TİTREŞİM SİNYALLERİNDEN RULMAN ARIZALARIN TESPİT EDİLMESİ

Öz

Döner makinelerdeki rulmanlar, sistem güvenliği ve operasyonel süreklilik açısından kritik öneme sahip mekanik bileşenlerdir. Bu çalışmada, Case Western Reserve University (CWRU) rulman veri seti kullanılarak dört farklı makine öğrenmesi algoritması olan Random Forest, XGBoost, Support Vector Machine (SVM) ve Naive Bayes modelleri ile rulman arıza sınıflandırması gerçekleştirilmiştir. Zaman alanında çıkarılan istatistiksel öznitelikler temelinde, her modelin doğruluk (accuracy), kesinlik (precision), duyarlılık (recall) ve F1-skoru metrikleri ile performansları değerlendirilmiş ve karşılaştırılmıştır. Bulgular, özellikle Random Forest ve XGBoost algoritmalarının %95.73 doğruluk ve %96 precision, recall ve F1-score ile üstün performans sergilediğini ortaya koymuştur. SVM modeli %93.73 doğrulukla güvenilir bir alternatif olarak değerlendirilirken, Naive Bayes algoritması %92.40 doğrulukla nispeten daha düşük performans göstermiştir. Ayrıca, istatistiksel özniteliklerin tekil sınıflandırma başarımı incelenmiş ve özellikle standart sapma (sd) ve RMS gibi özniteliklerin yüksek katkı sunduğu belirlenmiştir. Bu çalışma, geleneksel makine öğrenmesi algoritmalarının farklı öznitelik yapılarına göre performanslarını detaylı biçimde analiz ederek, rulman arızalarının erken ve doğru tespiti için karar vericilere yol gösterici bir referans sunmaktadır.

Anahtar Kelimeler

• Rulman arıza teşhisi
• Makine öğrenmesi
• Random Forest
• XGBoost
• Destek Vektör Makineleri
• Naive Bayes
• CWRU veri seti.

Detection of Bearing Faults from Vibration Signals

Öz

Bearings are critical mechanical components in rotating machinery, playing a vital role in system safety and operational continuity. In this study, the Case Western Reserve University (CWRU) bearing dataset is used to perform fault classification using four machine learning algorithms: Random Forest, XGBoost, Support Vector Machine (SVM), and Naive Bayes. Based on statistical features extracted in the time domain, the performance of each model is evaluated using accuracy, precision, recall, and F1-score metrics. The results reveal that Random Forest and XGBoost algorithms achieved superior performance with 95.73% accuracy and 96% in precision, recall, and F1-score. The SVM model, with 93.73% accuracy, stands out as a robust alternative, while the Naive Bayes algorithm shows relatively lower performance with 92.40% accuracy. Additionally, an individual feature-based classification analysis indicates that standard deviation (sd) and root mean square (RMS) features contribute most significantly to model performance. This study provides a comprehensive performance analysis of traditional machine learning algorithms, offering a valuable reference for early and accurate detection of bearing faults.

Anahtar Kelimeler

• Bearing fault diagnosis
• Machine learning
• Random Forest
• XGBoost
• Support Vector Machine
• Naive Bayes
• CWRU dataset.

Kaynakça

1. [1] Kaya, Y., Kuncan, M., Kaplan, K., Minaz, M. R., & Ertunç, H. M. (2021). A new feature extraction approach based on one dimensional gray level co-occurrence matrices for bearing fault classification. Journal of Experimental & Theoretical Artificial Intelligence, 33(1), 161-178.
2. [2] Kaya, Y., Kuncan, F., & ERTUNÇ, H. M. (2022). A new automatic bearing fault size diagnosis using time-frequency images of CWT and deep transfer learning methods. Turkish Journal of Electrical Engineering and Computer Sciences, 30(5), 1851-1867.
3. [3] Zhang, X., Zhang, M., Wan, S., He, Y., & Wang, X. (2021). A bearing fault diagnosis method based on multiscale dispersion entropy and GG clustering. Measurement, 185, 110023.
4. [4] Du, Y., Geng, X., Zhou, Q., & Cheng, S. (2024). A fault diagnosis method for offshore wind turbine bearing based on adaptive deep echo state network and bidirectional long short term memory network in noisy environment. Ocean Engineering, 312, 119101.
5. [5] Wang, P., Xiong, H., & He, H. (2023). Bearing fault diagnosis under various conditions using an incremental learning-based multi-task shared classifier. Knowledge-Based Systems, 266, 110395.
6. [6] Wang, Z., Shi, D., Xu, Y., Zhen, D., Gu, F., & Ball, A. D. (2023). Early rolling bearing fault diagnosis in induction motors based on on-rotor sensing vibrations. Measurement, 222, 113614.
7. [7] Gu, X., Yu, Y., Guo, L., Gao, H., & Luo, M. (2023). CSWGAN-GP: A new method for bearing fault diagnosis under imbalanced condition. Measurement, 217, 113014
8. [8] Li, F., Wang, L., Wang, D., Wu, J., & Zhao, H. (2023). An adaptive multiscale fully convolutional network for bearing fault diagnosis under noisy environments. Measurement, 216, 112993.

9. [9] Wen, L., Xie, X., Li, X., & Gao, L. (2022). A new ensemble convolutional neural network with diversity regularization for fault diagnosis. Journal of Manufacturing Systems, 62, 964-971.
10. [10] Su, Z., Zhang, J., Tang, J., Wang, Y., Xu, H., Zou, J., & Fan, S. (2023). A novel deep transfer learning method with inter-domain decision discrepancy minimization for intelligent fault diagnosis. Knowledge-Based Systems, 259, 110065.
11. [11] Meng, Z., He, H., Cao, W., Li, J., Cao, L., Fan, J., ... & Fan, F. (2023). A novel generation network using feature fusion and guided adversarial learning for fault diagnosis of rotating machinery. Expert Systems with Applications, 234, 121058.
12. [12] Li, X., Xiao, S., Zhang, F., Huang, J., Xie, Z., & Kong, X. (2024). A fault diagnosis method with AT-ICNN based on a hybrid attention mechanism and improved convolutional layers. Applied Acoustics, 225, 110191.
13. [13] Borghesani, P., Herwig, N., Antoni, J., & Wang, W. (2023). A Fourier-based explanation of 1D-CNNs for machine condition monitoring applications. Mechanical Systems and Signal Processing, 205, 110865.
14. [14] Xu, Y., Li, Z., Wang, S., Li, W., Sarkodie-Gyan, T., & Feng, S. (2021). A hybrid deep-learning model for fault diagnosis of rolling bearings. Measurement, 169, 108502.
15. [15] Li, Y. (2024). An accurate lightweight algorithm for bearings fault diagnosis based on DPW ATTCNN model. Physical Communication, 66, 102383.
16. [16] Wu, Z., Guo, J., Liu, Y., Li, L., & Ji, Y. (2024). An Iterative Resampling Deep Decoupling Domain Adaptation method for class-imbalance bearing fault diagnosis under variant working conditions. Expert Systems with Applications, 252, 124240.
17. [17] Gu, J., Peng, Y., Lu, H., Chang, X., & Chen, G. (2022). A novel fault diagnosis method of rotating machinery via VMD, CWT and improved CNN. Measurement, 200, 111635.
18. [18] Huang, Z., & Zhao, X. (2024). A novel multi-scale competitive network for fault diagnosis in rotating machinery. Engineering Applications of Artificial Intelligence, 128, 107441.
19. [19] Zhang, X., Ma, Y., Pan, Z., & Wang, G. (2024). A novel stochastic resonance based deep residual network for fault diagnosis of rolling bearing system. ISA transactions, 148, 279-284.
20. [20] Han, S., & Jeong, J. (2020). An weighted CNN ensemble model with small amount of data for bearing fault diagnosis. Procedia Computer Science, 175, 88-95.
21. [21] Chen, S., Zheng, W., Xiao, H., Han, P., & Luo, K. (2023). A residual convolution transfer framework based on slow feature for cross-domain machinery fault diagnosis. Neurocomputing, 546, 126322
22. [22] Gupta, A., Onumanyi, A. J., Ahlawat, S., Prasad, Y., Singh, V., & Abu-Mahfouz, A. M. (2024). DAT: A robust Discriminant Analysis-based Test of unimodality for unknown input distributions. Pattern Recognition Letters, 182, 125-132.
23. [23] Hou, Y., Wang, J., Chen, Z., Ma, J., & Li, T. (2023). Diagnosisformer: An efficient rolling bearing fault diagnosis method based on improved Transformer. Engineering Applications of Artificial Intelligence, 124, 106507.
24. [24] Liu, X., Wang, J., Meng, S., Qiu, X., & Zhao, G. (2023). Multi-view rotating machinery fault diagnosis with adaptive co-attention fusion network. Engineering Applications of Artificial Intelligence, 122, 106138.
25. [25] Zhong, J., Zheng, Y., Ruan, C., Chen, L., Bao, X., & Lyu, L. (2025). M-IPISincNet: an explainable multi-source physics-informed neural network based on improved SincNet for rolling bearings fault diagnosis. Information Fusion, 115, 102761.
26. [26] Liu, S., Li, J., Zhou, N., Chen, G., Lu, K., & Wu, Y. (2025). Intelligent fault diagnosis of rotating machine via Expansive dual-attention fusion Transformer enhanced by semi-supervised learning. Expert Systems with Applications, 260, 125398.
27. [27] Zhang, W., Yu, B., Li, G., Zhuang, P., Liang, Z., & Zhao, W. (2024). Unified multi-color-model-learning-based deep support vector machine for underwater image classification. Engineering Applications of Artificial Intelligence, 138, 109437.
28. [28] Akinola, I. T., Sun, Y., Adebayo, I. G., & Wang, Z. (2024). Daily peak demand forecasting using pelican algorithm optimised support vector machine (POA-SVM). Energy Reports, 12, 4438-4448.
29. [29] Du, Q., & Zhai, J. (2024). Application of artificial intelligence Sensors based on random forest algorithm in financial recognition models. Measurement: Sensors, 33, 101245.
30. [30] Chen, Y., Li, T., Fu, B., Xia, Q., Liu, Q., Li, T., ... & Huang, Y. (2024). Deposit type discrimination of Jiaodong gold deposits using random forest algorithm: Constraints from trace elements of pyrite. Ore Geology Reviews, 175, 106343.
31. [31] Xi, W. (2024). Research on E-learning interactive English vocabulary recommendation education system based on naive Bayes algorithm. Entertainment Computing, 51, 100732.
32. [32] Shang, Y. (2024). Prevention and detection of DDOS attack in virtual cloud computing environment using Naive Bayes algorithm of machine learning. Measurement: Sensors, 31, 100991.
33. [33] Raj, S., Vishnoi, A., & Srivastava, A. (2024). Classify Alzheimer genes association using Naïve Bayes algorithm. Human Gene, 41, 201309.
34. [34] Kan, X., Fan, Y., Zheng, J., Chi, C. H., Song, W., & Kudreyko, A. (2023). Data adjusting strategy and optimized XGBoost algorithm for novel insider threat detection model. Journal of the Franklin Institute, 360(16), 11414-11443
35. [35] Wang, T., Bian, Y., Zhang, Y., & Hou, X. (2023). Classification of earthquakes, explosions and mining-induced earthquakes based on XGBoost algorithm. Computers & Geosciences, 170, 105242.
36. [36] Kaya, Y., Kuncan, M., Akcan, E., & Kaplan, K. (2024). An efficient approach based on a novel 1D-LBP for the detection of bearing failures with a hybrid deep learning method. Applied Soft Computing, 155, 111438.
37. [37] Akcan, E., Kuncan, M., Kaplan, K., & Kaya, Y. (2024). Diagnosing bearing fault location, size, and rotational speed with entropy variables using extreme learning machine. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 46(1), 4.