RubenDeur
Mechanical
- Jul 31, 2017
- 2
Dear all,
For the simulation of Acoustic Emission (acoustic stress waves that form during crack growth), I am trying to prestress a model in Abaqus/Implicit, until the point where crack growth would occur. After this step, I restart the simulation with a Dynamic Explicit simulation, in which I release one of the boundary conditions at the crack tip. Although this works well, I do not have the computational capacity to run the Implicit simulation with the desired mesh resolution for the Explicit simulation.
I have tried replacing the static preload step with a quasi-static one (using Abaqus/Explicit), but still end up with very long computational times, because of the very fine mesh.
I also tried to perform the static preload step using a coarse mesh. Here I would load the model with a force, and extract the displacements for each node. With MATLAB I interpolate the results to a finer mesh, and generate a .inp file to run the dynamic simulation. When releasing these displacements, but not the load, the model obtains it prestressed state. This however also doesn't seem to work, because I need to apply displacements (in 3 directions) at every single node in the input file. It takes a very long time before the model is generated. (I am using " *boundary, _pickednode, dof,dof, value " for each nodal displacement).
Does anyone have a better solution than the ones I am trying? Or maybe there is a way to do the input of the interpolated displacements in a more efficent way? Any help would be very much appreciated!
Kind regards,
Ruben
For the simulation of Acoustic Emission (acoustic stress waves that form during crack growth), I am trying to prestress a model in Abaqus/Implicit, until the point where crack growth would occur. After this step, I restart the simulation with a Dynamic Explicit simulation, in which I release one of the boundary conditions at the crack tip. Although this works well, I do not have the computational capacity to run the Implicit simulation with the desired mesh resolution for the Explicit simulation.
I have tried replacing the static preload step with a quasi-static one (using Abaqus/Explicit), but still end up with very long computational times, because of the very fine mesh.
I also tried to perform the static preload step using a coarse mesh. Here I would load the model with a force, and extract the displacements for each node. With MATLAB I interpolate the results to a finer mesh, and generate a .inp file to run the dynamic simulation. When releasing these displacements, but not the load, the model obtains it prestressed state. This however also doesn't seem to work, because I need to apply displacements (in 3 directions) at every single node in the input file. It takes a very long time before the model is generated. (I am using " *boundary, _pickednode, dof,dof, value " for each nodal displacement).
Does anyone have a better solution than the ones I am trying? Or maybe there is a way to do the input of the interpolated displacements in a more efficent way? Any help would be very much appreciated!
Kind regards,
Ruben