Turbidite sandstone reservoir modeling method

Abstract

The invention relates to a turbidite sandstone reservoir modeling method, and the method comprises the specific steps: S1, performing the analysis of the geometric elements of a turbidite sandstone and the building of a prototype model; S2, performing the inversion of a well-to-seismic integration turbidite sandstone, converting the conventional continuous amplitude and frequency attribute into adiscrete attribute code to obtain obtining a training image, correcting the distribution mode of discrete attributes through combining with the prototype model, readjusting a time window for an errorsegment of the training image to extract the attributes, performing the repeated optimization, and enabling the discrete attributes to be matched with the prototype model; S3, performing the modelingof a turbidite sandstone multi-point geological reservoir, building a turbidite sandstone reservoir geological model of a research area, performing the comprehensive utilization of the dynamic data ina production dynamic database, performing the error analysis for an unreasonable simulation region, repeatedly adjusting the scheme for optimizing the training image, and building a verifying model combining the static information with the dynamic information. The method is good in modeling effect of sedimentary microfacies, can achieve the quantitative prediction and precise description of the turbidite sandstone, and achieves the accurate representation of the spatial distribution of the turbidite sandstone.

Description

technical field [0001] The invention belongs to the technical field of oil and gas exploration, and relates to beach bar sand body prediction technology, in particular to a turbidite sand body reservoir modeling method. Background technique [0002] Geostatistics was initially applied in the mining industry, mainly to solve various reserve calculation and error estimation problems in the whole process of deposit survey and exploration, mine design and mine exploitation. Later, it developed rapidly in the petroleum industry and was mainly used in reservoirs. Representation and modeling. Traditional modeling methods mainly use the variogram to describe the correlation and variability of geological variables, and then the variogram can only grasp the correlation between two points in space, that is, under the premise of second-order stationary or intrinsic assumptions The correlation between any two points in space makes it difficult to characterize complex spatial structures ...

AI Technical Summary

Problems solved by technology

This method requires high consistency between seismic training images and well information training images. However, the multi-solution of seismic data determines the reliability difference between seismic information and well logging information, so it is difficult to obtain accurate training images.

(2) Except for fluvial facies, the application of multi-point geostatistics in other sedimentary facies is seldom

The traditional stochastic modeling method based on variogram is difficult to accurately characterize its spatial distribution pattern

Images