Adaptive-Historical Energy-Efficient Temperature Control for Tropical Greenhouses

Maintaining an optimal microclimate is essential for efficient operation of tropical greenhouses, particularly under fluctuating weather conditions. This study proposes an adaptive energy-efficient model for regulating air temperature in tropical greenhouses using historical climate data. The model optimizes the fan rotation speeds via an inverter to meet the temperature targets while minimizing energy consumption. Key methodologies include climate data analysis, development of a predictive model for indoor air temperature using Artificial Neural Networks, and optimization of fan speed control. The model achieved high predictive accuracy, with an RMSE of 0,02 and an R² of 0,96. The practical implementation demonstrated effective temperature control, with fan speeds ranging between 30 and 40 Hz during cloudy periods and 50 Hz in sunny conditions. Notably, the system reduced electricity consumption by 33,93% during cloudy weather and 18,54% in sunny weather, showing its potential for significant energy savings. This data-driven adaptive model approach is highly suited for tropical greenhouses experiencing dynamic climatic variations and offers a sustainable and efficient solution for greenhouse microclimate management.