Reservoir evaluation model construction method and reservoir identification method

A technology for evaluating models and construction methods, applied in the fields of artificial intelligence and geophysical exploration, can solve the problems of lack of sample support, lack of reservoir identification sample data, low coincidence rate, etc., to improve utilization efficiency, improve accuracy and efficiency, The effect of improving the matching rate

Pending Publication Date: 2021-05-28
CHINA PETROLEUM & CHEM CORP +1
View PDF1 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of this application is to provide a reservoir evaluation model construction method and reservoir identification method to solve the lack of reservoir identification sample data in the research target area, resulting in the lack of sample support when performing machine learning and deep learning on reservoir evaluation samples. And the problem of low coincidence rate when the model is applied after learning with too few samples

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Reservoir evaluation model construction method and reservoir identification method
  • Reservoir evaluation model construction method and reservoir identification method
  • Reservoir evaluation model construction method and reservoir identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] Taking the oil test data and logging data of a certain oil well as an example, the present invention is further described,

[0054] 1. Table 1 is the logging data table of the starting and ending depth intervals of the oil testing data collected in a well target formation in a certain research area of ​​Zhongyuan Oilfield. The oil testing results of this depth section are oil layers. It can be seen from the table that the logging data includes Medium-induction RILM, deep-induction RILD, compensated density DEN, compensated neutron CNL and compensated acoustic AC data; the logging data are actually continuous data, and Table 1 only gives some of them.

[0055] Table 1

[0056]

[0057]

[0058] 2. The well logging data (continuous data) determined in Table 1 is divided into the data of a depth point row by depth, forming the logging row data of a single depth point as shown in Table 2;

[0059] Table 2

[0060]

[0061]

[0062] 3. Sort the logging line dat...

Embodiment 2

[0078] In this embodiment, the process of expanding the capacity of a small sample is taken as an example to further explain the solution of this application within the start and end depths of oil testing data in a target formation of a certain well in a certain research area, and when the oil testing result is a water layer.

[0079] 1) Table 7 is the well logging data table after standardization of the starting and ending depth intervals of the oil testing data collected in the target formation of a certain well in a certain research area. The oil testing results of this depth section are water layers. Includes medium induction, deep induction, compensated density, compensated neutron and compensated acoustic data;

[0080] Table 7

[0081]

[0082] 2) The logging data determined in Table 7 is divided into data of a row of depth points according to depth, forming the logging row data of a single depth point as shown in Table 8;

[0083] Table 8

[0084]

[0085]

...

Embodiment 3

[0102] The difference between this embodiment and the above-mentioned specific embodiments only lies in:

[0103] In this embodiment, there is no need to sort the logging row data, and the logging row data of the set data are randomly extracted. Through this random sampling method, the efficiency of sample expansion is effectively improved.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a reservoir evaluation model construction method and a reservoir identification method, and belongs to the field of geophysical exploration and artificial intelligence. Logging data of corresponding starting and ending depth sections of different types of reservoirs is acquired; a new reservoir evaluation sample is formed by segmenting, recombining and calculating well logging data of well sections of an existing test oil well, so that expansion of the reservoir evaluation sample is achieved, the reservoir evaluation sample is trained, a reservoir evaluation model is obtained, and reservoir prediction is conducted through the reservoir evaluation model. The problems that due to the fact that reservoir identification sample data is lacked in a research target area, sample support is lacked when machine learning and deep learning are conducted on reservoir evaluation samples, and the coincidence rate is low when a model is applied after few sample learning is conducted are solved.

Description

technical field [0001] The application relates to a reservoir evaluation model building method and a reservoir identification method, belonging to the fields of geophysical exploration and artificial intelligence. Background technique [0002] Reservoir identification is the process of identifying reservoir fluids. The conventional reservoir identification method is to realize the reservoir identification by calculating the porosity, permeability and oil saturation of the reservoir through the interpretation model of the reservoir and the logging data. However, with the deepening of oil and gas field development, the development of oil and gas fields is becoming more and more difficult. Conventional reservoir identification methods have low reservoir identification accuracy due to complex geological conditions and insignificant differences in reservoir fluid response characteristics. [0003] Using machine learning methods to identify oil and gas layers in reservoirs can ex...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62
CPCG06F18/217G06F18/214
Inventor 王波董顺勇耿淑亚郝宁马博王守认刘蓉陆江莲
Owner CHINA PETROLEUM & CHEM CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap