期刊论文

Application of deep learning model based on decomposition to grain temperature prediction

中文

中南民族大学学报(自然科学版)

1672-4321

2023

42

05

696-701

10.20056/j.cnki.ZNMDZK.20230516

220900786024216:教育部卓越工程师教育培养计划项目产学合作协同育人项目 2018368:湖北省教学研究资助项目 whpu-2021-kj-762:湖北省研究生工作站校企合作项目 whpu-2022-kj-1586:武汉轻工大学横向科研项目

本校为第一机构

在粮食储存过程中，存储温度过高或过低都会造成粮食的损失，准确检测粮堆温度对粮食有效存储至关重要.提出了一种基于分解和深度学习方法的SSA-BiGRU-MLP组合模型来预测粮堆中层温度.首先针对粮堆的外温、外湿、仓温、仓湿和中层温度等多因素变量，使用奇异谱分析（SSA）方法对粮堆的中层温度进行分解，得到本征模态函数（IMF）；然后运用多层感知机（MLP）优化的双向门控循环神经网络（BiGRU）对粮堆温度进行预测.实验结果表明：所提出的SSA-BiGRU-MLP模型与SSA-BiLSTM-MLP,SSA-LSTM-MLP,SSA-GRU-MLP相比，不论是在绝对平均误差、均方误差，还是均方根误差上所提出的SSA-BiGRU-MLP方法都优于其...

In the process of grain storage,the storage temperature is too high or too low will cause grain loss,accurate detection of grain pile temperature is very important for effective grain storage.A combined SSA-BiGRU-MLP model based on decomposition and deep learning method...MORE In the process of grain storage,the storage temperature is too high or too low will cause grain loss,accurate detection of grain pile temperature is very important for effective grain storage.A combined SSA-BiGRU-MLP model based on decomposition and deep learning method is proposed to predict the middle temperature of gr...