Research On Fatigue Life Prediction Method Based On Ductile Dissipation Model

As one of the main failure modes of most of the mechanical structures,fatigue failure often appears in many fields.It results in an endless stream of accidents.Therefore,in-depth study of fatigue problems,accurate prediction and evaluation of mechanical structure fatigue life can protect its safe and reliable operation within the service period.As a result,major accidents and disasters can be avoided or reduced.This is of great significance for promoting China’s social and economic construction and sustainable development.Due to the complexity and randomness of fatigue failure process,the traditional life prediction theory is not perfect enough.There are still many unsolved problems and shortcomings.In this paper,Ye Duyi model and its modified model of nonlinear cumulative damage theory are studied.The Miner’s Law is systematically expounded and analyzed.This paper points out that it did not consider the loading order and the interaction effect between loads.Aiming at this defect,the definition of nonlinear damage is given,and the damage definitions of several typical models are compared and analyzed.Taking power function form as the example,the key factors for considering load sequence effect in the nonlinear damage accumulation model are pointed out.The influence of the interaction between different loads on the damage evolution law is analyzed and discussed in detail.The Ye Duyi model and its correction model have been compared and analyzed in terms of formula derivation and residual life prediction.It is verified that the stress ratio can effectively reflect the effects of mutual effects when it used as characterization amount.Based on that,the coefficient correction model is further improved.The validity of the model is verified by comparing the experimental data of typical materials.Based on the polynomial chaos theory which has a better ability to deal with the uncertainty,a probabilistic fatigue life prediction model considering the uncertainty of S-N curve was established by the exponential correction model and the coefficient simplification model,.The calculation analysis of typical material’s experimental proves the rationality of the proposed modal.The fatigue life prediction method based on generalized polynomial chaos method provides another effective calculation method for the fatigue life prediction of components.It improves the fatigue life prediction method based on S-N curve and cumulative damage theory.That the fatigue life prediction method based on generalized polynomial chaos method is similar to the Monte Carlo method and more efficient,which proves the rationality of the model.Meanwhile,the fatigue prediction method based on S-N curve and accumulation damage theory is improved.This paper takes the weld joint of intercity EMU bogie as the research object.Combined with the exponential correction model and the coefficient simplification model put forward by this paper,the calculation for the fatigue life of the key weld joints is carried out by applying the equivalent structural stress method.The results show that the prediction results of the exponential modified model and the simplified coefficient model proposed in this paper can well reflect the influence of load sequence and interaction between loads on fatigue damage,and overcome the problem that the Miner model’s prediction results are conservative.Based on that,chaotic polynomial method is introduced to characterize the uncertainty of the main S-N curve method and the probabilistic life of the key weld joints are predicted.Compared with the fatigue life prediction results of the main S-N curve method with a given probability,the fatigue life prediction results based on the chaotic polynomial method can give the probability of different damage degrees.It is more useful for engineering guidance.Through the research,the exponential modified model and the simplified coefficient model proposed in this paper are combined with the uncertainty of S-N curve,which provides a new idea for the fatigue life prediction of components.