High-Fidelity Modeling Optimizes Infill-Well Development in Shale Assets

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 230635, “Optimizing Infill-Well Development in Shale Assets With High-Fidelity Modeling of Fracture-Driven Interactions,” by Osama Hasan Khan, SPE, Olga Kresse, SPE, and Xiaowei Weng, SPE, SLB, et al. The paper has not been peer-reviewed. _ This paper presents a high-fidelity, reservoir-centric stimulation-to-production modeling workflow to prevent fracture-driven interactions. To prevent detrimental fracture-driven interactions (FDIs), which may reduce production rates and ultimate recovery from infill (child) wells or negatively affect parent wells, a high-fidelity, reservoir-centric stimulation-to-production modeling workflow is essential. This approach enables early quantification of FDI risks and supports the design of effective mitigation strategies. In this paper, a multiphysics modeling workflow is applied that integrates fit-for-purpose physics simulators to provide a practical approach to infill-well planning without compromising fidelity. Integration of Physics Engines Into FDI Modeling Workflow The workflow starts with a fracture-propagation simulation of the first-generation parent wells using an unconventional fracture model (UFM). At this stage, the fracture geometry can be calibrated to field observations. The calibrated fractures can be seamlessly input to the reservoir simulator using the embedded-discrete-fracture-model (EDFM) approach for production simulation and history matching until the next generation of wells is to be stimulated. The stress state of the reservoir before stimulation of the child wells is then computed by the geomechanics simulator. The updated stress field is used as input to the UFM to simulate fracture propagation in the child wells. The workflow can be repeated for any number of parent-/child-well generations. The different steps in this workflow coincide with real field events; the sequential nature is, therefore, useful for calibration with data acquired at each step. In this work, the UFM has been enhanced further to model the FDIs. When the fracture networks created during child-well stimulation do not directly connect with the existing parent-well fractures, the interaction between them mainly happens through the stress shadow effect. The modeling of fracture reopening has been developed and implemented in this work. The enhancements include modeling the propagation of open fracture tips along existing propped fractures and fluid diffusion ahead of the propagating tips into the closed network. The enhancements are detailed in the complete paper.