Analysis of the possibility of modeling local problems in hydrodynamic simulators for cyclic steam stimulation

This paper evaluates the capability of the tNavigator software package as a tool for modeling cyclic steam stimulation in the bottomhole formation zone during the development of high-viscosity oil fields. The purpose of the study is to conduct numerical analysis of the parameters involved in the cyclic steam stimulation technology using this package. The simulator is based on the approaches of mechanics of multiphase systems, which proved to be successful in solving the problems of subsurface hydromechanics. The system of equations of mechanics of multiphase systems is solved by the IMPES method. Lax's idea that, for evolutionary equations, a change in a small parameter can lead to different solutions is developed further. To assess the possibility of applying the simulator to solving local problems, 17 variants of the hydrodynamic model with different technological parameters were constructed. In the framework of each simulation, 3 numerical experiments were conducted with intent to identify the physical consistency of the calculated results using the hydrodynamic simulator, as well as to evaluate the degree of influence of changes in the time of steam-water mixture injection into the reservoir and the soak time on the dynamics of cumulative oil production. The results of calculations indicate the suitability of hydrodynamic simulators for solving local problems. Optimal oil production time period was established to achieve maximum cumulative oil recovery. Simulations show a good coincidence between the calculated dynamics of water cut in extracted products obtained with the application of cyclic steam stimulation at the Sho-Vel-Tum field and the field data. It is established that the tNavigator software package reliably reproduces the physical processes which occur during the hot carrier injection into the reservoir, as well as during the hot oil production; however, for the steam soak phase, these processes are reproduced incorrectly because phase transitions are described by a simplified model.