Every single mechanical engineer has to study a subject called mechanics. Some call it MOM, some call it SOM, some call it EM. But whatever you may call it, it's always there and it's always the same. This subject tells us how to decide whether our design will fail or not. And it is a very complex subject. And the bad news is, we only study this until we get familiar with the concept of Simulation. Once you start doing simulations, there is no going back. I had a glimpse of simulation very early in my college life, way before it would have according to my syllabus. And then I got myself involved in another project that made me understand simulation to a very great extend. This was the design and manufacturing of HAWT. Basically, we were trying to make a hybrid wind turbine that can produce electricity safely inside a crowded city. And the very first step included the Fluent based simulations. Actually, the first step was a literature review, but yeah you got the point. So, my s...
One of the most time-consuming types of simulations is Explicit simulations. Simulations can either be implicit or explicit. Explicit simply means that the simulation is run for a very small time period where energy change is very high and instantaneous. Many different studies are done using Explicit methods. One of these is the simulation of an explosion. This is what I did in my six-month internship. Since this internship was under a government organization, I can't describe the specifications of the process, but basically, it was the simulation of an explosion. Here's an image of how the failure occurred just after 0.1 s of the explosion.
Image: Failed sheet due to explosion on Ansys Explicit.
While using Explicit methods, it is very important to define the material properties properly. Unlike Implicit models where a few properties like density and elastic modulus can fully define the material, Explicit models require much more data. Thus, one of the most important steps is to find the required data. In case of an explosion, we need to find the equation of state for every material. However, the equation of state is not available for every material and needs to be manually derived and applied in the software.
Another important thing for Explicit models is convergence. Since the change in energy is happening instantaneously, so a very small error can cause the simulation to fail. The error may be due to a single mesh element, an under-defined material, or even an improperly defined contact. This further implies that any explicit study should be studied analytically and every single interaction and material property must be completely understood before starting the simulation on the software.
The Explicit simulation also takes much more time as compared to an Implicit one. This further reduces the feasibility of the hit and trial method. So, doing the simulations again and again to remove errors can be much more time-consuming. The mesh optimization also becomes a much harder task due to the same reason. Further, in an explosion simulation, the stresses are usually very high and hence the simulation usually can't be completed without defining the failure criteria. This means that the failure criteria for individual material also needs to be defined and understood.
All this makes the Explicit simulations a good way to understand the FEM based Softwares. The Explicit method also helps one in understanding the physical equations which are causing the material to behave in the way they are behaving. In simpler terms, Explicit simulations are the best way to completely understand the FEM software. If you are interested in gaining in-depth information about every single part of a simulation, Explicit methods are the best way to go.
I used both Abaqus CAE and Ansys Workbench to simulate an explosion and compared the results. And after nearly six months of continuous simulations, weeks of studying of tens of different papers, and hours of the derivation of the required equations, I was finally able to get comparable results on both the software. And throughout this period, I was able to learn much more about the simulations than what I had learned in the last couple of years.
Comments
Post a Comment