基于有限元和光滑粒子流体动力学的硬质工具钢切屑形貌预测方法研究（英文）

Chip morphology predictions while machining hardened tool steel using finite element and smoothed particles hydrodynamics methods

目的:获得有效的高精度切屑形貌仿真方法。创新点:通过比较不同切削参数下采用光滑粒子流体动力学模型和有限元模型仿真获得的切屑形貌,证明光滑粒子流体动力学模型可以很好地实现对节状切屑的仿真,而不需要额外的几何或基于网格的切屑分离准则。方法:基于有限元和光滑粒子流体动力学的切削形貌仿真方法。结论:通过比较不同切削参数下采用光滑粒子流体动力学模型和有限元模型仿真获得的切屑形貌,证明了基于裂纹形成与扩展理论,采用合理疲劳参数的标准Johnson-Cook模型完全可以实现对节状切屑形成过程的仿真,也即无需采用修正的Johnson-Cook模型。同时证明了有限元模型和光滑粒子流体动力学方法均可满足不同切削速度和进给量条件下的切削力和切屑形貌仿真。

Chip morphology predictions in metal cutting have always been challenging because of the complexity of the various multiphysical phenomena that occur across the tool-chip interface. An accurate prediction of chip morphology is a key factor in the assessment of a particular machining operation with regard to both tool performance and workpiece quality. Although finite element（FE） models are being developed over the last two decades, their capabilities in modeling correct material flow around the tool tip with shear localization are very limited. FE models with an arbitrary Lagrangian Eulerian（ALE） approach are able to simulate correct material flow around the tool tip. However, these models are unable to predict any shear localization based on material flow criteria. On the other hand, FE models with a Lagrangian formulation can simulate shear localization in the chip segments; they need to make use of a mesh-based chip separation criterion that significantly affects material flow around the tool tip. In this study a mesh-free method viz. smoothed particles hydrodynamics（SPH） is implemented to simulate shear localization in the chip while machining hardened steel. Unlike other SPH models developed by some researchers, this model is based on a renormalized formulation that can consider frictional stresses along the tool-chip interface giving a realistic chip shape and material flow. SPH models with different cutting parameters are compared with the traditional FE models and it has been found that the SPH models are good for predicting shear localized chips and do not need any geometric or mesh-based chip separation criteria.