- For ode45 you give numerical functions, not symbolic ones
- There were some variables that were used in the function but initialized after the loop
- If you have time dependent variables, you ideally should define them inside the function
- Your variable order in the function was inverted
- You didn't declared v0 anywhere
Error. I want to plot this ode45 pressure versus time
1 Ansicht (letzte 30 Tage)
Ältere Kommentare anzeigen
Ous Chkiri
am 3 Nov. 2019
Kommentiert: Ous Chkiri
am 3 Nov. 2019
R=8.314/32;
T0=2930;
a=(12)/((2.027*10^6)^0.45);
rhoP=1920;
Astar=pi*0.25^2;
k=1.35;
n=0.45;
P0=101325;
syms P(t)
for t = [0,0.1]
dP=@(P,t)(Ab*a*P^n*(rhoP-rhoO)-P*Astar*sqrt(k/(R*T0))*(2/(k+1))^((k+1)/(2*(k-1))))*R*T0/v0;
[P,t]=ode45(dP, [0,0,1], P0);
end
if t==0 %at beginning of the integration set initial values for the persistent variables
rp=0.35; %initial port radius
t1=0; %initial time step
end
Ab=2*pi*rp*8;%burn area
rhoO=P/(R*T0); %gas density
rp=min(rp+((a*P^n)*10^-3)*(t-t1),0.7);
figure(2)
plot(t,y)
xlabel("Time (s)")
ylabel("Chamber Pressure (Pa)")
title("Chamber Pressure vs Time (Start-Up)")
0 Kommentare
Akzeptierte Antwort
Thiago Henrique Gomes Lobato
am 3 Nov. 2019
Bearbeitet: Thiago Henrique Gomes Lobato
am 3 Nov. 2019
There were some errors in your code:
I choosed a random value for v0 and fix the other issues of your code, you just have to make sure that the considerations you made for your function and variables are right to fully trust the results.
R=8.314/32;
T0=2930;
a=(12)/((2.027*10^6)^0.45);
rhoP=1920;
Astar=pi*0.25^2;
k=1.35;
n=0.45;
P0=101325;
P = P0;
%syms P(t)
%at beginning of the integration set initial values for the persistent variables
rp=0.35; %initial port radius
t1=0; %initial time step
rhoO=P/(R*T0); %gas density
rp=min(rp+((a*P^n)*10^-3)*(t1-t1),0.7);
Ab=2*pi*rp*8;%burn area
v0 = 0.1;
dP=@(t,P)Fun(t,P,R,T0,rp,a,n,t1,Ab,rhoP,Astar,k,v0);%@(t,P)(Ab.*a.*P.^n.*(rhoP-rhoO)-P.*Astar.*sqrt(k/(R.*T0)).*(2/(k+1)).^((k+1)/(2.*(k-1)))).*R.*T0./v0;
[t,P]=ode45(dP, [0,0.1], P);
y = P;
figure(2)
plot(t,y)
xlabel("Time (s)")
ylabel("Chamber Pressure (Pa)")
title("Chamber Pressure vs Time (Start-Up)")
function dP = Fun(t,P,R,T0,rp,a,n,t1,Ab,rhoP,Astar,k,v0)
rhoO=P/(R*T0); %gas density
rp=min(rp+((a*P^n)*10^-3)*(t-t1),0.7);
Ab=2*pi*rp*8;%burn area
dP = (Ab.*a.*P.^n.*(rhoP-rhoO)-P.*Astar.*sqrt(k/(R.*T0)).*(2/(k+1)).^((k+1)/(2.*(k-1)))).*R.*T0./v0;
end
0 Kommentare
Weitere Antworten (1)
Ous Chkiri
am 3 Nov. 2019
2 Kommentare
Thiago Henrique Gomes Lobato
am 3 Nov. 2019
If you have conditions that make a discontinuity you will have to perform the integration in a piece-wise matter and add the conditions in your integration function ("Fun" in the example I gave you). This answer and comments shows an example about how you could do it and also possible pitfalls https://de.mathworks.com/matlabcentral/answers/487643-adding-a-piecewise-time-dependent-term-in-system-of-differential-equation#answer_398394?s_tid=prof_contriblnk
Siehe auch
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!