A random forest-based shear wave prediction method, system, and device

By constructing a shear wave prediction model using well logging attribute parameters and rock elastic parameters based on a random forest method, the problem of limited applicability of shear wave prediction is solved, and high-precision shear wave prediction under different geological conditions is achieved.

CN117008195BActive Publication Date: 2026-06-26CHENGDU UNIVERSITY OF TECHNOLOGY
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
CHENGDU UNIVERSITY OF TECHNOLOGY
Filing Date
2023-07-19
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing shear wave prediction methods have limited applicability and cannot be widely applied in different regions. In particular, due to the complex pore structure in carbonate reservoirs, the commonly used Biot-Gassmann and Castagna methods cannot be applied to all regions.

Method used

A random forest-based approach was adopted. By acquiring well logging attribute parameters, parameters with a correlation coefficient with shear waves greater than a set threshold were selected to construct a shear wave prediction parameter set. Combined with rock elastic parameters, a sample dataset was constructed, and a random forest regression model was established to predict shear waves.

Benefits of technology

It achieves broad applicability of shear wave prediction, improves prediction accuracy, and is applicable to shear wave prediction under various geological conditions, especially in unconventional tight reservoirs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117008195B_ABST
    Figure CN117008195B_ABST
Patent Text Reader

Abstract

The application provides a kind of random forest-based shear wave prediction method, system and equipment, it is related to oil and gas exploration field, the method comprises: obtaining well logging attribute parameters;The well logging attribute parameters include compensated acoustic wave, density, natural gamma, longitudinal wave and porosity;Select the well logging attribute parameter in the well logging attribute parameter, and the well logging attribute parameter is related to the coefficient greater than the set correlation coefficient threshold value, to construct the shear wave prediction parameter set;Rock elastic parameters are obtained, and according to the shear wave prediction parameter set and rock elastic parameters, sample data set is constructed;The sample data set is divided into training data set and test data set;According to the training data set, random forest regression model is constructed;According to the random forest regression model, the shear wave of current area is predicted.The application is more widely applicable.
Need to check novelty before this filing date? Find Prior Art