensors, 23(5), 2542, 2023. https://doi.org/10.3390/s23052542.
26. F. Chen, X. Zhou, B. Xu, Z. Yang, Z. Qu, “Instantaneous square current signal analysis for motors using vision transformer for the fault diagnosis of rolling bearings”, Applied Sciences, 13(16), 9349, 2023. https://doi.org/10.3390/app13169349.
27. G. Niu, X. Dong, Y. Chen, “Motor fault diagnostics based on current signatures: A review”, IEEE Transactions on Instrumentation and Measurement, 72, 2023. https://doi.org/10.1109/TIM.2023.3285999.
28. Z. Wang, C. Guan, S. Shi, G. Zhang, X. Gu, “Motor Bearing Fault Diagnosis Based on Current Signal Using Time–Frequency Channel Attention”, World Electric Vehicle Journal, 15(7), 281, 2024. https://doi.org/10.3390/wevj15070281.
29. M. A. Abbasi, S. Huang, A. S. Khan, “Hybrid fault diagnostics and prognostics system development for motor bearings health monitoring”, Quality Engineering, 37(2), 292–307, 2025. https://doi.org/10.1080/08982112.2024.2385920.
30. P. Nectoux, R. Gouriveau, K. Medjaher, E. Ramasso, B. Chebel-Morello, N. Zerhouni, C. Varnier, “PRONOSTIA: An experimental platform for bearings accelerated degradation tests”, IEEE International Conference on Prognostics and Health Management (PHM'12), Denver, Colorado, USA, June 2012, https://hal.science/hal-00719503v1.
31. W. A. Smith, R. B. Randall, “Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study”, Mechanical Systems and Signal Processing, 64-65, 100–131, 2015. https://doi.org/10.1016/j.ymssp.2015.04.021.
32. Y. Su, L. Shi, K. Zhou, G. Bai, Z. Wang, “Knowledge-informed deep networks for robust fault diagnosis of rolling bearings”, Reliability Engineering & System Safety, 244, 109863, 2024. https://doi.org/10.1016/j.ress.2023.109863.
33. V. Pandhare, J. Singh, J. Lee, “Convolutional neural network based rolling-element bearing fault diagnosis for naturally occurring and progressing defects using time-frequency domain features”, Proceedings of the Prognostics and System Health Management Conference (PHM-Paris), Paris, France, 02-05 May 2019, 320–326. https://doi.org/10.1109/PHM-Paris.2019.00061.
34. Case Western Reserve University, “Bearing data center test seeded fault test data”, https://engineering.case.edu/bearingdatacenter/download-data-file (14.09.2026).
35. E. Irgat, E. Çinar, A. Ünsal, A. Yazıcı, “An IoT-based monitoring system for induction motor faults utilizing deep learning models”, Journal of Vibration Engineering & Technologies,11(7), 3579–3589, 2023. https://doi.org/10.1007/s42417-022-00769-5.
36. W. Chen, J. Zhuang, W. Yu, Z. Wang, “Measuring complexity using fuzzyen, apen, and sampen”, Medical Engineering & Physics, 31(1), 61–68, 2009. https://doi.org/10.1016/j.medengphy.2008.04.005.
37. C. S. Park, Y. C. Choi, Y. H. Kim, “Early fault detection in automotive ball bearings using the minimum variance cepstrum”, Mechanical Systems and Signal Processing, 38(2), 534–548, 2013. https://doi.org/10.1016/j.ymssp.2013.02.017
38. J. Zheng, J. Cheng, Y. Yang, S. Luo, “A rolling bearing fault diagnosis method based on multi-scale fuzzy entropy and variable predictive model-based class discrimination”, Mechanism and Machine Theory, 78, 187–200, 2014. https://doi.org/10.1016/j.mechmachtheory.2014.03.014.
39. Y. Li, M. Xu, R. Wang, W. Huang, “A fault diagnosis scheme for rolling bearing based on local mean decomposition and improved multiscale fuzzy entropy”, Journal of Sound and Vibration, 360, 277–299, 2016. https://doi.org/10.1016/j.jsv.2015.09.016.
40. M. Costa, A. L. Goldberger, C. K. Peng, “Multiscale entropy analysis of complex physiologic time series”, Physical Review Letters, 89(6), 068102, 2002. https://doi.org/10.1103/PhysRevLett.89.068102.
41. F. C. Morabito, D. Labate, F. L. Foresta, A. Bramanti, G. Morabito, I. Palamara, “Multivariate multi-scale permutation entropy for complexity analysis of Alzheimer’s disease EEG”, Entropy, 14(7), 1186–1202, 2012. https://doi.org/10.3390/e14071186.
42. Y. Li, Q. Gao, B. Miao, W. Zhang, J. Liu, Y. Zhu, “Application of the refined multiscale permutation entropy method to fault detection of rolling bearing”, Journal of the Brazilian Society of Mechanical Sciences and Engineering, 43(5), 280, 2021. https://doi.org/10.1007/s40430-021-02986-7.
43. M. W. Flood, B. Grimm, “Entropyhub: An opensource toolkit for entropic time series analysis”, PloS One, 16(11), e0259448, 2021. https://doi.org/10.1371/journal.pone.0259448.
44. F. Pedregosa et al., “Scikit-learn: Machine learning in Python”, Journal of Machine Learning Research, 12, 2825–2830, 2011.
45. J. Wang, P. Neskovic, L. N. Cooper, “Neighborhood size selection in the k-nearest-neighbor rule using statistical confidence”, Pattern Recognition, 39(3), 417–423, 2006. https://doi.org/10.1016/j.patcog.2005.08.009.
46. S. R. Madeti, S. N. Singh, “Modeling of PV system based on experimental data for fault detection using kNN method”, Solar Energy, 173, 139–151, 2018. https://doi.org/10.1016/j.solener.2018.07.038.
47. N. Cristianini, J. Shawe-Taylor, “An introduction to support vector machines and other kernel-based learning methods”, Cambridge University Press, 2000. https://doi.org/10.1017/CBO9780511801389.
48. U. B. Parikh, B. Das, R. Maheshwari, “Fault classification technique for series compensated transmission line using support vector machine”, International Journal of Electrical Power & Energy Systems, 32(6), 629–636, 2010. https://doi.org/10.1016/j.ijepes.2009.11.020.
49. P. Barandier, M. Mendes, A. J. M. Cardoso, “Comparative analysis of four classification algorithms for fault detection of heat pumps”, Energy and Buildings, 316, 114342, 2024. https://doi.org/10.1016/j.enbuild.2024.114342.
50. Y. Li, W. Dong, Q. Yang, J. Zhao, L. Liu, S. Feng, “An automatic impedance matching method based on the feedforward-backpropagation neural network for a WPT system”, IEEE Transactions on Industrial Electronics, 66(5), 3963–3972, 2019. https://doi.org/10.1109/TIE.2018.2835410.
51. G. B. Huang, Q. Y. Zhu, C. K. Siew, “Extreme learning machine: theory and applications”, Neurocomputing, 70(1-3), 489–501, 2006. https://doi.org/10.1016/j.neucom.2005.12.126.
52. A. V. N. Reddy, C. P. Krishna, P. K. Mallick, “An image classification framework exploring the capabilities of extreme learning machines and artificial bee colony”, Neural Computing and Applications, 32(8), 3079–3099, 2020. https://doi.org/10.1007/s00521-019-04385-5.
53. X. Gao, G. Yuan, M. Zhang, “Fault detection of electric vehicle charging piles based on extreme learning machine algorithm”, Proceedings of the 2020 Fourth International Conference on Computing Methodologies and Communication (ICCMC), Erode, India, 11-13 March 2020, 849–852. https://doi.org/10.1109/ICCMC48092.2020.ICCMC-000157.
54. M. Jiang, Y. Liang, X. Feng, X. Fan, Z. Pei, Y. Xue, R. Guan, “Text classification based on deep belief network and softmax regression”, Neural Computing and Applications, 29(1), 61–70, 2018. https://doi.org/10.1007/s00521-016-2401-x.