Intelligent prediction method and apparatus for reservoir sensitivity

Patent metadata

Field	Value
Publication number	US-11326450-B2
Application number	US-202117345904-A
Country	US
Kind code	B2
Filing date	Jun 11, 2021
Priority date	Jun 11, 2020
Publication date	May 10, 2022
Grant date	May 10, 2022

Abstract

The embodiments of the invention provide an intelligent prediction method and apparatus for reservoir sensitivity, belonging to the technical field of reservoir sensitivity prediction. The method includes: acquiring a reservoir sensitivity influence factor item related to a reservoir sensitivity result to be predicted and numerical values of corresponding reservoir sensitivity influence factors; determining a corresponding type of database according to the reservoir sensitivity influence factor item; determining whether numerical values of reservoir sensitivity influence factors corresponding to core parameters in the numerical values of the reservoir sensitivity influence factors include a first upper boundary value or a first lower boundary value; and using, according to whether the first upper boundary value or the first lower boundary value is included, different intelligent sensitivity prediction models to calculate the reservoir sensitivity result to be predicted.

First claim

The invention claimed is:

1. A method for exploiting oil and gas resources with improved protection effect of oil and gas reservoirs, comprising:
acquiring a reservoir sensitivity influence factor item and numerical values of corresponding reservoir sensitivity influence factors related to the oil and gas reservoirs;
determining a corresponding type of database according to the reservoir sensitivity influence factor item;
determining whether numerical values of reservoir sensitivity influence factors corresponding to core parameters in the numerical values of the reservoir sensitivity influence factors include a first upper boundary value or a first lower boundary value;
inputting the numerical values of reservoir sensitivity influence factors into a first intelligent prediction model to obtain a reservoir sensitivity result of the oil and gas reservoirs when the first upper boundary value is included, wherein the first intelligent prediction model is pre-established according to an expert system method using an upper boundary database divided from the corresponding type of database as a sample;
inputting the numerical values of reservoir sensitivity influence factors into a second intelligent prediction model to obtain a reservoir sensitivity result of the oil and gas reservoirs when the first lower boundary value is included, wherein the second intelligent prediction model is pre-established according to the expert system method using a lower boundary database divided from the corresponding type of database as the sample; and
inputting the numerical values of reservoir sensitivity influence factors into a third intelligent prediction model to obtain a reservoir sensitivity result of the oil and gas reservoirs when the first upper boundary value and the first lower boundary value both are not included, wherein the third intelligent prediction model is pre-established according to an artificial neural network method using an intermediate database divided from the corresponding type of database as the sample;
formulating protection measures of the oil and gas reservoirs according to the reservoir sensitivity result of the oil and gas reservoirs;
exploiting the oil and gas resources wherein the protection effect of the oil and gas reservoirs is improved by the protection measures.

2. The method according to claim 1 , wherein:
the corresponding type of database comprises: a first database including all core parameters; a second database including all the core parameters and at least a part of important parameters; and a third database including all the core parameters, all important parameters, and at least a part of auxiliary parameters;
the first database is divided into a first upper boundary database including reservoir sensitivity data with a reservoir sensitivity result located within an upper boundary, a first intermediate database including reservoir sensitivity data with a reservoir sensitivity result located within an intermediate value, and a first lower boundary database including reservoir sensitivity data with a reservoir sensitivity result located within a lower boundary;
the second database is divided into a second upper boundary database including reservoir sensitivity data with a reservoir sensitivity result located within the upper boundary, a second intermediate database including reservoir sensitivity data with a reservoir sensitivity result located within the intermediate value, and a second lower boundary database including reservoir sensitivity data with a reservoir sensitivity result located within the lower boundary; and
the third database is divided into a third upper boundary database including reservoir sensitivity data with a reservoir sensitivity result located within the upper boundary, a third intermediate database including reservoir sensitivity data with a reservoir sensitivity result located within the intermediate value, and a third lower boundary database including reservoir sensitivity data with a reservoir sensitivity result located within the lower boundary.

3. The method according to claim 2 , wherein the first database, the second database or the third database is used as follows: when new reservoir sensitivity data is collected, and a reservoir sensitivity result in the new reservoir sensitivity data is known and one reservoir sensitivity influence factor is missing relative to one of the first database, the second database or the third database, the one of the first database, the second database or the third database is used to perform inversion calculation to obtain a numerical value of the missing reservoir sensitivity influence factor,
wherein the numerical value of the missing reservoir sensitivity influence factor obtained by inversion calculation and the new reservoir sensitivity data are stored in the one of the first database, the second database or the third database as one piece of new reservoir sensitivity data.

4. The method according to claim 2 , wherein a range of the upper boundary is not less than 0.95 and not greater than 1.00; a range of the lower boundary is not less than 0 and not greater than 0.10; and a range of the intermediate value is greater than 0.10 and less than 0.95.

5. The method according to claim 1 , wherein the determining whether the numerical values of the reservoir sensitivity influence factors corresponding to the core parameters in the numerical values of the reservoir sensitivity influence factors comprise the first upper boundary value or the first lower boundary value comprises:
normalizing the numerical values of the reservoir sensitivity influence factors; and
determining whether the numerical values of the reservoir sensitivity influence factors corresponding to the core parameters in the numerical values of the reservoir sensitivity influence factors comprise the first upper boundary value or the first lower boundary value.

6. The method according to claim 1 , wherein a range of the first upper boundary value is not less than 0.90 and not greater than 1.00; and a range of the first lower boundary value is not less than 0 and not greater than 0.05.

7. The method according to claim 1 , wherein the database is established in advance according to the following steps:
collecting reservoir sensitivity data, wherein the reservoir sensitivity data comprises a new reservoir sensitivity influence factor item, new numerical values of reservoir sensitivity influence factors and a new reservoir sensitivity result;
comparing the new reservoir sensitivity factor item with a standard reservoir sensitivity factor item to obtain a comparison result;
performing inversion calculation to obtain a numerical value of the missing reservoir sensitivity influence factor, when the comparison result indicates that one reservoir sensitivity influence factor is missing in the new reservoir sensitivity factor item;
determining whether the numerical value of the missing reservoir sensitivity influence factor obtained by inversion calculation is accurate by:
obtaining a first reservoir sensitivity result according to a first reservoir sensitivity prediction model generated by using a database that does not comprise the piece of new reservoir sensitivity data as a sample, and numerical values of tested reservoir sensitivity factors;
obtaining a second reservoir sensitivity result according to a second reservoir sensitivity prediction model generated by using a database that comprises the piece of new reservoir sensitivity data as a sample, and the numerical values of the tested reservoir sensitivity factors;
calculating a first prediction accuracy of the first reservoir sensitivity result compared to an actual reservoir sensitivity result corresponding to the numerical values of the tested reservoir sensitivity f

Assignees

• Univ China Petroleum Beijing

Inventors

• Jiang Guancheng
• Wu Xiongjun
• Yang Lili
• He Yinbo
• Dong tengfei
• Quan Xiaohu
• Peng Chunyao
• Luo Xuwu
• Tan Bin
• Cai Jun
• Liang Xing
• You Zhiliang
• Wang Yong
• Li Bin
• Guo Yongbin
• Guan Shen

Classifications

• G06N3/048

  Activation functions · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• E21B2200/22

  Fuzzy logic, artificial intelligence, neural networks or the like · CPC title

• E21B47/003

  Determining well or borehole volumes · CPC title