A Multiscale Computational Model Combining a Single Crystal Plasticity Constitutive Model with the Generalized Method...

Area of Science:

• Computational Materials Science
• Solid Mechanics
• Materials Engineering

Background:

• Predicting the elasto-plastic behavior of polycrystal metals requires understanding microstructural stress fields.
• Existing methods may face computational challenges when analyzing complex material structures.

Purpose of the Study:

• To develop and validate a multiscale computational model for elasto-plastic behavior in polycrystal metals.
• To integrate a single crystal plasticity constitutive model within a finite element analysis (FEA) framework using generalized method of cells (GMC) for homogenization.
• To assess the computational efficiency and accuracy of the developed model for microstructural analysis.

Main Methods:

• A multiscale model was developed by coupling FEA with a generalized method of cells (GMC) micromechanics model.
• The model employs a single crystal plasticity constitutive law to capture microstructural stress.
• Stand-alone GMC was used for simple microstructures (e.g., RUCs) and verified against FEA models before application to larger systems.

Main Results:

• The GMC homogenization combined with crystal plasticity accurately predicts von Mises stress at both average and individual grain levels.
• Significant computational cost savings (two to three orders of magnitude) were achieved with GMC.
• The model successfully analyzed a real-life engine disc component, revealing microstructural field details.

Conclusions:

• The developed multiscale model offers a computationally efficient and accurate approach for the elasto-plastic analysis of polycrystal metals.
• This method is promising for structural failure analysis, providing insights into microstructural behavior.
• The model demonstrates capability in solving real-world engineering problems at component scale.