车间调度基于matlab候鸟算法MBO求解柔性作业车间调度问题【含Matlab源码 2354期】.zip

clc;clear %% 下载数据 % 加工数据包括加工时间，加工机器，机器数，各机器权重，工件数，各工件对应的工序数 load data operation_time operation_machine num_machine machine_weight num_job num_op %% 基本参数 MAXGEN=200; % 最大迭代次数 sizepop=201; % 种群规模 e=0.5; % 目标值权重 N_size=30; % 邻域解数量 S_size=15; % 共享解数量 G=5; % 巡回次数 G1=20; % 竞争机制1参数 G2=10; % 竞争机制2参数 trace=zeros(2,MAXGEN); chrom_best=[]; %% ===========================种群初始化============================ total_op_num=sum(num_op); chroms=initialization(num_op,num_job,total_op_num,sizepop,operation_machine,operation_time); [Z,~,~,~,~]=fitness(chroms,num_machine,e,num_job,num_op); % 将最好的解划分为领飞鸟 [Z_leader,ind]=min(Z); leader=chroms(ind,:); % 从chroms中移出领飞鸟，然后划分左右两个跟飞鸟种群 chroms(ind,:)=[]; Z(ind)=[]; sp=(sizepop-1)/2; lefts=chroms(1:sp,:); Z_left=Z(1:sp); rights=chroms(sp+1:end,:); Z_right=Z(sp+1:end); %% ============================迭代过程============================= for gen=1:MAXGEN fprintf('当前迭代次数：'),disp(gen) %% 巡回阶段 for i=1:G %% 领飞鸟进化 [leader,Z_leader,share,Z_share]=bird_evolution(leader,Z_leader,[],[],... N_size,S_size,total_op_num,num_machine,e,num_job,num_op); %% 跟飞鸟进化 % 初始化左右队列的共享解集 share_left=share; Z_share_left=Z_share; share_right=share; Z_share_right=Z_share; for j=1:sp % 左队列 [lefts(j,:),Z_left(j),share_left,Z_share_left]=bird_evolution(lefts(j,:),Z_left(j),share_left,Z_share_left,... N_size-S_size,S_size,total_op_num,num_machine,e,num_job,num_op); % 右队列 [rights(j,:),Z_right(j),share_right,Z_share_right]=bird_evolution(rights(j,:),Z_right(j),share_right,Z_share_right,... N_size-S_size,S_size,total_op_num,num_machine,e,num_job,num_op); end %% 竞争机制2：队间交叉 % 随机产生G2对位置相同的个体 ind=randperm(sp,G2); [rights(ind,:),Z_right(ind),lefts(ind,:),Z_left(ind)]= crossover(lefts(ind,:),rights(ind,:),... Z_left(ind),Z_right(ind),total_op_num,num_machine,e,num_job,num_op); end %% 竞争机制1：队内竞争 ind=randperm(sp,G1); % 左队列 [~,ind1]=sort(Z_left(ind)); lefts(ind,:)=lefts(ind(ind1),:); Z_left(ind)=Z_left(ind(ind1)); % 右队列 [~,ind2]=sort(Z_right(ind)); rights(ind,:)=rights(ind(ind2),:); Z_right(ind)=Z_right(ind(ind2)); %% 领飞鸟替换 if rand<0.5 % 选择左队列首只跟飞鸟 [leader,Z_leader,lefts,Z_left]=update_leader(leader,Z_leader,lefts,Z_left); else % 选择右队列首只跟飞鸟 [leader,Z_leader,rights,Z_right]=update_leader(leader,Z_leader,rights,Z_right); end %% 记录相关数据 % 记录每代的最优适应度与平均适应度 Z=[Z_leader,Z_left,Z_right]; [val,ind]=min(Z); trace(1,gen)=val; trace(2,gen)=mean(Z); % 更新全局最优适应度 if gen==1 || MinVal>trace(1,gen) MinVal=trace(1,gen); end end %% ============================输出结果============================= %% 输出最优适应度 fprintf('最优适应度：'),disp(MinVal) %% 绘制最优适应度与平均适应度的迭代曲线图 figure(1) plot(trace(1,:)); hold on; plot(trace(2,:),'-.');grid; legend('解的变化','种群均值的变化'); %% 绘制全局最优解的甘特图 [Z,~,~,machine_time,Pvals]=fitness(leader,num_machine,e,num_job,num_op); Pval1=Pvals{1,1}; figure(2); for i=1:total_op_num mText=leader(total_op_num+i); b=leader(i); x1=Pval1(1,i); x2=Pval1(2,i); y1=mText-0.2; y2=mText; hold on; fill([x1,x2,x2,x1],[y1,y1,y2,y2],[1-1/b,1/b,b/num_job]); text((x1+x2)/2,mText-0.1,num2str(b)); end