基于直角切削的高温合金John-Cook本构参数逆向识别

With the development of computer technology, finite element simulation is increasingly used in the fields of parameter optimization, cutting mechanism research etc. John-Cook (J-C) constitutive model,the foundation of superalloy cutting simulation, is immediately related to the accuracy and reliability of simulation results. Centring on the problem that the material flow characteristics obtained by traditional Split Hopkinson pressure bar (SHPB) test cannot accurately describe the thermoplastic deformation of materials in cutting process and the results of parameter identification are quite different from the actual cutting conditions, a reverse identification method for J-C constitutive parameters of superalloy based on OXLEY cutting theory and orthogonal cutting test is put forward. Firstly, considering the model of unequal shear zone, the distribution regularity of strain, strain rate and shear temperature in cutting process is brought out; secondly, in view of the impossibility about directly comparing traditional turning with its two-dimensional simulation model, an orthogonal cutting experimental platform is designed and built; thirdly, based on the unequal shear zone model and the orthogonal cutting test, the theoretical and experimental values of the flow stress on the main shear plane are calculated respectively, and the least square identification of the five J-C constitutive parameters (initial yield stress A, strain strengthening coefficient B, strain rate strengthening coefficient C, thermal softening coefficient m, work hardening index n) is realized by searching the optimal combination of constitutive parameters with genetic algorithm by setting the minimum gap between the theoretical value and the experimental value as the objective function under the constraint condition of the constitutive parameters to search for its optimal combination. Eventually, the comparison between the simulation results of J-C constitutive parameters and the experimental ones is made to verify the feasibility of the reverse identification method and the accuracy of the constitutive model.