【ELM回归预测】粒子群算法优化极限学习机PSO-ELM回归预测【含Matlab源码 1722期】.zip

clc clear all %% 加载数据 load matlab3.mat AllSamIn = data_train'; AllSamOut =lab_train'; AllTestIn=data_test'; AllTestOut=lab_test'; %% 训练样本归一化 global minAllSamOut; global maxAllSamOut; [AllSamInn,minAllSamIn,maxAllSamIn,AllSamOutn,minAllSamOut,maxAllSamOut] = premnmx(AllSamIn,AllSamOut); TrainSamIn=AllSamInn; TrainSamOut=AllSamOutn; global Ptrain; Ptrain = TrainSamIn; global Ttrain; Ttrain = TrainSamOut; %% 测试样本归一化 global minAllTestOut; global maxAllTestOut; [AllTestInn,minAllTestIn,maxAllTestIn,AllTestOutn,minAllTestOut,maxAllTestOut] = premnmx(AllTestIn,AllTestOut); TestIn=AllTestInn; TestOut=AllTestOutn; global Ptest; Ptest = TestIn; global Ttest; Ttest = TestOut; %% 加载网络的初始变量 global indim; indim=6; global hiddennum; hiddennum=8; global outdim; outdim=1; %% 加载PSO模型的相关参数 vmax=1; minerr=0.001; wmax=0.80; wmin=0.40; global itmax; itmax=100; c1=2.8; c2=1.3; for iter=1:itmax W(iter)=wmax-((wmax-wmin)/itmax)*iter; end a=-1; b=1; m=-1; n=1; global N; N=100; global D; D=(indim+1)*hiddennum+(hiddennum+1)*outdim; rand('state',sum(100*clock)); X=a+(b-a)*rand(N,D,1); V=m+(n-m)*rand(N,D,1); global fvrec; MinFit=[]; BestFit=[]; %% PSO优化ELM模型 过程 1 fitness=fitcal(X,indim,hiddennum,outdim,D,Ptrain,Ttrain,minAllSamOut,maxAllSamOut); fvrec(:,1,1)=fitness(:,1,1); [C,I]=min(fitness(:,1,1)); MinFit=[MinFit C]; BestFit=[BestFit C]; L(:,1,1)=fitness(:,1,1); B(1,1,1)=C; gbest(1,:,1)=X(I,:,1); for p=1:N G(p,:,1)=gbest(1,:,1); end for i=1:N; pbest(i,:,1)=X(i,:,1); end V(:,:,2)=W(1)*V(:,:,1)+c1*rand*(pbest(:,:,1)-X(:,:,1))+c2*rand*(G(:,:,1)-X(:,:,1)); for ni=1:N for di=1:D if V(ni,di,2) > vmax V(ni,di,2) = vmax; elseif V(ni,di,2) < -vmax V(ni,di,2) = -vmax; else V(ni,di,2) = V(ni,di,2); end end end X(:,:,2)=X(:,:,1)+V(:,:,2); for ni=1:N for di=1:D if X(ni,di,2) > 1 X(ni,di,2) = 1; elseif V(ni,di,2) < -1 X(ni,di,2) = -1; else X(ni,di,2) = X(ni,di,2); end end end %% PSO优化ELM模型 过程 2 for j=2:itmax disp('Iteration and Current Best Fitness') disp(j-1) disp(B(1,1,j-1)) fitness=fitcal(X,indim,hiddennum,outdim,D,Ptrain,Ttrain,minAllSamOut,maxAllSamOut); fvrec(:,1,j)=fitness(:,1,j); [C,I]=min(fitness(:,1,j)); MinFit=[MinFit C]; BestFit=[BestFit min(MinFit)]; L(:,1,j)=fitness(:,1,j); B(1,1,j)=C; gbest(1,:,j)=X(I,:,j); [C,I]=min(B(1,1,:)); if B(1,1,j)<=C gbest(1,:,j)=gbest(1,:,j); else gbest(1,:,j)=gbest(1,:,I); end if C<=minerr, break, end if j>=itmax, break, end for p=1:N G(p,:,j)=gbest(1,:,j); end for i=1:N; [C,I]=min(L(i,1,:)); if L(i,1,j)<=C pbest(i,:,j)=X(i,:,j); else pbest(i,:,j)=X(i,:,I); end end V(:,:,j+1)=W(j)*V(:,:,j)+c1*rand*(pbest(:,:,j)-X(:,:,j))+c2*rand*(G(:,:,j)-X(:,:,j)); for ni=1:N for di=1:D if V(ni,di,j+1)>vmax V(ni,di,j+1)=vmax; elseif V(ni,di,j+1)<-vmax V(ni,di,j+1)=-vmax; else V(ni,di,j+1)=V(ni,di,j+1); end end end X(:,:,j+1)=X(:,:,j)+V(:,:,j+1); for ni=1:N for di=1:D if X(ni,di,j+1) > 1 X(ni,di,j+1) = 1; elseif V(ni,di,j+1) < -1 X(ni,di,j+1) = -1; else X(ni,di,j+1) = X(ni,di,j+1); end end end end disp('Iteration and Current Best Fitness') disp(j) disp(B(1,1,j)) disp('Global Best Fitness and Occurred Iteration') [C,I] = min(B(1,1,:)) for t=1:hiddennum x2iw(t,:)=gbest(1,((t-1)*indim+1):t*indim,j); end for r=1:outdim x2lw(r,:)=gbest(1,(indim*hiddennum+1):(indim*hiddennum+hiddennum),j); end x2b=gbest(1,((indim+1)*hiddennum+1):D,j); x2b1=x2b(1:hiddennum).'; x2b2=x2b(hiddennum+1:hiddennum+outdim).'; IWbest1=x2iw; IWbest2=x2lw; IBbest1=x2b1; IBbest2=x2b2; %% Adaboost %训练样本 %input_train=Ptrain; %output_train=Ttrain; %测试样本 %input_test=Ptest; %output_test=Ttest; %样本权重 [mm,nn]=size(Ptrain); DD(1,:)=ones(1,nn)/nn; %训练样本归一化 %[inputn,inputps]=mapminmax(input_train); %[outputn,outputps]=mapminmax(output_train); K=10; for i=1:K %弱预测器训练 %net=newff(inputn,outputn,5); %net.trainParam.epochs=20; %net.trainParam.lr=0.1; %net=train(net,inputn,outputn); %弱预测器预测 %an1=sim(net,inputn); %BPoutput=mapminmax('reverse',an1,outputps); T_ELM1 = ELMfun2(IWbest1,IBbest1,Ptrain,Ttrain,hiddennum,IWbest2,IBbest2); ELMout1 = postmnmx(T_ELM1,minAllSamOut,maxAllSamOut); %ELMout1 = postmnmx(T_ELM1,minAllTestOut,maxAllTestOut); %预测误差 erroryc(i,:)=AllSamOut-ELMout1 ; %测试数据预测 T_sim2= ELMfun2(IWbest1,IBbest1,Ptest,Ttest,hiddennum,IWbest2,IBbest2); test_simu(i,:) = postmnmx(T_sim2,minAllTestOut,maxAllTestOut); %test_simu(i,:)=mapminmax('reverse',an2,outputps); %调整D值 Error(i)=0; for j=1:nn if abs(erroryc(i,j))>0.2 %较大误差 Error(i)=Error(i)+DD(i,j); DD(i+1,j)=DD(i,j)*1.1; else DD(i+1,j)=DD(i,j); end end %计算弱预测器权重 at(i)=0.5/exp(abs(Error(i))); %D值归一化 DD(i+1,:)=DD(i+1,:)/sum(DD(i+1,:)); end %% 强预测器预测 at=at/sum(at); %% 结果统计 %强分离器效果 output=at*test_simu; T_sim_test = ELMfun2(IWbest1,IBbest1,Ptest,Ttest,hiddennum,IWbest2,IBbest2); testsamout = postmnmx(T_sim_test,minAllTestOut,maxAllTestOut); realtesterr=mse(testsamout-AllTestOut); err1=norm(testsamout-AllTestOut); disp(['优化模型测试样本的仿真误差:',num2str(err1)]) N = length(Ttest); R1 = (N*sum(T_sim_test.*Ttest)-sum(T_sim_test)*sum(Ttest))^2/((N*sum((T_sim_test).^2)-(sum(T_sim_test))^2)*(N*sum((Ttest).^2)-(sum(Ttest))^2)); %% 不使用PSO优化ELM算法 [IW,B,LW,TF,TYPE] = elmtrain2(Ptrain,Ttrain,5,'sig',0); T_sim_test1 = elmpredict(Ptest,IW,B,LW,TF,TYPE); testsamout2 = postmnmx(T_sim_test1,minAllTestOut,maxAllTestOut); realtesterr2=mse(testsamout2-AllTestOut); err2=norm(testsamout2-AllTestOut); R2 = (N*sum(T_sim_test1.*Ttest)-sum(T_sim_test1)*sum(Ttest))^2/((N*sum((T_sim_test1).^2)-(sum(T_sim_test1))^2)*(N*sum((Ttest).^2)-(sum(Ttest))^2)); disp(['原始模型测试样本的仿真误差:',num2str(err2)]) %% 输出参数图 % PSO 优化迭代图 figure(1); P0 = plot(1:itmax,BestFit); grid on xlabel('PSO items'); ylabel('MSE rate'); string = {'PSO-ELM模型100次迭代误差变化';['决定系数:R^2=' num2str(R1) ' ' ];['模型仿真均方误差:mse=' num2str(realtesterr) ' ']}; title(string) bestErr=min(BestFit); fprintf(['PSO-ELM模型:\n决定系数R^2=',num2str(R1),'\n模型仿真均方误差:mse=',num2str(realtesterr),'\n']) set(P0,'LineWidth',1.5); % PSO与ELM结果比较图 figure(2) P1=plot(1:76,AllTestOut,'r-square',1:76,testsamout,'b-o',1:76,testsamout2,'g-diamond',1:76,output,'m-.') grid on legend('真实值','PSO-ELM预测值','单纯ELM预测值','Adaboost+PSO-ELM预测值') xlabel('样本编号') ylabel('样本数据分类号') string = {'测试集-预测分类-结果对比(真实值 & PSO-ELM & ELM)';['ELM:(mse = ' num2str(err2) ' R^2 = ' num2str(R2) ')'];['PSO-ELM:(mse = ' num2str(err1) ' R^2 = ' num2str(R1) ')']}; title(string) set(P1,'LineWidth',1.0);