A shale reservoir compressibility evaluation method based on support vector machine technology

Abstract

The invention relates to a shale reservoir compressibility evaluation method based on support vector machine technology. It includes the following steps: (1) Quantitative analysis of mineral components and uniaxial compression test are carried out first; (2) Logging characteristic parameters used to identify fractures are extracted; (3) Maximum and minimum values ​​of different layers are calculated by combined spring model (4) Establish the compressibility evaluation model; (5) Establish the compressibility evaluation sample space of the discrete layer strata, and compare the compressibility evaluation results with different response logging curves. Correlation analysis of the relationship between them. The beneficial effects are: the present invention more fully considers the factors influencing the compressibility, and establishes the compressibility model of the shale gas reservoir from the three aspects of the brittleness index, the degree of natural fracture development and the stress sensitivity, which makes up for the existing evaluation method's Insufficient consideration of this factor better reflects the comprehensive ability to distinguish complex fracture networks formed during volume fracturing of shale.

Description

Technical field[0001]The present invention relates to a very normally oily gas field development, in particular to a shaft reservoir compressibility evaluation method based on support vector machine technology.Background technique[0002]With the rapid development and progress of society, it is very urgent to find resources other than conventional coal, oil and natural gas resources. The rock gas has the characteristics of rich reserves, extensive distribution, long production period. The huge success of commercial mining in the US shale gas in the past decade has aroused extensive attention from other countries in the world. It is estimated that my country's rock gas resources are widely distributed in Sichuan, Erdos, Tarim, Junggar, Songliao, etc., and estimated mining reserves up to 36 trillion cubic meters. Therefore, efficient and scientific exploration and development shale gases have become a national strategy to ensure our energy security and achieve economic stability and rap...

AI Technical Summary

Problems solved by technology

However, no scholars have applied it to the calculation of reservoir compressibility evaluation at present. Using the strong nonlinear learning ability of support vector machine, the relationship between well logging data and indoor petrophysical experiment data is deeply excavated, and the whole well interval can be calculated. The continuous section interpretation of pressure will establish a new method and open up a new approach for the optimization of reservoir oil and gas stimulation

Images