Intelligent health evaluation method of slewing bearing adopting multiple types of signals from monitoring system

Slewing bearing, which is widely applied in tank, excavator and wind turbine, is a critical component of rotational machines. Standard procedure for bearing life calculation and condition assessment has been established for general rolling bearings. Nevertheless, relatively less literatures in regard to the health condition assessment of slewing bearing has been published in past. Real time health condition assessment for slewing bearing is used for avoiding catastrophic failures by detectable and preventative measurement. In this paper, a new strategy is presented for health evaluation of slewing bearing based on multiple characteristic parameters, and ANN (Artificial Neural Network ) and ANFIS(Adaptive Neuro-Fuzzy Inference System ) models are demonstrated to predict the health condition of slewing bearings. The prediction capabilities offered by ANN and ANFIS are shown by data obtained from full life test of slewing bearings in NJUT test System. Various statistical performance indexes have been utilized to compare the performance of two predicted models. The results suggest that ANFIS-based prediction model outperforms ANN models.