A Connectivity-Graph Approach to Optimising Well Locations in Geothermal Reservoirs

Abstract

In this work we address the problem of optimal location of injection and production wells in fractured-based geothermal reservoirs. The optimisation is based on a distance distribution function, and the length and aperture of pathways between the two wells. The initial locations of the two candidate wells are chosen at random and the fracture pathways between the wells are determined using graph theory concepts. The weighted shortest pathway incorporates the equivalent aperture and total length of pathway elements (i.e., linked fractures). The method is efficient and effective for generating final optimal well locations (as coordinates) and also provides a map of optimality for any given fracture network. The sampling scheme used can incorporate any constraint including technical, topographical and or design. Furthermore, stochastic modelling of fracture networks can be used to extend the use of the proposed method to deal with the uncertainty involved in estimated or simulated fracture networks.