Abstract: As a supplement or alternative metering method for physical metering devices, Virtual Flow Metering(VFM) is more and more widely used in the upstream metering scenario of oil and gas fields. To enhance the accuracy of single well VFM system and improve the convenience of system maintenance, a new hybrid model based VFM method that combines the mechanism model and machine learning model was proposed on the basis of the traditional mechanism model of dynamic multi-phase flow. Meanwhile, the interactive logic between machine learning model of data on well condition and mechanism model of well, as well as the system architecture, was established. Then, the stable and accurate operation of the new hybrid model based VFM system driven by real-time measurement data was realized. The research results show that the hybrid model based VFM system could provide the stable output of each-phase metering with the input of realtime measurement data. Additionally, the field well test data were used to evaluate the new VFM results, which shows that the overall errors of liquid flow and gas flow are less than 5% and 3% respectively, and the metering accuracy is improved to a certain extent. Compared with the virtual metering based on the mechanism model, the new hybrid model based VFM system is more convenient in later maintenance, only with the model setting adjusted according to the change of water cut and gas oil ratio in the well testing data. Further, the metering is applicable to a wider scope. Generally, the research results could provide feasible directions for the research on traditional virtual metering in terms of the integrated development of multiple technologies.