Matlab基于粒子群算法优化径向基神经网络(PSO-RBF)的多输入单输出回归预测，含优化前后(Matlab完整源码和数据）

%% 清空环境变量 warning off % 关闭报警信息 close all % 关闭开启的图窗 clear % 清空变量 clc % 清空命令行 %% 导入数据 % 训练集——190个样本 P_train = xlsread('data','training set','B2：G191')'; T_train= xlsread('data','training set','H2：H191')'; % 测试集——44个样本 P_test=xlsread('data','test set','B2:G45')';T_test=xlsread('data','test set','H2:H45')'; N = size(P_test, 2); % 测试集样本数 M = size(P_train, 2); % 训练集样本数 %% 数据归一化 [p_train, ps_input] = mapminmax(P_train, 0, 1); p_test = mapminmax('apply', P_test, ps_input); [t_train, ps_output] = mapminmax(T_train, 0, 1); t_test = mapminmax('apply', T_test, ps_output); %% 优化算法 pop = 20; %种群数量 Max_iter = 20; %最大迭代次数 lb = 0.001; %下边界 ub = 1; %上边界 dim = 1; %维度 fobj=@(X)fobj(X,p_train,t_train,p_test,t_test); [ Best_score, Best_P,curve] = PSO(pop, Max_iter, lb, ub, dim, fobj); %% 重新训练 disp(['最佳扩散值为',num2str(Best_P)]) %% 采用最佳方法建立RBFN网络 % 径向基函数的扩展速度 net = newrbe(p_train, t_train, Best_P); t_sim1=sim(net,p_train); %训练集预测 t_sim2=sim(net,p_test); %测试集预测 %% 数据反归一化 T_sim1 = mapminmax('reverse', t_sim1, ps_output); T_sim2 = mapminmax('reverse', t_sim2, ps_output); %% 均方根误差 error1 = sqrt(sum((T_sim1 - T_train).^2) ./ M); error2 = sqrt(sum((T_sim2 - T_test ).^2) ./ N); %% 适应度曲线 figure plot(1 : length(curve), curve,'r-', 'LineWidth', 1.5); title('PSO-RBF', 'FontSize', 13); xlabel('迭代次数', 'FontSize', 10); ylabel('适应度值', 'FontSize', 10); grid off %% 绘图 figure plot(1: M, T_train, 'r-^', 'LineWidth', 1,'MarkerFaceColor', 'r','MarkerSize', 4); hold on plot(1: M, T_sim1, 'b-o', 'LineWidth', 1,'MarkerFaceColor', 'b','MarkerSize', 4); legend('真实值','预测值') xlabel('预测样本') ylabel('预测结果') string = {'PSO-RBF训练集预测结果对比'; ['RMSE=' num2str(error1)]}; title(string) xlim([1, M]) grid off figure plot(1: N, T_test, 'r-^', 'LineWidth', 1,'MarkerFaceColor', 'r','MarkerSize', 4); hold on plot(1: N, T_sim2, 'b-o', 'LineWidth', 1,'MarkerFaceColor', 'b','MarkerSize', 4); legend('真实值','预测值') xlabel('预测样本') ylabel('预测结果') string = {'PSO-RBF测试集预测结果对比';['RMSE=' num2str(error2)]}; title(string) xlim([1, N]) grid off %% 误差图 figure [AX,H1,H2]=plotyy(1:N,(T_sim2-T_test),1:N,(T_sim2-T_test)./T_test,'plot')%plot函数图像 set(AX(1),'XColor','k','YColor','b');%设置左边y轴 set(AX(2),'XColor','k','Ycolor','r');%设置右边y轴 xlabel('测试集样本编号'); %设置右边y轴 HH1=get(AX(1),'Ylabel'); set(HH1,'String','绝对误差');%设置左边y坐标轴标题 set(HH1,'color','b');%设置左边y坐标轴图样颜色为‘b’，蓝色 HH2=get(AX(2),'Ylabel'); set(HH2,'String','相对误差');%设置右边y坐标轴标题 set(HH2,'color','r');%设置右边y坐标轴图样颜色为‘r’红色 set(H1,'LineStyle','-','marker','o', 'LineWidth', 1,'MarkerFaceColor', 'b','MarkerSize', 4);%设置函数y1的线形状 set(H1,'color','b');%设置函数y1的线颜色 set(H2,'LineStyle',':','marker','^', 'LineWidth', 1,'MarkerFaceColor', 'r','MarkerSize', 4);%设置函数y2的线形状 set(H2,'color','r');%设置函数y2的线颜色 set(gca, 'FontSize', 10); figure [AX,H1,H2]=plotyy(1:M,(T_sim1-T_train),1:M,(T_sim1-T_train)./T_train,'plot');%plot函数图像 set(AX(1),'XColor','k','YColor','b');%设置左边y轴 set(AX(2),'XColor','k','Ycolor','r');%设置右边y轴 xlabel('训练集样本编号'); %设置右边y轴 HH1=get(AX(1),'Ylabel'); set(HH1,'String','绝对误差');%设置左边y坐标轴标题 set(HH1,'color','b');%设置左边y坐标轴图样颜色为‘b’，蓝色 HH2=get(AX(2),'Ylabel'); set(HH2,'String','相对误差');%设置右边y坐标轴标题 set(HH2,'color','r');%设置右边y坐标轴图样颜色为‘r’红色 set(H1,'LineStyle','-','marker','o', 'LineWidth', 1,'MarkerFaceColor', 'b','MarkerSize', 4);%设置函数y1的线形状 set(H1,'color','b');%设置函数y1的线颜色 set(H2,'LineStyle',':','marker','^', 'LineWidth', 1,'MarkerFaceColor', 'r','MarkerSize', 4);%设置函数y2的线形状 set(H2,'color','r');%设置函数y2的线颜色 set(gca, 'FontSize', 10); %% 测试集结果 figure; plotregression(T_test,T_sim2,['回归图']); figure; ploterrhist(T_test-T_sim2,['误差直方图']); %% 相关指标计算 disp(['训练集数据误差：']) [mae_train,mse_train,rmse_train,mape_train,error_train,errorPercent_train,R_train]=calc_error(T_train,T_sim1); % disp(['测试集数据误差：']) [mae_test,mse_test,rmse_test,mape_test,error_test,errorPercent_test,R_test]=calc_error(T_test,T_sim2); % %% 训练集拟合效果图 R1 = 1 - norm(t_train - t_sim1)^2 / norm(t_train - mean(t_train))^2; R2 = 1 - norm(t_test - t_sim2)^2 / norm(t_test - mean(t_test ))^2; figure plot(T_train,T_sim1,'ok'); xlabel('真实值') ylabel('预测值') string = {'训练集效果图';['R^2_c=' num2str(R1) ' RMSEC=' num2str(error1) ]}; title(string) hold on ;h=lsline; set(h,'LineWidth',1,'LineStyle','-','Color','r') %% 预测集拟合效果图 figure plot(T_test,T_sim2,'ob'); xlabel('真实值') ylabel('预测值') string1 = {'测试集效果图';['R^2_p=' num2str(R2) ' RMSEP=' num2str(error2) ]}; title(string1) hold on ;h=lsline(); set(h,'LineWidth',1,'LineStyle','-','Color','r') %% 全部数据集拟合效果图 figure plot([T_train T_test],[T_sim1 T_sim2],'oc'); xlabel('真实值') ylabel('预测值') string1 = {'所有样本拟合预测图';['R^2_p=' num2str(mean([R2 R1])) ' RMSEP=' num2str(mean([R2 R1]))]}; title(string1) hold on ;h=lsline(); set(h,'LineWidth',1,'LineStyle','-','Color','r') %% 采用随机参数建立RBF网络 net1=newgrnn(p_train,t_train); %未优化RBF t_sim1_rand=sim(net1,p_train); %训练集预测 t_sim2_rand=sim(net1,p_test); %测试集预测 %% 数据反归一化 T_sim1_rand = mapminmax('reverse', t_sim1_rand, ps_output); T_sim2_rand = mapminmax('reverse', t_sim2_rand, ps_output); %% 均方根误差 error1_rand = sqrt(sum((T_sim1_rand - T_train).^2) ./ M); error2_rand = sqrt(sum((T_sim2_rand - T_test ).^2) ./ N); %% 绘图 figure plot(1: M, T_train, 'r-^', 'LineWidth', 1,'MarkerFaceColor', 'r','MarkerSize', 4); hold on plot(1: M, T_sim1_rand, 'b-o', 'LineWidth', 1,'MarkerFaceColor', 'b','MarkerSize', 4); legend('真实值','预测值') xlabel('预测样本') ylabel('预测结果') string = {'原始RBF训练集预测结果对比'; ['RMSE=' num2str(error1_rand)]}; title(string) xlim([1, M]) grid off figure plot(1: N, T_test, 'r-^', 'LineWidth', 1,'MarkerFaceColor', 'r','MarkerSize', 4); hold on plot(1: N, T_sim2_rand, 'b-o', 'LineWidth', 1,'MarkerFaceColor', 'b','MarkerSize', 4); legend('真实值','预测值') xlabel('预测样本') ylabel('预测结果') string = {'原始RBF测试集预测结果对比';['RMSE=' num2str(error2_rand)]}; title(string) xlim([1, N]) grid off