Current Advances in Agricultural Sciences(An International Journal)

In recent times, the significance of protected cultivation has surged rapidly owing to population growth and declining agricultural productivity. Adopting protected cultivation systems mitigates the impact of external environmental factors on crops by establishing a favorable microclimate around them. When maintained optimally, this microclimate exponentially boosts crop productivity. Controlling the microclimate necessitates an understanding of the potential variations, driving researchers to develop models for studying greenhouse microclimates. These models can be approached using physically based models and computational models, including process-based models like Computational Fluid Dynamics (CFD) and data-driven models such as Machine Learning methods. This paper aims to explore microclimate dynamics, control strategies, and diverse modelling techniques. The efficacy of CFD and Machine Learning models in microclimate prediction is assessed. While CFD proves effective, it is time-consuming compared to Machine Learning (ML) methods. It is concluded that combining CFD and ML models can accurately predict microclimates by simulating and analyzing data patterns. Integration with IoT Sensor system paves the way for greenhouse automation.