I'm using Coefficient Form PDE. I have a cylinder that changes the angle is constant (the two-dimensional space) and I need the divergence in cylindrical coordinates.
Just for general suggestion: if your geometry is cylindrical (as for your problem) only cylindrical co-ordinate system is the optimized co-ordinate for your problem. Do not go for Cartesian. Change everything in to cylindrical system in terms of r, ,phi, z.. If you see your target (you are looking for) physical parameter is not changing with respect to phi (angle co-ordinate) you can further reduce your PDE into a two dimensional PDE in cylindrical system leaving to only r,z co-ordinates. In this two dimensional case you can use 2D-axio symmetry as Kevin Du clos already said. If you are still in need of phi dependence then your only option is to go for 3D problem. In that case, you need to draw base geometry and then you have to extrude it. I suggest draw regular shape first then you can introduce deformation.
Hi, can you help me how to define a prescribed displacement relative to a angular/spherical coordinate system? in prescribed displacement/rotation, just there is GLOBAL coordinate system, I can not add any extra coordinate system. thanks.