In this work, we present a model parameter calibration procedure for a physics-based crystal plasticity model. The calibration process utilizes a powerful statistics-based Bayesian calibration method. Calibration of the crystal plasticity parameters makes use of experimentally-measured data, i.e. compressive stress–strain response, from $〈100〉$ and $〈123〉$ single crystal copper dynamically loaded via Kolsky bar tests. The calibration of damage parameters is achieved using experimentally-measured free-surface velocity history data from plate impact test on $〈100〉$ and $〈110〉$ single crystal copper, which generates shock compression followed by dynamic tensile failure. A validation assessment is then carried out by comparing the calibrated model predictions and experimental measurements of the dynamic tensile damage generated in an impacted bicrystal copper plate. Lastly, a model-informed rationale for the experimentally-observed dependence of the spatial distribution of ductile damage (porosity) on crystallography is provided.

在这项工作中，我们提出了基于物理的晶体可塑性模型的模型参数校准程序。校准过程利用了强大的基于统计的贝叶斯校准方法。晶体可塑性参数的校准利用了实验测量的数据，即压缩应力-应变响应，来自$〈100〉$ 和 $〈123〉$通过Kolsky条形测试动态加载的单晶铜。损伤参数的校准是使用实验测量的自由表面速度历史数据（来自板撞击测试）完成的$〈100〉$ 和 $〈110〉$单晶铜，会产生冲击压缩，随后发生动态拉伸破坏。然后，通过比较经校准的模型预测值和在受冲击的双晶铜板中产生的动态拉伸损伤的实验测量结果，进行验证评估。最后，提供了模型学上的理论依据，以实验观察到的延性破坏（孔隙度）的空间分布对晶体学的依赖性。