It has recently been demonstrated that complex fracture networks (CFN) especially activated natural fractures (ANF) play an important role in unconventional reservoir development. However, traditional rate transient analysis (RTA) methods barely investigate the impact of CFN or ANF. Furthermore, the influence of CFN on flow regime is still ambiguous. Failure to consider these effects could lead to misdiagnosis of flow regimes and underestimation of original oil in place (OOIP). A novel numerical RTA method is therefore presented herein to improve the quality of reserves assessment.
A new methodology is introduced. Propagating hydraulic fractures (HF) can generate different stress perturbations to allow natural fractures (NF) to fail, forming various ANF pattern. An embedded discrete fracture model (EDFM) of ANF is stochastically generated instead of local grid refinement (LGR) method to overcome the time-intensive computation time. These models are coupled with reservoir models using non-neighboring connections (NNCs).
Results show that except for simplified models used in previous studies subjected to traditional concept of stimulated reservoir volume (SRV), in our study, the ANF region has been discussed to emphasis the impact of NF on simulation results. Henceforth, ANF could be only concentrated around the near-wellbore region, and it may also cover the whole simulation area. Obvious distinctions could be viewed for different kinds of ANF on diagnostic plots. Instead of SRV-dominated flow mentioned in previous studies, ANF-dominated flow developed in this work is shown to be more reasonable. Also, new flow regimes such as interference flow inside and outside activated natural fracture flow region (ANFR) are found. In summary, better evaluation of reservoir properties and reserves assessment such as OOIP are achieved based on our proposed model compared with conventional models. The novel RTA method considering CFN presented herein is an easy-to-apply numerical RTA technique that can be applied for characterizing reservoir and fracture information as well as assessing OOIP.