写一个灰狼算法的matlab程序

灰狼算法是一种优化算法，主要用于解决优化问题。以下是一个基本的灰狼算法的MATLAB程序的示例：

%% 灰狼算法
function [x_best, f_best] = Grey_Wolf_Algorithm(f_obj, n_dim, n_wolves, lb, ub, max_iter)
% 初始化种群
alpha = zeros(1, n_dim);
beta = zeros(1, n_dim);
delta = zeros(1, n_dim);
position = zeros(n_wolves, n_dim);
for i = 1 : n_wolves
position(i, :) = lb + (ub - lb) * rand(1, n_dim);
end
% 计算初始适应度
fitness = zeros(n_wolves, 1);
for i = 1 : n_wolves
fitness(i) = f_obj(position(i, :));
end
% 迭代
for iter = 1 : max_iter
a = 2 - 2 * iter / max_iter;
for i = 1 : n_wolves
for j = 1 : n_dim
A = 2 * a * rand - a;
C = 2 * rand;
D = abs(C * delta(j) - position(i, j));
X1 = delta(j) - A * D;
A = 2 * a * rand - a;
C = 2 * rand;
D = abs(C * beta(j) - position(i, j));
X2 = beta(j) - A * D;
A = 2 * a * rand - a;
C = 2 * rand;
D = abs(C * alpha(j) - position(i, j));
X3 = alpha(j) - A * D;
position(i, j) = (X1 + X2 + X3) / 3;
% 边界处理
if position(i, j) < lb(j)
position(i, j) = lb(j);
end
if position(i, j) > ub(j)
position(i, j) = ub(j);
end
end
% 计算适应度
fitness(i) = f_obj(position(i, :));
end
% 更新alpha, beta, delta
[f_min, idx_min] = min(fitness);
[f_max, idx_max] = max(fitness);
if f_min < f_best
x_best = position(idx_min, :);
f_best = f_min;
end
if f_max > f_best
alpha = position(idx_max, :);
end
for i = 1 : n_wolves
if i ~= idx_max
r = norm(position(idx_max, :) - position(i, :)) / norm(position(idx_max, :) - alpha);
A = 2 * a * rand - a;
a1 = 2 * A * r - A;
C = 2 * rand;
b1 = C;
D = abs(C * delta - position(i, :));
X1 = delta - a1 * D;
A = 2 * a * rand - a;
a2 = 2 * A * r - A;
C = 2 * rand;
b2 = C;
D = abs(C * beta - position(i, :));
X2 = beta - a2 * D;
A = 2 * a * rand - a;
a3 = 2 * A * r - A;
C = 2 * rand;
b3 = C;
D = abs(C * alpha - position(i, :));
X3 = alpha - a3 * D;
position(i, :) = (X1 + X2 + X3) / 3;
% 边界处理
for j = 1 : n_dim
if position(i, j) < lb(j)
position(i, j) = lb(j);
end
if position(i, j) > ub(j)
position(i, j) = ub(j);
end
end
% 计算适应度
fitness(i) = f_obj(position(i, :));
end
end
% 更新alpha, beta, delta
[f_min, idx_min] = min(fitness);
if f_min < f_best
x_best = position(idx_min, :);
f_best = f_min;
end
[f_max, idx_max] = max(fitness);
if f_max > f_best
alpha = position(idx_max, :);
delta = beta;
beta = position(idx_max, :);
else
m = randi(n_wolves);
alpha = position(m, :);
m = randi(n_wolves);
beta = position(m, :);
while all(beta == alpha)
m = randi(n_wolves);
beta = position(m, :);
end
m = randi(n_wolves);
delta = position(m, :);
while all(delta == alpha) || all(delta == beta)
m = randi(n_wolves);
delta = position(m, :);
end
end
end
end