Data-Driven Digital Twin Based Energy Flexibility Investigation for Commercial Greenhouse Production Process

Commercial greenhouses require sophisticated energy management strategies to ensure optimal plant growth while minimizing operational costs. The ability to adapt energy usage flexibly in response to varying market prices and environmental conditions is crucial. This paper introduces a data-driven digital twin, incorporating discrete event simulations and symbolic regression, to model and optimize real-world processes within a commercial greenhouse. This approach enables precise control and adjustment of energy usage, enhancing flexibility. A case study of one of Denmark's largest commercial greenhouses is applied to demonstrate the digital twin's applicability and effectiveness. Implementation of the digital twin significantly enhanced the greenhouse's energy flexibility, allowing for adaptive energy consumption that aligns with fluctuating energy prices and availability without compromising plant growth. The results illustrate that digital twins can substantially improve energy flexibility, providing a valuable tool for greenhouse operators to optimize energy usage dynamically.