Runge-Kutta 4th order method
1,112 ビュー (過去 30 日間)
古いコメントを表示
% It calculates ODE using Runge-Kutta 4th order method
% Author Ido Schwartz
clc; % Clears the screen
clear;
h=5; % step size
x = 0:h:100; % Calculates upto y(3)
Y = zeros(1,length(x));
y(1) = [-0.5;0.3;0.2];
% initial condition
F_xy = @(t,r) 3.*exp(-t)-0.4*r; % change the function as you desire
for i=1:(length(x)-1) % calculation loop
k_1 = F_xy(x(i),y(i));
k_2 = F_xy(x(i)+0.5*h,y(i)+0.5*h*k_1);
k_3 = F_xy((x(i)+0.5*h),(y(i)+0.5*h*k_2));
k_4 = F_xy((x(i)+h),(y(i)+k_3*h));
y(i+1) = y(i) + (1/6)*(k_1+2*k_2+2*k_3+k_4)*h; % main equation
end
display(Y(i+1));
if i run the programme i get answer =0;
how can i solve this problem if i have three initial condition -0.5 ,0.3 and 0.2
while x=0:5:100
and how i can plot the answer with respect to x?
採用された回答
David Wilson
2019 年 5 月 6 日
編集済み: MathWorks Support Team
2023 年 4 月 18 日
First up, you will need a much smaller step size to get an accurate solution using this explicit RK4 (with no error control). I suggest h = 0.05. Validate using say ode45 (which does have error control).
Then you will need to run your ode above three separate times, once starting from y(1) = -0.5, again with y(1) = 0.3, etc.
Then finally plot the result with plot(x,y,'o-').
h=0.05; % step size
x = 0:h:100; % Calculates upto y(3)
y = zeros(1,length(x));
%y(1) = [-0.5;0.3;0.2];
y(1) = -0.5; % redo with other choices here.
% initial condition
F_xy = @(t,r) 3.*exp(-t)-0.4*r; % change the function as you desire
for i=1:(length(x)-1) % calculation loop
k_1 = F_xy(x(i),y(i));
k_2 = F_xy(x(i)+0.5*h,y(i)+0.5*h*k_1);
k_3 = F_xy((x(i)+0.5*h),(y(i)+0.5*h*k_2));
k_4 = F_xy((x(i)+h),(y(i)+k_3*h));
y(i+1) = y(i) + (1/6)*(k_1+2*k_2+2*k_3+k_4)*h; % main equation
end
% validate using a decent ODE integrator
tspan = [0,100]; y0 = y(1);
[tx, yx] = ode45(F_xy, tspan, y0);
plot(x,y,'o-', tx, yx, '--')
4 件のコメント
Walter Roberson
2022 年 11 月 14 日
the code would need to be adjusted slightly if the ode function has more than one state (and so returns a vector.)
その他の回答 (6 件)
Sandip Das
2021 年 7 月 28 日
%Published in 25 July 2021
%Sandip Das
clc;
clear all;
dydt=input('Enter the function : \n');
t0=input('Enter the value of t0 : \n');
y0=input('Enter the value of y0 : \n');
tn=input('Enter the value of t for which you want to find the value of y : \n');
h=input('Enter the step length : \n');
i=0;
while i<tn
k_1 = dydt(t0,y0);
k_2 = dydt(t0+0.5*h,y0+0.5*h*k_1);
k_3 = dydt((t0+0.5*h),(y0+0.5*h*k_2));
k_4 = dydt(((t0)+h),(y0+k_3*h));
nexty = y0 + (1/6)*(k_1+2*k_2+2*k_3+k_4)*h;
y0=nexty
t0=t0+h
i=i+h;
end
fprintf('The value of y at t=%f is %f',t0,y0);
0 件のコメント
mahmoud mohamed abd el kader
2020 年 10 月 29 日
function [x,y] = rk4th(dydx,xo,xf,yo,h)
x = xo:h:xf ;
y = zeros(1,length(x));
y(1)= yo ;
for i = 1:(length(x)-1)
k_1 = dydx(x(i),y(i));
k_2 = dydx(x(i)+0.5*h,y(i)+0.5*h*k_1);
k_3 = dydx((x(i)+0.5*h),(y(i)+0.5*h*k_2));
k_4 = dydx((x(i)+h),(y(i)+k_3*h));
y(i+1) = y(i) + (1/6)*(k_1+2*k_2+2*k_3+k_4)*h;
end
dydx = @(x,y) 3.*exp(-x)-0.4*y;
%[x,y] = rk4th(dydx,0,100,-0.5,0.5);
%plot(x,y,'o-');
end
3 件のコメント
soham roy
2022 年 12 月 8 日
What modifications do we need to make in this code to solve 3 ODEs with different initial conditions?
Walter Roberson
2022 年 12 月 8 日
y = zeros(1,length(x));
would change to
y = zeros(length(x), length(y0));
and below that, each y(INDEX) would be replaced with y(INDEX,:)
Mj
2020 年 11 月 7 日
Hello everyone!
I have to solve this second order differential equation by using the Runge-Kutta method in matlab:
can anyone help me please? and how can i plot the figure?(a against e)
d2a/de2=(((((2+c2)*(Fu^2))/(1+c2))+1)*(a^c2)-((2+c2/1+c2)*(Fu^2/a))-a^(2+(2*c2)))/(((2+c2)*Fu^2)/(1+c2)*(3+c2));
Fu=1
c2=0 , 0.5 , 1 (there are 3 values for c2)
initial conditions are: a=0.8 , d_a=
David Wilson
2021 年 5 月 6 日
Wow, you haven't given us too much to go on, so that makes a real challenge.
First up, your 2nd order ODE is needlessly complex given that Fu=1, and c2 =0 say. (I'm not sure what the other valuesare for, Are you solving this 3 seprate times? (Be good to know if that is the case.)
If you have the symbolic toolbox, it makes it easy to simplify your problem to something doable. First up, I'm going to try and solve it analytically.
syms Fu c2 real
syms a(t)
f2 = (((((2+c2)*(Fu^2))/(1+c2))+1)*(a^c2)-((2+c2/1+c2)*(Fu^2/a))-a^(2+(2*c2)))/(((2+c2)*Fu^2)/(1+c2)*(3+c2));
f2_a = subs(f2,Fu,1)
f2_b = subs(f2_a,c2,0) % subs c2 for 0
Da = diff(a);
D2a = diff(a,2);
% Now attempt to solve analytically
dsolve(D2a == f2_b, a(0) == 0.8, Da(0) == 1)
Well that didn't work, but no real suprise there.
Let's try a numerical method:
syms Fu c2 real
syms a real
f2 = (((((2+c2)*(Fu^2))/(1+c2))+1)*(a^c2)-((2+c2/1+c2)*(Fu^2/a))-a^(2+(2*c2)))/(((2+c2)*Fu^2)/(1+c2)*(3+c2));
f2_a = subs(f2,Fu,1); f2_b = subs(f2_a,c2,0); pretty(f2_b)
We need to encode this as a system of 2 ODES. (Convert to Cauchy form)
aprime = @(t,a) [a(2); ...
0.5 - a(1).^2/6 - 1./(a(1)*3)]
Now we are ready to solve the ODE. I'll use ode45, and guess a t-span, and guess one of the initial conditions since you forgot to help us out there.
aprime = @(t,a) [a(2); ...
0.5 - a(1).^2/6 - 1./(a(1)*3)]
a0 = [0.8; 0]
[t,a] = ode45(aprime, [0,4], a0)
plot(t,a)
0 件のコメント
Amr Mohamed
2021 年 5 月 9 日
how can we write the code for this problem :
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/612505/image.jpeg)
2 件のコメント
Moneeb Ur Rehman
2021 年 5 月 27 日
get the y on other side, integrate then to find 1st derivative. Now apply R.k method to solve. Hope you understood;
monsef
2023 年 7 月 17 日
y=x^2-2yx
h=0.2
y0=0
x0=1
wriet program im mathlab
1 件のコメント
Ahmed J. Abougarair
2024 年 3 月 24 日
clc;
clear all;
F = @(t,y) 4*exp(0.8*t)-0.5*y
t0=input('Enter the value of t0 : \n');
y0=input('Enter the value of y0 : \n');
tn=input('Enter the value of t for which you want to find the value of y : \n');
h=input('Enter the step length : \n');
i=0;
while i<tn
k_1 = F(t0,y0);
k_2 = F(t0+0.5*h,y0+0.5*h*k_1);
k_3 = F((t0+0.5*h),(y0+0.5*h*k_2));
k_4 = F(((t0)+h),(y0+k_3*h));
nexty = y0 + (1/6)*(k_1+2*k_2+2*k_3+k_4)*h;
y0=nexty;
t0=t0+h;
i=i+h;
end
fprintf('The value of y at t=%f is %f \n',t0,y0)
% validate using a decent ODE integrator
tspan = [0,1]; Y0 = 2;
[tx,yx] = ode45(F, tspan, Y0);
fprintf('The true value of y at t=%f is %f \n',tspan(end),yx(end))
Et= (abs(yx(end)-y0)/yx(end))*100;
fprintf('The value of error Et at t=%f is %f%% \n',tspan(end),Et)
参考
カテゴリ
Help Center および File Exchange で Calculus についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!