Intelligent prediction method for meandering river sedimentary microfacies

Abstract

The invention discloses an intelligent prediction method for meandering river sedimentary microfacies, which comprises the following steps of: extracting waveform, amplitude and wave impedance data indifferent frequency band ranges, and generating five earthquake combination characteristics according to the protocol that a given time window is divided into three sections to carry out combinationcharacteristic on the earthquake attribute of a current point; constructing a machine learning sample library through selection of an optimal time window; according to the method, by means of a randomforest MDI method, evaluating the sample features, and performing fuzzy probability prediction on the microfacies of the optimized sample features through a random forest algorithm, then performing overlapping display on the high-probability results of each microphase, thereby acquiring a prediction result of the microfacies. Compared with the prior art, the method has the advantages that the problem that the distribution rule of each sedimentary microfacies on a plane cannot be accurately reflected due to a large well distance is solved, a sedimentary microfacies prediction model is established through sedimentary microfacies sample data construction and a random forest machine learning method, and intelligent prediction of the sedimentary microfacies of the meandering river is realized.

Description

technical field [0001] The invention relates to the field of geophysical exploration seismic interpretation and comprehensive research, in particular to an intelligent prediction method for meandering river sedimentary microfacies. Background technique [0002] Sedimentary microfacies are the most basic sedimentary units formed under basically consistent depositional conditions, and their types and spatial distribution have always been one of the important research contents of oil and gas exploration and development. Through the study of sedimentary microfacies, we can understand the distribution and evolution of framework sand bodies in each depositional event, and predict the heterogeneity and changes inside the reservoir. At the same time, sedimentary microfacies also affect the horizontal distribution of remaining oil. main factor. [0003] The traditional sedimentary microfacies prediction method mainly studies the sedimentological characteristics of coring wells to ac...

AI Technical Summary

Problems solved by technology

At present, the method of seismic waveform clustering is often used to classify seismic facies, and then determine the plane distribution of sedimentary microfacies after calibration of sedimentary microfacies at well points. This method only uses seismic information and lacks well control. , the prediction accuracy is low, and it is often only used for the identification of sedimentary subfacies

Images