matlab-基于SA模拟退火算法的TSP问题优化仿真,一个起点,多辆车,各自搜索最优路径-源码

clc; clear; close all; warning off; addpath(genpath(pwd)); T0 = 500 ; % initial temperature r = 0.997 ; % temperature damping rate Ts = 1 ; % stop temperature iter = 300; model = initModel(); set(gcf,'unit','normalized','position',[0,0.35,1,0.7]); flag = 0; % initialization while(1) route = randomSol(model); if(isFeasible(route,model)) break; end end cost = calculateCost(route,model); T = T0; cnt = 1; minCost = cost; minRoute = route; maxIterate = 2100; costArray = zeros(maxIterate,1); % SA while(T > Ts) for k = 1:iter mode = randi([1 3]); newRoute = createNeibor(route,model,mode); newCost = calculateCost(newRoute,model); delta = newCost - cost; if(delta < 0) cost = newCost; route = newRoute; else p=exp(-delta/T); if rand() <= p cost = newCost; route = newRoute; end end end costArray(cnt) = cost; if cost<minCost minCost = cost; minRoute = route; flag = 1; end T = T*r; % annealing disp(['Iteration ' num2str(cnt) ': BestCost = ' num2str(minCost) ': CurrentCost = ' num2str(cost) ' T = ' num2str(T)]); cnt = cnt+1; figure(1); if(flag == 1) plotSolution(minRoute,model); flag = 0; end % figure(2); subplot(1,2,2) plot(costArray); pause(0.0001); end