遗传算法matlab程序

以下是一个简单的遗传算法MATLAB程序示例：

% 遗传算法参数设置
pop_size = 50; % 种群大小
n_generations = 100; % 进化代数
n_variables = 2; % 变量数量
lb = [-5,-5]; % 变量下限
ub = [5,5]; % 变量上限

% 初始化种群
pop = repmat(lb,pop_size,1) + repmat((ub-lb),pop_size,1).*rand(pop_size,n_variables);

% 进化循环
for i=1:n_generations
    % 计算适应度
    fitness = sum(pop.^2,2);
    % 选择父代
    [~,idx] = sort(fitness);
    parents = pop(idx(1:round(pop_size/2)),:);
    % 交叉
    children = zeros(size(parents));
    for j=1:size(children,1)
        p1 = parents(randi(size(parents,1)),:);
        p2 = parents(randi(size(parents,1)),:);
        children(j,:) = (p1+p2)/2;
    end
    % 变异
    mutation_rate = 0.1;
    mutation_amount = (ub-lb).*mutation_rate;
    mutation = mutation_amount.*randn(size(children));
    children = children + mutation;
    % 合并父代和子代
    pop = [parents;children];
    % 限制变量范围
    pop = max(pop,lb);
    pop = min(pop,ub);
end

% 打印最优解
[~,idx] = min(fitness);
disp(['Optimal solution: x1=',num2str(pop(idx,1)),', x2=',num2str(pop(idx,2))]);

这个程序用于优化一个二元函数，其目标是最小化 x1^2+x2^2。程序中的遗传算法使用了经典的轮盘赌选择、单点交叉和高斯变异操作。