Machine Learning-Enabled Smart Greenhouse Environmental Control Service Model

Extreme weather often causes crop losses with sharp fluctuations in agricultural prices, which imposes negative impacts on sustainable agricultural development. Greenhouse farming is regarded as an effective measure against extreme weather. Thus, it requires a better understanding of the growing complexity of agri-food systems involving greenhouse environmental and societal tradeoffs under climate variations. Considering high energy consumption of greenhouses, this study aims at adopting machine learning with IoT-big data mining to innovatively develop a smart greenhouse environmental control service model under the nexus between meteorology, water, energy, food, and greenhouse environmental control while exploring pathways to low-carbon greenhouse cultivation. The proposed model will be applied to greenhouses in Taiwan for evaluating cross-sectoral synergies and environmental benefits. The results are expected to support greenhouse owners and authorities to make the best use of resources of water, energy, and food through the optimal environmental operation on greenhouse cultivation under extreme climatic events for achieving sustainable development goals (SDGs) and move towards green economy.