Evaluation of time-convolution approach for modelling heat exchange between wellbore and formation during multiphase CO2 injection with T2Well-ECO2M

During CO2 injection operation start-up, strong transient flow occurs. It may lead to sudden changes in CO2 flowing pressure and temperature profiles, and possible phase transition. Moreover, wellbore flow behaviour affects the temperature at which CO2 enters the reservoir posing operational risks such as hydrates precipitation in the reservoir, ice formation around the wellbore, and thermal shocks of the casing. This indicates the importance of a reliable simulation of the coupled wellbore and reservoir flow to design the injection operations. The T2Well-ECO2M coupled wellbore-reservoir simulator allows the modelling of three-phase flow with the possible coexistence of liquid and gaseous CO2. T2Well-ECO2M offers a numerical and a semi-analytical approach to calculate heat exchange between the wellbore and surrounding formation. The full numerical approach is the most accurate, however, it requires building and solving spatial grids with many additional discretisation elements just for the modelling of heat conduction. The semi-analytical approach for wellbore heat exchange is based on the Ramey’s method with or without time-convolution. The numerical and the semi-analytical option with time-convolution approach options were compared with reference to the simulation of dry CO2 injection at constant wellhead rate and enthalpy in a simple 1D radial reservoir containing CO2 and immobile brine. The conceptual model, considered in both approaches, assumes a reservoir at depth of 3067 m with a radial extension of 2000 m. The evolution of flowing pressure and temperature profiles as well as the radial distribution of reservoir thermodynamic conditions are compared. The computation time to solve the time-convolution semi-analytical case was approximately 2.5 % of the time necessary to solve the numerical case with, overall, a good approximation of the numerical solution.