Shale gas well type classification method

Patent metadata

Field	Value
Publication number	US-2026055699-A1
Application number	US-202519309649-A
Country	US
Kind code	A1
Filing date	Aug 26, 2025
Priority date	Aug 26, 2024
Publication date	Feb 26, 2026
Grant date	—

Abstract

A shale gas well type classification method includes: S1, acquiring raw data of evaluation parameters for stable production wells and trial production wells; S2, performing standardization on the raw data; S3, calculating distance coefficients; S4, classifying the stable production wells; S5, after determining a class number of the stable production wells, calculating comprehensive evaluation parameters S of the stable production wells based on standardized data; S6, for each class of the stable production wells, calculating average values of standardized data of the evaluation parameters for the stable production wells to construct a standard fuzzy set A, and constructing a to-be-identified fuzzy set B for each of the trial production wells, S7, calculating a closeness degree between each to-be-identified fuzzy set and each standard fuzzy set, and classifying a trial production well corresponding to the to-be-identified fuzzy set into a class of the stable production wells with a highest closeness degree.

First claim

What is claimed is:

1 . A shale gas well type classification method, comprising the following steps:
S1, acquiring raw data of evaluation parameters for production wells and trial production wells, wherein the evaluation parameters comprise six parameters: an effective thickness, a gas saturation, a formation pressure, an open flow rate, a gas production per unit casing pressure drop, and a peak daily gas production; S2, performing standardization on the raw data of the evaluation parameters to obtain standardized data of the evaluation parameters for the production wells and the trial production wells; S3, calculating, based on the standardized data of the evaluation parameters for the production wells, a distance coefficient d ij between every two production wells of the production wells as per the following formula:

d

i
⁢
j

=

∑

k
=
1

u

(

x

i
⁢
k

-

x

j
⁢
k

)

2

,
i
,

j
=
1

,
2
,
3
,
…

,
m

where x ik represents a parameter value at an i-th row and a k-th column in the standardized data for the evaluation parameters of the production wells, x jk represents a parameter value at a j-th row and the k-th column in the standardized data for the evaluation parameters of the production wells, m represents a number of the production wells, and u represents a number of the evaluation parameters;
S4, classifying, based on a principle that a correlation between two wells will be stronger when a distance coefficient between the two wells becomes small, the production wells, comprising the following steps:
S41, sorting the distance coefficient between every two production wells of the production wells except for 0 in an ascending order to obtain sorted distance coefficients, merging two production wells of the production wells corresponding to a distance coefficient ranked first in the sorted distance coefficients into one class to form a first new entity to thereby obtain first-merged production wells comprising the first new entity and remaining production wells of the production wells for subsequent merging and classifying;
S42, merging two of the first-merged production wells corresponding to a distance coefficient ranked second in the sorted distance coefficients into one class to form a second new entity to thereby obtain second-merged production wells comprising the second new entity and remaining production wells of the first-merged production wells for subsequent merging and classifying; and
S43, performing, according to the above steps, a merging and classification operation on two of the second-merged production wells corresponding to a distance coefficient ranked third in the sorted distance coefficients, and repeating the merging and classification operation until a class number of the production wells is q as required for production, thereby stopping the merging and classification operation to obtain q classes of the production wells, where 1≤q≤m, and q and m are positive integers;
S5, after determining the class number of the production wells, calculating, based on the standardized data of the evaluation parameters for the production wells, comprehensive evaluation parameters S of the production wells, wherein each of the comprehensive evaluation parameters S is a sum of standardized values of the six evaluation parameters; and calculating an average value of the comprehensive evaluation parameters S of the production wells in each of the q classes, and designating the q classes of the production wells sequentially in a descending order of the average value of the comprehensive evaluation parameters S as Class I gas wells, Class II gas wells, Class III gas wells, . . . , Class q gas wells;
S6, for each of the q classes of the production wells, calculating average values of the standardized data of the evaluation parameters for the production wells to construct a standard fuzzy set A, thereby obtaining standard fuzzy sets A 1 , A 2 , A 3 , . . . ,A q , where A=(an average effective thickness value, an average gas saturation value, an average formation pressure value, an average open flow rate value, an average gas production value per unit casing pressure drop, an average peak daily gas production value); and for each of the trial production wells, calculating average values of the standardized data of the evaluation parameters for the trial production wells to construct a to-be-identified fuzzy set B, thereby obtaining to-be-identified fuzzy sets B 1 , B 2 , B 3 , . . . ,B n , where n represents a number of the trial production wells; and
S7, calculating a closeness degree between one of the to-be-identified fuzzy sets and each of the standard fuzzy sets, and classifying, based on a principle that a similarity between the standard fuzzy set and the to-be-identified fuzzy set will be higher when the closeness degree becomes greater, one of the trial production wells corresponding to the one of the to-be-identified fuzzy sets into one of the q classes of the production wells corresponding to a highest closeness degree, wherein a formula for calculating the closeness degree is expressed as follows:

σ
⁡
(

A
,
B

)

=
def

∑

k
=
1

z

[

A
⁡
(

x
c

)

∧

B
⁡
(

x
c

Assignees

• Univ Southwest Petroleum
• Cnpc National Oil & Gas Drilling Equipment Engineering & Tech Research Center Co Ltd
• Univ China Geosciences Wuhan

Inventors

• He Xinyang
• Zhang Kun
• Yin Lei
• Jiang Shu
• Lin Yanhua
• Zhang Yunbo
• Wang Xueying
• Niu Hulin
• Liu Weiwei
• Cheng Sihong
• Wan Songyang
• Zhi Tianyou
• Wu Daiyu
• Song Yan
• Jiang Zhenxue
• Jiang Lin
• Ding Xiong
• Liu Pei
• Su Jie
• Chen Yuxiang
• Zhang Yi
• Qiao Yiting
• Liu Jiayi
• Peng Jun
• Li Bin
• Liu Jinhua
• Zheng Binsong
• Chen Lei
• Yang Xuefei
• Han Fengli
• Tang Weishi
• Ruan Jingru
• Gou Hengfeng
• Xiao Yi
• Yang Zilong
• Wang Shouqian
• Peng Yanchen
• Li Lintao
• Liu Yipeng
• Liu Ping

Classifications

• E21B43/00

  Methods or apparatus for obtaining oil, gas, water, soluble or meltable materials or a slurry of minerals from wells (applicable only to water E03B) · CPC title

• E21B2200/22

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

• E21B2200/20

  Computer models or simulations, e.g. for reservoirs under production, drill bits · CPC title

• E21B49/008 Primary

  by injection test; by analysing pressure variations in an injection or production test, e.g. for estimating the skin factor (measuring pressure E21B47/06) · CPC title

• G06Q50/02

  Agriculture; Fishing; Forestry; Mining · CPC title