Residual fatigue life prediction of natural rubber components under variable amplitude loads

In the present study, the residual fatigue life (RFL) of natural rubber (NR) components under variable amplitude loads is predicted. To this end, a support vector machine (SVM) model is established to estimate the fatigue life of NR specimens under a constant amplitude load. Here, the strain amplitude and strain mean of NR specimens are used as input variables while the rubber fatigue life is considered as the output variable of the SVM model. It is found the regression results and predicted life distribution of the SVM model are promising. Then the load sequence and load interaction are considered in the calculations and a nonlinear fatigue cumulative damage (NFCD) model is developed. Finally, the RFL of rubber specimens under variable amplitude load is predicted using the Miner criterion and the NFCD model. The performed experiments demonstrate that the prediction accuracy of the NFCD model is higher than that of the Miner criterion.