Integrated conditional global optimisation for discrete fracture network modelling

This paper presents a methodology to simulate discrete fracture networks for naturally fractured reservoirs by combining statistical and spatial analyses, object-based modelling and conditional global optimisation. The methodology examines and utilises both continuum and discrete fracture information, such as spatial distribution of fracture density, statistical and geostatistical distributions of fracture size and orientation. The output is a network of discrete fractures, with their corresponding details of location, size and orientation. The methodology is illustrated by a case study on the surface fault system in New York region. The results show that it is able to produce discrete fracture network that match closely to the target fault map, even in the case where data are limited. The results show that it is also able to improve results of several recent fracture models, such as integrated stochastic simulations as well as grid-based simulations.