Chen Renxiang

Bearings in simulated space conditions running state detecton based on Tsallis entropy-KPCA and optimized fuzzy c-means model

Aiming at achieving the bearing running condition and fault recognition for different types and levels especially for bearings used in satellites in space conditions, this research proposes a method based on the Tsallis entropy and fuzzy c-means model optimized by the particle swarm optimization for bearing fault diagnosis research. The local mean decomposition method is used to decompose the signal into different product functions to separate the collected space bearing non-linear vibration signal in the simulated space condition. Then the Tsallis entropy is used to calculate the entropy of the product functions to obtain the entropy features of the decomposed product function signals, and the extracted features are processed by the kernel principal component analysis method to reduce the dimension and the obtained features for input to the fuzzy c-means model. A test data of Case Western Reserve University and some bearings vibration data in simulated space conditions were used to validate the proposed method. The results proved that the method can fulfill the demand of the space bearing faults level detection.