This work proposes a methodology for detecting plant stress induced by water or nutrient deficiencies through the integration of a Digital Twin within a greenhouse Cyber-Physical System. The proposed methodology leverages heterogeneous data collected from environmental and plant-level sensors, which are processed as multi-source inputs within a Multi-Input Multi-Output modeling framework. These inputs continuously feed the Digital Twin, enabling real-time monitoring and state estimation of plant physiological conditions. By comparing the simulated and observed plant states, the Digital Twin can detect early signs of stress related to insufficient water or nutrient availability. When such conditions are identified, the system activates feedback control actions aimed at adjusting irrigation parameters, thereby increasing the supply of water and nutrients to the plant. This closed-loop architecture enables proactive stress mitigation and supports more efficient and adaptive greenhouse management strategies. The proposed approach was validated through a case study involving tomato plants (Solanum lycopersicum) cultivated in a controlled greenhouse environment.
An AI-driven Digital Twin Methodology To Optimize Irrigation in Greenhouses
Ficili, IleniaPrimo
;Giacobbe, Maurizio;Tricomi, Giuseppe
;Puliafito, Antonio
2026-01-01
Abstract
This work proposes a methodology for detecting plant stress induced by water or nutrient deficiencies through the integration of a Digital Twin within a greenhouse Cyber-Physical System. The proposed methodology leverages heterogeneous data collected from environmental and plant-level sensors, which are processed as multi-source inputs within a Multi-Input Multi-Output modeling framework. These inputs continuously feed the Digital Twin, enabling real-time monitoring and state estimation of plant physiological conditions. By comparing the simulated and observed plant states, the Digital Twin can detect early signs of stress related to insufficient water or nutrient availability. When such conditions are identified, the system activates feedback control actions aimed at adjusting irrigation parameters, thereby increasing the supply of water and nutrients to the plant. This closed-loop architecture enables proactive stress mitigation and supports more efficient and adaptive greenhouse management strategies. The proposed approach was validated through a case study involving tomato plants (Solanum lycopersicum) cultivated in a controlled greenhouse environment.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


