پایش وضعیت یاتاقان‌های غلتشی به روش ارتعاشی با بهره‌گیری از مدل یادگیری ماشین

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشکده مهندسی مکانیک، دانشگاه صنعتی اصفهان، اصفهان، ایران.

2 پژوهشکده سیستمهای مکاترونیک، دانشگاه علمی کاربردی زوریخ، زوریخ، سوئیس.

چکیده

در سال‌های اخیر با ظهور انقلاب صنعتی چهارم و توسعه فناوری‌های هوش مصنوعی، رویکردهای نوینی در زمینه نگه‌داری و تعمیر افزارگان معرفی‌شده‌اند؛ که از آن جمله می‌توان به همزاد دیجیتال اشاره نمود. اولین گام برای ایجاد همزاد دیجیتال یک افزار، ساخت شاخص‌های کمّی و کیفی است که برای توصیف لحظه‌ای افزار در طی مدت بهره‌برداری به کار می‌رود. در این پژوهش یک روش نوین برای ساخت شاخص سلامت افزارگان براساس اندازه‌گیری ارتعاشات و مدل‌های یادگیری عمیق معرفی‌شده است. برای این منظور داده‌های ارتعاشی خام تجهیز با استفاده از تبدیل موجک پیوسته به تصاویر دوبعدی تبدیل خواهند شد. سپس با استفاده از یک مدل یادگیری عمیق، میزان تفاوت تصاویر وضعیت سالم و معیوب افزار تشخیص داده‌شده و شاخص سلامت ایجاد می‌شود. مدل ارائه‌شده می‌تواند به‌صورت خودکار شاخص سلامت را ایجاد نماید و نیازمند دانش متخصص خبره برای تفسیر نتایج آنالیز ارتعاشی نیست. همچنین، آموزش مدل یادگیری عمیق به‌صورت بدون نظارت بوده و تنها با استفاده از داده‌های ارتعاشی وضعیت سالم تجهیز صورت می‌پذیرد و بنابراین نیازمند داده‌های خرابی پیشین نیست. عملکرد مدل پیشنهادشده توسط داده‌های ارتعاشی یاتاقان مورد ارزیابی قرارگرفته که نشان از توانایی قابل‌قبول شاخص سلامت در تشخیص آغاز خرابی و چگونگی روند رشد آن دارد.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

A New Machine Learning Method for Ball Bearing Condition Monitoring Based on Vibration Analysis

نویسندگان [English]

  • mohammadreza kaji esfahani 1
  • Jamshid Parvizian 1
  • Mohammad Silani 1
  • Hans Wernher van de Venn 2
1 Mechanical En.g Department,, Isfahan University of Technology, Isfahan, Iran
2 Institute of Mechatronic Systems, Zurich University of Applied Sciences
چکیده [English]

In recent years, with the advent of the Fourth Industrial Revolution concepts and the development of artificial intelligence technologies, new approaches such as the digital twin have been introduced. In a digital twin, a virtual counterpart of the physical system during its whole life is created, with abilities such as analyzing, evaluating, optimizing, and predicting. The first step in creating a digital twin model is to construct a (multi) digital health indicator that describes different aspects of the physical component state during the whole life of the component. In this research, a new method for constructing health indicators based on vibration measurement and a deep learning model has been introduced. For this purpose, the Continuous Wavelet Transform was used to convert the raw vibration signals into two-dimension images; Then, the deep learning model was used to extract features from the images and the health indicator is constructed based on the differences of the images in normal and failure stages. In this article, various Autoencoder architectures are discussed, and it is demonstrated that the Convolutional Autoencoder has better performance in terms of detecting incipient faults. The performance of the proposed model is evaluated by the vibration data of the bearing, and the constructed health indicator exhibited a monotonically increasing degradation trend and had good performance in terms of detecting incipient faults.

کلیدواژه‌ها [English]

  • Condition monitoring
  • Artificial intelligence
  • Deep learning
  • Vibration analysis
  • Digital twin
[1] L.Z. Zepeng Liu, A Review of Failure modes, Condition Monitoring and Fault Diagnosis Methods for Large-Scale Wind Turbine Bearings, Measurement, 149 (2020).
[2] K.-L.T. Dong Wang, Qiang Miao Prognostics and Health Management: A Review of Vibration Based Bearing and Gear Health Indicators, IEEE Access, 6 (2017) 665 - 676.
[3] B.S. Gandhare, Maintenance Strategy Selection, in:  Ninth AIMS International Conference on Management, 2012.
[4] R.K. Mobley, An introduction to predictive maintenance Butterworth-Heinemann, America, 2002.
[5] B.S. Gandhare, Maintenance Strategy Selection, presented at the Ninth AIMS International Conference on Management,  (2012).
[6] B.A. Gandhare, Milind, Maintenance Strategy Selection, in:  The 9th AIMS International Conference on Management, 2012.
[7] N.L. Liang Guo, Feng Jia, Yaguo Lei, Jing Lin, A Recurrent Neural Network Based Health Indicator for Remaining Useful Life Prediction of Bearings, Neurocomputing, 240 (2017) 98–109.
[8] S.H.U. Akhand Rai, A Aeview on Signal Processing Techniques Utilized in the Fault Diagnosis of Rolling Element Bearings, Tribol. Int., 96 (2016) 289-306.
[9] K.-C.L.G.G. Yen, Wavelet Packet Feature Extraction for Vibration Monitoring, IEEE Transactions on Industrial Electronics, 47(3) (2000) 650 - 667.
[10] M.S.H. J. L. Won Gi Lee, Sung-Ho Nam, YongHo Jeon, andMoon G. Lee, Failure Diagnosis System for a Ball-Screw by Using Vibration Signals, Hindawi Publishing Corporation Shock and Vibration,  (2015).
[11] G.P.C. Christopher Torrence, A Practical Guide to Wavelet Analysis, American Meteorological Society, 79(1) (1998) 61-78.
[12] Y.H. C. C. P. Tsai, Ball Screw Preload Loss Detection Using Ball Pass Frequency, Mechanical Systems and Signal Processing, 48 (2014) 77-91.
[13] J.L. Y. L. Feng Jia, Xin Zhou, and Na Lu, Deep Neural Networks: A Promising Tool for Fault Characteristic Mining and Intelligent Diagnosis of Rotating Machinery with massive Data, Mechanical Systems and Signal Processing, 72 (2016) 303-315.
[14] W.-L.Q. Z.-Y. W. Chen Lu, Jian Ma, Fault Diagnosis of Rotary Machinery Components Using a Stacked Denoising Autoencoder-Based Health State Identification, Signal Processing, 130 (2017) 377–388.
[15] J.-G.B. Youngji Yoo, A Novel Image Feature for the Remaining Useful Lifetime Prediction of Bearings Based on Continuous Wavelet Transform and Convolutional Neural Network, Appl. Sci., 8(7) (2018).
[16] S.S. Wathiq Abed, Robert Sutton, Amit Motwani A Robust Bearing Fault Detection and Diagnosis Technique for Brushless DC Motors Under Non-stationary Operating Conditions, J. Control. Autom. Electr. Syst., 26 (2015) 241–254.
[17] M.O. Jun He, Chen Yong, Danfeng Chen, Jing Guo, Yan Zhou, A Novel Intelligent Fault Diagnosis Method for Rolling Bearing Based on Integrated Weight Strategy Features Learning, sensors, 20(6) (2020).
[18] Z.L. Hang Yin, Jiankai Zuo, Hedan Liu, Kang Yang, Fei Li, Wasserstein Generative Adversarial Network and Convolutional Neural Network (WG-CNN) for Bearing Fault Diagnosis, Math. Probl. Eng., 2020 (2020).
[19] S.Z. Shen Zhang, Bingnan Wang, Thomas G. Habetler, Deep Learning Algorithms for Bearing Fault Diagnostics—A Comprehensive Review, IEEE Access 8(2020) 29857 - 29881.
[20] T.Y. Samir Khan, A Review on the Application of Deep Learning in System Health Management, Mech. Syst. Signal Process, 107 (2018) 241-265.
[21] J.P.a.H.W.v.d.V. Mohammadreza Kaji, Constructing a Reliable Health Indicator for Bearings Using Convolutional Autoencoder and Continuous Wavelet Transform, Appl. Sci.,  (2020).
[22] T.G.H. Wei Zhou, Ronald G. Harley, Bearing Condition Monitoring Methods for Electric Machines: A General Review, in:  2007 IEEE International Symposium on Diagnostics for Electric Machines, Power Electronics and Drives, Cracow, Poland, 2007.
[23] S.V.K. Prashant P. Kharche, Review of Fault Detection in Rolling Element Bearing, International Journal of Innovative Research in Advanced Engineering, 1(5) (2014) 169-174.
[24] S.C.S. P.K. Kankar, S.P. Harsha, Fault Diagnosis of Ball Bearings Using Continuous Wavelet Transform, Appl. Soft Comput., 11 (2011) 2300–2312.
[25] PCoE Datasets, Bearing Data Set, Intelligent Maintenance Systems (IMS), University of Cincinnati, in.
[26] A.B. Andrea Borghesi, Luca Benini, Anomaly Detection using Autoencoders in High Performance Computing Systems, in:  The Thirty-First AAAI Conference on Innovative Applications of Artificial Intelligence, Hawaii, USA, 2019.
[27] G.G. Fangyi Wan, Chunlin Zhang, Qing Guo, Jie Liu, Outlier Detection for Monitoring Data Using Stacked Autoencoder, IEEE Access 7(2019) 173827 - 173837.
[28] M.S.L. Ngui Wai Keng, Lim Meng Hee, Ahmed. M. Abdelrhman, Wavelet Analysis: Mother Wavelet Selection Methods, Applied Mechanics and Materials, 393 (2013) 953-958.
[29] K.-C.L. G.G. Yen, Wavelet packet feature extraction for vibration monitoring, IEEE Transactions on Industrial Electronics 47(3) (2000) 650 - 667.
[30] H.Y. Yasi Wang, Sicheng Zhao, Auto-Encoder Based Dimensionality Reduction, Neurocomputing, 184 (2015) 232-242.
[31] F. Chollet, Keras: Deep Learning Library for Theano and Tensorflow.
[32] H.T. Ahmed Ali Mohammed Al-Saffar, Mohammed Ahmed Talab, Review of Deep Convolution Neural Network in Image Classification, in:  International Conference on Radar, Antenna, Microwave, Electronics, and Telecommunications, IEEE, Jakarta, Indonesia 2018.
[33] A.M. Saleh Albelwi, A Framework for Designing the Architectures of Deep Convolutional Neural Networks, Entropy, 19(6) (2017).
[34] M.H.L. Jeongyoun Ahn, Jung Ae Lee, Distance-Based Outlier Detection for High Dimension, Low Sample Size Data, J. Appl. Stat., 46(1) (2019) 13-29.