A smoother surfplot, maybe with pchip interpolation

hello everyone,

I have made a surfplot using linear interpolation with x,y,z data from a csv. However i want to make the plot smoother. I have tried cubic interpolation and it does make it better. However i have heard that pchip would make it even smoother, though i have not been able to use that. So looking for a way to utilise pchip or any other interpolation method to make my surfplot smoother. and yes i am aware that increasing N would make it better, but looking for other ways.

T1 = readtable('solarcol3.csv');
VarNames = T1.Properties.VariableNames; 
x = T1{:,1};
y = T1{:,2};
z = T1{:,3};

[Ux,iax,ixx] = unique(x);
[Uy,iay,ixy] = unique(y);

N = 25;
xv = linspace(min(x), max(x), N);
yv = linspace(min(y), max(y), N);

[Xm,Ym] = ndgrid(xv, yv);

Zm = griddata(x, y, z, Xm, Ym, 'cubic'); % specify cubic interpolation

figure
surfc(Xm, Ym, Zm)
colorbar
grid on
xlabel('Wind Velocity (m/s)')
ylabel('Heat Flux (W/m2)')
zlabel('Mass Flow Rate (kg/s)')

Answer 1

[Mathieu NOE]

hello

I used this excellent FEX submission to get this result :

smoothn - File Exchange - MATLAB Central (mathworks.com)

T1 = readtable('solarcol3.csv');
VarNames = T1.Properties.VariableNames; 
x = T1{:,1};
y = T1{:,2};
z = T1{:,3};
[Ux,iax,ixx] = unique(x);
[Uy,iay,ixy] = unique(y);
N = 25;
xv = linspace(min(x), max(x), N);
yv = linspace(min(y), max(y), N);
[Xm,Ym] = ndgrid(xv, yv);
Zm = griddata(x, y, z, Xm, Ym, 'cubic'); % specify cubic interpolation

Zm = smoothn(Zm,1000); % Fex : https://fr.mathworks.com/matlabcentral/fileexchange/25634-smoothn?s_tid=ta_fx_results

figure
surfc(Xm, Ym, Zm)
colorbar
grid on
xlabel('Wind Velocity (m/s)')
ylabel('Heat Flux (W/m2)')
zlabel('Mass Flow Rate (kg/s)')

[Mohammed Saifuddin Ustad]

Hey thanks so much, that works beautifully. Also is the picture that you uploaded for s=1000? because mine looks a different, plus the formatting of your axis text is different too. Can you also upload your .fig file. thanks

[Mathieu NOE]

hello again

yes the picture was done with s = 1000 as in the code I posted

now you can of course adapt s to your own taste

fyi the fig (with s = 1000) in attachment)

[Mohammed Saifuddin Ustad]

hey thanks so much, that works beautifully. however is there any way to specifically set the aspect ratio of the window/plot. i wanted a 5:3 ratio, width:height that is. any way to set that?

[Mathieu NOE]

hello again

yes you can specify the size and position of any figure , like in this example :

%%%Matlab convention [left bottom width height]%%%
set(0,'Units','Pixels');
scrsz = get(0,'ScreenSize');
scr_width = scrsz(3);
scr_heigth = scrsz(4);

% create a large 5/3 format figure
height = scr_heigth/1.25;
width = 5/3*height;
left = (scr_width - width)/2;
bottom = (scr_heigth - height)/2;


f = figure('Position', [left bottom width height]);  %%%Matlab convention [left bottom width height]%%%

[Mohammed Saifuddin Ustad]

heyy thanks veryy muchh

any idea how to make an equation out of my xyz data??

[Mohammed Saifuddin Ustad]

i want to solve for z inserting x and y

[Mohammed Saifuddin Ustad]

or maybe i can formulate an equation from the 3D plot

[Mathieu NOE]

this is a job either for the curve fitting toolbox ( I don't have it) or some available files on the FEX section , like :

2D Weighted Polynomial Fitting and Evaluation - File Exchange - MATLAB Central (mathworks.com)

Fit 2D polynomial surface to scattered data - File Exchange - MATLAB Central (mathworks.com)

Fast Polynomial Curve and Surface Fitting - File Exchange - MATLAB Central (mathworks.com)

[Mathieu NOE]

forgot this one :

polyfitn - File Exchange - MATLAB Central (mathworks.com)

hope it helps

[Mathieu NOE]

and this one :

2D polynomial fitting with SVD - File Exchange - MATLAB Central (mathworks.com)

[Mathieu NOE]

Hello

Problem solved ?

would you mind accepting my answer ? thanks !