Abstract
A gear transmission system is a complex nonstationary and nonlinear time-varying coupling system. When faults occur on gear system, it is difficult to extract the fault feature. In this paper, a novel fault diagnosis method based on ensemble empirical mode decomposition (EEMD) and Deep Briefs Network (DBN) is proposed to treat the vibration signals measured from gearbox. The original data is decomposed into a set of intrinsic mode functions (IMFs) using EEMD, and then main IMFs were chosen for reconstructed signal to suppress abnormal interference from noise. The reconstructed signals were regarded as input of DBN to identify gearbox working states and fault types. To verify the effectiveness of the EEMD-DBN in detecting the faults, a series of gear fault simulate experiments at different states were carried out. Results showed that the proposed method which coupled EEMD and DBN can improve the accuracy of gear fault identification and it is capable of applying to fault diagnosis in practical application.
Highlights
Gearbox is an indispensable part in the modern industry, especially for lots of large equipment [1]
The original vibration signals were decomposed into several intrinsic mode functions (IMFs) components with ensemble empirical mode decomposition (EEMD) method
A comparative analysis was carried out between coupled EEMD with Deep Briefs Network (DBN) and DBN without EEMD to verify the availability of the coupled method
Summary
Gearbox is an indispensable part in the modern industry, especially for lots of large equipment [1]. Failure of the gearbox leads to shutting down or threatening personal safety and causes considerable economic losses. It is very important for engineers and researchers to monitor the gear conditions to prevent this kind of malfunction of the plants. A lot of factors need to be considered when evaluating the performance status of the equipment, such as vibration, noise temperature, the debris contaminants in the oil and grease, torque of the power input and output, and stress distribution on the tooth surface. There are many methods for gearbox fault diagnosis, including vibration analysis [3], noise analysis [4], and oil analysis. Real-time, and nondamage, vibration signal analysis shows many advantages which make it widely used for gearbox fault diagnosis
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
