System of Nonlinear Equations are failing to be solved

1 view (last 30 days)
Hello Altruists,
I am trying to solve the system of nonlinear equations using both fsolve() and vpasolve(). However, they are failing to provide any result and end up in some empty symbolic variables (for vpasolve()) or just stopped iteration at some point (for fslove()). The code given below is a simplified version of my code. In this case, I have to solve two equations (in reality 5 equations) eq1 and eq2 to find out the roots Kp, Ki. How can I solve this problem?
syms Kp ;
syms Ki ;
Kd = 0.2023;
lm = 1.1477;
mu = 0.9973;
phi_m = 80;
w_gc = 0.3;
L = 0.2694;
K = 0.42456;
p2zw = 1.2157;
wn = 0.50679;
a = 2.109;
b = 1.015;
P_mag = K / ((w_gc^(a)*cos(a*pi/2) + p2zw*w_gc^(b)*cos(b*pi/2) + wn^2)^2 + (w_gc^(a)*sin(a*pi/2) + p2zw*w_gc^(b)*sin(b*pi/2))^2);
P_arg = -w_gc*L-atan((w_gc^a*sin(a*pi/2))+(p2zw*w_gc^b*sin(b*pi/2))/(w_gc^a*cos(a*pi/2))+(p2zw*w_gc^b*cos(b*pi/2)+w_gc^2));
C_mag = sqrt((Kp + (Ki*w_gc^(-lm)*cos(lm*pi/2)) + Kd*w_gc^mu*cos(mu*pi/2))^2+((-Ki*w_gc^(-lm)*cos(lm*pi/2)) + Kd*w_gc^mu*sin(mu*pi/2))^2);
C_arg = atan(((-Ki*w_gc^(-lm)*cos(lm*pi/2)) + Kd*w_gc^mu*sin(mu*pi/2))/(Kp + (Ki*w_gc^(-lm)*cos(lm*pi/2)) + Kd*w_gc^mu*cos(mu*pi/2)));
eq1 = C_arg + P_arg == pi-phi_m*pi/180;
eq2 = C_mag*P_mag == 1;
[sol1, sol2] = vpasolve([eq1 eq2],[Kp Ki])
sol1 = 
0.40751782462346212168748687200968
sol2 = 
0.1065888515318568220223455502278

Accepted Answer

Torsten
Torsten on 25 Sep 2022
In my opinion, it should be
eq1 = C_arg + P_arg == pi-phi_m*pi/180;
instead of
eq1 = C_arg + P_arg == pi-phi_m*180/pi;
(see above)

More Answers (0)

Community Treasure Hunt

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

Start Hunting!