Improved Reservoir Simulation With Seismically Derived Fracture Models

Abstract

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractHistory matching and simulation of naturally fractured reservoir is a recurring challenge to many oil and gas companies seeking to manage and develop fractured reservoirs. Several techniques have been applied in the past to match past production and pressure history that have been proven unreliable. This paper describes a methodology to improve the simulation of fractured reservoir using seismically driven reservoir characterization. The methodology presented in this paper uses the integration of geophysical, geologic, and engineering data simultaneously to improve the reservoir description. At the root of the reservoir characterization lays the more and more accurate seismic data collected on most of the reservoirs around the world. The initial use of this seismic information is made possible through high-resolutioninversion and spectral imaging. These two processes allow a better imaging of key reservoir properties that have an important impact on fracturing. Based on this seismically driven reservoir characterization, the reservoir properties necessary as inputs to the reservoir simulator, i.e. fracture porosity and permeability, are generated using artificial intelligence tools and core measuremenst as fracture indicators. The usefulness of the derived seismic attributes is illustrated on a specific reservoir where a new well was recently drilled. The drilling results indicate that the derived seismic attributes can be used successfully to locate highly fractured areas. Using the generated seismic attributes in an integrated fracture modeling approach allows for a better modeling of the "plumbing" of the reservoir through a correct estimation of the fracture permeability and porosity. These improved fracture properties lead to a history match of the well performances. Examples of such history matches are given for illustration purposes.