Abstract
Kashagan field contains complex geologic features with complex flow phenomena near bottomholes, e.g., well productivity index (PI) and injectivity index (II) are time-dependent (well scaling, workover, and stimulations etc.) and/or rate-dependent (RD geomechanics/turbulence). Using the built-in functions of commercial simulators cannot give satisfactory solutions for the field. This paper proposes a systematic approach for bottomhole pressure history matching that is automated and capable to handle rate-dependent PI(RDPI) for producers and rate-dependent II(RDII) for injectors.
After achieved acceptable history match qualities on static pressure/GOR/Water-cut for the Kashagan field, the next step of the history match process was to do bottomhole pressure (BHP) match. There were two major challenges during BHP history match. The first one was time-dependent skin caused by bottomhole environment changes, e.g., well scaling, workovers, and stimulations. Traditionally, a time-consuming, trial-and-error process was used to match BHPs. The second challenge was well RDPI and RDII that may cause significant BHP mismatches in trend. RDPI for producers can decrease as production rate increases while RDII for injectors can increase or decrease as injection rate increases depending on in a geomechanic flow regime or turbulent flow regime. In the most commercial simulators, non-Darcy’s flow function for RDPI is only available for gas producers rather than oil producers while RDII cannot be handled satisfactorily for gas injectors. In this situation, the team created an automated skin adjustment Python script to help BHP HM time reduction. Meanwhile, the team also created new RDPI/RDII non-Darcy’s equations and Python scripts implemented as an additional build-in function. Using the Python scripts for the automated skin adjustment, RDPI, and RDII has significantly improved BHP match efficiency (minutes vs days/weeks) and accuracy (satisfactory trends).
