Analysis of early-time production data from multi-fractured shale gas wells by considering multiple transport mechanisms through nanopores

Abstract

Analysis of early-time production and flowback data from multi-fractured shale gas wells is of critical importance in the characterization of hydraulic fractures (HF). However, the accurate estimation of HF properties in the current models is challenged by the complexity in multiple transport mechanisms and two-phase flow of shale gas reservoir. This study presents a new semi-analytical method to estimate HF attributes for shale gas wells exhibiting two-phase flow based on straight line analysis. The proposed method considers two-phase infinite acting linear flow (IALF) and boundary dominated flow (BDF) for both HF and matrix domains. In addition, matrix flow considers desorption, diffusion, slip flow, continuum flow, stress dependence rock, and adsorbed water film in the nanopores. A modified material balance equation is proposed to calculate average pressure in the fracture and distance of investigation (DOI) in the matrix by considering gas desorption and diffusion in the matrix domain. The accuracy of the proposed method is tested against numerical results obtained from commercial software (IMEX-CMG). The validation results confirm that the developed method can closely estimate initial fracture volume, permeability, and permeability modulus from early-time production data exhibiting two-phase IALF and BDF regimes. Furthermore, the results indicate that the consideration of desorption, diffusion, and slip flow plays an important role in the modeling of gas transport in shale matrix. The impact of these transport mechanisms on production modeling as well as characterization of HF properties becomes significant when the matrix experiences substantial pressure drops. For the purpose of field application, the proposed method is used to analyze flowback data from a multi-fractured horizontal well. The field results reveal that the calculated fracture properties are consistent between the proposed method and the long-term production data analysis method.