Message Boards Message Boards

0
|
11889 Views
|
3 Replies
|
0 Total Likes
View groups...
Share
Share this post:

How can I solve this ImplicitRegion Error in NDSolve?

Posted 10 years ago

I'd like to solve a differential equation with some initial conditions to get an eclipse curve for Earth's orbit using Newton's Law in a cylindrical coordinate.

NDSolve gave me two more useless solutions, so I added ImplicitRegion option. But, if I run the following code, I get an error message like this.

"The function [Rho][t] does not have the same number of arguments as independent variables (3)". >>

x[t_] := rho[t] Cos[theta[t]]
y[t_] := rho[t] Sin[theta[t]]

r[t] = ( {
      {Cos[theta[t]], Sin[theta[t]]},
      {-Sin[theta[t]], Cos[theta[t]]}
     } ).( {
      {x[t]},
      {y[t]}
     } ) // Simplify // Flatten

vCylinderical[t] = ( {
      {Cos[theta[t]], Sin[theta[t]]},
      {-Sin[theta[t]], Cos[theta[t]]}
     } ).( {
      {D[x[t], {t, 1}]},
      {D[y[t], {t, 1}]}
     } ) // Simplify // Flatten

aCylinderical[t] = ( {
      {Cos[theta[t]], Sin[theta[t]]},
      {-Sin[theta[t]], Cos[theta[t]]}
     } ).( {
      {D[x[t], {t, 2}]},
      {D[y[t], {t, 2}]}
     } ) // Simplify // Flatten

Thread[({x[t], y[t]} /. t -> 0) == {149.6 10^6, 0}]
Thread[(D[{x[t], y[t]}, t] /. t -> 0) == {0, 29.786 3600 }]
initialConditions = Union[%, %%];

G = 3600^2 6.673 10^-20;
M = 1.989 10^30;
interval = 9000;
diffEqs = Thread[aCylinderical[t] == {-((G M)/rho[t]^2), 0}];

Rgn = ImplicitRegion[
  0 <= rho[t] <= 149.6 10^6 && 0 <= theta[t] <= 2 \[Pi], {rho[t], 
   theta[t]}]


NDSolve[{diffEqs, initialConditions}, {rho, theta}, {t, 0, 
  interval}, {rho[t], theta[t]} \[Element] Rgn]
POSTED BY: Changsoo Lee
3 Replies

The easiest way to set up the equations of motion in polar co-ordinates is to use the Euler-Lagrange equation.

http://en.wikipedia.org/wiki/Euler%E2%80%93Lagrange_equation

The Mathematica function EulerEquations will generate them from the Lagrangian, which is the kinetic energy minus the potential energy, in the coordinates of interest.

The analytic solution, an ellipse, is obtained by a change of variable u = 1/r.

POSTED BY: Frank Kampas

Well that's true but I think the Hamiltonian is the more powerful representation and perhaps we should consider the symmetry elements? Functionals and extreemals?

Any thoughts on my random jottings, fae bonnie scotland?

POSTED BY: jonathan lister

enter image description here

POSTED BY: Simon Cadrin
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract