Engineered geothermal reservoirs usually include fractures, either in the form of discrete fracture(s) or highly fractured rock bodies. These fractures provide the main conduit for fluid transport and interface for solid-fluid heat exchange. The aperture width and transmissivity of rock fractures in a geothermal reservoir evolve remarkably during reservoir stimulation and heat production, owing to combined thermal-hydrological-mechanical (THM) mechanisms. A competent reservoir simulation code should faithfully model fractures’ responses to various THM factors as well as the effects of such fracture responses on reservoir behavior and field observables. Through the US Department of Energy (DOE) Geothermal Technologies Office (GTO)’s geothermal code comparison project, we have designed three problems to evaluate geothermal simulation codes’ ability in these aspects. The current paper documents the problem design and analyzes the results collected in this effort. The first problem involves a poroelastic response to water injection into a geothermal reservoir, and is loosely based on recent observations from well RRG-9 at the Raft River EGS demonstration in southern Idaho. The observed reservoir behavior includes a strong non-linear response between the injection rates and pressures over a multiple day, variable rate injection tests. The behavior is simplified to make an initial test case, robust yet reasonable, with the permeability of an inferred fault zone being assigned by an exponential function based on pressure. The model successfully solves the geomechanical governing equations and provides results for development of a consistent dialogue between reservoir engineers, hydrogeologists, and other team members. The second problem considers the responses of a single planar fracture in the rock based on the first experimental EGS in the U.S. at Fenton Hill, Phase I. The THM responses of this fracture during a 24-day injection and production period are considered. The rock mechanics model component includes the thermo-elastic response of the self-propped fracture layer coupled to the thermal model of the reservoir during coolant injection. The goal of the simulation study is to match measured production temperature at the extraction point and pressure loss across the fracture for the 24-day time period of constant fluid injection rate at Fenton Hill. A question is raised as to why the constant-extension fracture model of the first 24 days cannot match the measured temperature and pressure data under a doubled coolant injection flow rate between days 25 through 75 also measured at Fenton Hill. The need for a self-opening, self-propped fracture model is identified, leading to the topic of a future challenge problem. The third problem entails the calculation of ground surface deformation caused by a pressurized subsurface fracture. Surface deformation, usually measured using interferometric synthetic aperture radar (InSAR) or tiltmeter, provides an important means to infer reservoir condition, particularly fracture extent development. A rectangle-shape fracture with various dipping angles (0°, 45°, and 90°) is considered, and both 2D and 3D solutions are compared. More than ten simulation codes from various US institutions participated in the comparison. The compiled results provide a timely overview of the state-of-the-art in modeling fracture-opening phenomena in engineered geothermal systems.

Press the Back button in your browser, or search again.

Copyright 2015, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.