Integrated reservoir flow and geomechanical model to generate type curves for pressure transient responses of a hydraulically-fractured well in shale gas reservoirs

• Stress-dependent correlations are used to compute deformation of shale reservoir.
• Discussions are made regarding impacts of reservoir and geomechanical properties.
• Reservoir properties are estimated using the type curve matching technique.
• History matching is performed to verify the proposed model with Barnett Shale data.

Shale gas reservoirs have ultra-low matrix permeability and consist of matrix and natural fracture systems. Horizontal wells with hydraulic fractures are necessary to economically exploit shale gas reservoirs. Due to these complex features of shale gas reservoirs, understanding pressure transient characteristics is of importance. Through coupling of flow and geomechanical models for well/reservoir/hydraulic fracture systems, a new approach has been established to predict the pressure transient behavior and to generate type curves. To account for geomechanical effects with stress dependent permeability, linear elastic geomechanical models and two stress dependent permeability correlations, exponential and power law correlations, from previous experimental researches were implemented in the reservoir model. A number of cases considering a geomechanical model, exponential correlation, and power law correlation were compared. These processes showed roughly a 1–5% variation of productivity in shale gas reservoirs. Based on the extensive numerical simulations with the updated model, a series of type curves were developed in terms of dimensionless pseudopressure and derivative of dimensionless pseudopressure versus dimensionless time considering various reservoir and fracture properties. Using the type curves as a guide, we then presented flow regimes that are expected for different values of geomechanical properties, number of fractures, length of the fractures, spacing between fractures, and properties of stimulated reservoir volume (SRV). Using the characteristics of new dimensionless parameters and the developed type-curve set, a simple and practical procedure was presented to estimate reservoir properties in multi-fractured horizontal wells. Application of type curve matching to synthetic data was shown to be useful in estimating effective fracture and matrix permeability and porosity. A history match of a field example in Barnett Shale data indicates that the model with stress-dependent properties matches the actual production data better than the model with constant properties. This study provides insight into the characteristics of a hydraulically-fractured horizontal well, and the newly developed type curves yield more unique and accurate results.