Classification model channel extracting seismic facies clustering analysis method

Abstract

The invention discloses a classification model channel extracting seismic facies clustering analysis method, which comprises the steps of acquiring an initial classification model channel through neural network and particle swarm clustering calculation based on seismic waveform data; acquiring a new classification model channel through correlation analysis based on the initial classification model channel, carrying out clustering analysis on original data by using the new classification model channel, and acquiring a seismic facies classification figure; and predicting distribution of oil gas underground based on the seismic facies classification figure. According to the invention, the classification model channel is optimized through a two-step method, thereby effectively ensuring the stability of a classification result and effective discrimination of classes of neurons, finally realizing objective analysis for the seismic facies, and enabling a large amount of information in the seismic data to be interpreted accurately.

Description

technical field [0001] The invention relates to the technical field of geophysical exploration seismic facies analysis, in particular to a seismic facies cluster analysis method for extracting classified model traces. Background technique [0002] Seismic facies generally refer to facies belts with similar shapes formed by seismic reflection features formed in a specific sedimentary environment. The depositional environment includes marine or continental facies, which can control the distribution and development of reservoirs. Seismic facies analysis technology can characterize the morphology of sedimentary facies from the seismic point of view, and has become an indispensable method in the technology of seismic reservoir characterization and detection. The seismic data obtained by monitoring contains relevant information such as the amplitude, phase and frequency of the waveform, which is the comprehensive response of the underground rock mass, physical properties, fluid p...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is to provide a seismic phase clustering analysis method for extracting classification model traces, which can solve the uncertainty of classification results and the inability to effectively distinguish similar types of neurons The problem

Images