摘要
Gear transmissions are widely used in industrial drive systems.Fault diagnosis of gear transmissions is important for maintaining the system performance,reducing the maintenance cost,and providing a safe working environment.This paper presents a novel fault diagnosis approach for gear transmissions based on convolutional neural networks(CNNs)and decision-level sensor fusion.In the proposed approach,a CNN is first utilized to classify the faults of a gear transmission based on the acquired signals from each of the sensors.Raw sensory data is sent directly into the CNN models without manual feature extraction.Then,classifier level sensor fusion is carried out to achieve improved classification accuracy by fusing the classification results from the CNN models.Experimental study is conducted,which shows the superior performance of the developed method in the classification of different gear transmission conditions in an automated industrial machine.The presented approach also achieves end-to-end learning that ean be applied to the fault elassification of a gear transmission under various operating eonditions and with signals from different types of sensors.
Gear transmissions are widely used in industrial drive systems.Fault diagnosis of gear transmissions is important for maintaining the system performance,reducing the maintenance cost,and providing a safe working environment.This paper presents a novel fault diagnosis approach for gear transmissions based on convolutional neural networks(CNNs) and decision-level sensor fusion.In the proposed approach,a CNN is first utilized to classify the faults of a gear transmission based on the acquired signals from each of the sensors.Raw sensory data is sent directly into the CNN models without manual feature extraction.Then,classifier level sensor fusion is carried out to achieve improved classification accuracy by fusing the classification results from the CNN models.Experimental study is conducted,which shows the superior performance of the developed method in the classification of different gear transmission conditions in an automated industrial machine.The presented approach also achieves end-to-end learning that ean be applied to the fault elassification of a gear transmission under various operating eonditions and with signals from different types of sensors.
基金
supported byan ENGAGE Grant from the Natural Sciences and Engineering Research Council of Canada(NSERC),[funding reference number 11R01296].
作者简介
Min XIA received the B.S.degree from Southeast University in 2009,the o M.S.degree from the University of Science and Technology of China in 2012,and Ph.D.degree.in mechanical engineering from the University of British Columbia(UBC)in 2017.He is currently a postdoctoral research fellow at UBC.His re-search interests include machine condition monitoring,deep neural networks,wireless sensor network,and sensor fusion.E-mail:minx ia@mech.ubc.ca;Clarence W.DE SILVA received the Ph.D.degree in mechanical engineering from the Massachusetts I nstitute of Technology,Cambridge,MA,USA,in 1978,and also the Ph.D.degree in information engineering from the Uni-versity of Cambridge,Cambridge,U.K.,in 1998,and the honorary D.Eng:degree from the Uni-versity of Waterloo,Waterloo,ON,Canada,in 2008.Since 1988,he has been a Professor of mechanical engi-neering at UBC.His other appointments include the Tier Canada Research Chair in Mechatronics and Industrial Auto-mation,a Professorial Fellow,the Peter Wall Scholar,the Mobil Endowed Chair Professor,and the NSERCBC Packers Chair in Industrial Automation.He has authored 24 books and approximately 550 papers,approximately half of which are in journals.His recent books include Modeling of Dynamic Sys-tems-With Engineering Applications(Taylor&Francis/CRC,2018),Sensor Systems(Taylor&Francis/CRC Press,2017),Sensors and Actuators Engineering System Instru-mentation,2nd edition(Taylor&Francis/CRC Press,2016),Mechanics of Materials(Taylor&Francis/CRC Press,2014),Mechatronics-A Foundation Course(Taylor&Francis/CRC Press,2010),Modeling and Control of Engi-neering Systems(Taylor&Francis/CRC Press,2009),,VIBRATION-Fundamentals and Practice 2nd edition(Taylor&Francis/CRC Press,2007),and Soft Computing and Intelligent Systems Design-Theory,Tools,and Appli-cations(with F.Karray:Addison Wesley,2004).Dr.de Silva is a fellow of:The Institute of Electrical and Electronics Engineers(IEEE),the American Society of Mechanical Engineers(ASME),the Canadian Academy of Engineering,and the Royal Society of Canada.E-mail:desilva@mech.ubc.ca
