Developing and applying OEGOA-VMD algorithm for feature extraction for early fault detection in cryogenic rolling bearing

Abstract

For ensuring the safe transportation of Liquefied natural gas (LNG), a cryogenic bearing failure detection test platform is designed. Meanwhile, an adaptive extraction algorithm (OEGOA-VMD) based on optimized grasshopper optimization algorithm (OEGOA) and variational modal decomposition (VMD) is proposed. Firstly, the search process of grasshopper optimization algorithm population is combined with differential evolution. Meanwhile, a new exponentially optimized mean characteristic energy ratio (OCFER) is introduced as the fitness function of OEGOA. Then, the optimal parameters of the VMD are obtained with the maximization of OCFER as the objective function, and the optimized VMD is used to decompose the bearing signal. Finally, Hilbert algorithm based on fast Fourier transform (FFT) is used to process the envelope of the decomposed and reassembled signal. Compared with other popular methods, the OEGOA-VMD algorithm shows better adaptability in fault feature extraction of LNG cryogenic rolling bearings, avoiding the occurrence of local optimal phenomenon.