Estimating the inside greenhouse temperature using artificial neural networks : a case study of Ghardaia

Abstract

Agriculture in desert regions, such as the Ghardaïa area, faces climatic challenges that require smart and sustainable solutions. This study aims to predict the temperature inside greenhouses using Artificial Neural Networks (ANN) to improve thermal stability and support intelligent climate control. The experiment was conducted on a single greenhouse, where internal and external climate data were collected to train and test the ANN model. The results showed good performance, with the model demonstrating acceptable accuracy in temperature prediction, which supports its use as a decision- Page | IV support tool for greenhouse climate management. This study highlights the effectiveness of artificial intelligence—particularly Artificial Neural Networks—as an innovative solution to improve greenhouse climate conditions in desert areas.