MATLAB 变色龙优化算法代码

MATLAB 变色龙优化算法代码如下：

function [best_position, best_fitness] = chameleon_optimization(fitness_function, dim, lb, ub, num_iterations, num_population)
% fitness_function: 适应度函数
% dim: 变量维度
% lb: 变量下界
% ub: 变量上界
% num_iterations: 迭代次数
% num_population: 种群数量

% 初始化种群
population = initialize_population(num_population, dim, lb, ub);

% 初始化最优解和最优适应度
best_position = population(1,:);
best_fitness = fitness_function(best_position);

% 开始迭代
for i = 1:num_iterations
    % 计算每个个体的适应度
    fitness = zeros(num_population, 1);
    for j = 1:num_population
        fitness(j) = fitness_function(population(j,:));
        % 更新最优解
        if fitness(j) < best_fitness
            best_fitness = fitness(j);
            best_position = population(j,:);
        end
    end
    
    % 变色龙优化算法
    for j = 1:num_population
        % 随机选择三个不同的个体
        r1 = randi([1,num_population]);
        r2 = randi([1,num_population]);
        while r2 == r1
            r2 = randi([1,num_population]);
        end
        r3 = randi([1,num_population]);
        while r3 == r1 || r3 == r2
            r3 = randi([1,num_population]);
        end
        
        % 计算变色龙方向
        direction = population(r1,:) + rand(1,dim).*(population(r2,:)-population(r3,:));
        % 随机选择另一个个体
        r4 = randi([1,num_population]);
        while r4 == j
            r4 = randi([1,num_population]);
        end
        
        % 更新个体位置
        if fitness(j) > fitness(r4)
            population(j,:) = population(j,:) + rand(1,dim).*(best_position - population(j,:)) + rand(1,dim).*(direction - population(j,:));
        else
            population(j,:) = population(j,:) + rand(1,dim).*(best_position - population(j,:)) - rand(1,dim).*(direction - population(j,:));
        end
        
        % 限制个体位置在边界内
        population(j,:) = max(population(j,:), lb);
        population(j,:) = min(population(j,:), ub);
    end
end

end

function population = initialize_population(num_population, dim, lb, ub)
% 初始化种群
population = zeros(num_population, dim);
for i = 1:num_population
    population(i,:) = lb + rand(1,dim).*(ub-lb);
end
end

使用方法：

1. 定义适应度函数 `fitness_function`，例如：

function y = sphere(x)
y = sum(x.^2);
end

2. 调用 `chameleon_optimization` 函数进行优化，例如：

best_position = chameleon_optimization(@sphere, 10, -100, 100, 1000, 50);

其中，`10` 为变量维度，`-100` 和 `100` 分别为变量的下界和上界，`1000` 为迭代次数，`50` 为种群数量。函数的返回值 `best_position` 为最优解。