Research on Fault Diagnosis of Planetary Gearbox Based on Hierarchical Extreme Learning Machine

Abstract

Currently, the planetary gear box health monitoring system has collected a huge amount of data, and the data needs to be quickly learned and real-time monitoring diagnostic requirements. The traditional fault diagnosis methods mostly need a complex signal processing process in advance and there are fewer layers, the feature extraction and classification effect are not ideal. In order to diagnose the planetary gearbox effectively, this paper presents a fault diagnosis method for planetary gearbox based on hierarchical extreme learning machine (H-ELM). This method analyses the time domain signal of fault vibration instead of the frequency domain signal, thus eliminates the time for complex signal processing to adaptively mine available fault characteristics and automatically identify machinery health conditions. The Stacked Denoising Auto-encoders (SDAE) and the Deep Belief Network (DBN) were used to test the diagnosis data of planetary gearbox, and make the comparison with hierarchical extreme learning machine methods. The experimental results show that the method has good effect and application value in the fault diagnosis of planetary gearbox.