基于cnn-lstm模型的负荷预测

[数学建模加油站]

 利用CNN提取负荷的特征输入到LSTM中进行负荷预测。采用MATLAB搭建模型如下：       

           

 2、部分代码 

layers4=  [ sequenceUnfoldingLayer('name','unfold')
            flattenLayer('name','flatten')
            lstmLayer(numhidden_units1,'Outputmode','sequence','InputWeightsInitializer',func,'name','lstm1') 
            dropoutLayer(0.1,'name','dropout_1')
            lstmLayer(numhidden_units2,'Outputmode','last','InputWeightsInitializer',func,'name','lstm2') 
            dropoutLayer(0.1,'name','drdiopout_2')
            fullyConnectedLayer(outputSize,'WeightsInitializer',func,'name','fullconnect')   % 全连接层设置（影响输出维度）（cell层出来的输出层） %
            leakyReluLayer('name','ReluLayer')
            regressionLayer('name','output')];
            
lgraph = addLayers(lgraph,layers4);
lgraph = connectLayers(lgraph,'concat','unfold/in');
lgraph = connectLayers(lgraph,'fold/miniBatchSize','unfold/miniBatchSize');

figure
plot(lgraph)


%%
% 网络训练
tic

net = trainNetwork(Train_xNorm,Train_yNorm,lgraph,opts);




%% 测试
% close all
figure
Predict_Ynorm = net.predict(Test_xNorm);
Predict_Y  = mapminmax('reverse',Predict_Ynorm',yopt);
Predict_Y = Predict_Y';

plot(Predict_Y,'g-')
hold on 

plot(Test_y);    
legend('预测值','实际值')

3、结果展示

      

完整代码见 基于cnn-lstm模型的负荷预测-数据 (mianbaoduo.com)