Yunnan is an important production area of fresh-cut flowers in China. The cold chain circulation covers postharvest pre-cooling, cold storage, cold storage transportation and terminal distribution. The existing management mode lacks continuous tracking of the operation status of the cold chain, the node connection efficiency and the change of carbon emissions; the monitoring results are scattered, and the regulation response is lagging, which is difficult to meet the low carbon operation demand under the requirements of high timeliness and high fresh-keeping of fresh cut flowers. Focusing on the entity process of Yunnan fresh cut flowers cold chain, this paper constructed the framework of carbon emission dynamic monitoring and an intelligent regulation system driven by a digital twin. The research content includes the overall architecture design of the system, digital mapping of entity process, multi-source data acquisition and synchronisation of virtual and real, the construction of carbon emission dynamic monitoring model, and the design of intelligent regulation mechanism. In the experimental part, typical cold chain circulation scenarios are selected to test and analyse the pre-cooling, storage, transportation and distribution links. The results show that the system can control the monitoring error of carbon emission of key nodes within 6.8%, the abnormal state recognition rate reaches 93.4%, the early warning response time is reduced by 27.6%, the comprehensive energy consumption per unit batch is reduced by 11.2%, and the carbon emission of transportation is reduced by 9.7%. The results show that the framework can improve the state perception ability, carbon emission monitoring accuracy and regulation efficiency of the fresh cut flower cold chain operation process, which can provide reference for the low-carbon and digital management of Yunnan fresh cut flower cold chain.
