I want to solve the equation within the loop and plot it, but it seems not work

clear;
close all;
Ls = 75e-6;
Lr = 75e-6;
Lm = 2e-3;
rs = 20e-3;
rr = 20e-3;
Vbat = 400;
nph = 3;
B = 0.01;
Kload = 50e-7;
P = 4;%poles
for nr = 0:6000
    wr = 2*pi/60*nr;
    ns = 120*fe/P;
    Vs_rms = Vbat/2/2^0.5;
    syms fe;
    Xms = fe*Lm;
    Xls = fe*Ls;
    Xlr = fe*Lr;
    Vth = Xms*Vs_rms/(rs^2+(Xms+Xls)^2)^0.5;
    rth = rs*(Xms/(Xms+Xls));
    Xth = Xms*Xls/(Xms+Xls);
    we = 2/P*2*pi*fe;
    s = (we-wr)/we;
    eqn = nph*Vth^2*rr/(s*we)/((rth+rr/s)^2+(Xth+Xlr)^2)== Kload*nr^2+B*wr; %Te
    [solfe] = solve(eqn,fe);
    fe = double(solfe);
    Te = Kload*nr^2+B*wr
    plot (wr, Te);
    plot (we, Te);
    grid on;
    hold on;
    xlabel('rad/s');
    ylabel('T');
    legend;
end

Answer 1

[Jon]

Maybe you can explain further what you mean by it doesn't seem to work. What is it doing that you don't want it to do or what isn't it doing that you want it to do.

In any case, without getting into all of the details of your program I can see two possible issues.

I see that you issue two plot commands one right after the other. In this case the second plot will overwrite the first. You should call the figure function between the calls to plot to open a new figure. So it should look something like this instead

    plot (wr, Te);
    figure
    plot (we, Te);

Also, it doesn't look like you "hold on" is place correctly. It looks like this would make a curve and keep it for every loop iteration, so 6000 curves. Is that what you intended?

Finally, for when you want to include code it is a good idea to use the code button in the toolbar where you edit your question. That way it is nicely formatted and easier for other people to view and use.

[Xueshen Zhang]

I want to get the solution of fe within each loop, and plot speed versus torque. But it seems like the solve function always give me four answers(including a negative value, positive value, x+yj, x-yj) each loop, which makes plot function not work. I solve it by extracting the second answer each time. But are there any better ideas?

[Walter Roberson]

Perhaps it would make sense to use vpasolve(), and give it a range of [0 inf] so that it does not give imaginary or negative results.