| Abstract |
The Rock Mechanics Group at MIT has developed a stochastic fracture pattern model, GEOFRAC. This is based on statistical input on fracture patterns from the field. GEOFRAC has been applied and tested by estimating the fracture intensity and estimated fracture size from tunnel records and from borehole logs. Since its original development, GEOFRAC has been expanded by including an intersection algorithm, a flow model and, most recently, a thermal model. The fracture-, flow- and thermal models have been tested, and a parametric study was conducted in order to check the sensitivity of the output results to the inputs. The aim of this paper is to show the applicability of GEOFRAC to a real case. Data of a geothermal power plant obtained from the energy company, Landsvirkjun, Iceland, are implemented in GEOFRAC. Assumptions are made in order to determine the unavailable parameters to input in GEOFRAC, such as the fracture intensity P32 (fracture area per volume) and the best estimate fracture size E[A]. The assumptions and the results of the simulation are presented in this paper. The analysis shows that GEOFRAC produces reasonably close results to the real geothermal conditions. The estimation of important factors such as flow rate and energy extraction rate are reasonable. However, GEOFRAC has still limitations that need to be assessed. The analysis conducted and reported in this paper allowed our group to detect limitations of our model but as well as its strong points. |