Azzera filtri
Azzera filtri

Inconsistency of the figure and value on the curve

3 visualizzazioni (ultimi 30 giorni)
Hi all. I wrote a code to find the curve for trace and determinant. The problem is that when I zoom and choose one value on the curve and put it into the code, the value obtained for T is different from what is shown on the curve.
For example, here, for c=0.05826 on the curve is -4.06892e-07, while when code runs for this specific value, gives me T =-2.4350e-04.
Could you please let me know what the problem is that this happened and how I can solve it? Thanks in advance for any help
vplc=0.19;
delta=2.5;
tau_max=0.18;
Ktau=0.045;
kc=0.1;
kh=0.05;
Kbar=0.000015;
kp=0.15;
kb=0.4;
Kpm=0.15;
Ke=7;
vs=0.002;
ks=0.1;
kplc=0.055;
ki=2;
gamma=5.5;
kipr=0.18;
c=0.05826
%c =0:0.000001:0.3; %first I plotted the curves for this interval, zoom and choose a point on the curve and run the code for that specific vale
p=(vplc./ki).*(c.^2./((kplc).^2+c.^2));
h=kh.^4./(c.^4+kh.^4);
Po=(p.^2.*c.^4.*h)./(p.^2.*c.^4.*h.*(1+kb)+kb.*kp.^2.*(kc.^4+c.^4-(c.^4.*kh.^4./(c.^4+kh.^4))));
ct=(vs./(gamma.*kipr.*Po)).*(c.^2./(c.^2+ks.^2))+c.*(1+(1/gamma));
Podenominator=(p.^2.*c.^4.*h.*(1+kb)+kb.*kp.^2.*(kc.^4+c.^4-((c.^4.*kh.^4)./(c.^4+kh.^4))));
Dcnumerator=(8.*c.^7.*kh.^4*vplc.^2)./(ki.^2*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^2) - (4.*c.^9.*kh.^4.*vplc.^2)./(ki.^2.*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^3) - (4.*c.^11.*kh.^4*vplc.^2)./(ki.^2.*(c.^4 + kh.^4).^2.*(c.^2 + kplc.^2).^2);
Dcdenominator=kb.*kp.^2.*(4.*c.^3 - (4*c.^3*kh.^4)./(c.^4 + kh.^4) + (4.*c.^7*kh.^4)./(c.^4 + kh.^4).^2) + (8.*c.^7*kh.^4.*vplc.^2.*(kb + 1))./(ki.^2.*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^2) - (4*c.^9.*kh.^4.*vplc.^2.*(kb + 1))/(ki.^2.*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^3) - (4*c.^11.*kh^4.*vplc^2.*(kb + 1))./(ki^2.*(c.^4 + kh^4).^2.*(c.^2 + kplc.^2).^2);
DcPo=(Dcnumerator.*Podenominator-Dcdenominator.*(p.^2.*c.^4.*h))./(Podenominator.^2);
DhPo=((c.^4.*p.^2.*(Podenominator))-(c.^8.*p.^4.*(kh.^4./(c.^4+kh.^4)).*(1+kb)))./(Podenominator.^2);
Dcf1=kipr.*DcPo.*gamma.*ct-kipr.*DcPo.*(1+gamma).*c-kipr.*Po.*(gamma+1)-((2*vs.*c.*(ks^2))./(c.^2+ks^2).^2);
Dhf1=kipr.*DhPo.*(gamma.*ct-(1+gamma).*c);
Dcf2=(4.*c.^3./tau_max).*((kh.^8./(c.^4+kh.^4).^2))-(4.*c.^3./tau_max).*(kh.^4./(c.^4+kh.^4));
Dhf2=-c.^4./tau_max;
T = Dcf1 + Dhf2
D = Dcf1.*Dhf2-Dhf1.*Dcf2
eig1=(T+sqrt(T.^2-4.*D))./2;
eig2=(T-sqrt(T.^2-4.*D))./2;
figure
plot(c,T,'b')
hold on;
plot(c,D,'r')
hold on
plot(c,T.^2-4.*D,'g')
xlabel('c')
ylabel('T & D')
xlim([0 0.3])
ylim([-0.001 0.003])

Risposta accettata

Paul
Paul il 18 Ott 2022
Modificato: Paul il 18 Ott 2022
Hi @M
The equation for Dcdenominator is not properly vectorized. In one spot it uses /, which should be ./
% was
% Dcdenominator=kb.*kp.^2.*(4.*c.^3 - (4*c.^3*kh.^4)./(c.^4 + kh.^4) + (4.*c.^7*kh.^4)./(c.^4 + kh.^4).^2) + (8.*c.^7*kh.^4.*vplc.^2.*(kb + 1))./(ki.^2.*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^2) - (4*c.^9.*kh.^4.*vplc.^2.*(kb + 1))/(ki.^2.*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^3) - (4*c.^11.*kh^4.*vplc^2.*(kb + 1))./(ki^2.*(c.^4 + kh^4).^2.*(c.^2 + kplc.^2).^2);
%
% should be
% xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx^use./
Dcdenominator=kb.*kp.^2.*(4.*c.^3 - (4*c.^3*kh.^4)./(c.^4 + kh.^4) + (4.*c.^7*kh.^4)./(c.^4 + kh.^4).^2) + (8.*c.^7*kh.^4.*vplc.^2.*(kb + 1))./(ki.^2.*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^2) - (4*c.^9.*kh.^4.*vplc.^2.*(kb + 1))./(ki.^2.*(c.^4 + kh.^4).*(c.^2 + kplc.^2).^3) - (4*c.^11.*kh^4.*vplc^2.*(kb + 1))./(ki^2.*(c.^4 + kh^4).^2.*(c.^2 + kplc.^2).^2);
  2 Commenti
M
M il 18 Ott 2022
Excellent @Paul, it works perfectly... thanks
I have a question, I would appreciate if you can answer it.
With another method that I trust, I found a point where T~0 (when c~0.091), but even though all my calculations (DcPo, DhPo, T and D...)and parameters are correct here (I checked them in Maple), it contradicts with what I got (for example here for c~0,058, T becomes almost zero). you don't have any suggestion how to find the problem?

Accedi per commentare.

Più risposte (0)

Community Treasure Hunt

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

Start Hunting!

Translated by