LSTM神经网络Matlab编程实例

1 简介

基于自编LSTM神经网络实现空调能耗数据预测。

2 部分代码

%% 程序说明
%  1、数据为7天，四个时间点的空调功耗，用前三个推测第四个训练，依次类推。第七天作为检验
%  2、LSTM网络输入结点为12，输出结点为4个，隐藏结点18个

clear all;
clc;
%% 数据加载，并归一化处理
[train_data,test_data]=LSTM_data_process();
data_length=size(train_data,1);
data_num=size(train_data,2);
%% 网络参数初始化
% 结点数设置
input_num=12;
cell_num=18;
output_num=4;
% 网络中门的偏置
bias_input_gate=rand(1,cell_num);
bias_forget_gate=rand(1,cell_num);
bias_output_gate=rand(1,cell_num);
% ab=1.2;
% bias_input_gate=ones(1,cell_num)/ab;

    end
    if(Error_Cost(1,iter)<cost_gate)
        break;
    end
end
%% 绘制Error-Cost曲线图
% for n=1:1:iter
%     text(n,Error_Cost(1,n),'*');
%     axis([0,iter,0,1]);
%     title('Error-Cost曲线图');   
% end
for n=1:1:iter
    semilogy(n,Error_Cost(1,n),'*');
    hold on;
    title('Error-Cost曲线图');   
end
%% 使用第七天数据检验
%数据加载
test_final=[0.4557 0.4790 0.7019 0.8211 0.4601 0.4811 0.7101 0.8298 0.4612 0.4845 0.7188 0.8312]
test_final=test_final/sqrt(sum(test_final.^2));
test_output=test_data(:,4);
%前馈
m=4;
gate=tanh(test_final'*weight_input_x+h_state(:,m-1)'*weight_input_h);
input_gate_input=test_final'*weight_inputgate_x+cell_state(:,m-1)'
*weight_inputgate_c+bias_input_gate;
forget_gate_input=test_final'*weight_forgetgate_x+cell_state(:,m-1)'
*weight_forgetgate_c+bias_forget_gate;
output_gate_input=test_final'*weight_outputgate_x+cell_state(:,m-1)'
*weight_outputgate_c+bias_output_gate;
for n=1:cell_num
    input_gate(1,n)=1/(1+exp(-input_gate_input(1,n)));
    forget_gate(1,n)=1/(1+exp(-forget_gate_input(1,n)));
    output_gate(1,n)=1/(1+exp(-output_gate_input(1,n)));
end
cell_state_test=(input_gate.*gate+cell_state(:,m-1)'.*forget_gate)';
pre_h_state=tanh(cell_state_test').*output_gate;
h_state_test=(pre_h_state*weight_preh_h)'
test_output

3 仿真结果

免责声明：本文系网络转载或改编，未找到原创作者，版权归原作者所有。如涉及版权，请联系删