5th Degree Polynomials
- To: mathgroup at smc.vnet.net
- Subject: [mg37129] 5th Degree Polynomials
- From: jdhouse4 at mac.com
- Date: Sat, 12 Oct 2002 05:04:54 -0400 (EDT)
- Sender: owner-wri-mathgroup at wolfram.com
Hey folks,
I have been working on a problem that seems to not lend itself to a
solution. The following Mathematica code begins with the expression
that I am trying to solve. For the curious, it's a degree 2 zonal and
sectoral harmonics problem where I am trying to calculate and plot the
geoid of earth as compared to an ellipse to see how well the geoid is
approximated as an ellipse. In any case, we have the following relation
ship,
U =GM/r( 1 - (ae/r)^2 ( J2 (3/2 Sin[t]^2 - 1/2) - 3 Cos[t]^2 (C22
Cos[2 x] + S22 Sin[2 x]));
Ur =1/2 we^2 (r Cos[t])^2;
W[x_] =U + Ur;
In trying to reorder W to become a function r wrt t, that is r[t_], I
tried, among others,
Solve[W[t], r]
which returned
({{r ->
Root[ae^2 GM J2 + 6 ae^2 C22 GM Cos[ [t]] ^2 -
3 ae^2 GM J2 Sin[ [t]] ^2 + 2 GM #1 ^2 -
2 W0 #1 ^3 + we^2 Cos[ [t]] ^2 #1 ^5 &,
1]}, {r ->
Root[ae^2 GM J2 + 6 ae^2 C22 GM Cos[ [t]] ^2 -
3 ae^2 GM J2 Sin[ [t]] ^2 + 2 GM #1 ^2 -
2 W0 #1 ^3 + we^2 Cos[ [t]] ^2 #1 ^5 &,
2]}, {r ->
Root[ae^2 GM J2 + 6 ae^2 C22 GM Cos[ [t]] ^2 -
3 ae^2 GM J2 Sin[ [t]] ^2 + 2 GM #1 ^2 -
2 W0 #1 ^3 + we^2 Cos[ [t]] ^2 #1 ^5 &,
3]}, {r ->
Root[ae^2 GM J2 + 6 ae^2 C22 GM Cos[ [t]] ^2 -
3 ae^2 GM J2 Sin[ [t]] ^2 + 2 GM #1 ^2 -
2 W0 #1 ^3 + we^2 Cos[ [t]] ^2 #1 ^5 &,
4]}, {r ->
Root[ae^2 GM J2 + 6 ae^2 C22 GM Cos[ [t]] ^2 -
3 ae^2 GM J2 Sin[ [t]] ^2 + 2 GM #1 ^2 -
2 W0 #1 ^3 + we^2 Cos[ [t]] ^2 #1 ^5 &, 5]}} )
which wasn't too much help, though it is a list of 5 Root functions.
But in order to plot, I need a function r(t) so I can plot r wrt
t...right?
ParametricPlot[r[t], {t, 0, Pi}]
So, I guess my questions are as follows:
1. How do I get Solve[ ] to output numbers, as //N and NSolve did
nothing to Solve[r[t], ...] to get any numbers instead of just r ->
Root[...]?
2. Is there a way to use ParametricPlot[ W[t], {t, 0.0, Pi}] instead of
using r[t] and negating the whole issue of solving W[t] for r[t]?
I have read that Solve only works for up to 4th order polynomials. I
have been unable to find anything that works on my problem, having
tried SolveAlways[ ] and other, and combination of others.
Any help is welcome. I'll be glad to forward my Notebook if someone
asks. Thanks ahead of time.
Jim Hillhouse
jdhouse4 at mac.com
Ph.D. Graduate Student
Aerospace Engineering
University of Texas at Austin
jdhouse4 at mail.utexas.edu
512-784-3205