CHARACTERIZATION OF FORMATION FRACTURES WITH MULTICOMPONENT INDUCTION LOGGING BASED ON BIAXIAL ANISOTROPY MODELS: METHOD AND CASE STUDIES

摘要

Borehole multicomponent induction (MCI) logging hasbeen successfully used throughout the last decade forevaluation of various types of anisotropic formations,such as laminated shaly sands and low-resistivityreservoirs, by means of the determined resistivityanisotropy (horizontal and vertical resistivity), dip, anddip azimuth. Recently, a new application of the MCIlogging was reported for characterization of formationfractures through fracture detection and assessment offracture parameters. This new application is a veryimportant step in oil/gas exploration and developmentbecause it can significantly help improve drilling andcompletion decisions made regarding fracturedformations. However, to date, literature publishedrelated to this new application has been minimal.Currently, MCI data processing and interpretationschemes are usually based on a transversely isotropic(TI) formation model for fracture characterization. Theprimary reason is because formations are commonlybedded or laminated, so the TI-model assumption isreasonable under these conditions. In reality, subsurfaceformations frequently contain different types of (naturalor non-natural; open or closed) fractures (or faults),where fractured formations very often displayazimuthal resistivity anisotropy in the bedding plane,which leads to biaxial anisotropy (BA) in formationresistivity (or conductivity). Therefore, the MCI dataprocessing and interpretation based on the TI models isno longer valid in complex formations where BA iscaused by fractures. Moreover, the hydrocarbonexploration and development is currently in need ofnew MCI processing and interpretation methods basedon the BA formation models so that a more accuratedescription of fractured laminated reservoirs can beprovided. It is well known that fractures cansignificantly affect fluid flow in formations and a moreaccurate model of fractures can lead to optimizedoil/gas development and production, particularly inunconventional reservoirs. Based on the BA formationmodels and through exploitation of the equivalencebetween fractured formation models and BA models, apractical integrated method and workflow weredeveloped with multi-array MCI tensor data foreffective determination of formation BA anisotropy (ortriaxial resistivities: Rx, Ry, and Rz), dip, and dipazimuth in fractured formations. Because of thecomplexity of the fracture characterization, a few newfracture identification (FID) functions with acombination of multiple log-data inputs have beenproposed for detection of fracture existence, followedby an approach for quantitative estimation of fractureorientation/azimuth, dip, density, etc. along a wellbore.The newly developed method and workflow wereapplied to synthetic and field data sets for validationand case studies. The application results clearlydemonstrate that more accurate triaxial formationanisotropy and dip can be obtained using the newmethod based on the BA models, compared to the TIprocessed logs. This allows fractures to be moreaccurately characterized by integrating corrected tensormeasurements (e.g., borehole-corrected MCI logs) andmultiple inverted log data, such as two horizontalresistivities (Rx and Ry), vertical resistivity, andformation dips based on the combination of both BAand TI models. Furthermore, this new application alsopaves the way to more accurately evaluate differenttypes of fractures in formations, such as shaly sands,carbonates, and unconventional reservoirs using MCImeasurements and integration of other sensor log data(e.g., imaging and sonic data).