AutoDyn 3D Explosion Simulations in Traditional GUI

Hi, this is Mingyao from Ozen Engineering. In this video, I'll show you how to set up a 3D simulation to run on parallel in the AutoDyn GUI. So here's a continuation of the model we had before. We did a 2D simulation. Let me change that name there.

So this is my 2D simulation in explicit dynamics, and this is my 3D simulation in explicit dynamics.

Now, as you see, we can do most of the analysis in explicit dynamics, but some of the explosive-specific post-processing and multimaterial Eulerian post-processing in general is more convenient in AutoDyn.

In fact, it may make sense for you to run the whole simulation in AutoDyn because then we can monitor the progress of the simulation. There are a lot of interactive operations. Unfortunately, for AutoDyn, the simulation is very simple.

You can run a simulation in AutoDyn, and then you can run a simulation in AutoDyn. So you can run a simulation in AutoDyn, but for AutoDyn right now in the workbench environment, this AutoDyn, if I just link these two cells together, this simulation is only going to run in a single-core.

And for any realistic 3D simulation, that's not going to be enough. But there's a pretty easy way to set up a multi-core simulation, and we do that by generating a batch file. So I have a batch file here, put into a temporary directory. If I look at what this batch file says, it looks like this.

So there is an AutoDyn wrapper, Win64, there's an executable called AutoDyn wrapper. It's in the Ansys installation directory, your program files, Ansys Inc. The version, AISOL, AutoDyn, Win 64. If you're using Linux, it'll be Linux64, autodynwrapper.exe.

And then there's a few inputs, -no input, -MPI. You have to specify the MPI name, and then the machine. I'm only going to run this on my computer, which has this name, and I'm going to run it on four cores. So that's pretty much it.

If we go to the installation directory here, under Ansys Inc, I have the latest release available as well, so I can update this to AISOL, autodyn164, and you can see there's an AutoDyn wrapper right there. So all I have to do to update to the latest release is change my script. To this value.

Let's save it and we'll go ahead and run this script. I put it in my temp directory. And this will allow you to run multiple machines, as well. We're running a single machine. So we want to load a...

It's still very helpful to set up the whole simulation in Workbench, so usually I go up to the Setup option. Position and then if I go to view files or I can do an update here.

You can see that I have a bunch of files available in E - that's those are all the results, nothing in F yet - if I do a refresh here, it will be linked to that file. So I can just grab one of the other dive files in E here, you can right-click and open continue folder.

This will tell you where the file is located. So I can go to autodyne here and say I want to open this folder here and open the zero cycle. Would bring up the Setup. So now we can we're in the usual autodyne environment and we can set up the simulation.

So parallel controls are already set up here for the output. We want to specify how often we want to refresh this - always tend to be a little bit low. So maybe every 10 iterations we want to update of what's happening and how often do we want to save this - we probably want to save it.

Put a zero there. So 50 times I'm only going to run this for a few iterations, but let's go ahead and run the simulation. You you you you danke for for out exactly till I can apply this. And this is really the convenience of running it in AutoDyne GUI is you can easily stop, start, make changes.

You can also extract different types of information from the simulation. So let's go to contour and look at the pressure. Okay, and run it for a few more iterations. Okay, we have a stop, save, and restart to do this in parallel. So stop, save. And we have to restart.

Okay, let's do a quick restart here. It's in my temp folder. And we can save the model somewhere else so we can run this independent of the other AutoDyne simulation here. That's probably what I should have done. So let's go to contour, pressure. So everything is constant right now.

Let's check on the output. Okay, everything is set up the way we want it to be. So let's go ahead and run this. The simulation time is expected to be the same. The setup is the same. It's using the exact same solver.

But the main difference here is that we have these additional post-processing and dynamic post-processing tools that we can use on the fly. We can run a little bit of UMD in these touch training loves for a lot of different performances, so we've got the touch Tibetan.

So we can hit the specific values that we want to strike a function in this pixel. Okay. We also want to make our symbol set to the external level to the southern state of our model. Okay. All right. So I want to give you a quick idea of how to set this up in Git run going in Autodyne.

Again, the command is you go and find this Autodyne wrapper. We have the extra input of -no input MPI. There's Intel MPI and Microsoft MPI on Windows machines. Then the machine name and how many cores. So that's all you need to run these simulations. Once again, thank you for watching this video.

If you have any questions, feel free to contact us at [www.ozinink.com](http://www.ozinink.com). Thank you. Take care.