چكيده به لاتين
Improving food security and production efficiency can be achieved by controlling climatic conditions through the regulation of temperature, humidity, carbon dioxide concentration, and light intensity in greenhouses. In this study, comprehensive modeling and optimization were conducted to determine the optimal environmental conditions, including temperature, humidity, and carbon dioxide concentration, for greenhouse tomato cultivation with the aim of minimizing operational costs while achieving the best crop performance.
By introducing optimal environmental conditions, the needs for heating, water for humidification and irrigation, dehumidification, carbon dioxide injection, and supplementary lighting, alongside the dynamic behavior of crop growth, were simultaneously examined. Optimization, as a single objective, was performed by defining the operational cost per unit weight of the crop as the objective function and temperature during the day (both natural and artificial lighting), night temperature (dark period), humidity, and carbon dioxide concentration as the four design variables for a 200-day tomato cultivation period in Tehran. Since this modeling and optimization required significant time, for the first time, an artificial neural network was employed for comprehensive modeling to approximate very well. With the desired design variables as input, the objective function was calculated and optimization was performed. The optimal values for the objective function and the aforementioned design variables were calculated as 23.7°C, $0.47/kg for lighting time, 16°C for the dark period, 68.2% relative humidity, and 627.7 ppm CO₂. The use of the neural network significantly reduced the time required for modeling and optimization by 99%.
