Rock reservoir structure characterization method, device, computer readable storage medium and electronic equipment

Abstract

The application discloses a rock reservoir structure characterization method, device, computer-readable storage medium, and electronic equipment. The method includes: acquiring a three-dimensional seismic data volume of the rock reservoir to be characterized, and decomposing it; Transform the intrinsic mode function components to obtain the time spectrum, and add the time spectra of all components to obtain the time spectrum of the seismic data; through the time frequency components of each time spectrum of the side-hole seismic trace and the synthetic seismic trace of the logging data Cross-correlation screens out the sensitive components with the highest correlation as input features, and performs fuzzy C-means clustering and spatial smoothing on them to obtain seismic facies with set standard division; characterize the rock reservoir to be characterized, and obtain the rock reservoir to be characterized Structural representation. The means, media and equipment can be used to carry out the method. They can solve the problem that the existing data-driven machine learning methods to characterize reservoirs cannot guarantee the validity of input features, clustering reliability and noise resistance.

Description

technical field [0001] The invention relates to a rock reservoir structure characterization method, in particular to a rock reservoir structure characterization method, device, computer-readable storage medium and electronic equipment. Background technique [0002] In the prior art, there is a rock reservoir structure characterization method through empirical mode decomposition and KNN clustering algorithm, which has the following technical defects: different seismic traces have different numbers of IMF components, and the corresponding seismic sections of each IMF component No lateral continuity; no quantitative constraints of actual logging response; each data in the KNN clustering method only belongs to one sedimentary facies, without its degree of membership to each sedimentary facies, which is not conducive to further smooth analysis of clustering results . In the prior art, there is also a rock reservoir structure characterization method through SST and K-Means cluste...

AI Technical Summary

Problems solved by technology

[0003] In view of this, the present invention provides a rock reservoir structure characterization method, device, computer-readable storage medium and Electronic equipment, which can solve the problem that the existing data-driven machine learning method for describing reservoirs cannot guarantee the validity of input features and at the same time ensure the reliability and noise resistance of clustering, so it is more suitable for practical use

Images