Unlock instant, AI-driven research and patent intelligence for your innovation.

Oil and gas reservoir characterization method based on convolutional neural network

A convolutional neural network and a technology for oil and gas reservoirs, which are applied in the fields of devices, oil and gas reservoir characterization methods, equipment and readable storage media, can solve the problems of low oil and gas reservoir development efficiency, time-consuming and labor-intensive, and heavy workload, etc. Achieve the effect of improving development efficiency, enhancing generalization ability, improving speed and accuracy

Active Publication Date: 2019-10-29
CHINA UNIV OF PETROLEUM (BEIJING)
View PDF9 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of this application is to provide a convolutional neural network-based oil and gas reservoir characterization method, device, equipment, and readable storage medium to solve the problem that traditional high-frequency processing schemes need to Manual corrections and revisions required huge workload, time-consuming and labor-intensive, resulting in low efficiency of oil and gas reservoir development

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
  • Oil and gas reservoir characterization method based on convolutional neural network
  • Oil and gas reservoir characterization method based on convolutional neural network
  • Oil and gas reservoir characterization method based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] The following is an introduction to Embodiment 1 of a convolutional neural network-based oil and gas reservoir characterization method provided by this application, see figure 1 , embodiment one includes:

[0054] S101. Obtain seismic trace data;

[0055] S102. Intercepting the seismic trace data corresponding to the preset order of sedimentary cycles;

[0056] Specifically, the interception operation on the seismic trace data can be realized through the interception unit, which is located outside the convolutional neural network and is used to limit the input data to ensure that the input data is the seismic trace data corresponding to the depositional cycle of the preset order.

[0057] Among them, the sedimentary cycle refers to the regular and periodic repetition of several rocks with similar lithology and lithofacies on the vertical bottom section. According to the different scales, the sedimentary cycle is divided into levels, such as figure 2 As shown, the sc...

Embodiment 2

[0066] see Figure 4 , embodiment two specifically includes:

[0067] S401. Obtain seismic trace data;

[0068] S402. Intercepting the seismic trace data corresponding to the lowest order sedimentary cycle identifiable by seismic;

[0069] S403. Input the intercepted seismic trace data into the pre-trained convolutional neural network to obtain high-frequency synthetic records;

[0070] S404. According to the high-frequency synthetic records, determine the development position, geometric shape, scale and superposition relationship of small-scale geological bodies, so as to realize reservoir characterization.

[0071] This embodiment provides a convolutional neural network-based reservoir characterization method for oil and gas reservoirs, using convolutional neural networks to learn the relationship between relatively low-frequency seismic trace data and high-resolution synthetic records, corresponding to specific levels of depositional cycles The seismic trace data is the ...

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 present application discloses an oil and gas reservoir characterization method, device and apparatus based on a convolutional neural network and a readable storage medium. The method comprises thesteps of: acquiring seismic trace data; intercepting seismic trace data corresponding to a preset order sedimentary cycle; inputting the intercepted seismic data into a pre-trained convolutional neural network to obtain a high-frequency synthetic record; and according to the high-frequency synthetic record, determining the distribution law of a small-scale geological body to achieve accurate reservoir characterization. Since the convolutional neural network sharply increases the frequency of the seismic data and has an automatic learning ability, the method has a large frequency increase, high processing accuracy, and high processing efficiency. In addition, the method, in view of a large difference in different sedimentary cycles, intercepts the seismic trace data of the preset order sedimentary cycle as an input so as to achieve good pertinence, further improve the accuracy of high-frequency processing, and ultimately improve the development efficiency of oil and gas reservoirs.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a convolutional neural network-based oil and gas reservoir characterization method, device, equipment and readable storage medium. Background technique [0002] Most of my country's onshore oil and gas reservoirs have entered the late stage of development with high water cut and high recovery degree. Therefore, accurate characterization of small-scale geological bodies is the key to improving oil and gas reservoir recovery. Seismic data are the only direct and full-coverage observation data of cross-well geological bodies, but the vertical resolution of existing seismic data is almost all greater than 10m. The horizontal resolution is about 20m, and it is difficult to support fine and accurate characterization of the seismic response characteristics of interwell geological bodies. [0003] In recent years, many methods and theories have been developed for high-resolu...

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): G01V1/30G01V1/36G06N3/04G06N3/08
CPCG01V1/306G01V1/36G06N3/08G01V2210/52G01V2210/624G06N3/045
Inventor 徐朝晖方惠京孙盼科徐怀民
Owner CHINA UNIV OF PETROLEUM (BEIJING)