写一段用matlab遗传算法解决实际问题的示例

当然！下面是一个使用遗传算法解决旅行商问题（Traveling Salesman Problem，TSP）的示例代码：

% 遗传算法解决旅行商问题
function [bestPath, bestDistance] = tspGeneticAlgorithm(cityLocations, popSize, numGenerations, mutationRate)
    % 城市数量
    numCities = size(cityLocations, 1);
    
    % 初始化种群
    population = initializePopulation(popSize, numCities);
    
    for generation = 1:numGenerations
        % 计算适应度
        fitness = calculateFitness(population, cityLocations);
        
        % 选择父代
        parents = selection(population, fitness);
        
        % 生成子代
        offspring = crossover(parents, popSize);
        
        % 变异
        offspring = mutation(offspring, mutationRate);
        
        % 合并父代和子代
        population = [parents; offspring];
    end
    
    % 计算最佳路径和最佳距离
    [bestDistance, index] = min(fitness);
    bestPath = population(index, :);
end

% 初始化种群
function population = initializePopulation(popSize, numCities)
    population = zeros(popSize, numCities);
    
    for i = 1:popSize
        population(i, :) = randperm(numCities);
    end
end

% 计算适应度（路径长度）
function fitness = calculateFitness(population, cityLocations)
    popSize = size(population, 1);
    fitness = zeros(popSize, 1);
    
    for i = 1:popSize
        path = population(i, :);
        distance = 0;
        
        for j = 1:length(path)-1
            city1 = path(j);
            city2 = path(j+1);
            distance = distance + calculateDistance(cityLocations(city1, :), cityLocations(city2, :));
        end
        
        % 回到起始城市
        distance = distance + calculateDistance(cityLocations(path(end), :), cityLocations(path(1), :));
        
        fitness(i) = distance;
    end
end

% 选择父代（锦标赛选择）
function parents = selection(population, fitness)
    popSize = size(population, 1);
    tournamentSize = round(popSize / 10); % 锦标赛规模，可根据问题调整
    parents = zeros(popSize/2, size(population, 2));
    
    for i = 1:popSize/2
        % 随机选择锦标赛的个体
        tournamentIndices = randperm(popSize, tournamentSize);
        
        % 选择最佳个体作为父代
        [~, index] = min(fitness(tournamentIndices));
        
        parents(i, :) = population(tournamentIndices(index), :);
    end
end

% 生成子代（顺序交叉）
function offspring = crossover(parents, popSize)
    numCities = size(parents, 2);
    offspring = zeros(popSize/2, numCities);
    
    for i = 1:popSize/2
        parent1 = parents(i, :);
        parent2 = parents(mod(i, popSize/2)+1, :);
        
        % 随机选择交叉点
        crossoverPoint = randi([2, numCities-1]);
        
        % 顺序交叉
        offspring(i, :) = [parent1(1:crossoverPoint), parent2(crossoverPoint+1:end)];
    end
end

% 变异（交换两个位置）
function offspring = mutation(offspring, mutationRate)
    numCities = size(offspring, 2);
    
    for i = 1:size(offspring, 1)
        if rand < mutationRate
            % 随机选择两个位置进行交换
            swapIndices = randperm(numCities, 2);
            
            % 交换位置
            offspring(i, swapIndices) = offspring(i, swapIndices(end:-1:1));
        end
    end
end

% 计算两城市之间的距离（欧氏距离）
function distance = calculateDistance(city1, city2)
    distance = sqrt(sum((city1 - city2).^2));
end

请注意，这只是一个简化的遗传算法实现，用于解决旅行商问题。在实际应用中，你可能需要根据具体问题对初始化种群、计算适应度、选择父代、生成子代和变异等操作进行相应的调整和优化。希望能对你有所帮助！