Research On Modeling Of Continuous Random Fracture Network Based On Seismic Attributes

The detection and characterization of strata fractures are very important for the positioning of exploration wells,the design of drilling directions,and the selection of hydraulic fracturing and perforation positions.Due to the stress fields applied on underground rock formations in multiple periods of geological tectonic movements,the fracture system of a reservoir often becomes very complicated,and its exploration,detection and characterization are extremely challenging.At present,the oil and gas exploration industry’s modeling of reservoir fracture networks is limited to describing isolated discrete fractures,which cannot reflect the characteristics of connectivity between fractures.In addition,due to the blind areas between the effective fracture scales that can be covered by various exploration methods,some scale fractures cannot be effectively described.To solve these problems,this paper has carried out research on the modeling method of reservoir random fracture network based on reflection seismic attributes,and successfully developed a set of random continuous fracture network modeling technology,which can meet the demand for random modeling of fracture-scale blind areas.To extract fracture parameters required for the modeling from seismic data,the AVAZ fracture parameter inversion technique based on pre-stack seismic data,and a fracture parameter extraction scheme based on post-stack seismic data,are developed.On successfully completing the development of the AVAZ fracture parameter inversion software,in-depth modelling analysis on the limitations and inherent errors of the AVAZ fracture parameter inversion method have been performed.It was found that the errors of the fracture direction and fracture strength obtained by AVAZ inversion are quite different.In the range of 15°～35° incident angle,the error of crack direction is less than 1°,while the error of crack strength is as high as 461%.These suggeste that the fracture azimuth parameters inverted by AVAZ are effective,but the fracture strength parameters are not suitable for direct use in random fracture modeling.The development of the fracture parameter extraction scheme based on post-stack seismic data realizes the whole process from the qualitative analysis of post-stack seismic attributes to the quantitative interpretation of fracture parameters.This provides a method and technical means to obtain the required parameters for the random fracture modeling.With the above tchniques,a continuous fracture network modeling method based on seismic attributes was finally developed.Using the fracture parameters from seismic data as a constraint,a random discrete fracture network model is first established,and then the discrete fractures are connected to form a random continuous fracture network.The feasibility and correctness of the method have been verified by numerical experiments.Finally,a comprehensive modeling application test was done using actual industrial seismic data.A random fracture network model of the target layer within the work area and the random network model of small-scale fractures around drilling sites in the area were obtained.This verified the feasibility and applicability on industrial data of the methods and techniques developed in this paper.