ORIGINAL RESEARCH article

Front. Earth Sci.

Sec. Georeservoirs

Volume 13 - 2025 | doi: 10.3389/feart.2025.1605793

Dynamic and static integrated classification model of gas well based on XGBoost algorithm-An Example from block S of Sulige tight sa ndstone gas field

Provisionally accepted
Shuangshuang  ZhangShuangshuang Zhang1,2*Xiangdong  YuXiangdong Yu1Xiuli  GaoXiuli Gao1Donglin  LiDonglin Li1Shijun  HuangShijun Huang2
  • 1China National Petroleum Corporation Bohai Drilling Engineering Co. Ltd, Tianjin, Tianjin Municipality, China
  • 2Ministry of Education Key Laboratory for Petroleum Engineering, College of Petroleum Engineering, China University of Petroleum, Beijing, Beijing, China

The final, formatted version of the article will be published soon.

Classification of gas wells is an important part of optimizing development strategies and increasing the recovery. The original classification standard of gas wells in the Sulige gas field has weak regularity of each parameter, large overlapping range of classification results, serious discrepancy between the dynamic and static, and low efficiency of manual classification. Aiming at this problem, this paper establishes a set of dynamic and static integrated classification model of tight sandstone gas wells in Sulige based on XGBoost algorithm. After comparison and verification, it is proved to be accurate and reliable. The model can be substituted into the static and dynamic characteristic parameters at the same time to complete the importance ranking of classification features and model training, and realize the dynamic and static integration classification of Sulige gas well. The model is applied to 553 gas wells in S block, and it is concluded that the main factors affecting the classification of gas wells are initial daily production, effective thickness of a gas layer, formation permeability, original formation pressure, and porosity. The main factors affecting the classification of class I and class II wells are initial daily production and permeability, and the main factors affecting the classification of class III wells are initial daily production and the effective thickness of the gas layer. This method improves the effectiveness of gas well classification, reduces subjectivity, and the classification results are in line with the actual situation of the field, which has guiding significance for the classification management of gas wells and the formulation of development countermeasures.

Keywords: Gas well classification model1, static and dynamic integration2, XGBoost algorithm3, tight sandstone gas reservoir4, correlation analysis of characteristics5

Received: 04 Apr 2025; Accepted: 01 Jul 2025.

Copyright: © 2025 Zhang, Yu, Gao, Li and Huang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

* Correspondence: Shuangshuang Zhang, China National Petroleum Corporation Bohai Drilling Engineering Co. Ltd, Tianjin, Tianjin Municipality, China

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.