Optimised fracture network model for Habanero reservoir

Abstract

Fracture networks and their connectivity are the principal factors affecting fluid flow in hot dry rock (HDR) geothermal reservoirs. Largely because of the complexity of the problem models of HDR reservoirs tend to be over-simplified using either a very limited number of fractures or an equivalent porous media approach. This paper describes a Markov Chain Monte Carlo (MCMC) conditioning technique for reservoir fracture modelling by taking into account the seismic events collected during the fracture stimulation process. Using the technique, the fracture model “evolves” during the simulation process and eventually converges to a predefined optimal criterion. The proposed method is tested using seismic data collected during the hydraulic fracture stimulation processes of the Habanero wells in Geodynamics’ Cooper Basin project.