Journals / Konya Mühendislik Bilimleri Dergisi / 2022 / Cilt: 10 Sayı: 2

Application of Deep Learning Method for Condition Monitoring and Fault Diagnosis from Vibration Data in Bearings

RULMANLARDA TİTREŞİM VERİLERİNDEN DURUM İZLEME VE ARIZA TEŞHİSİ İÇİN DERİN ÖĞRENME YÖNTEMİNİN UYGULANMASI

Pages
346–365
Publication Date
2022-06-01
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

As a result, it is seen that the convolutional neural network method with different approaches provides high classification accuracy in the most basic bearing fault detection and is a promising method for fault diagnosis.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Since bearings are machine elements that are frequently used in several industry due to their load carrying capacity, they are subjected to wear or breakage such as adhesion, abrasion and creep under overloading conditions.
📊
Key Findings & Data:For this reason, condition monitoring and fault detection are an important issue for sustainability, high performance and reliability.
💡
Conclusion & Impact:As a result, it is seen that the convolutional neural network method with different approaches provides high classification accuracy in the most basic bearing fault detection and is a promising method for fault diagnosis.
🔬Design:Research Article
📍Scope:Comprehensive Sample
⚙️Variables:Arıza teşhisi · Rulmanlar · Derin Öğrenme

Abstract

Since bearings are machine elements that are frequently used in several industry due to their load carrying capacity, they are subjected to wear or breakage such as adhesion, abrasion and creep under overloading conditions. For this reason, condition monitoring and fault detection are an important issue for sustainability, high performance and reliability. Feature selection is a difficult task, hence, some features may change due to changing working conditions. Therefore, in this study, convolutional neural networks (ESA), which is a deep learning method in which features are determined by internal dynamics, are used for the detection of healthy bearings (SR) and bearing failures (outer ring failure-AR1, inner ring failure-AR2, rolling element failure-AR3). In order to train ESA approaches with different architectures, spectrograms of vibration signals using Short-Time Fourier Transform were obtained. The results of GoogleNet, ResNet-50, EfficientNet-B0 and AlexNet approaches that are trained with spectograms are comparatively examined. It has been seen that ESAs with complex architectures (GoogleNet, ResNet-50, EfficientNet-B0 ) detect failures with 100% accuracy and AlexNet with 90% accuracy, but it has been observed that the training time increases as the network structure changes and the number of layers increases. It is observed that the results of the study are far better than the similar papers in the literature. As a result, it is seen that the convolutional neural network method with different approaches provides high classification accuracy in the most basic bearing fault detection and is a promising method for fault diagnosis.

Özet

Rulmanlar, yük taşıma kapasiteleri nedeniyle endüstride pek çok alanda sıklıkla kullanılan makine elemanları olduklarından, aşırı yükleme durumlarında adhezyon, abrazyon ve sürünme gibi aşınma türlerine ya da kırılmalara maruz kalabilirler. Bu nedenle, rulmanlarda durum izlemesi yapılması ve arızaların teşhis edilmesi, sürdürülebilirlik, yüksek performans ve güvenlik açılarından önemli bir husustur. Arıza türlerinin ayırt edilmesinde belirleyici özniteliklerin seçilmesi, farklı çalışma koşullarında bir takım öznitelikler de değişebildiğinden zor bir süreçtir. Bu nedenle, bu çalışmada sağlıklı rulmanların (SR) ve rulman arızalarının (dış bilezik arızası-AR1, iç bilezik arızası-AR2, yuvarlanma arızası-AR3) tespiti için özniteliklerin içsel dinamiklerle belirlendiği derin öğrenme yöntemi olan olan evrişimli sinir ağları (ESA) kullanılmıştır. Birbirinden farklı mimarilere sahip ESA yaklaşımlarını eğitmek için Kısa Zamanlı Fourier Dönüşümü uygulanan titreşim sinyallerinin spektrogramları elde edilmiştir. Spektogram verileri ile eğitilen GoogleNet, ResNet-50, EfficientNet-B0 ve AlexNet yaklaşımlarının sonuçları karşılaştırmalı olarak incelenmiştir. Karmaşık mimariye sahip ESA’ların (GoogleNet, ResNet-50, EfficientNet-B0 ) arızaları %100 doğrulukla, AlexNet’in ise %90 doğrulukla tespit ettiği görülmüştür, ancak ağ yapısı değiştikçe ve katman saysı arttıkça eğitim süresinin de uzadığı görülmüştür. Elde edilen sonuçların literatürdeki çalışmaların sonuçlarından üstün olduğu gözlenmiştir. Sonuç olarak, farklı yaklaşımlara sahip evrişimli sinir ağları yönteminin en temel rulman arıza tespitinde yüksek sınıflandırma doğruluğu sağladığı ve arıza teşhisi için umut vadeden bir yöntem olduğu görülmektedir.

🔖

Cite This Article

KARABACAK, Y. E., ÖZMEN, N. G. (2022). Application of Deep Learning Method for Condition Monitoring and Fault Diagnosis from Vibration Data in Bearings. Konya Mühendislik Bilimleri Dergisi, Cilt: 10 Sayı: 2, 346–365. https://ajindex.com/articles/938486

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (8)

  1. Al Shorman, O., Irfan, M., Saad, N., Zhen, D., Haider, N., Glowacz, A., AlShorman, A., 2020, “A review of artificial intelligence methods for condition monitoring and fault diagnosis of rolling element bearings for induction motor”, Shock and Vibration, Cilt 2020.
  2. Amarnath, M., Sugumaran, V., Kumar, H., 2013, “Exploiting sound signals for fault diagnosis of bearings using decision tree”, Measurement, Cilt 46, Sayı 3, ss. 1250-1256.
  3. Caesarendra, W., Tjahjowidodo, T., 2017, “A review of feature extraction methods in vibration-based condition monitoring and its application for degradation trend estimation of low-speed slew bearing”, Machines, Cilt 5, Sayı 4, ss. 21. Cakir, M., Guvenc, M. A., Mistikoglu, S., 2021, “The experimental application of popular machine learning algorithms on predictive maintenance and the design of IoT based condition monitoring system”, Computers & Industrial Engineering, Cilt 151, ss. 106948. Chen, Z., Gryllias, K., Li, W., 2019, “Mechanical fault diagnosis using convolutional neural networks and extreme learning machine”, Mechanical systems and signal processing, Cilt 133, ss. 106272. Chen, Z., Li, C., Sanchez, R. V., 2015, “Gearbox fault identification and classification with convolutional neural networks”, Shock and Vibration, Cilt 2015.
  4. Cheng, H., Kong, X., Chen, G., Wang, Q., Wang, R., 2021, “Transferable convolutional neural network based remaining useful life prediction of bearing under multiple failure behaviors”, Measurement, Cilt 168, ss. 108286. Choudhary, A., Shimi, S. L., Akula, A., 2018, “Bearing fault diagnosis of induction motor using thermal imaging”, 2018 International Conference on Computing, Power and Communication Technologies (GUCON), ss. 950-955,
  5. IEEE. Correa, J. C. A. J., Guzman, A. A. L., 2020, Mechanical Vibrations and Condition Monitoring, Academic Press. CWRU Rulman Veri Merkezi web sitesi. http://csegroups.case.edu/bearingdatacenter)(14.12.2022)
  6. Dong, S., Luo, T., Zhong, L., Chen, L., Xu, X., 2017, “Fault diagnosis of bearing based on the kernel principal component analysis and optimized k-nearest neighbour model”, Journal of Low Frequency Noise, Vibration and Active Control, 36(4), 354-365.
  7. Duan, Z., Wu, T., Guo, S., Shao, T., Malekian, R., Li, Z., 2018, “Development and trend of condition monitoring and fault diagnosis of multi-sensors information fusion for rolling bearings: a review”, The International Journal of Advanced Manufacturing Technology, 96(1), 803-819.
  8. Elforjani, M., Shanbr, S., 2017, “Pr