Protected agriculture faces major climatic challenges, particularly the significant temperature
difference between day and night. This requires advanced technological solutions to ensure a
suitable indoor climate and enhance productivity. This study aims to develop an integrated
system combining efficient design, thermal storage techniques, and artificial intelligence to
predict the internal temperature. A solar greenhouse was constructed at the Applied Research
Unit for Renewable Energy in Ghardaïa (32.36° North, 3.51° West), incorporating a north
wall designed to store solar energy. This wall, with an area of 1.62 m2, was built using locally
sourced stones selected for their excellent thermal properties. It stores excess heat during the
day and releases it at night, thereby reducing nighttime temperature fluctuations.
Experimental measurements showed that the air temperature inside the greenhouse equipped
with the thermal wall was about 2.7 °C higher than the outside air temperature at night,
demonstrating the system’s effectiveness.
This system was coupled with an artificial intelligence model based on artificial neural
networks to accurately predict the internal temperature, using indoor and outdoor climatic
data collected from the experimental setup. The model showed high performance and good
agreement between predicted and measured values, with a correlation coefficient exceeding
98%, confirming its efficiency and reliability as a sustainable and effective solution to the
challenges of protected agriculture in harsh environments.
Protected agriculture faces major climatic challenges, particularly the significant temperature
difference between day and night. This requires advanced technological solutions to ensure a
suitable indoor climate and enhance productivity. This study aims to develop an integrated
system combining efficient design, thermal storage techniques, and artificial intelligence to
predict the internal temperature. A solar greenhouse was constructed at the Applied Research
Unit for Renewable Energy in Ghardaïa (32.36° North, 3.51° West), incorporating a north
wall designed to store solar energy. This wall, with an area of 1.62 m2, was built using locally
sourced stones selected for their excellent thermal properties. It stores excess heat during the
day and releases it at night, thereby reducing nighttime temperature fluctuations.
Experimental measurements showed that the air temperature inside the greenhouse equipped
with the thermal wall was about 2.7 °C higher than the outside air temperature at night,
demonstrating the system’s effectiveness.
This system was coupled with an artificial intelligence model based on artificial neural
networks to accurately predict the internal temperature, using indoor and outdoor climatic
data collected from the experimental setup. The model showed high performance and good
agreement between predicted and measured values, with a correlation coefficient exceeding
98%, confirming its efficiency and reliability as a sustainable and effective solution to the
challenges of protected agriculture in harsh environments.
