Classification of High Frequency Impact Signal in Vibrational Analysis of Spur Gears by using Convolutional Neural Networks
Keywords:
Periodic Preventive Maintenance, Fault Analysis, Gear Fault, Convolutional Neural NetworksAbstract
Spur gears are one of the widely used gears in a gearbox assembly. They often require lubrication and replacement of pinion and gears as prone to damage in high speed shafts with heavy loads and adverse working conditions. These creates spalling and breakage of gear tooth due to material fatigue from excessive loads and also forms pitting corrosion due to reduced lubrication and higher input shaft speeds. Vibrational analysis of these rolling elements is necessary for monitoring the health of gears periodically. These graphs provide a pattern waveform over time to study the characteristic high frequency impact noise signal peaks due to increased vibrations from faulty sections. This paper depicts about the implementation of convolutional neural networks to analyze the vibrational graphs obtained at different rotating speed of shafts for various gear ratios to plot the high frequency impact noise and train the neural networks to identify the peaks and classify among the faulty and healthy spur gears and pinions for a better way to reduce time in estimating the remaining average working life of gears and perform adequate maintenance of components.
Downloads
References
M. Watson, J. Sheldon, S. Amin, H. Lee, C. Byington, M. Begin, A Comprehensive High Frequency Vibration Monitoring System for Incipient Fault Detection and isolation of Gears, Bearings and Shafts/Couplings in Turbine Engines and Accessories, in: Volume 5: Turbo Expo 2007, ASMEDC, 2007. https://doi.org/10.1115/gt2007-27660.
D.C.H. Yang, J.Y. Lin, Hertzian Damping, Tooth Friction and Bending Elasticity in Gear Impact Dynamics, Journal of Mechanisms, Transmissions, and Automation in Design. 109 (1987) 189-196. https://doi.org/10.1115/1.3267437
S.Theodossiades, M. Gnanakumarr, H. Rahnejat, M. Menday, Mode identification in impact-induced high frequency vehicular driveline vibrations using an elasto-multi-body dynamics approach, Proceedings of the Institution of Mechanical Engineers, Part K: Jpurnal of Multi-Body Dynamics. 218 (2004) 81-94. https://doi.org/10.1243/146441904323074549.
F.K. Choy, V. Polyshchuk, J.J. Zakrajsek, R.F. Handschuh, D.P. Townsend, Analysis of the effects of surface pitting and wear on the vibration of a gear transmission system, Tribology International. 29 (1996)77-83. https://doi.org/10.1016/0301- 679x(95)00037-5.
G. DALPIAZ, A. RIVOLA, R. RUBINI, EFFECTIVENESS AND SENSITIVITY OF VIBRATION PROCESSING TECHNIQUES FOR LOCAL FAULT DETECTION IN GEARS, Mechanical Systems and Signal Processing. 14 (2000) 387-412. https://doi.org/10.1006/mssp.1999.1294
M. Izciler, M. Tabur, Abrasive wear behavior of different case depth gas carburization AISI 8620 gear steel, Wear. 260 (2006) 90-98. https://doi.org/10.1016/j.wear.2004.12.034
I.B. Goldman, Corrosive wear as a failure mode in lubricated gear contacts, Wear. 14 (1969) 431-444. https://doi.org/10.1016/0043-1648(69)90006-4
Vara Prasad, P. (2015). Detection of Gear Fault Using Vibration Analysis. International Journal of Research in Engineering and Science (IJRES). 3. 2320-9356.
V. Sze, Y.H. Chen, T.J. Yang, J.S. Emer, Efficient Processing of Deep neural Networks: A Tutorial and Survey, Proceedings of the IEEE. 105(2017) 2295- 2329, https://doi.org/10.1109/jproc.2017.2761740.
[Dataset] Juvith Ghosh, 2020. https://www.kaggle.com/dataset/889ac6ca70a0f06f52
b59e7964ead95a9c2a55a040ab1799194a443b008 01
Nikhil B., Nicholas L., 2017. Fundamentals of Deep Learning, first ed. United States of America, O.Reilly. [12]Batch normalization in Neural Networks: https://towardsdatascience.com/batch-normalization in-neural-networks-1ac91516821c.
Batch Normalization – Speed up Neural Network Training, Ilango R – Medium: https://medium.com/@ilango100/batch-normalization speed-up-neural-network-training-245e39a62f85.
Mark Hudson B., Martin T.H., Howard B.D., Deep Learning Toolbox – User’s Guide, MATLAB 2018b. https://www.mathworks.com/products/deep
learning.html
What are Max Pooling, Chris – MACHINECURVE: https://www.machinecurve.com/index.php/2020/01/30 /what-are-pooling-average-pooling-global-max pooling-and-global-average-pooling/
Fully Connected Layer: The brute force layer of a Machine Learning model, Surya Pratap Singh, OpenGenus IQ: https://iq.opengenus.org/fully connected-layer/
Softmax Layer by DeepAI: https://deepai.org/machine-learning-glossary-and terms/softmax-layer
Phil K. 2017. MATLAB Deep Learning with Machine Learning, Neural Networks and Artificial Intelligence, APress.
Kingma, Diederik & Ba, Jimmy. (2014). Adam: A Method for Stochastic Optimization, International Conference on Learning Representations.
Vincent V., Andrew S., Mark Z.M., 2011. Improving the speed of neural networks on CPUs. Deep Learning and Unsupervised Feature Learning Workshop, NIPS, California.
