PSO粒子群算法解决旅行商问题的MATLAB源码

%粒子群算法求解旅行商问题
% By lReij

close all;
clear all;

PopSize=500;%种群大小
CityNum = 14;%城市数

OldBestFitness=0;%旧的最优适应度值

Iteration=0;%迭代次数
MaxIteration =2000;%最大迭代次数
IsStop=0;%程序停止标志
Num=0;%取得相同适应度值的迭代次数

c1=0.5;%认知系数
c2=0.7;%社会学习系数
w=0.96-Iteration/MaxIteration;%惯性系数,随迭代次数增加而递减

%节点坐标
node=[16.47 96.10; 16.47 94.44; 20.09 92.54; 22.39 93.37; 25.23 97.24;...
22.00 96.05; 20.47 97.02; 17.20 96.29; 16.30 97.38; 14.05 98.12;...
16.53 97.38; 21.52 95.59; 19.41 97.13; 20.09 94.55];

%初始化各粒子，即产生路径种群
Group=ones(CityNum,PopSize);
for i=1:PopSize
Group(:,i)=randperm(CityNum)';
end
Group=Arrange(Group);

%初始化粒子速度（即交换序）
Velocity =zeros(CityNum,PopSize);
for i=1:PopSize
Velocity(:,i)=round(rand(1,CityNum)'*CityNum); %round取整
end

%计算每个城市之间的距离
CityBetweenDistance=zeros(CityNum,CityNum);
for i=1:CityNum
for j=1:CityNum
CityBetweenDistance(i,j)=sqrt((node(i,1)-node(j,1))^2+(node(i,2)-node(j,2))^2);
end
end

%计算每条路径的距离
for i=1:PopSize
EachPathDis(i) = PathDistance(Group(:,i)',CityBetweenDistance);
end

IndivdualBest=Group;%记录各粒子的个体极值点位置,即个体找到的最短路径
IndivdualBestFitness=EachPathDis;%记录最佳适应度值,即个体找到的最短路径的长度
[GlobalBestFitness,index]=min(EachPathDis);%找出全局最优值和相应序号

%初始随机解
figure;
subplot(2,2,1);
PathPlot(node,CityNum,index,IndivdualBest);
title('随机解');

%寻优
while(IsStop == 0) & (Iteration < MaxIteration)
%迭代次数递增
Iteration = Iteration +1;

%更新全局极值点位置,这里指路径
for i=1:PopSize
GlobalBest(:,i) = Group(:,index);
end

%求pij-xij ,pgj-xij交换序，并以概率c1，c2的保留交换序
pij_xij=GenerateChangeNums(Group,IndivdualBest);
pij_xij=HoldByOdds(pij_xij,c1);
pgj_xij=GenerateChangeNums(Group,GlobalBest);
pgj_xij=HoldByOdds(pgj_xij,c2);

%以概率w保留上一代交换序
Velocity=HoldByOdds(Velocity,w);

Group = PathExchange(Group,Velocity); %根据交换序进行路径交换
Group = PathExchange(Group,pij_xij);
Group = PathExchange(Group,pgj_xij);
for i = 1:PopSize % 更新各路径总距离
EachPathDis(i) = PathDistance(Group(:,i)',CityBetweenDistance);
end

IsChange = EachPathDis<IndivdualBestFitness;%更新后的距离优于更新前的，记录序号
IndivdualBest(:, find(IsChange)) = Group(:, find(IsChange)
);%更新个体最佳路径
IndivdualBestFitness = IndivdualBestFitness.*( ~IsChange) + EachPathDis.*IsChange;%更新个体最佳路径距离
[GlobalBestFitness, index] = min(Eac

hPathDis);%更新全局最佳路径,记录相应的序号

if GlobalBestFitness==OldBestFitness %比较更新前和更新后的适应度值;
Num=Num+1; %相等时记录加一;
else
OldBestFitness=GlobalBestFitness;%不相等时更新适应度值，并记录清零;
Num=0;
end
if Num >= 20 %多次迭代的适应度值相近时程序停止
IsStop=1;
end

BestFitness(Iteration) =GlobalBestFitness;%每一代的最优适应度

end

%最优解
subplot(2,2,2);
PathPlot(node,CityNum,index,IndivdualBest);
title('优化解');
%进化曲线
subplot(2,2,3);
plot((1:Iteration),BestFitness(1:Iteration));
grid on;
title('进化曲线');
%最小路径值
GlobalBestFitness


function Group=Arrange(Group)

[x y]=size(Group);
[NO1,index]=min(Group',[],2); %找到最小值1
for i=1:y
pop=Group(:,i);
temp1=pop([1: index(i)-1]);
temp2=pop([index(i): x]);
Group(:,i)=[temp2' temp1']';
end

function ChangeNums=GenerateChangeNums(Group,BestVar);

[x y]=size(Group);
ChangeNums=zeros(x,y);
for i=1:y
pop=BestVar(:,i);%从BestVar取出一个顺序
pop1=Group(:,i);%从粒子群中取出对应的顺序
for j=1:x %从BestVar的顺序中取出一个序号
NoFromBestVar=pop(j);
for k=1:x %从对应的粒子顺序中取出一个序号
NoFromGroup=pop1(k);
if (NoFromBestVar==NoFromGroup) && (j~=k) %两序号同且不在同一位置
ChangeNums(j,i)=k; %交换子
pop1(k)=pop1(j);
pop1(j)=NoFromGroup;
end
end
end
end

function Hold=HoldByOdds(Hold,Odds)
[x,y]=size(Hold);
for i=1:x
for j=1:y
if rand>Odds
Hold(i,j)=0;
end
end
end

function SumDistance=PathDistance(path,CityBetweenDistance)

L=length(path); %path为一个循环的节点顺序
SumDistance=0;
for i=1:L-1
SumDistance=SumDistance+CityBetweenDistance(path(i),path(i+1));
end
SumDistance=SumDistance+CityBetweenDistance(path(1),path(L)); %加上首尾节点的距离

function Group=PathExchange(Group,Index)

[x y]=size(Group);
for i=1:y
a=Index(:,i); %取出其中一组交换序
pop=Group(:,i); %取出对应的粒子
for j=1:x %取出其中一个交换算子作交换
if a(j)~=0
pop1=pop(j);
pop(j)=pop(a(j));
pop(a(j))=pop1;
end
end
Group(:,i)=pop;
end



function PathPlot(node,CityNum,index,EachBest);


for i=1:CityNum
NowBest(i,:)=nod
e((EachBest(i,index)),:);
end
NowBest(CityNum+1,:)=NowBest(1,:);
plot(node(:,1),node(:,2),'*');
line(NowBest(:,1),NowBest(:,2));
grid on;