A Comprehensive Method for Diverter-Performance Evaluation during Stimulation of Long-Interval Heterogeneous Reservoirs: A Case Study

Abstract

Summary The accomplishment of matrix stimulation in highly contrasted permeability reservoirs is critically dependent on diversion. Consequently, assessment of the diversion performance is a key to determine the success of stimulation. However, there are still doubts on the evaluation of diversion effectiveness, especially in long-interval heterogeneous reservoirs. When a diverter enters the formation, a hump in the surface pressure curve is usually expected. Then, it can be interpreted as supporting evidence for diversion. However, this is a simplification of the fluid-diversion process. It could be possible that a hump is not observed during a diversion stage, although it is effective. Therefore, what should be done? To overcome this challenge, we propose a more-accurate diversion-evaluation method and validate it with available matrix-stimulation data. Three methods were introduced in the literature to evaluate matrix-stimulation performance: Paccaloni, Prouvost, and Chan [inverse injectivity (i.e., Iinv)] methods (Prouvost and Economides 1987, 1989; Paccaloni and Tambini 1993; Chan et al. 2003). The latter is easy to use and accurate, which accounts for transient flow effects. In this paper, the inverse injectivity method is modified and validated with the real data of two matrix-acidizing operations in a gas/condensate field. The performed modifications in the evaluation process include a bottomhole-pressure-calculation procedure, which is validated with available drillstem-test (DST) matrix-stimulation data, and simultaneous utilization of Iinv and its derivative plot. Humps in the Iinv plot, which can be interpreted as diverter performance, are sometimes so small that it is difficult to distinguish the diverter effect from possible noises in the data. Here, the derivative plot of Iinv is used as a complementary tool to improve the interpretation process. Results indicate that for both wells in this study, the modified Iinv shows clear humps when diverters enter the reservoir. In addition, exactly when Iinv builds up, a sign change in the derivative plot is observed. This shows that these two parameters have a confirming behavior. Finally, pre/post-stimulation production data were used to practically prove the calculations behind the method. Here, the target of design was to divert stimulation fluids to the low-permeability bottom layer because it was both a high-pressure and high-hydrocarbon reserve. Per production-logging data, the majority of production before stimulation was originated from a sublayer. In the first operation, with the rare appearance of surface pressure humps, Iinv and its derivative showed satisfactory outputs of diversion occurrence. After stimulation, production logging confirmed the diversion of flow and nearly uniform production across the targeted interval. Hence, this indicates that the modified method accurately demonstrates the performance of the diversion system in acidizing operations with long perforated intervals, even if there is a rare distinct pressure hump in the surface. Therefore, this could be adapted either for cases where there is no access to the production logging or for the cases in which the hump in surface pressure is not observed.