Method using internal samples at reservoirs to carry out weighting and compile logging crossplot

Abstract

A method using internal samples at reservoirs to carry out weighting and compile a logging crossplot includes the steps of fully considering the difference of quantity of core analysis samples of the reservoirs in oildom target stratums and the difference between logging information resolution ratio and core analysis sampling resolution ratio, and carrying out weighting processing on testing values of all the core analysis samples in the reservoirs. Through the processing, the method fully reflects the difference of thickness of each reservoir and the internal sample quantity, accords with requirements for sample space in statistics, and obviously improves drawing precision of the logging crossplot.

Description

technical field [0001] The invention belongs to the field of exploration geophysics, specifically belongs to the technical branch of well logging methods, and in particular relates to a method for preparing well logging intersection diagrams by weighting internal samples of each layer. Background technique [0002] With the continuous improvement of oil and gas field exploration and development, the complexity of geological targets and the difficulty of logging interpretation are becoming more and more difficult. Therefore, it is necessary to analyze various geological features and their frequency of occurrence in detail. In this way, not only the geological characteristics of reservoirs with various attributes need to be analyzed in detail, but also supporting technical methods are needed to extract and reflect their attributes. Logging crossplot is a key technology commonly used to characterize reservoir lithology, physical properties, oiliness and other characteristics. ...

AI Technical Summary

Problems solved by technology

[0003] It is assumed that based on the comprehensive analysis of well logging data and mud logging data, there are n reservoirs with core physical property analysis and test data in the target interval in the well in an oil area, and there are m core physical property analysis and test data in these n reservoirs samples, m≥n, then the number of samples in n reservoirs is m 1 , m 2 ,...,m n , then: m 1 +m 2 +…+m n = m, since the conventional logging sampling interval is 8 points / m, that is, the measurement interval of 0.125m, the minimum thickness of the strata that can be distinguished by the conventional logging series data is generally 0.5m, and there is an interface effect, while the actual core analysis small column sample diameter It is 2.5cm, and the length is 3~5cm. It can be seen that the sampling accuracy of the two data does not match. The logging data near the interface cannot reflect the logging parameter values ​​corresponding to the analysis values ​​of each core sample one by one, that is, they cannot be strictly followed. The core physical property analysis data adopts the acoustic logging data corresponding to the depth of single point data to compile the intersection map. How to organically correspond the core physical property analysis data and logging data to achieve the purpose of core calibration logging is a problem

Generally, according to the characteristics of well logging data, layer point analysis method is currently used to carry out arithmetic weighted average of the sample analysis and test values ​​inside n reservoirs, for example, the m in the first, second, ..., n layers 1 , m 2 ,...,m n The porosity values ​​of samples are respectively calculated as their arithmetic mean value, and the arithmetic mean value is recorded as p 1 ,p 2 ,...,p n , and then combine the acoustic time difference Δt of each layer corresponding to the n average values 1 , Δt 2 ,...,Δt n , to compile a sonic transit time-porosity logging crossplot, but the number of samples used in such a sonic transit time-porosity logging crossplot is n. In the process of compiling the crossplot, the difference in thickness of each layer and the The sample space information carried by the difference in the number of samples in layers does not fully reflect the typicality and representativeness of the samples.

Images