RIME-BP多输出回归预测 (多输入多输出)，基于霜冰优化算法（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 res = xlsread('data.xlsx'); %% 划分训练集和测试集 temp = randperm(600); P_train = res(temp(1: 500), 1 : 10)'; T_train = res(temp(1: 500), 11: 13)'; M = size(P_train, 2); P_test = res(temp(501: end), 1 : 10)'; T_test = res(temp(501: end), 11: 13)'; N = size(P_test, 2); %% 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); %% 节点个数 inputnum = size(p_train, 1); % 输入层节点数 hiddennum = 15; % 隐藏层节点数 outputnum = size(t_train, 1); % 输出层节点数 %% 构建网络 net = newff(p_train, t_train, hiddennum); %% 设置训练参数 net.trainParam.epochs = 50; % 训练次数 net.trainParam.goal = 1e-4; % 目标误差 net.trainParam.lr = 0.01; % 学习率 net.trainParam.showWindow = 0; % 关闭窗口 %% 参数设置 fun = @getObjValue; % 目标函数 dim = inputnum * hiddennum + hiddennum * outputnum + ... hiddennum + outputnum; % 优化参数个数 lb = -1 * ones(1, dim); % 优化参数目标下限 ub = 1 * ones(1, dim); % 优化参数目标上限 pop = 20; % 数量 Max_iteration = 20; % 最大迭代次数 %% 优化算法 [Best_score,Best_pos,curve] = RIME(pop, Max_iteration, lb, ub, dim, fun); %% 把最优初始阀值权值赋予网络预测 w1 = Best_pos(1 : inputnum * hiddennum); B1 = Best_pos(inputnum * hiddennum + 1 : inputnum * hiddennum + hiddennum); w2 = Best_pos(inputnum * hiddennum + hiddennum + 1 : inputnum * hiddennum + hiddennum + hiddennum*outputnum); B2 = Best_pos(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} = B2'; %% 网络训练 net.trainParam.showWindow = 1; % 打开窗口 net = train(net, p_train, t_train); %% BP网络预测 t_sim1 = sim(net, p_train); t_sim2 = sim(net, p_test); %% 数据反归一化 T_sim1 = mapminmax('reverse',t_sim1,ps_output); T_sim2 = mapminmax('reverse',t_sim2,ps_output); %% 优化曲线 figure plot(curve, 'linewidth',1.5); title('RIME') xlabel('The number of iterations') ylabel('Fitness') grid on; for i = 1: 3 %% 均方根误差 error1(i, :) = sqrt(sum((T_sim1(i, :) - T_train(i, :)).^2) ./ M); error2(i, :) = sqrt(sum((T_sim2(i, :) - T_test (i, :)).^2) ./ N); %% 绘图 figure subplot(2, 1, 1) plot(1: M, T_train(i, :),'-+','Color',[255 0 0]./255,'linewidth',1,'Markersize',3,'MarkerFaceColor',[255 0 0]./255) hold on plot(1: M, T_sim1(i, :),'o','Color',[0 0 255]./255,'linewidth',1,'Markersize',3,'MarkerFaceColor',[0 0 255]./255) legend('真实值','预测值') xlabel('预测样本') ylabel('预测结果') string = {['输出', num2str(i),'训练集预测结果对比']; ['RMSE=' num2str(error1(i, :))]}; title(string) xlim([1, M]) set(gca,'FontSize',12) set(gca,'TickDir','out'); set(gca,'LineWidth',0.8); drawnow subplot(2, 1, 2) plot(1: N, T_test(i, :),'-+','Color',[255 0 0]./255,'linewidth',1,'Markersize',3,'MarkerFaceColor',[255 0 0]./255) hold on plot(1: N, T_sim2(i, :),'o','Color',[0 0 255]./255,'linewidth',1,'Markersize',3,'MarkerFaceColor',[0 0 255]./255) legend('真实值','预测值') xlabel('预测样本') ylabel('预测结果') string = {['输出', num2str(i),'测试集预测结果对比'];['RMSE=' num2str(error2(i, :))]}; title(string) xlim([1, N]) set(gca,'FontSize',12) set(gca,'TickDir','out'); set(gca,'LineWidth',0.8); drawnow %% 分割线 disp('**************************') disp(['下列是输出', num2str(i)]) disp('**************************') %% 相关指标计算 % 决定系数 R2 R1(i, :) = 1 - norm(T_train(i, :) - T_sim1(i, :))^2 / norm(T_train(i, :) - mean(T_train(i, :)))^2; R2(i, :) = 1 - norm(T_test (i, :) - T_sim2(i, :))^2 / norm(T_test (i, :) - mean(T_test (i, :)))^2; disp(['训练集数据的R2为：', num2str(R1(i, :))]) disp(['测试集数据的R2为：', num2str(R2(i, :))]) % 平均绝对误差 MAE mae1(i, :) = sum(abs(T_sim1(i, :) - T_train(i, :))) ./ M ; mae2(i, :) = sum(abs(T_sim2(i, :) - T_test (i, :))) ./ N ; disp(['训练集数据的MAE为：', num2str(mae1(i, :))]) disp(['测试集数据的MAE为：', num2str(mae2(i, :))]) % 平均相对误差 MBE mbe1(i, :) = sum(T_sim1(i, :) - T_train(i, :)) ./ M ; mbe2(i, :) = sum(T_sim2(i, :) - T_test (i, :)) ./ N ; disp(['训练集数据的MBE为：', num2str(mbe1(i, :))]) disp(['测试集数据的MBE为：', num2str(mbe2(i, :))]) end