Application of EEMD and Cloud Similarity Measurement in Gearbox Failure Feature Extraction

The gearbox is widely used in mechanical equipment. Due to the complexity of their working environment, gears, bearings, etc. will have varying degrees of damage, fracture and other problems, and they will not work. In this paper, the algorithm combining EEMD and cloud similarity is used to filter the sensitive IMF components in the gearbox fault signal to help feature extraction achieve better results. First, the gearbox signal is decomposed by EEMD to obtain a series of intrinsic mode function (IMF) with frequencies ranging from high to low. Then use the proposed cloud model similarity measurement method to identify the false IMF components and delete them, and the sensitive IMF is obtained. The simulation results show that the modified method is feasible.