Agius Dylan, Mamun Abdullah Al, Truman Christopher, Mostafavi Mahmoud, Knowles David
Solid Mechanics Research Group, Department of Mechanical Engineering, University of Bristol, United Kingdom.
Nuclear Futures Institute, Bangor University, Gwynedd LL57 2DG, United Kingdom.
MethodsX. 2022 Jun 20;9:101763. doi: 10.1016/j.mex.2022.101763. eCollection 2022.
A tool to implement a length scale dependency to classical crystal plasticity simulations is presented. Classical crystal plasticity models do not include a size effect; therefore, the size of the grain does not influence the simulated deformation. Classical crystal plasticity advancements have been through the inclusion of stress or strain gradient based constitutive models to improve the simulation of length scale dependent deformation. However, this tool presents an alternative to implementing a length scale, where the influence of slip pile-up in the form of dislocations at grain boundaries as a potential to explaining the Hall-Petch effect in materials. This is achieved by calculating the slip distance in adjacent grains for each slip system, by assuming the total slip length spans the grain in the slip direction. These calculations can occur in two ways. The first is the analysis occurs at the start of the simulation, therefore, only occurs once. If this approach is used, the computational cost of this tool is minute. However, if the simulations consider large deformations, during which it is expected that the grains are going to undergo large rotations, then it would be advantageous to the have the tool recalculate the information during the analysis. Consequently, the computational cost would depend on the resolution of the modelled geometry, the number of grains, and the number of slip systems. The tool also provides a capability to develop constitutive models based on complex grain boundary features which can be implemented in classical crystal plasticity models and gradient based crystal plasticity models. The described calculation process is implemented through a Fortran subroutine, which has been designed to be easily used in crystal plasticity simulations. The presented tool also includes Python code designed to link with microstructures built using DREAM.3D to extract the required input data to the Fortran subroutine. The proposed tool is not limited to classical crystal plasticity formulations, instead the data extracted and outputted from the Fortran subroutine can be used to serve alternative purposes in both stress and strain gradient crystal plasticity models. The proposed tool can be modified to extract additional data to that presented. The slip distance in the adjacent grain, the distance from the grain boundary of the current calculation point, and the interaction between slip systems between grains can be used in any crystal plasticity constitutive models.
本文提出了一种在经典晶体塑性模拟中实现长度尺度依赖性的工具。经典晶体塑性模型不包括尺寸效应;因此,晶粒尺寸不会影响模拟变形。经典晶体塑性的发展是通过纳入基于应力或应变梯度的本构模型来改善对长度尺度依赖性变形的模拟。然而,该工具提供了一种实现长度尺度的替代方法,其中以晶界处位错形式存在的滑移堆积影响有可能解释材料中的霍尔 - 佩奇效应。这是通过为每个滑移系计算相邻晶粒中的滑移距离来实现的,假设总滑移长度在滑移方向上跨越晶粒。这些计算可以通过两种方式进行。第一种是在模拟开始时进行分析,因此,只发生一次。如果使用这种方法,该工具的计算成本很小。然而,如果模拟考虑大变形,在此期间预计晶粒会发生大旋转,那么让工具在分析过程中重新计算信息将是有利的。因此,计算成本将取决于建模几何的分辨率、晶粒数量和滑移系数量。该工具还提供了基于复杂晶界特征开发本构模型的能力,这些模型可在经典晶体塑性模型和基于梯度的晶体塑性模型中实现。所描述的计算过程通过一个Fortran子程序实现,该子程序设计为易于在晶体塑性模拟中使用。所提出的工具还包括Python代码,旨在与使用DREAM.3D构建的微观结构链接,以提取Fortran子程序所需的输入数据。所提出的工具不限于经典晶体塑性公式,相反,从Fortran子程序中提取和输出的数据可用于应力和应变梯度晶体塑性模型中的其他目的。所提出的工具可以修改以提取比所呈现的更多的数据。相邻晶粒中的滑移距离、当前计算点到晶界的距离以及晶粒间滑移系之间的相互作用可用于任何晶体塑性本构模型。