Hello,
I'm looking to conduct simulations of RVEs (unidirectional discontinuous carbon fibers in resin matrix) utilizing the embedded finite elements technique within Abaqus. I've been successful in prescribing periodic boundary conditions (PBC) using the "*Equation" keyword and corresponding kinematic equations. However, I've encountered convergence issues specifically when running these simulations with Nlgeom = ON in Abaqus/Standard (interestingly, when Nlgeom = OFF, the convergence problems vanish entirely). Consequently, I'm considering employing Abaqus/Explicit to address these challenges.
The setback arises when attempting to implement PBCs with Abaqus/Explicit, triggering the following error message:
"Abaqus/Explicit has formed at least one very large implicit system of equations associated with enforcing constraints. Solving this system of equations would dramatically degrade performance."
While it's possible to proceed with the simulation by incorporating "*Diagnostics, Large Implicit System=Warning" into the model keywords, this approach comes at the cost of diminished performance. Even for small RVEs, the computational time drastically increases.
I'm curious if it exists a more effective method for prescribing PBCs when utilizing the Abaqus/Explicit Solver?
To address convergence issues in Abaqus/Standard with Nlgeom = ON, consider using Abaqus/Explicit solver for simulations with periodic boundary conditions (PBCs). However, implementing PBCs in Abaqus/Explicit may lead to large implicit systems of equations, affecting performance. Adjusting contact settings, reducing explicit time increments, and exploring alternative constraint equations can help improve convergence. Simplifying or refining the model and seeking expert advice are additional steps to consider for effective PBC implementation in Abaqus/Explicit.
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling