Development of inverse modeling techniques for geothermal applications
Stefan Finsterle and Karsten Pruess
Proceedings, DOE Geothermal Program Review XV, San Francisco, CA, March 24-26, 1997, p. 2-46 – 2-54
Lawrence Berkeley National Laboratory, Earth Sciences Division
University of California, Berkeley, CA 94720
Abstract. We have developed inverse modeling capabilities for the non-isothermal, multiphase, multicomponent numerical simulator TOUGH2 to facilitate automatic history matching and parameter estimation based on data obtained during testing and exploitation of geothermal fields. The iTOUGH2 code allows one to estimate TOUGH2 input parameters based on any type of observation for which a corresponding simula-tion output can be calculated. In addition, a detailed residual and error analysis is performed, and the uncer-tainty of model predictions can be evaluated. One of the advantages of inverse modeling is that it over-comes the time and labor intensive tedium of trial-and-error model calibration. Furthermore, the estimated parameters refer directly to the numerical model used for the subsequent predictions and opti-mization studies. This paper describes the methodol-ogy of inverse modeling and demonstrates an applica-tion of the method to data from a synthetic geother-mal reservoir. We also illustrate its use for the optimization of fluid reinjection into a partly depleted reservoir.