Develop a Smart Microclimate Control System for Greenhouses through System Dynamics and Machine Learning Techniques

Agriculture is extremely vulnerable to climate change. Greenhouse farming recognized as a promising measure against Nevertheless, greenhouse frequently encounters environmental adversity, especially greenhouses built protect typhoons. Short-term microclimate prediction challenging because meteorological variables are strongly interconnected and change rapidly. Therefore, this study proposes water-centric smart microclimate-control system (SMCS) that fuses dynamics machine-learning techniques in consideration of the internal hydro-meteorological process regulate micro-environment within canopy for cooling with improved resource-use efficiency. SMCS was assessed by situ data collected from tomato Taiwan. The results demonstrate proposed could save 66.8% water energy (electricity) used early spraying during entire cultivation period compared traditional greenhouse-spraying based mainly on operators’ experiences. suggests practicability niche machine-learning-enabled automation crop productivity This will increase agricultural resilience hydro-climate uncertainty promote resource preservation, which offers pathway towards carbon-emission mitigation sustainable water–energy–food nexus.