Araştırma Makalesi
BibTex RIS Kaynak Göster

Determination of Health Indicator in High Speed Bearings Using Machine Learning Methodology

Yıl 2021, Sayı: 22, 176 - 183, 31.01.2021
https://doi.org/10.31590/ejosat.843465

Öz

The machine health indicator shows the deterioration stages of a machine part and its failure. The main purpose of this study is to determine the machine health indicator from vibration data using machine learning. The determined health indicator will then be used in the remaining useful life estimation. The necessary steps to calculate the indicator are listed as feature extraction, normalization, and principal component analysis. In this study, vibration signals are transformed from time domain to frequency domain using the Welch method then the listed features kurtosis, skewness, effective value, crest factor and impact factor are calculated. The noise of these features has been reduced utilizing z-score and Savitzky-Golay methods. Subsequently, principal component analysis is applied to compute principal component vectors. Of these vectors, vectors that best match the monotone exponential upward trend are chosen as useful principal component vectors. The health indicator is the mean value of the useful principal component vectors and it has been used to estimate the remaining useful life. The success of the prediction is determined by the determination coefficient (R2) and root mean square error (RMSE) values. According to the results, R2 and RMSE values are: 0.6625 and 17.8925 for the prior training set, respectively; 0.9947 and 1.7688 for posterior training set, respectively; for the test set it is 0.9897 and 2.2025, respectively.

Kaynakça

  • Abdelli, K., Grießer, H., & Pachnicke, S. (2020). Machine Learning Based Data Driven Diagnostic and Prognostic Approach for Laser Reliability Enhancement. In 22nd International Conference on Transparent Optical Networks (ICTON) (pp. 1-4). IEEE.
  • Akçay, H., & Türkay, S. (2019). Power spectrum estimation in innovation models. Mechanical Systems and Signal Processing, 121, 227-245.
  • Ali Ben, J., Saidi, L., Harrath, S., Bechhoefer, E., & Benbouzid, M. (2018). Online automatic diagnosis of wind turbine bearings progressive degradations under real experimental conditions based on unsupervised machine learning. Applied Acoustics, 132 (2017), 167–181.
  • Banerjee, A., Gupta, S. K., & Datta, D. (2020). Remaining Useful Life as a Cognitive Tool in the Domain of Manufacturing. In Emotion and Information Processing (pp. 171-183). Springer, Cham.
  • Bektaş, O. (2020). Kestirimci Bakım İçin Döner Mekanizma Bozulma Eğrisinin Tanımlanması. European Journal of Science and Technology, 420–428.
  • Elasha, F., Shanbr, S., Li, X., & Mba, D. (2019). Prognosis of a wind turbine gearbox bearing using supervised machine learning. Sensors (Switzerland), 19(14), 1–17.
  • Guo, Y., Na, J., Li, B., & Fung, R. F. (2014). Envelope extraction based dimension reduction for independent component analysis in fault diagnosis of rolling element bearing. Journal of Sound and Vibration, 333(13), 2983–2994.
  • Jin, X., Wang, Y., & Hong, W. (2019). Power Spectrum Estimation Method Based on Matlab. In Proceedings of the 3rd International Conference on Vision, Image and Signal Processing (pp. 1-5).
  • Kappal, S. (2019). Data Normalization Using Median Median Absolute Deviation MMAD based Z-Score for Robust Predictions vs. Min–Max Normalization. London Journal of Research in Science: Natural and Formal.
  • Kızrak, M. A., & Bolat, B. (2019). Uçak Motoru Sağlığı için Uzun-Kısa Süreli Bellek Yöntemi ile Öngörücü Bakım. International Journal of InformaticsTechnologies, 12(2).
  • Kollmann, S., Estaji, A., Bratukhin, A., Wendt, A., & Sauter, T. (2020). Comparison of Preprocessors for Machine Learning in the Predictive Maintenance Domain. In IEEE 29th International Symposium on Industrial Electronics (ISIE) (pp. 49-54). IEEE.
  • Liu, H., & Han, M. (2014). A fault diagnosis method based on local mean decomposition and multi-scale entropy for roller bearings. Mechanism and Machine Theory, 75, 67–78.
  • Liu, Z., & Zhang, L. (2020). A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings. Measurement, 149, 107002.
  • Muratet, L., Doncieux, S., & Meyer, J. A. (2004). A biomimetic reactive navigation system using the optical flow for a rotary-wing UAV in urban environment. Proceedings of the International Session on Robotics, 2262-2270.
  • Orhan, S., Aktürk, N., & Çelik, V. (2003). Bir santrifüj pompa rulmanlarının çalışabilirliğinin titreşim analizi ile belirlenmesi. Gazi Üniversitesi Fen Bilimleri Dergisi, 16(3), 543-552.
  • Ozkat, E. C. (2019). “The Comparison of Machnine Learning Algorithms in Estimation of Remaining Useful Life.” In IX. International Maintenance Technologies Congress. pages, 614–619. Denizli, Turkey: UCTEA CHAMBER OF MECHANICAL ENGINEERS, ISBN NO: 978-605-01-1288-7
  • Saidi, L., Ben Ali, J., Bechhoefer, E., & Benbouzid, M. (2017). Wind turbine high-speed shaft bearings health prognosis through a spectral Kurtosis-derived indices and SVR. Applied Acoustics, 120, 1–8.
  • Sun, W., Yang, G. A., Chen, Q., Palazoglu, A., & Feng, K. (2013). Fault diagnosis of rolling bearing based on wavelet transform and envelope spectrum correlation. JVC/Journal of Vibration and Control, 19(6), 924–941.
  • Tao, X., Ren, C., Wu, Y., Li, Q., Guo, W., Liu, R., ... & Zou, J. (2020). Bearings fault detection using wavelet transform and generalized Gaussian density modeling. Measurement, 155, 107557.
  • Teng, W., Zhang, X., Liu, Y., Kusiak, A., & Ma, Z. (2017). Prognosis of the remaining useful life of bearings in a wind turbine gearbox. Energies, 10(1).
  • Tibaduiza, D. A., Mujica, L. E., & Rodellar, J. (2011). Structural Health Monitoring based on principal component analysis: damage detection, localization and classification. Advances in Dynamics, Control, Monitoring and Applications, Universitat Politècnica de Catalunya, Departament de Matemàtica Aplicada, 3(1), 8–17.
  • Villwock, S., & Pacas, M. (2008). Application of the welch-method for the identification of two- and three-mass-systems. IEEE Transactions on Industrial Electronics, 55(1), 457–466.
  • Yıldız, K., Çamurcu, Y., & Doğan, B. (2010). Veri madenciliğinde temel bileşenler analizi ve Negatifsiz matris çarpanlarına ayırma tekniklerinin karşılaştırmalı analizi. Akademik Bilişim, 10-12.

Makine Öğrenmesi Metodolojisi Kullanılarak Yüksek Hızlı Rulmanlarda Sağlık Göstergesinin Belirlenmesi

Yıl 2021, Sayı: 22, 176 - 183, 31.01.2021
https://doi.org/10.31590/ejosat.843465

Öz

Makine sağlığı göstergesi, bir makine parçasının bozulma aşamalarını ve ortaya çıkacak nihai arızasını gösterir. Bu çalışmanın temel amacı, makine öğrenmesi metodolojisi kullanarak titreşim verilerinden makine sağlığı göstergesini belirlemektir. Tespit edilen bu sağlık göstergesi daha sonra kalan faydalı ömür tahminde kullanılacaktır. Makine sağlık göstergesini hesaplamak için gerekli adımlar, özellik çıkarma, normalleştirme ve temel bileşen analizi olarak listelenmiştir. Bu çalışmada titreşim sinyalleri zaman alanından frekans alanına Welch yöntemi kullanılarak dönüştürülmüş ve ardından sıralanan öznitelikler basıklık, çarpıklık, etkin değer, tepe faktörü ve etki faktörü hesaplanmış. Özniteliklerin gürültüsü z-skor ve Savitzky-Golay yöntemleri kullanılarak azaltılmıştır. Sonrasında, temel bileşen vektörlerini hesaplamak için düzeltilmiş özniteliklere temel bileşen analizi uygulanmıştır. Bu vektörlerden monoton eksponansiyel artış eğilimine en iyi uyan vektörler kullanışlı temel bileşen vektörlerdir. Sağlık göstergesi, faydalı temel bileşen vektörlerinin ortalama değeridir ve kalan faydalı ömrü tahmin etmek için kullanılmıştır. Tahminin başarısı determinasyon katsayısı (R2) ve kök ortalama kare hata (RMSE) değerleri ile belirlenmiştir. Sonuçlara göre, R2 ve RMSE değerleri: prior eğitim seti için sırayıla 0.6625 ve 17.8925; posterior eğitim seti için sırayıla 0.9947 ve 1.7688; test seti için sırayıla 0.9897 ve 2.2025’tir.

Kaynakça

  • Abdelli, K., Grießer, H., & Pachnicke, S. (2020). Machine Learning Based Data Driven Diagnostic and Prognostic Approach for Laser Reliability Enhancement. In 22nd International Conference on Transparent Optical Networks (ICTON) (pp. 1-4). IEEE.
  • Akçay, H., & Türkay, S. (2019). Power spectrum estimation in innovation models. Mechanical Systems and Signal Processing, 121, 227-245.
  • Ali Ben, J., Saidi, L., Harrath, S., Bechhoefer, E., & Benbouzid, M. (2018). Online automatic diagnosis of wind turbine bearings progressive degradations under real experimental conditions based on unsupervised machine learning. Applied Acoustics, 132 (2017), 167–181.
  • Banerjee, A., Gupta, S. K., & Datta, D. (2020). Remaining Useful Life as a Cognitive Tool in the Domain of Manufacturing. In Emotion and Information Processing (pp. 171-183). Springer, Cham.
  • Bektaş, O. (2020). Kestirimci Bakım İçin Döner Mekanizma Bozulma Eğrisinin Tanımlanması. European Journal of Science and Technology, 420–428.
  • Elasha, F., Shanbr, S., Li, X., & Mba, D. (2019). Prognosis of a wind turbine gearbox bearing using supervised machine learning. Sensors (Switzerland), 19(14), 1–17.
  • Guo, Y., Na, J., Li, B., & Fung, R. F. (2014). Envelope extraction based dimension reduction for independent component analysis in fault diagnosis of rolling element bearing. Journal of Sound and Vibration, 333(13), 2983–2994.
  • Jin, X., Wang, Y., & Hong, W. (2019). Power Spectrum Estimation Method Based on Matlab. In Proceedings of the 3rd International Conference on Vision, Image and Signal Processing (pp. 1-5).
  • Kappal, S. (2019). Data Normalization Using Median Median Absolute Deviation MMAD based Z-Score for Robust Predictions vs. Min–Max Normalization. London Journal of Research in Science: Natural and Formal.
  • Kızrak, M. A., & Bolat, B. (2019). Uçak Motoru Sağlığı için Uzun-Kısa Süreli Bellek Yöntemi ile Öngörücü Bakım. International Journal of InformaticsTechnologies, 12(2).
  • Kollmann, S., Estaji, A., Bratukhin, A., Wendt, A., & Sauter, T. (2020). Comparison of Preprocessors for Machine Learning in the Predictive Maintenance Domain. In IEEE 29th International Symposium on Industrial Electronics (ISIE) (pp. 49-54). IEEE.
  • Liu, H., & Han, M. (2014). A fault diagnosis method based on local mean decomposition and multi-scale entropy for roller bearings. Mechanism and Machine Theory, 75, 67–78.
  • Liu, Z., & Zhang, L. (2020). A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings. Measurement, 149, 107002.
  • Muratet, L., Doncieux, S., & Meyer, J. A. (2004). A biomimetic reactive navigation system using the optical flow for a rotary-wing UAV in urban environment. Proceedings of the International Session on Robotics, 2262-2270.
  • Orhan, S., Aktürk, N., & Çelik, V. (2003). Bir santrifüj pompa rulmanlarının çalışabilirliğinin titreşim analizi ile belirlenmesi. Gazi Üniversitesi Fen Bilimleri Dergisi, 16(3), 543-552.
  • Ozkat, E. C. (2019). “The Comparison of Machnine Learning Algorithms in Estimation of Remaining Useful Life.” In IX. International Maintenance Technologies Congress. pages, 614–619. Denizli, Turkey: UCTEA CHAMBER OF MECHANICAL ENGINEERS, ISBN NO: 978-605-01-1288-7
  • Saidi, L., Ben Ali, J., Bechhoefer, E., & Benbouzid, M. (2017). Wind turbine high-speed shaft bearings health prognosis through a spectral Kurtosis-derived indices and SVR. Applied Acoustics, 120, 1–8.
  • Sun, W., Yang, G. A., Chen, Q., Palazoglu, A., & Feng, K. (2013). Fault diagnosis of rolling bearing based on wavelet transform and envelope spectrum correlation. JVC/Journal of Vibration and Control, 19(6), 924–941.
  • Tao, X., Ren, C., Wu, Y., Li, Q., Guo, W., Liu, R., ... & Zou, J. (2020). Bearings fault detection using wavelet transform and generalized Gaussian density modeling. Measurement, 155, 107557.
  • Teng, W., Zhang, X., Liu, Y., Kusiak, A., & Ma, Z. (2017). Prognosis of the remaining useful life of bearings in a wind turbine gearbox. Energies, 10(1).
  • Tibaduiza, D. A., Mujica, L. E., & Rodellar, J. (2011). Structural Health Monitoring based on principal component analysis: damage detection, localization and classification. Advances in Dynamics, Control, Monitoring and Applications, Universitat Politècnica de Catalunya, Departament de Matemàtica Aplicada, 3(1), 8–17.
  • Villwock, S., & Pacas, M. (2008). Application of the welch-method for the identification of two- and three-mass-systems. IEEE Transactions on Industrial Electronics, 55(1), 457–466.
  • Yıldız, K., Çamurcu, Y., & Doğan, B. (2010). Veri madenciliğinde temel bileşenler analizi ve Negatifsiz matris çarpanlarına ayırma tekniklerinin karşılaştırmalı analizi. Akademik Bilişim, 10-12.
Toplam 23 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Konular Mühendislik
Bölüm Makaleler
Yazarlar

Erkan Caner Özkat 0000-0003-0530-5439

Yayımlanma Tarihi 31 Ocak 2021
Yayımlandığı Sayı Yıl 2021 Sayı: 22

Kaynak Göster

APA Özkat, E. C. (2021). Makine Öğrenmesi Metodolojisi Kullanılarak Yüksek Hızlı Rulmanlarda Sağlık Göstergesinin Belirlenmesi. Avrupa Bilim Ve Teknoloji Dergisi(22), 176-183. https://doi.org/10.31590/ejosat.843465