Plotting a contour plot for a variable dependednt on the while loop
    2 views (last 30 days)
  
       Show older comments
    
Hello everyone,
I'm currently running a code to see how long does it take for a range of volume to reach 0 for different relative humidities and I'm trying to plot a contour plot for the values with Volume on the y-axis and Relative humidity on the x axis and the time taken on the right side. For some reason, my volume doesn't seem to converge and the code won't stop running. Does anyone know why?
Thanks in advance
clear all 
close all
clc
%% Inputs
IR = 1e-3; %initial radius [m]
Rho = 1000; %density [kg/m^3]
T = 23.5; %temperature [celsius]
% Loop 1
aInitial = 1;
aStep = 1;
aMax = 25;
VMin = 1e-12; %minimum volume [m^3]
VMax = 2e-9; %maximum volume [m^3]
% Loop 2
bInitial = 1;
bStep = 1;
bMax = 25;
ReHuMin = 0.0; %minimum relative humidity [m]
ReHuMax = 0.9; %maximum relative humidity [m]
% Loop 3
c = 1;
tInitial = 0; %initial time [s]
tStep = 1; %final time [s]
D_T = 2.5e-4*exp(-684.15/(T+273.15)); %coefficient [m^2/s]
c_sat = (9.99e-7)*T^3 - (6.94e-5)*T^2 + (3.2e-3)*T - 2.87e-2; %concentration [kg/m^3]
%% Computing
for a = aInitial:aStep:aMax
    for b = bInitial:bStep:bMax
        Vol(a,b) = VMin + (VMax - VMin)*(a-aInitial)/(aMax);
        h(a,b) = ((sqrt(pi^2*IR^6 + 9*(Vol(a,b))^2) + 3*(Vol(a,b)))^(2/3) - pi^(2/3)*IR^2)/(pi^(1/3)*(sqrt(pi^2*IR^6 + 9*(Vol(a,b))^2) + 3*(Vol(a,b)))^(1/3));
        CAR(a) =  2*atan(h(a,b)/IR);
        CAD(a) = CAR(a)*180/pi;
        RH(a,b) = ReHuMin + (ReHuMax - ReHuMin)*(b - bInitial)/(bMax);
        h(c,a,b) = h(a,b);
        CAR(c,a,b) = CAR(a);
        Vol(c,a,b) = Vol(a,b);
        timeSec(c,a,b) = tInitial;
        while Vol(c,a,b) > 0
            M_dot(c,a,b) = -pi*IR*D_T*(1 - RH(a,b))*c_sat*(0.27*CAR(c,a,b)^2+1.30); %mass flow
            Mkg(c,a,b) = M_dot(c,a,b)*tStep; %mass loss at each time step [kg]
            Vm3(c,a,b) = Mkg(c,a,b)/Rho; %volume loss [m^3]
            Vol(c+1,a,b) = Vol(c,a,b) + Vm3(c,a,b); %new volume [m^3]
            h(c+1,a,b) = ((sqrt(pi^2*IR^6 + 9*(Vol(c+1,a,b))^2) + 3*(Vol(c+1,a,b)))^(2/3) - pi^(2/3)*IR^2)/(pi^(1/3)*(sqrt(pi^2*IR^6 + 9*(Vol(c+1,a,b))^2) + 3*(Vol(c+1,a,b)))^(1/3)); %new height [m]
            CAR(c+1,a,b) =  2*atan(h(c+1,a,b)/IR); %new angle [radians]
            CAD(c+1,a,b) = CAR(c+1,a,b)*180/pi; %new angle [degrees]
            timeSec(c+1,a,b) = tInitial + tStep*(c - 1);
            c = c + 1;
        end
        EvaporationTime(a,b) = timeSec(c,a,b);
        c = 1;
    end
end
%% Plotting
contourf(RH*100,Vol*1e9,EvaporationTime)
c = colorbar;
c.Label.String = 'Evaporation Time (s)';
xlabel('Relative Humidity (%)')
ylabel('Volume (mm^3)')
0 Comments
Accepted Answer
  Alan Stevens
      
      
 on 8 Aug 2020
        I'm not sure if I've interpreted your requirements correctly, but what about the following:
%% Inputs
IR = 1e-3; %initial radius [m]
Rho = 1000; %density [kg/m^3]
T = 23.5; %temperature [celsius]
% Loop 1
aInitial = 1;
aStep = 1;
aMax = 25;
VMin = 1e-12; %minimum volume [m^3]
VMax = 2e-9; %maximum volume [m^3]
% Loop 2
bInitial = 1;
bStep = 1;
bMax = 25;
ReHuMin = 0.0; %minimum relative humidity [m]
ReHuMax = 0.9; %maximum relative humidity [m]
% Loop 3
c = 1;
tInitial = 0; %initial time [s]
tStep = 1; %final time [s]
D_T = 2.5e-4*exp(-684.15/(T+273.15)); %coefficient [m^2/s]
c_sat = (9.99e-7)*T^3 - (6.94e-5)*T^2 + (3.2e-3)*T - 2.87e-2; %concentration [kg/m^3]
%% Computing
for a = aInitial:aStep:aMax
    Va = VMin + (VMax - VMin)*(a-aInitial)/(aMax);
    for b = bInitial:bStep:bMax
        V(a,b) = Va;  %Initial volume
        Vol(a,b) = Va;
        h(a,b) = ((sqrt(pi^2*IR^6 + 9*(Vol(a,b))^2) + 3*(Vol(a,b)))^(2/3) - pi^(2/3)*IR^2)/(pi^(1/3)*(sqrt(pi^2*IR^6 + 9*(Vol(a,b))^2) + 3*(Vol(a,b)))^(1/3));
        CAR(a,b) =  2*atan(h(a,b)/IR);
        CAD(a,b) = CAR(a,b)*180/pi;
        RH(a,b) = ReHuMin + (ReHuMax - ReHuMin)*(b - bInitial)/(bMax);
        timeSec(a,b) = tInitial;
         Vold = 0; c = 0;
        while Vol(a,b)>VMin && c<10000
            Vold = Vol(a,b);
            M_dot(a,b) = -pi*IR*D_T*(1 - RH(a,b))*c_sat*(0.27*CAR(a,b)^2+1.30); %mass flow
            Mkg(a,b) = M_dot(a,b)*tStep; %mass loss at each time step [kg]
            Vm3(a,b) = Mkg(a,b)/Rho; %volume loss [m^3]
            Vol(a,b) = Vol(a,b) + Vm3(a,b); %new volume [m^3]
            Vol(a,b) = max(Vol(a,b), VMin);
            h(a,b) = ((sqrt(pi^2*IR^6 + 9*(Vol(a,b))^2) + 3*(Vol(a,b)))^(2/3) - pi^(2/3)*IR^2)/(pi^(1/3)*(sqrt(pi^2*IR^6 + 9*(Vol(a,b))^2) + 3*(Vol(a,b)))^(1/3)); %new height [m]
            CAR(a,b) =  2*atan(h(a,b)/IR); %new angle [radians]
            CAD(a,b) = CAR(a,b)*180/pi; %new angle [degrees]
            timeSec(a,b) = tStep*c;
            c = c + 1;
        end
        EvaporationTime(a,b) = timeSec(a,b);
    end
end
%% Plotting
contourf(RH*100,V*1e9,EvaporationTime)  % plot against Initial volume not final volume
k = colorbar;
k.Label.String = 'Evaporation Time (s)';
xlabel('Relative Humidity (%)')
ylabel('Initial Volume (mm^3)')
figure
surf(RH*100,V*1e9,EvaporationTime)
k = colorbar;
k.Label.String = 'Evaporation Time (s)';
xlabel('Relative Humidity (%)')
ylabel('Initial Volume (mm^3)')
zlabel('Evaporation Time (s)')
More Answers (1)
  Alan Stevens
      
      
 on 8 Aug 2020
        Your code doesn't stop because Vol never reaches zero.  
What does the c index do for you?
Your contour function won't plot while you have RH and EvaporationTime as 2d matrices with Vol as a 3d matrix.
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
