For the code below how would I make it ALWAYS select and set the upper left corner to be 1 always no matter what the n value is (10, 1000, 1000000)?

 n=10;           % number of nodes in the x-direction 
meshsize = n*n/2;
nodes = 1:meshsize;
nodetype = zeros(n/2, n);
delx = 2/n;
[X,Y] = meshgrid(delx:delx:2,delx:delx:1);

 for i = 1:length(nodes)
     if X(i) == .2 && Y(i) == .2                    %DOESNT SCALE 
         nodetype(i) = 1;
         %FOR TOP LEFT CORNER
 end

[KALYAN ACHARJYA]

Still, Question is not clear for me, can you elaborate more?

Be Specific?

[Josh Glenn]

So, my full code is posted below...It is a project to select and set each node in a heat transfer system and certain node points have different conditions and equations to use. I'm currently in the stage where I'm assigning each node type so that I can later evaluate each one based on its conditions. I'm currently stuck with how to set the corner nodes while allowing the n value to scale.

Does this help you to better understand the question?

 %%Clear, clc, close, format
clear
clc
close all
format short
%%knowns 
n=10;                               % number of nodes in the x-direction 
%Create Mesh based on n
meshsize = n*n/2;
nodes = 1:meshsize;
nodetype = zeros(n/2, n);
delx = 2/n;
[X,Y] = meshgrid(delx/2:delx:2,delx/2:delx:1); 
%Problem Givens
T1 = 28;                %C
T2 = 50;                %C
T3 = @(y) 10 + 14*y;    %C
T_inf = 32;             %C
h = 28;                 %W/m^2*K
k = 12;                 %W/m*K
q_dot = 47000;          %W/m^3
%%Node Stuff
 for i = 1:length(nodes)
     if X(i) == .1 && Y(i) == .1                        %DOESNT SCALE 
         nodetype(i) = 1;
         %FOR TOP LEFT CORNER
     elseif X(i) == 1.9 && Y(i) == .1                   %DOESNT SCALE OR WORK
         nodetype(i) = 2;
         %FOR TOP RIGHT CORNER
     elseif X(i) == .1 && Y(i) == .9                    %DOESNT SCALE
         nodetype(i) = 3;
         %FOR BOTTOM LEFT CORNER
     elseif X(i) == 1.9 && Y(i) == .9                   %DOESNT SCALE OR WORK
         nodetype(i) = 4;
         %FOR BOTTOM RIGHT CORNER
     elseif X(i) > .1 && X(i) < 1.9 && Y(i) == .1       %DOESNT SCALE
         nodetype(i) = 5;
         %FOR TOP EDGE
     elseif X(i) == .1 && Y(i) > .1 && Y(i) < .9        %DOESNT SCALE
         nodetype(i) = 6;
         %FOR LEFT EDGE
     elseif X(i) == 1.9 && Y(i) >.1 && Y(i) < .9
         nodetype(i) = 7;
         %FOR RIGHT EDGE
%     elseif 
%          nodetype(i) = 8;
%          %FOR BOTTOM LEFT EDGE
%      elseif
%          nodetype(i) = 9;
%          %FOR BOTTOM RIGHT EDGE
%      elseif
%          nodetype(i) = 10;
%          %FOR BOTTOM LEFT INNER CORNER
%      elseif
%          nodetype(i) = 11;
%          %FOR BOTTOM RIGHT INNER CORNER
%      elseif
%          nodetype(i) = 12;
%          %FOR BOTTOM LEFT INNER SIDE
%      elseif
%          nodetype(i) = 13;
%          %FOR BOTTOM RIGHT INNER SIDE
%      elseif
%          nodetype(i) = 14;
%          %FOR BOTTOM CENTER EDGE
%      elseif
%          nodetype(i) = 15;
%          %FOR FULL GENERATION
%      elseif
%          nodetype(i) = 16;
%          %FOR HALF GENERATION
%      elseif
%          nodetype(i) = 0;
%          %FOR NOTHING THERE (EMPTY)
     else
         nodetype(i) = 17;
         %Regular Nodes
    end

 end

Answer 1

[Dennis]

I am only guessing, but i assume you want nodetype(1,1) to be 1? The easiest way of doing this would be:

nodetype(1,1)=1;

actually this does the same:

nodetype(1)=1;

Now your loop does not set the top left corner to 1, but rather checks for 0.2 values. However depending on your n it is very possible that there are none in your matrix. This is obvious for n=98, but it is also true for n=100, although the variable inspector might tell you something different on the first look.

The reason for this is that not every floating point number can be represented exact in binary form. For more information on this topic please look here.

[Stephen23]

"... not every floating point number can be represented exact in binary form"

By definition, all binary floating point numbers are stored in a binary form, which exactly represents that floating point number... did you actually mean something like "not all decimal fractions are representable using (finite) binary floating point numbers"?