Author ORCID Identifier:
Date of Graduation
7-2026
Document Type
Thesis
Degree Name
Master of Science in Computer Science (MS)
Degree Level
Graduate
Department
Computer Science & Computer Engineering
Advisor/Mentor
Gauch, John
Committee Member
Gomez, Alejandro Martin
Second Committee Member
Dang, Tuan
Keywords
Greenhouses, Predictive Digital Twins, Physics-Informed Machine Learning, root mean squared error (RMSE)
Abstract
Greenhouses are widely used to create controlled environments for crop production, where indoor climate variables such as air temperature, relative humidity, CO2 concentration, radiation, and soil or substrate conditions directly affect plant growth, yield, energy consumption, and resource use. However, greenhouse climates are highly dynamic because they are influenced by complex interactions among outdoor weather, greenhouse structure, crop transpiration, heating systems, ventilation, screens, fans, irrigation, and other actuators. Building digital twins for greenhouses provides a powerful way to represent, monitor, and predict these climate dynamics. A predictive digital twin can simulate future indoor climate states based on current greenhouse conditions, weather forecasts, crop information, and control actions. In this work, we propose a predictive digital twin based on a model-based machine learning framework that leverages well-established physics knowledge to represent interactions among greenhouse components. Specifically, we formulate the indoor climate dynamics as a graph neural network, where nodes represent greenhouse states or components and edges encode physics-informed interactions. In addition, we apply proportional control to model actuator behavior, allowing the digital twin to predict future climate states under different control actions. We evaluate our models on two key indoor climate variables, namely air temperature and vapor pressure, using coefficient of determination R2 and root mean squared error (RMSE) as performance metrics. The experimental results show that the proposed models achieve reasonable predictive performance, providing a promising baseline for future research on predictive digital twins for greenhouse climate modeling, control optimization, and energy-efficient crop production.
Citation
Tran, H. (2026). Physics-Informed Machine Learning for Predictive Digital Twins in Greenhouses. Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/6329
