Abstract: For the usual failure mode of structural components under variable loading, fatigue life prediction is very important for the selection, design, and safety assessments of these components. The linear damage model (Miner’s rule) is used most widely for life prediction in practical engineering application. However, the fatigue damage caused by small loads cannot be ignored, especially when the load is near the constant amplitude of fatigue limit of a component. This article discusses in detail an acumulative fatigue damage model and the corresponding formula of fuzzy fatigue life based on the damage and coaxing effect caused by each loading cycle on a metallic material. The fatigue life of structures depends on both the damage and coaxing effect of the small load and a new approach is proposed. This model has improved the application of Miner’s rule not only by considering the coaxing effect caused by a stress lower than the fatigue limit, but also by introducing membership functions in the fuzzy accumulative damage caused by small loads. In order to apply the proposed model conveniently, different membership functions that affect the result of estimating fatigue life are investigated.Two examples are given in which the structural fatigue life is estimated by two different load spectrums. By comparing the results, the law of selecting membership functions for different load spectrums is found, and the errors of predicted fatigue life are reduced. The examples show that the prediction of fatigue life by the proposed method is more accurate than by the traditional method.

Key words: fatigue life, coaxing effect, fuzzy accumulative damage Miner’s rule, membership functions, life prediction