Articles
A simple model to predict air temperature inside a Mediterranean greenhouse
Article number
1182_11
Pages
95 – 104
Language
English
Abstract
The main purpose of this work was to develop a simplified dynamic energy balance model to predict the air temperature inside naturally ventilated greenhouses.
Simplification involved reducing the input parameters, using a linear regression to estimate the vegetation temperature from the calculated inside temperature and outside solar radiation.
The temperature of the plastic mulch was calculated from the simulated soil temperature at 0.2 m depth with a sine wave function varying around an average temperature.
The model was validated comparing simulated and experimentally measured temperatures of inside air, plants and soil.
Measurements were carried out on 76 days throughout two tomato crop cycles (autumn-winter and spring-summer) in a multi-span greenhouse equipped with insect-proof screens in the openings and variable cover transmittance (different whitewashing during the season). The current model was based on previous models that had been adapted for use in the climatic conditions of southern Spain.
The input data of this model were the outside weather conditions, physical crop parameters and greenhouse characteristics.
The outputs were the inside air, cover, crop and soil temperatures.
Ground temperature was obtained by calculating the heat flux through the woven polypropylene film covering the soil.
Furthermore, an empirical coefficient was included to consider the effect on light distribution inside the greenhouse caused by the diffusivity of the plastic cover.
The energy balance model was established by means of ordinary differential equations solved numerically using the LSODA algorithm.
A program was written in the statistical software R for solving these equations.
Statistical comparison between the inside air temperature predicted by the model and the measured results showed good agreement with a relative root mean squared error (RRMSE) of below 10%.
Simplification involved reducing the input parameters, using a linear regression to estimate the vegetation temperature from the calculated inside temperature and outside solar radiation.
The temperature of the plastic mulch was calculated from the simulated soil temperature at 0.2 m depth with a sine wave function varying around an average temperature.
The model was validated comparing simulated and experimentally measured temperatures of inside air, plants and soil.
Measurements were carried out on 76 days throughout two tomato crop cycles (autumn-winter and spring-summer) in a multi-span greenhouse equipped with insect-proof screens in the openings and variable cover transmittance (different whitewashing during the season). The current model was based on previous models that had been adapted for use in the climatic conditions of southern Spain.
The input data of this model were the outside weather conditions, physical crop parameters and greenhouse characteristics.
The outputs were the inside air, cover, crop and soil temperatures.
Ground temperature was obtained by calculating the heat flux through the woven polypropylene film covering the soil.
Furthermore, an empirical coefficient was included to consider the effect on light distribution inside the greenhouse caused by the diffusivity of the plastic cover.
The energy balance model was established by means of ordinary differential equations solved numerically using the LSODA algorithm.
A program was written in the statistical software R for solving these equations.
Statistical comparison between the inside air temperature predicted by the model and the measured results showed good agreement with a relative root mean squared error (RRMSE) of below 10%.
Authors
A. Reyes-Rosas, F.D. Molina-Aiz, A. López, D.L. Valera
Keywords
greenhouse, numerical model, temperature, natural ventilation, heat transfer
Online Articles (34)