Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
2 decades ago
Jul 24, 2009, 12:39 p.m. EDT
Hi
I'm not sure I understand you, the subdomain "element" item is one entity that you can pick (identified by a number), so if you want to study a part of it, cut it up in slices, cubes, circles, or whatever symmetry you need (easily done in i.e. SolidWorks see split ot cut commands), then you can integrate over one of these "sub-units". This often improves the meshing if you are cutting up along some symmetry lines.
Now you can always import it into Matlab / script and relly attack the nodes.
If really its someting simple you coud say integrate
(u^2+v^2+w^2)*(x^2+y^2*z^2<=1) no ?
(=0 when r^2=(x^2+y^2+z^2) >1) just check that you units are OK and that if you normalise you integrate the volume accordingly the formula above, when normalised correctly with the corresponding volume should give the rms deformation, squared. Having the possibility to play so nicely with the physics and the mathematics is one of the strong points of COMSOL
Try things out on simple geometry, when you trust the expression, go for the big jump, but always cross scheck your results
Good luck
Ivar
Hi
I'm not sure I understand you, the subdomain "element" item is one entity that you can pick (identified by a number), so if you want to study a part of it, cut it up in slices, cubes, circles, or whatever symmetry you need (easily done in i.e. SolidWorks see split ot cut commands), then you can integrate over one of these "sub-units". This often improves the meshing if you are cutting up along some symmetry lines.
Now you can always import it into Matlab / script and relly attack the nodes.
If really its someting simple you coud say integrate
(u^2+v^2+w^2)*(x^2+y^2*z^21) just check that you units are OK and that if you normalise you integrate the volume accordingly the formula above, when normalised correctly with the corresponding volume should give the rms deformation, squared. Having the possibility to play so nicely with the physics and the mathematics is one of the strong points of COMSOL
Try things out on simple geometry, when you trust the expression, go for the big jump, but always cross scheck your results
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
2 decades ago
Jul 28, 2009, 5:13 p.m. EDT
I like your idea, and I didn't know COMSOL could take logicals as parts of integration expressions.
Alternatively, you could also use posteval or postinterp on the desired quantity in the desired subset of the domain, and then sum (normalizing properly of course).
I like your idea, and I didn't know COMSOL could take logicals as parts of integration expressions.
Alternatively, you could also use posteval or postinterp on the desired quantity in the desired subset of the domain, and then sum (normalizing properly of course).