Vibration Characteristics Diagnosis and Estimation of Fault Sizes in Rolling Contact Bearings: A Model Based Approach
Abstract A novel model based technique is presented in this paper for the estimation of the fault size in different components of rolling contact bearings. A detailed dimensional analysis of the problem is carried out and experimental methodology using Box-Behnken design is applied to generate the experimental data set. First, analysis of the vibration acceleration amplitude at the fault frequency, its dependence on the bearing operating and fault parameters using the obtained vibration data set is carried out by statistical analysis of variance. Numerical equations are developed then using the experimental data set for the correlation of the vibration acceleration amplitude in frequency domain with the fault sizes based on the developed dimensionless terms. A hybrid Back propagation neural network integrating genetic algorithms is also developed so as to check the computational performance of the developed model equations. Validation of the proposed method is carried experimentally also for three seeded defect sizes on outer race, inner race and rolling element. The maximum model accuracy observed is for inner race defect case with predictive accuracy of 99.44 percentage and for the roller defect case it is 98.77 percentage. The deviance observed for the model predictive performance is maximum for outer race defect case with least accuracy of 90.47 percentage amongst all.