Study On Fatigue Elastic Life Prediction Based On Probabilistic Crystal Plastic Finite Element Method

Disposable machinery is mainly used in some military and aviation fields,which is a new concept proposed in recent years.Disposable machinery is different from conventional machinery,which mainly refers to a kind of non reusable machinery with self destruction as the main failure form.Fatigue elastic life is an important performance index of disposable machinery.In order to achieve high reliability design of disposable machinery,more accurate fatigue elastic life prediction results are needed.In this paper,the dislocation density is introduced as the model control parameter in the crystal plastic finite element method,and the probabilistic crystal plastic finite element model based on dislocation density is established,which can reduce the phenomenological effect of the complementary crystal plastic finite element method.Finally,the fatigue failure process of GCr15 bearing steel was simulated by using the established model,and the fatigue elastic life prediction results were obtained.The applicability and reliability of the model were verified by comparing with the experimental results.In order to obtain the actual fatigue elastic life of GCr15 bearing steel and the material parameters needed for simulation,the fatigue elastic life data of GCr15 bearing steel is obtained by stress controlled low cycle fatigue test.According to the stress-strain curve of bearing steel obtained from tensile test,the mechanical property parameters of GCr15 bearin g steel are determined.The grain size,grain orientation and initial dislocation density of GCr15 steel are measured by EBSD,which laid a foundation for the subsequent establishment of crystal plastic constitutive model.Then,the basic principle of crystal plasticity finite element method and dislocation density evolution law are studied,and the cyclic constitutive model of crystal plasticity based on dislocation density is established.The cumulative characteristics of dislocation dipoles under cyclic loading are simulated by abquas finite element analysis software combined with damask subroutine when the probability distribution of dislocation dipole spacing conforms to uniform distribution and Gaussian distribution respectively.The simulation results show that the constitutive model can simulate the accumulation process of dislocation dipoles under cyclic loading well when the distance between dislocation dipoles conforms to Gaussian distribution.Based on the single crystal model,a multi grain repre sentative volume element(RVE)model is established by further considering the grain size and grain number.In order to reflect the dispersion of mechanical properties and microstructure of metal materials,the probability simulation method based on dislocation density is adopted.The dislocation density in accordance with Gaussian distribution is randomly assigned to each grain of the crystal plastic finite element model,and the uncertainty of the microscopic model is used to reflect the dispersion of macro materials.Finally,the fatigue elastic failure process of GCr15 bearing steel is simulated and the fatigue elastic life is predicted.By comparing the predicted results with the experimental results,it is found that the constitutive model established in this paper can be well applied to the prediction of fatigue elastic life under unidirectional cyclic loading.