Euler's method to plot orbital trajectory of comet

I am trying to use eulers method to plot the trajectory of a comet but no matter what I i get linear plots for position,velocity, and acceleration and generally numbers that don't make sense. I am pretty sure it has something to do with my step size, but Ive tried hundreds f combinations and I dont know how exactly to fix that. The units are all in kilometers.
% Euler's Method
% Initial conditions and setup
h = 1; % step size
t = 0:h:100; % the range of x
y = zeros(size(t)); % allocate
x = zeros(size(t));
vx = zeros(size(t));
vy = zeros(size(t));
ax = zeros(size(t));
ay = zeros(size(t));
y(1) = 0; % the initial values
x(1) = 193929400;
vx(1) = -5.909;
vy(1) = 50.00294822;
[ax(1),ay(1)] = newaccel(x(1),y(1));
n = numel(y);
for i=1:n-1
vx(i+1) = vx(i) + h * ax(i);
vy(i+1) = vy(i) + h * ay(i);
y(i+1) = y(i) + h * vy(i+1);
x(i+1) = x(i) + h * vx(i+1);
[ax(i+1),ay(i+1)] = newaccel(x(i+1),y(i+1));
end
This is what my function to determine acceleration looks like.
function [a_newx a_newy] = newaccel(Xx_n,Xy_n)
X_n = [Xx_n * 1000,Xy_n * 1000];
G = 6.67408e-11;
ms = 1.989e30;
mag_X = sqrt(X_n(1).^2 + X_n(2).^2);
a_new = vpa((-(G*ms)/(mag_X)^3)*(X_n));
a_newx = a_new(1)/1000;
a_newy = a_new(2)/1000;
end

4 commentaires

Alan Stevens
Alan Stevens le 10 Juil 2020
Modifié(e) : Alan Stevens le 10 Juil 2020
Are your times in seconds (your G value suggests this)? If so, 100 seconds won't take you very far!
Yes and I've tried higher times, but the loop takes impracticly long (5+ hours) unless I increase the step size to equally as useless scales. What in my code is making anything over 1000 or so iterations take forever?
How did you get Vx? Please answer if you see this
I meant to say how did you get Vy?

Connectez-vous pour commenter.

Réponses (1)

It's the vpa function slowing everything. There is no need for it here. Try the following:
% Euler's Method
% Initial conditions and setup
h = 10; % step size
t = 0:h:10^7; % the range of x
y = zeros(size(t)); % allocate
x = zeros(size(t));
vx = zeros(size(t));
vy = zeros(size(t));
ax = zeros(size(t));
ay = zeros(size(t));
y(1) = 0; % the initial values
x(1) = 193929400;
vx(1) = -5.909;
vy(1) = 50.00294822;
[ax(1),ay(1)] = newaccel(x(1),y(1));
n = numel(y);
for i=1:n-1
vx(i+1) = vx(i) + h * ax(i);
vy(i+1) = vy(i) + h * ay(i);
y(i+1) = y(i) + h * vy(i+1);
x(i+1) = x(i) + h * vx(i+1);
[ax(i+1),ay(i+1)] = newaccel(x(i+1),y(i+1));
end
plot(x,y), grid
function [a_newx, a_newy] = newaccel(Xx_n,Xy_n)
X_n = [Xx_n * 1000,Xy_n * 1000];
G = 6.67408e-11;
ms = 1.989e30;
mag_X = sqrt(X_n(1).^2 + X_n(2).^2);
a_new = (-(G*ms)/(mag_X)^3)*(X_n); % No need for vpa here
a_newx = a_new(1)/1000;
a_newy = a_new(2)/1000;
end

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by