elman位移监测（代码完整，数据齐全）

%% 清空环境变量 clear clc close all num1 = xlsread('数据文件0.xlsx'); num = []; for ii = 1:length(num1)-1 num = [num;num1(ii,:) num1(ii+1,:)]; end %% 导入数据 n =randperm(length(num)); m=60; P_train=num(n(1:m),1:3)'; T_train=num(n(1:m),4)'; P_test=num((1:3:end),1:3)'; T_test=num((1:3:end),4)'; %% 归一化 % 训练集 [Pn_train,inputps] = mapminmax(P_train,-1,1); Pn_test = mapminmax('apply',P_test,inputps); % 测试集 [Tn_train,outputps] = mapminmax(T_train,-1,1); %Tn_test = mapminmax('apply',T_test,outputps); % Pn_train = Pn_train'; % Tn_train = Tn_train'; % tic %% ELM创建/训练 net=newelm(minmax(Pn_train),[3,1],{'logsig','tansig'},'traingdx');%建立网络模型，其中参数可以根据要求修改 net.trainparam.show=100;%每迭代100次显示1次 net.trainparam.epochs=2000;%最大迭代次数2000 net.trainparam.goal=0.001;%迭代目标 net=init(net);%初始化网络 [net,tr]=train(net,Pn_train,Tn_train);%训练网络 Tn_sim=sim(net,Pn_test );%仿真 % 反归一化 T_sim = mapminmax('reverse',Tn_sim,outputps); % T_sim1 = mapminmax('reverse',Tn_sim1,outputps); % toc %% 结果对比 result = [T_test T_sim]; % 均方误差 E = mse(T_sim - T_test); % 决定系数 N = length(T_test); R2 = (N*sum(T_sim.*T_test)-sum(T_sim)*sum(T_test))^2/((N*sum((T_sim).^2)-(sum(T_sim))^2)*(N*sum((T_test).^2)-(sum(T_test))^2)); [MSE ,RMSE ,MBE, MAE ,MAPE] =MSE_RMSE_MBE_MAE(T_test,T_sim) %% 绘图 figure (1) plot(T_test(1,:),'r-*') hold on plot(T_sim(1,:),'b-o') hold on %plot('aa') hold off xlabel('测试集样本编号') ylabel('测试集输出') title('Elman测试集输出') legend('期望输出','预测输出') error1 = T_test-T_sim; figure plot(error1,'k-*') ylabel('误差')