Dynamic Fracture Characterization Using Multiphase Rate Transient Analysis of Flowback and Production Data

Abstract

Abstract This study presented a comprehensive method for characterizing reservoir properties and hydraulic fracture (HF) closure dynamics using the Rate Transient Analysis (RTA) of flowback and production data. The proposed method includes straight-line analysis (SLA), type-curve analysis (TCA), and model history matching (MHM), which are developed for scenarios of two-phase flow in fracture, stimulated reservoir volume (SRV), and NSRV domains. HF closure dynamics are characterized by two key parameters: pressure-dependent permeability and porosity controlled by fracture permeability-modulus and compressibility. The above techniques are combined into a generalized workflow to iteratively estimate the five parameters (four optional parameters and one fixed parameter) by reconciling data in different domains of time (single-phase water flow, two-phase flow, and hydrocarbon-dominated flow), analysis methods (SLA, TCA, MHM), and phases (water and hydrocarbon phase). We used flowback and production data from a shale gas well in the US to verify the practicability of the method. The analysis results of the field cases confirm the good performance of the newly developed comprehensive method and verify the accuracy in estimating the static fracture properties (initial fracture pore volume and permeability) and the HF dynamic parameters using the proposed generalized workflow. The accurate prediction of the decreasing fracture permeability and porosity, fracture permeability-modulus, and compressibility demonstrates the applicability of the workflow in quantifying HF dynamics. The field application results suggest a reduction of the fracture pore volume by 30%, and a reduction of the fracture permeability by 98% for shale gas well. Instead of a single analysis method for RTA, this paper proposed a comprehensive analysis method that includes SLA, TCA, and MHM. The interpretation results of the three analysis methods are mutually constrained, which can reduce the non-uniqueness problem of inversion. Compared with the others fracture characterization workflow that need fixed input and output parameters. This proposes general workflow not only completely characterizes the fracture closure dynamics but also can select the unknown parameters (to be determined) according to the actual scenarios of a well and the demands of reservoir engineers.