A Discrete fracture–matrix model for pressure transient analysis in multistage fractured horizontal wells with discretely distributed natural fractures

Abstract

Analysis of transient pressure test has become a reliable and powerful tool for petroleum engineers to study the impacts of complex fracture networks on multistage fractured horizontal wells (MFHWs) in unconventional reservoirs. Although lots of models have been established in the past decades, such as well-known dual-porosity model (DPM), dual-permeability model (DKM) and stripe fracture model (SFM), they have been proven unsuitable for modeling discretely distributed natural fractures (NFs).In this study, to overcome the disadvantages of existing models, a Discrete fracture–matrix method based on numerical well testing model (DFM-WTM) is proposed to study the pressure transient responses of MFHW with discretely distributed NFs. In DFM-WTM, local grid refinement method, well model and transmissibility corrections, natural fracture generation and time-step size are improved based on the framework of Matlab Reservoir Simulation Toolbox (MRST) (Lie, 2019), ensuring that pressure transient responses of MFHW with complex fracture networks can be obtained accurately and efficiently. Then, sensitivity analysis is conducted to study the impacts of different fracture properties. Finally, a case study is performed to further demonstrate the reliability and practicality of DFM-WTM.The results show that MFHW with discretely distributed NFs can be divided into 8 flow regimes, including (1) wellbore storage and skin effect, (2) fracture linear flow, (3) bilinear flow, (4) NF–HF fluid supply, (5) second bilinear flow, (6) formation linear flow, (7) pseudo boundary flow and (8) pseudo radial flow. Besides, a “dip” is found in the pressure derivative curve which is caused by the NF–HF fluid supply. Through the sensitivity analysis, it is found that this “dip” is an important signal to identify the impacts of NFs. It can also help us fathom the properties of HFs and NFs because this “dip” can be mainly affected by the number and conductivity of both HFs and NFs.