• icon+90(533) 652 66 86
  • iconnwsa.akademi@hotmail.com
  • icon Fırat Akademi Samsun-Türkiye

Article Details

Article Details

  • Article Code : FIRAT-AKADEMI-501-5714
  • Article Type : Araştırma Makalesi
  • Publication Number : 1A0493
  • Page Number : 1-20
  • Doi : 10.12739/NWSA.2025.20.1.1A0493
  • Abstract Reading : 146
  • Download : 17
  • Share :

  • PDF Download

Issue Details

  • Year : 2025
  • Volume : 20
  • Issue : 1
  • Number of Articles Published : 1
  • Published Date : 1.01.2025

Cover Download Context Page Download
Dergi Kapağı
Engineering Sciences

Serial Number : 1A
ISSN No. : 1308-7231
Release Interval (in a Year) : 4 Issues

ASENKRON MOTORLARDA KIRIK ROTOR ÇUBUĞU ARIZA TEŞHİSİ İÇİN ÇOKLU SİNYAL FÜZYONU: SANAL SENSÖR TABANLI TOPLULUK ÖĞRENME YAKLAŞIMI

Özgür AYDIN1 , Erhan AKIN 2

Bu çalışma asenkron motorların arıza teşhisinde sanal sensör tabanlı ensemble learning yaklaşımının etkinliğini araştırmaktadır. Çalışmada özellikle kırık rotor çubuğu arızasının tespiti üzerine odaklanılmıştır. Akım (Ia, Ib, Ic) ve titreşim (Vib_acpe, Vib_acpi) sinyallerinden Fourier Dönüşümü (FFT) ve bant gücü analizi ile anlamlı özellikler çıkarılmıştır. Elde edilen özellikler, RNN, GRU ve LSTM gibi derin öğrenme modelleriyle bireysel olarak değerlendirilmiş ve ardından daha güçlü bir sınıflandırma performansı elde etmek amacıyla ensemble learning yaklaşımı uygulanmıştır. Sonuçlar, ensemble modelin %94.44 doğruluk ve %95.25 kesinlik oranı ile bireysel modelleri geride bıraktığını göstermektedir. Bu çalışma aynı veri setini kullanan literatürdeki diğer çalışmalara kıyasla önemli bir üstünlük sağlamaktadır. Literatürde, genellikle yalnızca tek bir akım (Ia) veya tek bir titreşim (Vib_acpi) sinyali ile sınıflandırma yapılırken, bu çalışmada çoklu akım ve titreşim sinyalleri birlikte kullanılarak daha kapsamlı bir veri temsili sağlanmıştır. Bu strateji yanlış pozitif ve yanlış negatif oranlarını düşürerek daha kararlı bir sınıflandırma performansı elde edilmesini sağlamıştır. Ayrıca ensemble yapısının, bireysel RNN, GRU ve LSTM modellerine göre daha stabil ve genellenebilir bir sonuç verdiği gözlemlenmiştir.

Keywords
Asenkron Motor Arıza Teşhisi, Sanal Sensör, Rotor Çubuğu Arızası, Topluluk Öğrenmesi, Çoklu Model Sınıflandırma,

MULTI-SIGNAL FUSION FOR FAULT DIAGNOSIS OF BROKEN ROTOR BARS IN INDUCTION MOTORS: A VIRTUAL SENSOR-BASED ENSEMBLE LEARNING APPROACH

Özgür AYDIN1 , Erhan AKIN 2

This study investigates the effectiveness of a virtual sensor-based ensemble learning approach for fault diagnosis in asynchronous motors. The study specifically focuses on the detection of broken rotor bar faults. Significant features were extracted from current (Ia, Ib, Ic) and vibration (Vib_acpe, Vib_acpi) signals using Fourier Transform (FFT) and band power analysis. The extracted features were individually evaluated using deep learning models such as RNN, GRU, and LSTM, and then the ensemble learning approach was applied to achieve stronger classification performance. The results demonstrate that the ensemble model outperforms individual models with an accuracy of 94.44% and a precision of 95.25%. This study provides a significant advantage compared to other studies in the literature that use the same dataset. In the literature, classification is typically performed using only a single current (Ia) or a single vibration (Vib_acpi) signal. However, in this study, all current and vibration signals were used together to achieve a more comprehensive data representation. This strategy enabled a more robust classification performance by reducing false positive and false negative rates. Additionally, it was observed that the ensemble structure provides a more stable and generalizable result compared to individual RNN, GRU, and LSTM models.

Keywords
Asynchronous Motor Fault Diagnosis, Virtual Sensor, Rotor Bar Fault, Ensemble Learning, Multi-Model Classification,

Details
   

Authors

Özgür AYDIN (1) (Corresponding Author)
Bingöl Üniversitesi - Enformatik Bölüm Başkanlığı
iamozguraydin@gmail.com | 0000-0001-8130-277X

Erhan AKIN (2)

Fırat Üniversitesi
eakin@firat.edu.tr | 0000-0001-6476-9255

Supporting Institution

 :

Project Number

 :

Thanks

 :
References
[1] Hughes, A. and Drury, B., (2019). Electric motors and drives: fundamentals, types and applications. Newnes.

[2] Sen, P.C., (2021). Principles of Electric Machines and Power Electronics, International Adaptation. John Wiley and Sons.

[3] Hosseinpoor, Z., Arefi, M.M., Razavi-Far, R., Mozafari, N., and Hazbavi, S., (2020). Virtual sensors for fault diagnosis: A case of induction motor broken rotor bar. IEEE Sensors Journal, 21(4):5044-5051.

[4] Yakhni, M.F., Cauet, S., Sakout, A., Assoum, H., Etien, E., Rambault, L., and El-Gohary, M., (2023). Variable speed induction motors’ fault detection based on transient motor current signatures analysis: A review. Mechanical Systems and Signal Processing, 184, 109737.

[5] Nandi, S., Toliyat, H.A., and Li, X., (2005). Condition monitoring and fault diagnosis of electrical motors—A review. IEEE transactions on energy conversion, 20(4):719-729.

[6] Tariq, M.F., Khan, A.Q., Abid, M., and Mustafa, G., (2018). Data-driven robust fault detection and isolation of three-phase induction motor. IEEE Transactions on Industrial Electronics, 66(6):4707-4715.

[7] Doğan, Z., (2012). Ayrıklaştırma yöntemleri ve yapay sinir ağı kullanarak asenkron motorlarda arıza teşhisi (Doctoral dissertation, Marmara Universitesi (Turkiye)).

[8] Boldea, I. and Nasar, S.A., (2002). Induction machines: An introduction. In The Induction Machine Handbook (p. 13). CRC Press.

[9] Siddiqui, K.M., Sahay, K., and Giri, V.K., (2014). Health monitoring and fault diagnosis in induction motor-a review. International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, 3(1):6549-6565.

[10] Ahmed, Q., Bhatti, A.I., and Iqbal, M., (2011). Virtual sensors for automotive engine sensors fault diagnosis in second-order sliding modes. IEEE Sensors Journal, 11(9):1832-1840.

[11] Gao, L., Li, D., Li, D., Yao, L., Liang, L., and Gao, Y., (2019). A novel chiller sensors fault diagnosis method based on virtual sensors. Sensors, 19(13):3013.

[12] Martin, D., Kühl, N., and Satzger, G., (2021). Virtual sensors. Business and Information Systems Engineering, 63:315-323.

[13] Madria, S., Kumar, V., and Dalvi, R., (2013). Sensor cloud: A cloud of virtual sensors. IEEE software, 31(2):70-77.

[14] Tegen, A., Davidsson, P., Mihailesc, R.C., and Persson, J.A., (2019). Collaborative sensing with interactive learning using dynamic intelligent virtual sensors. Sensors, 19(3):477.

[15] Serpas, M., Chu, Y., and Hahn, J., (2013). Fault detection approach for systems involving soft sensors. Journal of Loss Prevention in the Process Industries, 26(3):443-452.

[16] Mohammadi, A., Zarghami, R., Lefebvre, D., Golshan, S., and Mostoufi, N., (2019). Soft sensor design and fault detection using Bayesian network and probabilistic principal component analysis. Journal of Advanced Manufacturing and Processing, 1(4):e10027.

[17] Kadlec, P., Gabrys, B., and Strandt, S., (2009). Data-driven soft sensors in the process industry. Computers and chemical engineering, 33(4):795-814.

[18] Liu, Z., Huang, Z., Wang, J., Yue, C., and Yoon, S., (2021). A novel fault diagnosis and self-calibration method for air-handling units using Bayesian Inference and virtual sensing. Energy and Buildings, 250, 111293.

[19] Zhang, Z., Han, H., Cui, X., and Fan, Y., (2020). Novel application of multi-model ensemble learning for fault diagnosis in refrigeration systems. Applied Thermal Engineering, 164, 114516.

[20] Treml, A.E., Flauzino, R.A., Suetake, M., Maciejewski, N.R., and Afonso, N., (2020). Experimental database for detecting and diagnosing rotor broken bar in a three-phase induction motor. IEEE DataPort.

[21] Poudel, S., (2023, August 28). Recurrent neural network (RNN) architecture explained. Medium. URL: https://medium.com/@poudelsushmita878/recurrent-neural-network-rnn-architecture-explained-1d69560541ef, Son Erişim Tarihi: 10.12.2024.

[22] Akbari Asanjan, A., Yang, T., Hsu, K., Sorooshian, S., Lin, J., and Peng, Q., (2018). Short-term precipitation forecast based on the PERSIANN system and LSTM recurrent neural networks. Journal of Geophysical Research: Atmospheres, 123(22):12-543.

[23] Dey, P., Hossain, E., Hossain, M.I., Chowdhury, M.A., Alam, M.S., Hossain, M.S., and Andersson, K., (2021). Comparative analysis of recurrent neural networks in stock price prediction for different frequency domains. Algorithms, 14(8):251.

[24] Anishnama. (2023, May 4). Understanding gated recurrent unit (GRU) in deep learning. Medium. URL: https://medium.com/@anishnama20/understanding-gated-recurrent-unit-gru-in-deep-learning-2e54923f3e2, Son Erişim Tarihi: 10.12.2024.

[25] Rojas, C.A., (2024, April 8). What is LSTM (Long Short Term Memory)? Medium. URL: https://blog.ai-evergreen.club/what-is-lstm-long-short-term-memory-221099a981f7. Son Erişim Tarihi: 10.12.2024.

[26] Hamad, R. (2023, December 3). What is LSTM? Introduction to long short-term memory. Medium. URL: https://medium.com/@rebeen.jaff/what-is-lstm-introduction-to-long-short-term-memory-66bd3855b9ce, Son Erişim Tarihi: 10.12.2024.

[27] Elmaz, F., Eyckerman, R., Casteels, W., Latré, S., and Hellinckx, P., (2021). CNN-LSTM architecture for predictive indoor temperature modeling. Building and Environment, 206, 108327.

[28] Mienye, I.D. and Sun, Y., (2022). A survey of ensemble learning: Concepts, algorithms, applications, and prospects. IEEE Access, 10:99129-99149.

[29] Şirin, E., (2017, Aralık 19). Ensemble yöntemler (topluluk öğrenmesi): Basit teorik anlatım ve Python uygulama. URL: https://www.veribilimiokulu.com/ensemble-yontemler-topluluk-ogrenmesi-basit-teorik-anlatim-ve-python-uygulama, Son Erişim Tarihi: 10.12.2024.

[30] Dişli, F., Gedikpınar, M., and Sengur, A., (2023). Kırık Rotor Çubuğu Sayısının Ampirik Mod Ayrışımı ve Makine Öğrenmesi Yaklaşımları İle Belirlenmesi. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 35(2):783-795.

[31] Aydin, I. and Akin, E., (2024, July). Multi-sensory Fault Diagnosis of Broken Rotor Bars Using Transfer Learning. In International Conference on Intelligent and Fuzzy Systems (pp. 349-356). Cham: Springer Nature Switzerland.

[32] Misra, S., Kumar, S., Sayyad, S., Bongale, A., Jadhav, P., Kotecha, K., ... and Gabralla, L.A., (2022). Fault detection in induction motor using time domain and spectral imaging-based transfer learning approach on vibration data. Sensors, 22(21):8210.

[33] Aydın, Ö. and Akın, E., (2024). Kırık Rotor Çubuğu Arızalarının Belirlenmesinde Derin Öğrenme Yaklaşımları ve Motor Akım İmza Analizi. Türk Doğa ve Fen Dergisi, 13(3):1-7.

[34] Dişli, F., Gedikpınar, M., and Sengur, A., (2023). Deep transfer learning-based broken rotor fault diagnosis for Induction Motors. Turkish Journal of Science and Technology, 18(1):275-290.