simulated annealing simple code to minimize fitness function of 9 variables in matlab

main.m
% Define the fitness function
fitnessFunction = @(x) sum(x.^2); % Example fitness function (sum of squares)

% Set the number of variables (9 in this case)
numVariables = 9;

% Set the bounds for each variable
lowerBounds = zeros(1, numVariables); % Lower bounds
upperBounds = ones(1, numVariables); % Upper bounds

% Set the initial solution
initialSolution = rand(1, numVariables); % Random initial solution

% Set the initial temperature
initialTemperature = 100;

% Set the cooling schedule
coolingSchedule = @(temperature) 0.99 * temperature; % Example cooling schedule (multiply by 0.99)

% Set the number of iterations
numIterations = 1000;

% Initialize the current solution and fitness
currentSolution = initialSolution;
currentFitness = fitnessFunction(currentSolution);

% Start the simulated annealing loop
for i = 1:numIterations
    % Generate a random neighbor solution
    neighborSolution = currentSolution + (upperBounds - lowerBounds) .* (rand(1, numVariables) - 0.5);
    
    % Clip the neighbor solution within the bounds
    neighborSolution = max(lowerBounds, min(upperBounds, neighborSolution));
    
    % Calculate the neighbor fitness
    neighborFitness = fitnessFunction(neighborSolution);
    
    % Calculate the delta fitness
    deltaFitness = neighborFitness - currentFitness;
    
    % Accept the neighbor solution if it improves the fitness or according to the acceptance probability
    if deltaFitness < 0 || rand() < exp(-deltaFitness / initialTemperature)
        currentSolution = neighborSolution;
        currentFitness = neighborFitness;
    end
    
    % Update the temperature
    initialTemperature = coolingSchedule(initialTemperature);
end

% Print the final solution and fitness
finalSolution = currentSolution;
finalFitness = currentFitness;
disp("Final Solution:");
disp(finalSolution);
disp("Final Fitness:");
disp(finalFitness);
1886 chars
58 lines