SCORE! *Now with better graphs!* :D

[TecHNooB]

Yes, it's a very good feeling when you're able to figure something out. Sometimes, that's my only motivation as an engineering student.

 

[Fenixgoon]

Originally posted by: TecHNooB
Yes, it's a very good feeling when you're able to figure something out. Sometimes, that's my only motivation as an engineering student.

i'm not even in class right now (on co-op, actually). i'm just doing this cause i'm interested 😀

 

[Anonemous]

FEMLAB (Comsol Multiphysics) FTW

 

[Fenixgoon]

Originally posted by: Anonemous
FEMLAB (Comsol Multiphysics) FTW

gotta start somewhere 😀

 

[FoBoT]

Originally posted by: Printer Bandit
you should become brain surgeon!

fixed

 

[CycloWizard]

Originally posted by: Fenixgoon
yes technically it's not infinitely small, but that's the idea behind it. just like numerical analysis - it's an approximation that often is infinitely close, but not really the "real" thing.

FEA will give the exact solution to the model selected if the solution space is properly constructed. Like I said, you can achieve this in a number of ways, the most common of which are uniform h-extension (increasing the element density - the least efficient method) and p-extension (increasing the polynomial degree - more efficient, but still not ideal). The optimal method is a combination of the two, often called hp-extension, in which the polynomial degree is increased (often adaptively) and the mesh is refined in a geometric fashion rather than uniformly (again, often adaptively). So, if I were looking at the bending of a beam, I know that the exact solution is cubic (IIRC). I will therefore get the exact solution if I use a single element with a third-order polynomial. Or, I can get an approximate solution that is very close to the exact solution by making lots and lots of linear (first-order polynomial) elements. The first method will actually give the exact solution while the latter will not, but the latter will come close enough for any engineering calculation if I increase the mesh density sufficiently.

 

[Fenixgoon]

Originally posted by: CycloWizard

Originally posted by: Fenixgoon
yes technically it's not infinitely small, but that's the idea behind it. just like numerical analysis - it's an approximation that often is infinitely close, but not really the "real" thing.

FEA will give the exact solution to the model selected if the solution space is properly constructed. Like I said, you can achieve this in a number of ways, the most common of which are uniform h-extension (increasing the element density - the least efficient method) and p-extension (increasing the polynomial degree - more efficient, but still not ideal). The optimal method is a combination of the two, often called hp-extension, in which the polynomial degree is increased (often adaptively) and the mesh is refined in a geometric fashion rather than uniformly (again, often adaptively). So, if I were looking at the bending of a beam, I know that the exact solution is cubic (IIRC). I will therefore get the exact solution if I use a single element with a third-order polynomial. Or, I can get an approximate solution that is very close to the exact solution by making lots and lots of linear (first-order polynomial) elements. The first method will actually give the exact solution while the latter will not, but the latter will come close enough for any engineering calculation if I increase the mesh density sufficiently.

intaresting... i have much to learn 😀😀