SUMMARY
Metal forming processes (examples include rolling, extrusion, forging, deep drawing, and die forming) form the backbone of today’s modern manufacturing industries where the products are shaped and often strengthened by plastic deformation. Crystal plasticity models allow for incorporation of physics of plastic deformation at length scales smaller than a single crystal by accounting for the details of microstructure and the crystallography in a polycrystalline sample. However, crystal plasticity models are extremely computationally expensive, limiting their adoption into practice. In this thesis, this limitation is addressed by using a recently developed spectral database approach based on discrete Fourier transforms (DFTs). Significant improvements were made to the prior approach and a new spectral database and framework were developed. The efficiency of this fast and robust numerical tool is demonstrated by a significant reduction in the computational cost involved in crystal plasticity based forming limit diagram (FLD) predictions. FLD is very important safety tool in designing sheet metal forming process. The improved framework also involves the integration of spectral database with commercial FE package ABAQUS through a user materials subroutine, UMAT, to conduct more efficient crystal plasticity finite element simulations (CPFEM) of full multiscale forming processes widening the applicability of the approach. The framework is validated through example case studies on Face Centered Cubic (FCC) metals by comparing against the corresponding results from the conventional CPFEM. Proper utilization of toolsets presented in this thesis will be helpful to the materials development community and manufacturing industry for performing high fidelity and computationally efficient multiscale simulations helping in the accelerated insertion of new and improved materials into practice.