Logstic混沌映射初始化種群
Step 1:
隨機生成一個\(d\)維向量\({X_0}\),向量的每個分量在0-1之間。
Step 2:
利用Logistic映射生成N個向量。Logistic映射如下:
\[X_{i+1}=\mu{X_{i}.*(1-X_{i})} \]
Step 3:
將\(X\)的每個分量載波到變量的取值區間上
參數設置
Lb = -100; % 搜索空間下界
Ub = 100; % 搜索空間上界
N_iter = 1000; % 最大迭代次數
n_pop = 30; % 種群個數
d = 2; % 種群維度
利用混沌映射初始化種群
Z = zeros(n_pop, d);
% 隨機生成一個d維向量
Z(1, :) = rand(1, d);
% 利用logistic生成n_pop個向量
for i=2:n_pop
Z(i,:) = 4.0*Z(i-1,:).*(1-Z(i-1,:));
end
% 將z的各個分量載波到對應變量的取值區間
pop = zeros(n_pop, d);
for i=1:n_pop
pop(i,:) = Lb + (Ub - Lb)*Z(i,:);
end
figure
scatter(pop(:,1), pop(:,2), 'r*');
box on
Logistic map的第二種寫法:
particlePosition(1,:) = random('Uniform',-100, 100, 1, 2);
particlePosition(1,:) = (particlePosition(1,:) + 100)/200; %這是歸一化處理
for i=1:49
particlePosition(i+1,:) = 4*particlePosition(i,:).*(1 - particlePosition(i,:));
end
particlePosition = particlePosition.*200 - 100;
figure
scatter(particlePosition(:,1), particlePosition(:,2));
box on
隨機初始化種群
particlePosition = random('Uniform',-100, 100, 50, 2);
figure
scatter(particlePosition(:,1), particlePosition(:,2), 'r*');
box on