How can I use switch case function to get decimal numbers?

M% Temperatur_C = [13,6169 14,5432 15,2356 17,2456 18,1887 19.9874 21,2765]
% Coeff            -30C   -20   -10   0    10     20   30    40    50 
% nT                1.8   1.8   1.4   1.1  1.1    1    0.9   0.9   0.8

             for A=1:(a-1)
             switch (Temperatur_C(1,A))
             case num2cell(-10.0:0)
             disp('nT=1.4')
             nT_NO(1,A) = 1.4;
             case num2cell(0.1:4.99)
             disp('nT=1.1')
             nT_NO(1,A) = 1.1;
             end
             
             
             If I use round command I get the answer but that's not right for me because I want to be able to get the specific temperature not rounded.For example if the temp is 18,1887 it gets it as 20.
             
             for A=1:(a-1)
             switch round(Temperatur_C(1,A))
             case num2cell(-10.0:0)
             disp('nT=1.4')
             nT_NO(1,A) = 1.4;
             case num2cell(0.1:4.99)
             disp('nT=1.1')
             nT_NO(1,A) = 1.1;
             end
             

Answer 1

[Guillaume]

I'm curious if you understand what that num2cell(-10.0:0) actually do in the case statement? If you don't and it's something you've been given by someone else don't use it.

It's not clear exactly what you're trying to do. I suspect it is this (minus the disps which are a bit pointless):

Breakpoints = [-Inf -30  -10  0    20  30   50 +Inf];
nT_Values =   [NaN  1.8  1.4  1.1  1   0.9  0.8];
whichcolumn = discretize(Temperatur_C, Breakpoints);
nT_NO = nT_Values(whichcolumn);
%and if you want the equivalent to disp:
%celldisp(compose('nT=%g', nT_NO));

You can't do what you were trying to do with switch statements, you could do it with if...elseif

for idx = 1:numel(Temperature_C)
    if Temperatur_C(idx) >= -30 & Temperatur_C(idx) < -10
        nT_NO(idx) = 1.4;
    elseif Temperatur_C(idx) >= -10 & Temperatur_C(idx) < 0
        nT_NO(idx) = 1.1;
    elseif ...etc
    end
end

As you can see, using discretize is much simpler.

[Kofial]

Yes, it works! Sice I have to do this for different gasses, do you know a shorter way?

Thats how I did it for two gasses:

Temp_range = [-20 -10 0 10 20 30 40];
nT_OX_coeff = [0.9 0.9 1 1.3 1.5 1,7 2 2,5];
nT_NO_coeff = [1.8 1.8 1.4 1.1 1.1 1 0.9 0.9];

for idx = numel(Temperatur_C)
    if Temperatur_C(idx) >=-15 & Temperatur_C(idx) <-4.99
        nT_OX(idx)=1;
    elseif Temperatur_C(idx) >=-5 & Temperatur_C(idx) <4.99
        nT_OX(idx)=1.3;
    elseif Temperatur_C(idx) >= 5 & Temperatur_C(idx) <14.99
        nT_OX(idx)=1.5;
    elseif Temperatur_C(idx) >=15 & Temperatur_C(idx) <24.99
        nT_OX(idx)=1.7;
    elseif Temperatur_C(idx) >=25 & Temperatur_C(idx) <34.99
        nT_OX(idx)=2;
    elseif Temperatur_C(idx) >=35 & Temperatur_C(idx) <45
        nT_OX(idx)=2.5;
    end
end

for idx = numel(Temperatur_C)
    if Temperatur_C(idx) >=-15 & Temperatur_C(idx) <-4.99
        nT_NO(idx)=1.4;
    elseif Temperatur_C(idx) >=-5 & Temperatur_C(idx) <4.99
        nT_NO(idx)=1.1;
    elseif Temperatur_C(idx) >= 5 & Temperatur_C(idx) <14.99
        nT_NO(idx)=1.1;
    elseif Temperatur_C(idx) >=15 & Temperatur_C(idx) <24.99
        nT_NO(idx)=1.1;
    elseif Temperatur_C(idx) >=25 & Temperatur_C(idx) <34.99
        nT_NO(idx)=0.9;
    elseif Temperatur_C(idx) >=35 & Temperatur_C(idx) <45
        nT_NO(idx)=0.9;
    end
end

[Steven Lord]

Call discretize to determine which bin each element of Temperatur_C is in. Have a vector of nT values for each gas. Index into that vector of nT values for each gas using the bin numbers.

This requires two lines for the temperature data (one to define the data, one to discretize the data) and two lines per gas (one to define the nT values, one for the indexing.)

[Guillaume]

Yes, as I've explained use discretize it's so much simpler that your two loops with endless elseif, plus the discretize call, as I wrote it, has the advantage that you get defined behaviour even if the temperature is outside your defined [-15, 45] range. With your code, nothing is ever assigned to the output variables (which may cause error later on as the variable may not even exist).

Your code can be simplified to:

breakpoints = [-Inf, -15,  -5,   5,   15,   25,  35,  45, +Inf];  %temperature break points
nT_OX_Coeff = [    NaN,  1,  1.3,  1.5,  1.7,  2,  2.5, NaN];
nT_CO_Coeff = [    NaN, 1.4, 1.1,  1.1,  1.1, 0.9, 0.9, NaN];

nT_OX = discretize(Temperatur_C, breakpoints, nT_OX_Coeff);
nT_NO = discretize(Temperatur_C, breakpoints, nT_CO_Coeff);
%all done. No loop, no endless if...else...elseif

edit: lots of typos!

[Steven Lord]

In addition to the advantage Guillaume cited about what happens outside the bounds, this way if you need to add additional breakpoints (and the corresponding nT coefficients) you only need to modify the lines where the data was created. You don't need to modify the code at all (and run the risk of updating the large if / else / elseif / end statement for one gas but not the other.)