Thermal Effects of Reinjection in Geothermal Reservoirs With Major Vertical Fractures

Abstract

Introduction Reinjection of spent geothermal brines is the most attractive method for their disposal. Field experience and theoretical work have shown that properly designed injection systems can avoid premature thermal interference at production wells and may serve additional purposes, such as maintenance of reservoir pressures and enhancement of ultimate energy recovery. There has been some controversy in the literature regarding benefits and drawbacks of reinjection. At the present time it seems clear that the issues are largely site-specific, and that reservoir response to reinjection is strongly dependent on formation properties and the thermodynamic state of the reservoir fluids. Whether or not reinjection at a given site is deemed desirable from the standpoint of reservoir management, it appears that for environmental reasons most future geothermal projects will require full reinjection as a method of waste disposal. There is a need for methods of designing and testing reinjection systems to avoid premature breakthrough of colder fluids at the production wells. production wells. An advisory panel convened by the U.S. DOE identified the development of techniques for monitoring, prediction, and control of the migration of injected prediction, and control of the migration of injected waters as an urgent priority. Depending on the geological characteristics of a geothermal reservoir, this can be a rather difficult task. For reservoirs approaching the idealization of a porous medium of uniform thickness, porosity, and permeability, analytical solutions are available for estimating the migration of thermal (cold) fronts away from injection wells. However, most high-temperature geothermal reservoirs are situated in fractured volcanic rocks. There is plenty of field evidence showing that in fractured reservoirs injected water can migrate rather rapidly to production wells. The possibility of rapid thermal breakthrough along preferential pathways is of obvious concern to field developers. In the next section we review some field evidence for fast paths, and discuss possibilities and limitations for determining the thermal characteristics of fast paths by nonthermal means. We present idealized models of fast paths, and examine their applicability to field situations. paths, and examine their applicability to field situations. It is emphasized that tracer tests and pressure transient tests can provide only incomplete and ambiguous information on the thermal characteristics of fast paths. More reliable predictions of thermal interference are possible from nonisothermal injection tests. Numerical simulations incorporating approximate analytical solutions then are used to examine thermal breakthrough and recovery in vertical fractures. JPT p. 1567

TOUGH2 User's Guide Version 2

“…The present problem is designed to study thermal interference along such paths, by modeling nonisothermal injection into and production from a single vertical fracture, as illustrated in Fig. 23 (from Pruess and Bodvarsson, 1984). The fracture is bounded by semi-infinite half-spaces of impermeable rock, which provide a conductive heat supply.…”

TOUGH2: A general-purpose numerical simulator for multiphase fluid and heat flow

“…Time steps can be automatically adjusted (increased or reduced) during a simulation run, depending on the convergence of the iteration process. The linear equations arising at each iteration step are solved with the MA28 package from the Hawell program library, which implements a sparse version of LUdecomposition and backsubstitution (Duff, 1 9 7 7 ) : The accuracy of MULKOM has been tested by comparison with many d&rent analytical and numerical solutions, and with results from laboratory experiments (Pruess and Bodvarsson, 1984;Narasimhan, 1985: Verma, 1986 Different fluid mixtures can therefore be simulated with the same flow module, the thermophysical properties (or "PVT properties") of the specific fluid mixture of interest being provided by an appropriate "EOS" (equation-of-state) module.…”

“…designed to study thermal interference along such paths, by modeling nonisothermal injection into and production from a single vertical fracture, as illustrated in Fig. 16 * (from Pruess and Bodvarsson, 1984). The fracture is bounded by semi-infinite halfspaces of impermeable rock, which provide a conductive heat supply.…”

Hot Dry Rock Research A Compendium of Publications October 1989 - September 1990

“…Nicol and White (1983) proposed using tracer data in conjunction with the infinitely spaced fracture model, while Robinson and Jones (1987) used tracer data in conjunction with a volumetric heat extraction model. Pruess and Bodvarsson (1984) discussed the use of tracer data in predicting thermal interference from reinjection wells in geothermal fields. In this paper we use the available tracer data with the parallel fractures model, to see if any support can be given to the results obtained above.…”