基于遗传算法优化的神经网络预测代码

tic clear; clc; load data.mat inputnum=2; hiddennum=5; outputnum=1; % input_train=input(1:1500,:)'; % input_test=input(1501:2000,:)'; % output_train=output(1:1500)'; % output_test=output(1501:2000)'; input_train=input(1:600,:)'; input_test=input(601:623,:)'; output_train=output(1:600)'; output_test=output(601:623)'; [inputn,inputps]=mapminmax(input_train); [outputn,outputps]=mapminmax(output_train); net=newff(inputn,outputn,hiddennum,{'tansig','purelin'},'trainlm'); %%{'tansig','purelin'}为默认的激活函数（没记错的话，有兴趣的话可以试着进行调整，trainlm为默认的训练算法，Levenberg-Marquart算法) %% 遗传算法参数初始化 maxgen=10; %进化代数，即迭代次数 sizepop=300; %种群规模 pcross=0.3; %交叉概率选择，0和1之间 pmutation=0.1; %变异概率选择，0和1之间 numsum=inputnum*hiddennum+hiddennum+hiddennum*outputnum+outputnum; lenchrom=ones(1,numsum); bound=[-3*ones(numsum,1) 3*ones(numsum,1)]; %数据范围 individuals=struct('fitness',zeros(1,sizepop), 'chrom',[]); %将种群信息定义为一个结构体 avgfitness=[]; %每一代种群的平均适应度 bestfitness=[]; %每一代种群的最佳适应度 bestchrom=[]; %适应度最好的染色体 for i=1:sizepop %随机产生一个种群 individuals.chrom(i,:)=Code(lenchrom,bound); %编码 x=individuals.chrom(i,:); %计算适应度 individuals.fitness(i)=fun(x,inputnum,hiddennum,outputnum,net,inputn,outputn); %染色体的适应度 end [bestfitness bestindex]=min(individuals.fitness); bestchrom=individuals.chrom(bestindex,:); %最好的染色体 avgfitness=sum(individuals.fitness)/sizepop; %染色体的平均适应度 trace=[avgfitness bestfitness]; % 记录每一代进化中最好的适应度和平均适应度 for num=1:maxgen % 选择 individuals=select(individuals,sizepop); avgfitness=sum(individuals.fitness)/sizepop; %交叉 individuals.chrom=Cross(pcross,lenchrom,individuals,sizepop,bound); % 变异 individuals.chrom=Mutation(pmutation,lenchrom,individuals,sizepop,num,maxgen,bound); % 计算适应度 for j=1:sizepop x=individuals.chrom(j,:); %个体 individuals.fitness(j)=fun(x,inputnum,hiddennum,outputnum,net,inputn,outputn); end %找到最小和最大适应度的染色体及它们在种群中的位置 [newbestfitness,newbestindex]=min(individuals.fitness); [worestfitness,worestindex]=max(individuals.fitness); % 代替上一次进化中最好的染色体 if bestfitness>newbestfitness bestfitness=newbestfitness; bestchrom=individuals.chrom(newbestindex,:); end individuals.chrom(worestindex,:)=bestchrom; individuals.fitness(worestindex)=bestfitness; avgfitness=sum(individuals.fitness)/sizepop; trace=[trace;avgfitness bestfitness]; %记录每一代进化中最好的适应度和平均适应度 end figure(1) [r,c]=size(trace); plot([1:r]',trace(:,2),'b--'); title(['适应度曲线 ' '终止代数＝' num2str(maxgen)]); xlabel('进化代数');ylabel('适应度'); legend('平均适应度','最佳适应度'); disp('适应度 变量'); %% 把最优初始阀值权值赋予网络预测 % %用遗传算法优化的BP网络进行值预测 x=bestchrom; w1=x(1:inputnum*hiddennum); B1=x(inputnum*hiddennum+1:inputnum*hiddennum+hiddennum); w2=x(inputnum*hiddennum+hiddennum+1:inputnum*hiddennum+hiddennum+hiddennum*outputnum); B2=x(inputnum*hiddennum+hiddennum+hiddennum*outputnum+1:inputnum*hiddennum+hiddennum+hiddennum*outputnum+outputnum); net.iw{1,1}=reshape(w1,hiddennum,inputnum); net.lw{2,1}=reshape(w2,outputnum,hiddennum); net.b{1}=reshape(B1,hiddennum,1); net.b{2}=reshape(B2,outputnum,1); %% BP网络训练 %网络参数 net.trainParam.epochs=100; net.trainParam.lr=0.1; net.trainParam.goal=0.00001; net.divideParam.trainRatio = 75/100; %默认训练集占比 net.divideParam.valRatio = 15/100; %默认验证集占比 net.divideParam.testRatio = 15/100; %默认测试集占比 %网络训练 [net,per2]=train(net,inputn,outputn); %% BP网络预测 %数据归一化 inputn_test=mapminmax('apply',input_test,inputps); an=sim(net,inputn_test); test_simu=mapminmax('reverse',an,outputps); error=test_simu-output_test; figure(2) plot(test_simu,':og','LineWidth',1.5) hold on plot(output_test,'-*','LineWidth',1.5); legend('预测输出','期望输出') grid on set(gca,'linewidth',1.0); xlabel('X 样本','FontSize',15); %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% ylabel('Y 输出','FontSize',15); %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% set(gcf,'color','w') %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% title('GA-BP Network','Color','k','FontSize',15); toc