PSO和改进的PSO代码

%%%%%%%%%%%%%%%%Particle Swarm Optimization%%%%%%%%%%%%%%%%%%%%% %%%%自适应PSO算法，测试基准函数为Sphere clear all; clc; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%------初始化-------%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Dmin=0.4; Dmax=0.94; wmax=10; j=5; c1=2; c2=2; %%%加速常数 w=0.8; MaxDT=500; %%%迭代次数 D=10; %%%数据维度 N=40; %%%群体个数 rang_l=-100; rang_r=100; %%%%初始化种群和速度 x=rang_l + (rang_r - rang_l) * rand(N,D); v=rang_l*0.2 +(rang_r - rang_l)*0.2 * rand(N,D); %%%根据经验一般取空间范围的10%~20% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%------计算适应度-------%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% for i = 1:N %用测试函数输出每个数据的值 p(i) = fitness(x(i,:)); end y=x; pg=x(1,:); %%%初始化全局最优解 for i=2:N if p(i)<fitness(pg) pg=x(i,:); end end %%%%%%%%%%%%%%%%%%%%%%%%%%%%%------PSO算法开始-------%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% for t=1:MaxDT if(rand()<0.1) for i=1:N v(i,:)=w*v(i,:)+c1*rand*(y(i,:)-x(i,:))+c2*rand*(pg-x(i,:)); x(i,:)=x(i,:)+v(i,:); if fitness(x(i,:))<p(i) p(i)=fitness(x(i,:)); y(i,:)=x(i,:); end if p(i)<fitness(pg) pg=y(i,:); end end else for i=1:N D=((max(x(i,:)))-(min(x(i,:))))/abs((max(x(i,:)))+(min(x(i,:))));%%%分布向量惯性权重 if (D>Dmax) D=Dmax; else if (D<Dmin) D=Dmin; end end v(i,:)=D*(w*v(i,:)+c1*rand*(y(i,:)-x(i,:))+c2*rand*(pg-x(i,:))); x(i,:)=x(i,:)+v(i,:); if fitness(x(i,:))<p(i) p(i)=fitness(x(i,:)); y(i,:)=x(i,:); end if p(i)<fitness(pg) pg=y(i,:); end end end Gbest(t)=fitness(pg); %if(t>2) %if (Gbest(t)-Gbest(t-1)<j) %if(w<1) % w=w+1; %else % w=w+0; %end %end %end end disp('函数的全局最优位置为：') Solution=pg disp('最后得到的优化极值为：') Result=fitness(pg) plot(1:MaxDT,Gbest,'m') hold on