基于长短期神经网络LSTM的测量误差预测（代码完整，数据齐全）

%% LSTM network %% import data clc clear close all %读取double格式数据 a = xlsread('训练和测试数据.xlsx'); num = []; for ii = 1:572 num = [num;a(ii+4,2) a(ii:ii+4,1)']; end n = randperm(length(num)); m = 500; input_train =num(n(1:m),1:5);%训练数据输出数据 output_train = num(n(1:m),6);%训练数据输入数据 input_test = num((501:572),1:5);%测试数据输出数据 output_test = num((501:572),6);%测试数据输入数据 [inputn,inputps]=mapminmax(input_train',-1,1);%训练数据的输入数据的归一化 [outputn,outputps]=mapminmax(output_train',-1,1);%训练数据的输出数据的归一化de inputn_test=mapminmax('apply',input_test',inputps); %% Define Network Architecture % Define the network architecture. numFeatures = 5;%输入层维度 numResponses = 1;%输出维度 % 200 hidden units numHiddenUnits = 130;%第一层维度 % a fully connected layer of size 50 & a dropout layer with dropout probability 0.5 layers = [ ... sequenceInputLayer(numFeatures)%输入层 lstmLayer(numHiddenUnits,'OutputMode','sequence')%第一层 fullyConnectedLayer(130)%链接层 dropoutLayer(0.1)%遗忘层 fullyConnectedLayer(numResponses)%链接层 regressionLayer];%回归层 % Specify the training options. % Train for 60 epochs with mini-batches of size 20 using the solver 'adam' maxEpochs = 200;%最大迭代次数 miniBatchSize = 2;%最小批量 % the learning rate == 0.01 % set the gradient threshold to 1 % set 'Shuffle' to 'never' options = trainingOptions('adam', ... %解算器 'MaxEpochs',maxEpochs, ... %最大迭代次数 'MiniBatchSize',miniBatchSize, ... %最小批次 'InitialLearnRate',0.001, ... %初始学习率 'GradientThreshold',1, ... %梯度阈值 'Shuffle','every-epoch', ... %打乱顺序 'Plots','training-progress',... %画图 'Verbose',0); %不输出训练过程 %% Train the Network net = trainNetwork(inputn,outputn,layers,options);%开始训练 %% Test the Network y_pred = predict(net,inputn_test,'MiniBatchSize',20)';%测试仿真输出 y_pred=(mapminmax('reverse',y_pred',outputps))'; y_pred0 = predict(net,inputn,'MiniBatchSize',1);%训练拟合值 y_pred0=(mapminmax('reverse',y_pred0,outputps)); load maydata.mat y_pred=(double(y_pred)); figure%打开一个图像窗口 plot(y_pred(:,1),'k-o')%黑色实线，点的形状为* hold on%继续画图 plot(output_test(:,1),'r-*')%红色实线，点的形状为o hold off%停止画图 title('测试图')%标题 ylabel('上端差')%Y轴名称 legend('测试值','实际值')%标签 error1 = y_pred-output_test;%误差 figure plot(error1(:,1),'k-') title('测试误差图') ylabel('误差') [MSE,RMSE,MBE,MAE ] =MSE_RMSE_MBE_MAE(output_test,y_pred); result_table = table; result_table.sim = y_pred; result_table.true = output_test; writetable(result_table,'./结果.csv')