conversion to logical from sym is not possible error within a for loop

Hi everyone!

I had an issue regarding referring to my variables in a loop. Essentially, above my loop, I have a bunch of functions defined where Csns, Czns, Csnl and Cznl are symbols.

In the for loop, I wanted to input each distinct combination of the 4 variables into a system of 3 equations, and if the equations held true, record those values in solution.

However there is an error repeatedly coming up saying:

Conversion to logical from sym is not possible.

Error in mtrl575prob4attempt5Alyssa (line 183) if dGS_Csns - dGL_Csnl > -.0001

Is there any way to negate this error in the loop?

Far above the loop: GS (Csns, Czns) GL (Csnl, Cznl) dGS_Csns (Csnl, Cznl) dGS_Czns (Csnl, Cznl) dGL_Csnl (Csnl, Cznl) dGL_Cznl (Csnl, Cznl) are all set as complicated functions of the variables shown in brackets (they are really large so I did not include them).

%system of equations we want to solve:
eq1 = dGS_Csns - dGL_Csnl == 0;
eq2 = dGS_Czns - dGL_Cznl  == 0;
eq3 = dGL_Csnl*Csnl+dGL_Cznl*Cznl-dGS_Csns*Csns+dGS_Czns*Czns-(GL-GS) == 0;

solution = zeros(1,4);

for i = 0.001:0.001:1
    Csns = i;
    for j = 0.001:0.001:1;
        Czns = j;
        for k = 0.001:0.001:1
            Csnl = k;
            for z = 0.001:0.001:1
                Cznl = z;
                if eq1 > -.0001 && eq1 <.0001
                    if eq2  >-.0001 && eq2  <.0001
                        if eq3 >-.0001 && eq3 <.0001
                            n = 1;
                            solution(n,:) = [Csns,Czns,Csnl,Cznl];
                            n = n + 1;
                        end
                    end
                end
            end
        end
    end
end

Answer 1

[Walter Roberson]

[Csns, Czns, Csn, Cznl] = ndgrid( 0.001:0.001:1, 0.001:0.001:1, 0.001:0.001:1, 0.001:0.001:1);

TGS       = GS(Csns, Czns);
TGL       = GL(Csnl, Cznl);
TdGS_Csns = dGS_Csns(Csnl, Cznl);
TdGS_Czns = dGS_Czns(Csnl, Cznl);
TdGL_Csnl = dGL_Csnl(Csnl, Cznl);
TdGL_Cznl = dGL_Cznl (Csnl, Cznl);

eq1 = TdGS_Csns - TdGL_Csnl;
eq2 = TdGS_Czns - TdGL_Cznl;
eq3 = TdGL_Csnl .* Csnl + TdGL_Cznl .* Cznl - TdGS_Csns .* Csns + TdGS_Czns .* Czns - (TGL - TGS);

eq1holds = abs(eq1) < 0.001;
eq2holds = abs(eq2) < 0.001;
eq3holds = abs(eq3) < 0.001;

alleqhold = eq1holds & eq2holds & eq3holds;

n = nnz(alleqhold);    %number of non-zeros is count of trues

where_holds = [Csns(alleqhold), Czns(alleqhold), Csnl(alleqhold), Cznl(alleqhold)];

Expect this to take rather some time to calculate all 10^12 calculations. And hope you have a lot of memory.

[Alyssa]

Hi Walter,

I like the idea of the abs to clean it up a bit.

I am running into issues with ndgrid though. I had initially tried using it but encountered memory problems, which is why I resorted to looping.