4th order Runge Kutta Method Differential System
1 view (last 30 days)
Show older comments
4th order Runge Kutta Method Differential System. I don't know what's wrong with my code and I have been trying to figure out what went wrong. I'm trying to solve a system with 3 First-Order Differential Equations.
clear all
close all
clc
COne = 1000;
CTwo = 1000;
R = 50;
L = 0.1;
Vt = 0.026;
Is = 1*10^(-8);
a = 0.025/2500
b = abs(27.673314419-(2*tan(1.50)))
b = 5*sin(2*pi*50*0.017)
xprime_func = @(t,x,y,z) ((1/COne)*y-(1/(COne*R))*x);
yprime_func = @(t,x,y,z) ((1/L)*z-(1/L)*x);
zprime_func = @(t,x,y,z) ((1/CTwo)*Is*(exp((abs(10*sin(2*pi*50*t))-z)/(2*Vt))-1)-(1/CTwo)*y);
% Define time interval and step size
tmax=0.025; steps=2500; h=tmax/steps;
% Initial conditions:
x(1)=0; y(1)=0; z(1)=0; t(1)=0;
% Estimate of derivatives and marching in time.
for i=1:steps
t(i+1)=i*h;
K(1)=h*xprime_func(t(i),x(i),y(i),z(i));
L(1)=h*yprime_func(t(i),x(i),y(i),z(i));
M(1)=h*zprime_func(t(i),x(i),y(i),z(i));
K(2)=h*xprime_func(t(i)+h/2,x(i)+1/2*K(1),y(i)+1/2*L(1),z(i)+1/2*M(1));
L(2)=h*yprime_func(t(i)+h/2,x(i)+1/2*K(1),y(i)+1/2*L(1),z(i)+1/2*M(1));
M(2)=h*zprime_func(t(i)+h/2,x(i)+1/2*K(1),y(i)+1/2*L(1),z(i)+1/2*M(1));
K(3)=h*xprime_func(t(i)+h/2,x(i)+1/2*K(2),y(i)+1/2*L(2),z(i)+1/2*M(2));
L(3)=h*yprime_func(t(i)+h/2,x(i)+1/2*K(2),y(i)+1/2*L(2),z(i)+1/2*M(2));
M(3)=h*zprime_func(t(i)+h/2,x(i)+1/2*K(2),y(i)+1/2*L(2),z(i)+1/2*M(2));
K(4)=h*xprime_func(t(i)+h,x(i)+1*K(3),y(i)+1*L(3),z(i)+1*M(3));
L(4)=h*yprime_func(t(i)+h,x(i)+1*K(3),y(i)+1*L(3),z(i)+1*M(3));
M(4)=h*zprime_func(t(i)+h,x(i)+1*K(3),y(i)+1*L(3),z(i)+1*M(3));
x(i+1)=x(i)+1/6*(K(1)+2*K(2)+2*K(3)+K(4));
y(i+1)=y(i)+1/6*(L(1)+2*L(2)+2*L(3)+L(4));
z(i+1)=z(i)+1/6*(M(1)+2*M(2)+2*M(3)+M(4));
end
plot(t,x,t,y,t,z);
2 Comments
Torsten
on 17 Sep 2024
Edited: Torsten
on 17 Sep 2024
The Table 7.7 and figure 7.9 is what should my values for my x,y,z and it is insanely way far off.
Please include a mathematical description of your problem with equations and constants used, Table 7.7 and figure 7.9.
I would advice to solve the problem first with a sophisticated MATLAB integrator like ode45 to get a reference solution.
Answers (1)
Torsten
on 17 Sep 2024
Edited: Torsten
on 17 Sep 2024
Note that COne and CTwo are given in muF - thus they should be prescribed as 1000e-6 in your code, I guess.
COne = 1000e-6;
CTwo = 1000e-6;
R = 50;
L = 0.1;
Vt = 0.026;
Is = 1e-8;
xprime_func = @(t,x,y,z) ((1/COne)*y-(1/(COne*R))*x);
yprime_func = @(t,x,y,z) ((1/L)*z-(1/L)*x);
zprime_func = @(t,x,y,z) ((1/CTwo)*Is*(exp((abs(10*sin(2*pi*50*t))-z)/(2*Vt))-1)-(1/CTwo)*y);
f = @(t,x,y,z)[xprime_func(t,x,y,z);yprime_func(t,x,y,z);zprime_func(t,x,y,z)];
F = @(t,u)f(t,u(1),u(2),u(3));
% Define time interval and step size
tmax=0.025; steps=2500; tspan=linspace(0,tmax,steps);
% Initial conditions:
u0 = [0;0;0];
[T,U] = ode15s(F,tspan,u0);
figure(1)
plot(T,U(:,2))
grid on
figure(2)
plot(T,[U(:,1),U(:,3)])
grid on
0 Comments
See Also
Categories
Find more on Manage Products in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!