As the United States (US) solar capacity continues to expand, the demand for land has intensified, leading to increasing land use competition between solar photovoltaic (PV) installations and agriculture. This trend, along with the urgency of the climate crisis and the growing demand for food and energy, points to a potential solution of combining crop production and solar energy generation on the same land, known as agrivoltaics. However, solar developers currently lack the essential tools required to design and manage solar facilities that optimize both electricity generation and crop production. In this study we develop a digital twin to enable simulation and predictive control for single axis tracking solar facilities, improving system efficiency by considering electricity generation and crop production. Central to this study is the innovative use of both standard (i.e., solar panel remains perpendicular to the sun) and reverse (i.e., solar panel remains parallel to the sun) tracking schemes with an adaptive feedback control mechanism. This system targets optimal daily light integral (DLI) for cultivation between the panels based on predicted weather data and real-time sensor monitoring. The algorithm automatically adjusts standard and reverse tracking schedules to meet the DLI requirement; subsequent changes are then implemented in the physical panel system and synchronized in the Digital Twin user interface for visualization. Through extensive testing and design iterations, this solar farm digital twin model will equip developers with the tools to maximize solar energy generation while supporting crop production.