Simulink-Driven Digital Twin Implementation for Smart Greenhouse Environmental Control

Abstract

Sustainable food production must grow unprecedentedly in the face of the growing global hunger crisis. This proposal significantly reduces global hunger by creating an environmentally friendly approach to a smart greenhouse that aligns with zero hunger and sustainable development. This novel study is dissimilar to the conventional implementation of small-scale greenhouse farming as it implements modern sophisticated techniques applied specifically in greenhouses. The novelty of work lies in the integration of Simulink, the digital twin model into the smart greenhouse environment, capable of providing intelligent insights about plant growth patterns, enabling the farmers to make the right decision at the right time with remote monitoring capabilities, while maximizing the yield potential, trained via boosted trees algorithm with 8.4684 RMSE and 85% validation accuracy. Additionally, we have used state-of-the-art CNN model, Internet of Things (IoT) sensors and image-processing techniques to identify and classify diseases of crops in a greenhouse with 98.39% validation accuracy. The reason for this is quite long-term too as it involves not only dealing with the woes befalling greenhouse agriculture but reforming a more sustainable approach to food production.