How to Set Up a Genetic Algorithm to Minimize Goodness of Fit

I would like to use the genetic algorithm functionality to minimize the error between two sets of data, and determine the best values for two coefficients. The following is my code:

% Set initial paramters.
Cl_initial=3;
R=0.5;
u=0.5;
T=40;

t=[68.92 109.97 287.22 116.95 171.89 90.3 103.54];
Nt = length(t);

Clt = zeros(1,Nt);
THM = zeros(1, Nt);

result = zeros(2, Nt);

% Solve for Chlorine residual.
for i = 1:Nt
    Clt(i) = [Cl_initial*(1-R)/(1-R*exp(-u*t(i)))];
end

% Solve for THM formation.
NClt = length(Clt);
for j = 1:NClt
    THM(j) = T*(Cl_initial-Clt(j));
end

%Combine into single matrix.
result = [Clt; THM];

%Calculate mean square error between results and solution.
solution=[0.7075 0.9133 0.7125 0.86 0.7375 0.8225 0.984;
    16.4 24.7 27.4 22.7 27.2 20.6 22.9];

cost_func = 'MSE';
fit = goodnessOfFit(result, solution, cost_func);

I need to minimize "fit" while varying the coefficients of "R" and "u" to find the best values for those two coefficients. I also know that R and u have to be between 0 and 1.

I am a little overwhelmed as to where to start. Any help would be appreciated.

Thanks.

[Brendan Hamm]

Why would you think that you would need a Genetic Algorithm here?

This could be solved with basic Optimization Toolbox solvers like fmincon or lsqcurvefit .

[CJ]

Yes, you are correct. I was thinking I could use the ga function in preparation for something later, but it is much easier to use fmincon. Thanks.