Broken magnets fault detection in pmsm using a  convolutional neural network and SVM

Benkaihoul Said; Lakhdar Mazouz; Toufik Tayeb NAAS; Özüpak Yildirim; Ridha Djamel Mohammedi

doi:10.5935/jetia.v10i48.1185

Autores

Benkaihoul Said University of Djelfa, BP 3117, Djelfa, Algeria. http://orcid.org/0000-0002-8824-9761 (não autenticado)
Lakhdar Mazouz Renewable Energy Systems Applications Laboratory (LASER), Department of Electrical Engineering in the Faculty ofScience and Technology, Ziane Achour University of Djelfa, PO Box 3117, Djelfa 17000, Algeria. http://orcid.org/0000-0002-7664-9901 (não autenticado)
Toufik Tayeb NAAS Renewable Energy Systems Applications Laboratory (LASER), Department of Electrical Engineering in the Faculty ofScience and Technology, Ziane Achour University of Djelfa, PO Box 3117, Djelfa 17000, Algeria http://orcid.org/0000-0002-5539-6415 (não autenticado)
Özüpak Yildirim Department of Electrical Energy, Silvan Vocational School, Dicle University, 21002, Diyarbakır, Turkey http://orcid.org/0000-0001-8461-8702 (não autenticado)
Ridha Djamel Mohammedi Ziane Achour University of Djelfa, PO Box 3117, Djelfa 17000, Algeria http://orcid.org/0000-0003-4170-543X (não autenticado)

DOI:

https://doi.org/10.5935/jetia.v10i48.1185

Resumo

O motor síncrono de ímã permanente (PMSM) é um componente essencial em várias aplicações.
em várias aplicações, mas continua suscetível a uma série de falhas tanto no rotor quanto no estator.
Por isso, é imperativo resolver esses problemas de forma rápida e inteligente. Neste
estudo, foi adotada uma nova abordagem em que um projeto de PMSM foi conceituado no programa
programa Ansys Maxwell, seguido pela introdução deliberada de uma falha no nível magnético do rotor.
nível magnético do rotor. Especificamente, três cenários de falha distintos foram delineados com base no
número de ímãs quebrados (BM), a saber, 2, 3 e 4, localizados em áreas específicas do rotor.
Notavelmente, a densidade do fluxo magnético foi selecionada como o parâmetro focal para essa investigação.
Para detectar e diagnosticar com eficácia as falhas decorrentes de BM, uma arquitetura inovadora de rede neural convolucional
(CNN) inovadora foi desenvolvida. Aproveitando as imagens do projeto do PMSM
PMSM capturadas durante as fases operacionais em vários intervalos de tempo, a CNN demonstrou uma
notável eficácia no discernimento e na categorização de instâncias de falhas. Após a análise dos resultados derivados
resultados derivados, fica evidente que a CNN apresentou uma precisão sem igual na
detecção de falhas, alcançando uma notável taxa de sucesso de 100% quando comparada com
metodologias alternativas, como Support Vector Machines (SVM) e K-Nearest Neighbors (KNN),
que produziram taxas de precisão de 97%.

Downloads

Os dados de download ainda não estão disponíveis.

Referências

X. Sun, Z. Shi, G. Lei, Y. Guo, and J. Zhu, “Analysis and design optimization of a permanent magnet synchronous motor for a campus patrol electric vehicle”, IEEE Transactions on Vehicular Technology, vol. 68, no.11, pp.10535-10544, September 2019. DOI:10.1109/TVT.2019.2939794.

J. A. Luta, H. Liu, D. A. Moguel, S. D. Pekarek, and J. A. Weibel, “Validation of a Thermal Model for Electromagnetic Co-Optimization of Electric Machines”, in 2023 IEEE Energy Conversion Congress and Exposition (ECCE), IEEE, pp. 4099- 4104, October 2023. DOI: 10.1109/ECCE53617.2023.10362916.

Q. Huang, Q. Huang, H. Guo, and J. Cao, “Design and research of permanent magnet synchronous motor controller for electric vehicle,” Energy Science & Engineering, vol. 11, no. 1, pp. 112-126, 2023. DOI: https://doi.org/10.1002/ese3.1316.

P. N. Mueller, L. Woelfl, and S. Can, “Bridging the gap between AI and the industry—A study on bearing fault detection in PMSM-driven systems using CNN and inverter measurement”, Engineering Applications of Artificial Intelligence, vol. 126, p. 106834, November 2023. DOI: https://doi.org/10.1016/j.engappai.2023.106834.

A. Aggarwal, and E. G. Strangas, “Review of detection methods of static eccentricity for interior permanent magnet synchronous machine”, Energies, vol. 12, no. 21, October 2019. https://doi.org/10.3390/en12214105

M. Tang, A. Wang, and Z. Zhu, “Asynchronous Motor Fault Diagnosis Output Based on VMD-XGBoost,” IEEE 4th China International Youth Conference on Electrical Engineering (CIYCEE), pp. 1–7, December 2023. DOI: 10.1109/CIYCEE59789.2023.10401350.

S. Benkaihoul, L. Mazouz, and T. T. Naas, “Advanced Design of DC Machines Using Finite Element Method Simulator”, International Journal of Advanced Studies in Computers, Science and Engineering, vol. 11, no. 2, pp. 8–18, February 2022.

N. Praveen Kumar, and T. B. Isha, “Electromagnetic signature study of a closed loop speed controlled three-phase induction motor under broken rotor bar fault using finite element method”, journal of engineering science and technology, vol. 14, no. 5, pp. 2731–2745, October 2019.

K. Edomwandekhoe, and X. Liang, “Advanced feature selection for broken rotor bar faults in induction motors”, CIEEE/IAS 54th Industrial and Commercial Power Systems Technical Conference (I&CPS), vol. 2018-May, pp. 1–10, May 2018. DOI: 10.1109/ICPS.2018.8369981.

S. Muthukumaran, A. Rammohan, S. Sekar, M. Maiti, and K. Bingi, “Bearing fault detection in induction motors using line currents”, ECTI Transactions on Electrical Engineering, Electronics, and Communications, vol. 19, no. 2, pp. 209–219, June 2021. DOI: 10.37936/ecti-eec.2021192.244163.

E. Irgat, A. Ünsal, and H. T. Canseven, “Detection of Eccentricity Faults of Induction Motors Based on Decision Trees”, 13th International Conference on Electrical and Electronics Engineering (ELECO), pp. 435–439, November 2021. DOI: 10.23919/ELECO54474.2021.9677809.

S. E. Chehaidia, H. Cherif, M. Alraddadi, M. I. Mosaad, and A. M. Bouchelaghem, “Experimental Diagnosis of Broken Rotor Bar Faults in Induction Motors at Low Slip via Hilbert Envelope and Optimized Subtractive Clustering Adaptive Neuro-Fuzzy Inference System”, Energies, vol. 15, no. 18, September 2022.DOI: https://doi.org/10.3390/en15186746.

K. Barrera-llanga, and J. Burriel-valencia, “A Comparative Analysis of Deep Learning Convolutional Neural Network Architectures for Fault Diagnosis of Broken Rotor Bars in Induction Motors”, Sensors, vol. 23, no. 19, p. 8196, September, 2023. DOI: https://doi.org/10.3390/s23198196.

S. Ghosh, “Method for Fault Diagnosis and Speed Control of PMSM”, Computer System Science Engineering, vol. 45, no. 3, pp. 2391–2404, 2023. DOI: 10.32604/csse.2023.028931.

P. W. Thamke, N. M. Yewale, U. B. Vaidya, and P. G. Asutkar, “Faults associated with permanent magnet synchronous motor,” International Journal of Core Engineering & Management, vol. 2, no. 3, pp. 211–217, 2015.

F. Nian, and Y. Yu, “PMSM Demagnetization Fault Diagnosis Based on Back-EMF Signal”, in Journal of Physics: Conference Series, vol. 2395, no. 1, p. 012041, December 2022. DOI: 10.1088/1742-6596/2395/1/012041.

P. Ewert, T. Orlowska-kowalska, and K. Jankowska, “Effectiveness Analysis of PMSM Motor Rolling Bearing Fault Neural Networks”, Energies, vol. 14, no 3, p. 712. January 2021. DOI: https://doi.org/10.3390/en14030712.

E. Aslan, and Y. Özüpak, “Classification of Blood Cells with Convolutional Neural Network Model”, Bitlis Eren Üniversitesi Fen Bilimleri Dergisi, vol. 13, no. 1, pp. 314–326, March 2024.DOI: https://doi.org/10.17798/bitlisfen.1401294.

Q. Zhang, X. Wei, Y. Wang, and C. Hou, “Convolutional Neural Network with Attention Mechanism and”, Sensors, vol. 24, no. 6, p. 1831, March 2024. DOI: https://doi.org/10.3390/s24061831.

S. Benkaihoul, L. Mazouz, T. T. Naas, Ö. Yıldırım, and A. Regaz, “Magnetic rotor breakage study in permanent magnet synchronous motor at COMSOL multiphysics and fault detection using machine learning”, Studies in Engineering and Exact Sciences, vol. 5, no. 1, pp. 603–618, March 2024. DOI:

https://doi.org/10.54021/seesv5n1-034.

X. Zhang, B. Zhao, and Y. U. N. Lin, “Machine Learning Based Bearing Fault Diagnosis Using the Case Western Reserve University Data: A Review”, IEEE Access, vol. 9, p. 155598-155608, November 2021. DOI: 10.1109/ACCESS.2021.3128669.

M. C. Kim, J. H. Lee, D. H. Wang, and I. S. Lee, “Induction Motor Fault Diagnosis Using Support Vector Machine, Neural Networks, and Boosting Methods”, Sensors, vol. 23, no. 5, February 2023. DOI: https://doi.org/10.3390/s23052585.

P. Kumar, and A. S. Hati, “Review on Machine Learning Algorithm Based Fault Detection in Induction Motors”, Archives of Computational Methods in Engineering, vol. 28, no. 3, pp. 1929– 1940, 2021. DOI: https://doi.org/10.1007/s11831-020-09446-w.

L. V. Vaz, M. C. Gonçalves, I. C. P. Dias, and E. O. B. Nara, “Application of a production planning model based on linear programming and machine learning techniques”, ITEGAM-JETIA, vol. 10, no 45, p. 17-29, 2024. DOI: https://doi.org/10.5935/jetia.v10i45.920.

R. N. Zufar, and D. Banjerdpongchai, “Selection of Lightweight CNN Models with Limited Computing Resources for Drone Collision Prediction”, ECTI Transactions on Electrical Engineering, Electronics, and Communications, vol. 22, no. 1, pp. 1–10, February 2024. DOI: https://doi.org/10.37936/ecti-eec.2024221.251164.

A. Jenefa, E. Veemaraj, and A. Lincy, “ABM-OCD: Advancing ovarian cancer diagnosis with attention-based models and 3D CNNs”, ITEGAM-JETIA, vol. 9, no 43, p. 23-33, 2023. DOI: https://doi.org/10.5935/jetia.v9i43.904.

D. Ruan, J. Wang, J. Yan, and C. Gühmann, “CNN parameter design based on fault signal analysis and its application in bearing fault diagnosis”, Advanced Engineering Informatics, vol. 55, p. 101877, October 2022. DOI: https://doi.org/10.1016/j.aei.2023.101877.

A. F. Amiri, S. Kichou, H. Oudira, A. Chouder, and S. Silvestre, “Fault Detection and Diagnosis of a Photovoltaic System Based on Deep Learning Using the Combination of a Convolutional Neural Network (CNN) and Bidirectional Gated Recurrent Unit (BiGRU)”, Sustainability, vol. 16, no. 3, p. 1012, January 2024. DOI: 10.3390/su16031012.