[R-sig-dyn-mod] 2-D Cylindrical Model

[Karline Soetaert]

Stephenthis is what I would do:

 rgrid  <- setup.grid.1D(L = Lr, N = Nr)
 zgrid  <- setup.grid.1D(L = Lz, N = Nz)

 ri <- rgrid$x.int
 dz <- zgrid$dx

# Interface surfaces in z direction (along radius)
  A.z <- matrix(nrow =  Nr, ncol = Nz+1, data =
                rep(pi*(ri[2:(Nr+1)]^2 - ri[1:Nr]^2), Nz+1))

# Interface surfaces in r-direction (along z axis)
  A.r <- matrix(nrow = Nr+1, ncol = Nz,
                data = rep(2 * pi * ri, Nz) * dz)



Karline

-----Original Message-----
From: r-sig-dynamic-models-bounces at r-project.org [mailto:r-sig-dynamic-models-bounces at r-project.org] On Behalf Of stephen's mailinglist account
Sent: vrijdag 12 september 2014 23:54
To: R-sig-dynamic-models at r-project.org
Subject: [R-sig-dyn-mod] 2-D Cylindrical Model

I have been using ReacTran to solve the heat equation for self-reactive solid materials, starting with the 1-d infinite plane model, then a 1-d spherical model and now I am looking at solving for a finite (short
equicylinder) cylinder in 2-d coordinates as there will be no rotational variation in properties around the z axis, so discretising in z and r only.
I have been setting up a model using tran.2d for r and z, but I was just trying to clarify whether the areas used A.x and A.y should be 2*pi*r*dr and 2*pi*r*dz setup through setup.prop.1D.
As an alternative I am considering using tran.cylindrical, and using a single cell in the theta direction, I wondered whether this approach made more sense?

thanks for any help
Stephen

	[[alternative HTML version deleted]]

_______________________________________________
R-sig-dynamic-models mailing list
R-sig-dynamic-models at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-dynamic-models