A highly efficient multi-scale approach of locally refined nonlinear analysis for large composite structures

Shengya Li,Yiqi Mao,Wenyang Liu,Shujuan Hou
DOI: https://doi.org/10.1016/j.compstruct.2022.116578
IF: 6.3
2023-01-01
Composite Structures
Abstract:The behaviour of materials at the microscale is generally considered the main driver for the macroscopic response. However, it is still a great challenge in terms of computational cost and efficiency to carry out nonlinear multiscale analysis of advanced material structures, especially when dealing with large macro industrial structures. In fact, the materials nonlinear behaviour only occurs in a small part of structure and the microscopic evolution are mainly considered in this part. In this paper, the novelty lies in providing a multiscale finite element solution procedure based on the Nonlinear Criterion Function (NLCF) switching criteria for large composite structures along with the locally refined nonlinear analysis, to reduce computation time and storage costs. To this aim, a nonlinear multiscale analysis technique is performed based on the decomposition of the macroscopic integration point into linear stage and nonlinear stage during the calculation process. The two mature computational homogenization methods of finite element-based homogenization (FEH) approach and multi-level finite element approach (FE2) are used in linear stage and nonlinear stage, respectively, to perform a hybrid solution. In order to switch the two methods on-the-fly, the NLCF depending on the macroscopic integration point is established. The hybrid FEH + FE2 technique is implemented on ABAQUS via user-defined materials (UMAT) and the effectiveness and efficiency are verified by two key numerical examples of three-dimensional composite structures. Finally, the outstanding computational capability of the hybrid FEH + FE2 technique is further demonstrated through a large industrial application of an electric vehicle battery pack under the continuous impulse loads from the ground protrusions.
What problem does this paper attempt to address?