Fault severity assessment for rolling element bearings based on LMD and Lempel-Ziv index

For different rotation speeds and different inner or outer race defects severity, a fault severity assessment scheme based on local mean decomposition (LMD) and Lempel-Ziv index was put forward. In LMD method the bearing vibration signal was decomposed into several product functions (PF) with instantaneous frequency having definite physical meaning. The optimal PF component can be found according to kurtosis conditions. On the basis of the best PF, the normalized Lempel-Ziv values for the PF envelope were calculated. Then the values multiplied by given weights were summed up to form a final measure named the integrated Lempel-Ziv index. Different intervals of the index value correspond to different fault severity. At the same time, for making conclusions more universal, the bearing fault Lempel-Ziv index distribution rule of the inner and outer defects at different rotating speed was studied. The experiment results show that the algorithm can be effectively applied in rolling bearing fault diagnosis.