灰狼算法matlab代码

以下是灰狼优化算法的 MATLAB 代码：

function [best_score, best_pos] = grey_wolf_optimizer(fitness_func, dim, lb, ub, max_iter, pack_size)
% Grey Wolf Optimizer
% References: 
% S. Mirjalili, S. M. Mirjalili, A. Hatamlou, "Multi-objective grey wolf optimizer: A novel algorithm for multi-criterion optimization," Expert Systems with Applications, vol. 47, pp. 106-119, 2016.
% S. Mirjalili, "Grey Wolf Optimizer," Advances in Engineering Software, vol. 69, pp. 46-61, 2014.
% 
% Parameters:
% fitness_func - the fitness function to be optimized
% dim - number of dimensions of the search space
% lb - lower bound of the search space
% ub - upper bound of the search space
% max_iter - maximum number of iterations
% pack_size - number of wolves in a pack
%
% Output:
% best_score - the best fitness score obtained by the algorithm
% best_pos - the best solution obtained by the algorithm

% Set the parameters of the algorithm
alpha = 0.1; % Alpha parameter of the GWO algorithm
beta = 0.2; % Beta parameter of the GWO algorithm
delta = 0.5; % Delta parameter of the GWO algorithm
D = ub - lb; % Search space range

% Initialize the positions of the wolves
positions = lb + rand(pack_size, dim) .* D;
fitness = zeros(pack_size, 1);

% Evaluate the fitness of the initial positions
for i = 1:pack_size
    fitness(i) = fitness_func(positions(i, :));
end

% Initialize the best score and position
[best_score, best_idx] = min(fitness);
best_pos = positions(best_idx, :);

% Main loop of the algorithm
for iter = 1:max_iter
    % Update the positions of the wolves
    for i = 1:pack_size
        % Calculate the distance and direction of the alpha, beta, and delta wolves
        A = alpha * (2 * rand(dim) - 1);
        C = 2 * rand(dim);
        D = abs(C .* best_pos - positions(i, :));
        X1 = best_pos - A .* D;
        
        B = beta * (2 * rand(dim) - 1);
        D = abs(C .* best_pos - positions(i, :));
        X2 = best_pos - B .* D;
        
        E = delta * (2 * rand(dim) - 1);
        D = abs(C .* best_pos - positions(i, :));
        X3 = best_pos - E .* D;
        
        % Calculate the new position of the current wolf
        new_pos = (X1 + X2 + X3) / 3;
        
        % Check if the new position is within the search space
        new_pos = max(new_pos, lb);
        new_pos = min(new_pos, ub);
        
        % Evaluate the fitness of the new position
        new_fitness = fitness_func(new_pos);
        
        % Update the position and fitness of the current wolf
        if new_fitness < fitness(i)
            positions(i, :) = new_pos;
            fitness(i) = new_fitness;
            
            % Update the best score and position if necessary
            if new_fitness < best_score
                best_score = new_fitness;
                best_pos = new_pos;
            end
        end
    end
end
end

在上面的代码中，`fitness_func` 是要优化的目标函数，`dim` 是搜索空间的维度，`lb` 和 `ub` 是搜索空间的下界和上界，`max_iter` 是算法的最大迭代次数，`pack_size` 是狼群中的狼的数量。函数的输出是最优解的适应度值 `best_score` 和最优解的位置 `best_pos`。