基于GA-BP神经网络的温室温度预测研究  

Greenhouse temperature prediction based on GA-BP neural network

作　　者：李其操 董自健 LI Qicao;DONG Zijian(School of Electronic Engineering,Jiangsu Ocean University,Lianyungang Jiangsu 222005,China)

机构地区：[1]江苏海洋大学电子工程学院,江苏连云港222005

出　　处：《智能计算机与应用》2023年第9期168-171,共4页Intelligent Computer and Applications

摘　　要：温度对于温室内作物的生长起着重要的作用,为了更精准的管理和控制温室内的温度,提出了基于遗传算法优化的BP神经网络预测模型(GA-BP),对温室内温度进行预测。本文利用遗传算法对BP神经网络的权值和阈值进行优化,使模型避免出现局部最优,有效改善了传统BP神经网络预测模型的性能,使预测出的温度更加精准。实验证明,选择隐藏层节点数为7时,GA-BP神经网络预测模型的预测结果最佳,平均绝对误差(MAE)、均方误差(MSE)和平均绝对百分比误差(MAPE)分别为0.441、0.276、0.525。与传统BP神经网络预测模型相比分别提升了13.2%、38.4%、21.5%。Temperature plays an important role in the growth of crops in the greenhouse.In order to manage and control the temperature in the greenhouse more accurately,a genetic algorithm-optimized BP neural network prediction model(GA-BP)was proposed to predict the temperature in the greenhouse.In this paper,the genetic algorithm is used to optimize the weights and thresholds of the BP neural network,so that the model avoids local optimization,effectively improves the performance of the traditional BP neural network prediction model,and makes the predicted temperature more accurate.Experiments show that when the number of hidden layer nodes is selected to be 7,the prediction result of the GA-BP neural network prediction model is the best,and the mean absolute error(MAE),mean square error(MSE)and mean absolute percentage error(MAPE)are 0.441,0.276,and 0.525 respectively.Compared with the traditional BP neural network prediction model,it has increased by 13.2%,38.4%,and 21.5%respectively.

关 键 词：遗传算法 BP神经网络 温室温度 预测模型 

分 类 号：S625[农业科学—园艺学] TP183[自动化与计算机技术—控制理论与控制工程]