Proceedings, TOUGH Workshop '98, Lawrence Berkeley National Laboratory, Berkeley, CA, May 4-6, 1998, p. 281 - 287

Analysis of boiling experiment
using inverse modeling

Stefan Finsterle1, Cengiz Satik2, and Marilou Guerrero2

1Lawrence Berkeley National Laboratory, Earth Sciences Division
University of California, Berkeley, CA 94720

2Stanford Geothermal Program
Stanford University, Stanford, CA 94305

Abstract. Numerical predictions of geothermal reservoir behavior strongly depend on the assumed steam-water relative permeabilities, which are difficult and time-consuming to measure in the laboratory. This paper describes the estimation of the parameters of the relative permeability and capillary pressure functions by automatically matching simulation results to data from a transient boiling experiment performed on a Berea sandstone. A sensitivity analysis reveals the strong dependence of the observed system behavior on effects such as heat transfer from the heater to the core, as well as heat losses through the insulation. Parameters of three conceptual models were estimated by inverse modeling. Each calibration yields consistent effective steam permeabilities, but shape of the liquid relative permeability remains ambiguous.