Abstract

Interacting with the pre-existing natural fractures, multi-stage hydraulic fracturing along with long horizontal well creates complex fracture networks in gas-shale reservoirs. Field data indicate that only a small fraction of the hugely injected fracturing fluid can be recovered during the clean-up period, especially when the well undergo an extended shut-in period. Therefore, it is essential to investigate the interaction between the non-recovered fluid and the fracture to understand the behavior of fluid-loss and production performance.A series of mechanistic simulation models consisting of hydraulic fracture (HF) and natural/secondary fracture (NF) were constructed to simulate imbibition, fluid re-distribution, and flowback during well shut-in and cleanup periods. Firstly, the established 3-D model system was used to investigate the impact of shut-in immediately after the production and a certain period of flowback. Furthermore, sensitivity analysis was subsequently performed with corresponding stochastic 2-D model, which systematically quantified the impact of fracture uncertainties in 1) fracture complexity and, 2) fracture distribution.The results indicate that, after incorporated with the imbibition hysteresis, water saturation of fracture is close to 1 even if the well has been shut in for a long period before flowback, and the saturation increases with the extending of shut-in during flowback. In addition, the rate fluid imbibition into the reservoir is extremely slow, and the invasion depth is less than 3 m after 100 days’ shut-in. It is also concluded that shut-in duration is not proportional to gas production when we divide the shut-in situation into two cases. Therefore, it is inappropriate to judge that the longer the shut-in period is, the more favorable the production will be. For the fracture networks, we linked the density of NF with the half-length of HF to investigate the fracture complexities. The results show that more complicated fracture will greatly enhance gas production and reduce water recovery efficiency (flowback rate). Finally, different distribution patterns have different impact on well performance, but, the overall influence is much less than that of complexity.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call