3维粒子群算法matlab

以下是一个简单的3维粒子群算法的MATLAB代码示例：

% 3D Particle Swarm Optimization

% Define the objective function
fun = @(x) 100*(x(2)-x(1)^2)^2 + (1-x(1))^2 + 100*(x(3)-x(2)^2)^2 + (1-x(2))^2;

% Define the lower and upper bounds of the variables
lb = [-5,-5,-5];
ub = [5,5,5];

% Define the number of particles and iterations
nParticles = 50;
nIterations = 100;

% Initialize the particle positions and velocities
particlePositions = bsxfun(@plus,lb,bsxfun(@times,rand(nParticles,3),ub-lb));
particleVelocities = zeros(nParticles,3);

% Initialize the personal best positions and values
personalBestPositions = particlePositions;
personalBestValues = inf(nParticles,1);

% Initialize the global best position and value
globalBestPosition = zeros(1,3);
globalBestValue = inf;

% Run the main loop
for iIteration = 1:nIterations
    % Evaluate the objective function for each particle
    particleValues = arrayfun(fun,particlePositions);
    
    % Update the personal best positions and values
    needToUpdatePersonalBest = particleValues < personalBestValues;
    personalBestPositions(needToUpdatePersonalBest,:) = particlePositions(needToUpdatePersonalBest,:);
    personalBestValues(needToUpdatePersonalBest) = particleValues(needToUpdatePersonalBest);
    
    % Update the global best position and value
    [iterationBestValue,iterationBestIndex] = min(personalBestValues);
    if iterationBestValue < globalBestValue
        globalBestPosition = personalBestPositions(iterationBestIndex,:);
        globalBestValue = iterationBestValue;
    end
    
    % Update the particle velocities and positions
    inertiaWeight = 0.729; % Inertia weight
    cognitiveWeight = 1.49445; % Cognitive weight
    socialWeight = 1.49445; % Social weight
    r1 = rand(nParticles,3); % Random numbers
    r2 = rand(nParticles,3); % Random numbers
    cognitiveComponent = bsxfun(@times,cognitiveWeight*r1,bsxfun(@minus,personalBestPositions,particlePositions));
    socialComponent = bsxfun(@times,socialWeight*r2,bsxfun(@minus,globalBestPosition,particlePositions));
    particleVelocities = bsxfun(@plus,inertiaWeight*particleVelocities,cognitiveComponent+socialComponent);
    particlePositions = particlePositions + particleVelocities;
    
    % Enforce the lower and upper bounds of the variables
    particlePositions = max(particlePositions,lb);
    particlePositions = min(particlePositions,ub);
end

% Display the global best position and value
disp(['Global best position: ',num2str(globalBestPosition)]);
disp(['Global best value: ',num2str(globalBestValue)]);

这个代码实现了一个简单的3D粒子群算法来最小化一个测试函数。你可以将函数`fun`替换成你自己的目标函数，并调整粒子数量、迭代次数、变量的上下限等参数来适应你的问题。