how can i use the output of one genetic algorithm program in another GA programme.

lets say we have the equation

t = 122.14-0.84.*x(1)-0.42.*x(2)+0.34.*x(3)-0.09.*x(4)+363.77.*x(5);

It have 5 variables and have some lower and upper bounds. i perform genetic algorithm and got optimum values as answers. now whatever value I got for these 5 variables. I want to use them in another equation which is

s = - 463.21-3.05x(1) + 5.21x(2)+0.54x(3) +0.11x(4)-6541.17x(5) + 41.67x(6).

It have 6 variables ( variable x(1) to x(5) are same ) and x(6) is unknown and have some lower and upper bounds . here, i want the output from my previous equation(i.e. equation1 ) as an input of x(1) to x(5) variable in 2nd equation and x(6) will be treated as unknown variables having some bounds ofc.

Inshort, i want to use the answer of 5 variable from 1st equation in 2nd equation.

ps:- Since whenever I run a genetic algorithm, the output of first equation is going to change, i want your help in the code in which the output of x(1) to x(5) variable from 1st equation will automatically get assigned in my 2nd equation. now, i hope you got it. thank you for taking interest in question.

this general coding for 1st equation

function t=strength(x)

t = 122.14-0.84.*x(1)-0.42.*x(2)+0.34.*x(3)-0.09.*x(4)+363.77.*x(5);

end

function [state,options,optchanged] = stop40_80(options,state,flag)

optchanged = false;

switch flag

case 'init'

case 'iter'

% Find the best objective function, and stop if it is right range

mask = state.Score >= 40 & state.Score <= 80;

if any(mask)

state.StopFlag = 'range statisfied';

state.Score = state.Score(mask);

state.Population = state.Population(mask,:);

state.Best(end) = state.Score(end);

end

case 'done'

end

end

this is my calling function of GA for 1st equation:-

options = optimoptions('ga','MutationFcn', @mutationadaptfeasible, 'CrossoverFcn', {@crossoverintermediate, 0.7}, 'SelectionFcn', {@selectiontournament,3}, 'PlotFcn', {@gaplotselection, @gaplotbestf, @gaplotbestindiv, @gaplotrange}, 'FitnessLimit', -inf, 'MaxGenerations', 1e5);

A = []; b = []; Aeq = []; beq = []; lb = [29 160 30 0 0.040]; ub = [50 180 60 20 0.069]; nonlcon = [];

obj = @(x) (strength(x)-80).^2;

bestx = ga(obj, 5, A, b, Aeq, beq, lb, ub, nonlcon, options);

display(bestx)

display(strength(bestx))

All I want is use this data (output) from 1st equation( Basically output of variables x(1) to x(5)) for finding the answer of 2nd equation whose fitness function is

s= -463.21-3.05.*x(1)+5.21.*x(2)+0.54.*x(3)+0.11.*x(4)-6541.17.*x(5)+41.67.*x(6);

here variable x(1) to x(5) variable are same from 1st equation

lower and upper bound of x(6)-

lb=1.50

ub=9.0

Answer 1

[Ayush Aniket]

Hi Harsh,

I understand that you want to pass the result of one Genetic Algorithm execution to another. Here's how you can do it:

Create a function that will calculate "s".

function s = second_equation(x6, bestx)
    x = [bestx, x6]; % Combine the known best values with the new variable x6
    s = -463.21 - 3.05.*x(1) + 5.21.*x(2) + 0.54.*x(3) + 0.11.*x(4) - 6541.17.*x(5) + 41.67.*x(6);
end

This function should be called after your first execution of Genetic Algorithm is over. The results of that execution should be passed as the argument "bestx".

Now "s" contains only one unknown variable i.e. "x(6)". You can use this result for the next execution of the Genetic Algorithm in a way similar to the first execution.