Analysis and simulation of polymer injectivity

Polymer flooding is one of the most successful and mature chemical EOR methods. Qualification programs, which often include injectivity and/or performance pilots, are a prerequisite to reduce risk in further implementation. The former being typically pressure fall-off tests and the latter single well tracer tests (SWTT). However, limited work has been conducted on assessing interpretability and uncertainty associated with these tests. This is something of a paradox since investments to perform these pilots are large and implications of successful versus failed pilots are enormous. In field tests, only injection bottom-hole pressure (BHP) and volumetric injection rate are the available parameters to determine polymer in-situ rheology. In addition, analysis of pressure fall-off tests for polymer injections are far more complex compared to water and gas due to the non-Newtonian behavior of the polymer. The following question remains: what rheological information is actually obtainable based solely on BHP? Moreover, how sensitive is polymer rheology estimation to uncertainties in pressure measurements? Can high uncertainties in pressure data completely distort the rheological information obtained? Lastly, is pressure response from the near wellbore region sufficient to obtain accurate estimations of polymer injectivity in porous media? The aforementioned issues are investigated herein by modelling pressure fall-off tests using the STARS simulation tool (CMG). Generic field data were used to design a near-well sector and a high molecular weight partially hydrolyzed polyacrylamide (HPAM) was used as polymer reference. Here, the influence of different polymer rheological behaviors on BHP and BHP-transient was analyzed and identified. Influence of pressure measurement noise on polymer rheology was evaluated using the automated history-matching tool CMOST (CMG) for lab scale simulations. A recently developed history match method, based on pressure measurements from internal pressure taps distributed between injector and producer, was used. Even though results showed deviations from a generic (base case) rheology curve for individual rates, the arithmetic average of these curves displayed negligible deviation from its generic behavior below a threshold noise level. Moreover, simulations show that polymer injectivity is solely dependent on polymer behavior in the near wellbore region. Finally, two different flood experiments using the same HPAM polymer were history matched and results confirm the conclusions suggested in the simulation study. This paper provides additional interpretational anchoring for pressure fall-off test for polymer injectivity assessments. Additional methods and insights developed in this paper should both improve experimental design and reduce implementation risk for polymer flood projects.