Ode45 initial conditions iteration

I am trying to solve the following 2nd order differential equation, which works fine but I need to iterate the initial conditions, so everytime the ode is solved, the last values for q(:,1) and q(:,2) should be the new value for X and Y respectivley.

%solve_2nd_order_diff
duf=@(t,q)[q(2);f*w^2*cos(w*t)-(Cmech/J)*q(2)-(K1/J)*q(1)-(K3/J)*q(1)^3];


for   r=1:length(Rad_speed)%freq of harmonic fluctuations, rad/s
[t,q]=ode45(duf,Time,[X Y]);

Max_values=max(q);
plot(t,q(:,1),'k')
xlabel('TIme,s')
ylabel('Relative Displacement, Rad')
grid

end

Answer 1

[Alan Stevens]

Like so

...
    [X Y] = ...
for   r=1:length(Rad_speed)%freq of harmonic fluctuations, rad/s
[t,q]=ode45(duf,Time,[X Y]);
[X Y] = [q(end,1) q(end,2)];

...

[Adedayo Odeleye]

Thanks for the swift reply.

it says "Too many output arguments." in regards to the [X Y] = [q(end,1) q(end,2)]; line

[Alan Stevens]

Can you upload your full code?

[Alan Stevens]

Sorry, I should have written

X = q(end,1); Y = q(end,2);

rather than

[X Y] = [q(end,1) q(end,2)];

Incidentally, your plots will be overwritten unless you use the figure command or hold on (whichever is appropriate.

[Adedayo Odeleye]

full code is below, it is working fine now.

J=3.37e-5; %inertia of rotor (kg/m2)
K1=0.7243; %Linear stiffness coef (N.m/rad)
K3=1409.7; %Nonlinear stiffness coef (N.m/rad3)
f=0.005;
Zeta=0.05;%damping ratio
Cmech=2*Zeta*(sqrt(K1/J))*J; %mechanical damping coef
ECF=0.2; %electromagnetic coupling factor
Rint=82;
Celec=ECF^2/2*Rint;%Electrical damping coef
L=160;%Inductance (mH)
X = 0; 
Y = 0;

%shaft speed in RPM
Shaft_speed=2000;
%shaft speed in rad/s
Rad_speed=Shaft_speed*((2*pi)/60);
%freq of harmonic fluctuations 
w=2*Rad_speed;
w_Hz=w/2*pi; %convert to Hertz

nt=1/w_Hz; 
Time=0:nt/100:100*nt;

%solve_2nd_order_diff
duf=@(t,q)[q(2);f*w^2*cos(w*t)-(Cmech/J)*q(2)-(K1/J)*q(1)-(K3/J)*q(1)^3];

Iter=100;
for   r=1:Iter
[t,q]=ode45(duf,Time,[X Y]);

X=q(end,1);
Y=q(end,2);

Max_values=max(q)
plot(t,q(:,1),'k')
xlabel('TIme,s')
ylabel('Relative Displacement, Rad')
grid

end

[Alan Stevens]

Excellent, but notice that your plot and your Max_values will only apply for the last iteration of the loop (they are overwritten every time through the loop). If that is what you want it is probably best to put that piece of coding outside the loop at the end. To store the Max_values (rather than just have them printed in the command window every iteration), you could use

Max_values(r) = max(q);