Scholarlink

<p>The design of fracturing operating parameters of shale gas wells plays an important role in improving the fracturing productivity. This paper studies a deep Q-network-based optimization method for shale gas well operation parameters. In this paper, an optimization model for operating parameters of shale gas wells based on deep Q-network (DQN) is established by utilizing the sample data of 282 fractured shale gas wells in Weiyuan (WY) block in Sichuan Basin, and a proxy model based on light gradient boosting machine is used to simulate the real environment and give corresponding feedback to the DQN agent. The model optimizes the amounts of proppant and fracturing fluid per perforated meter in the way of unfixed step-size, and obtains globally optimal design strategy through training, so as to realize self-optimization of operating parameters. Fracturing effect evaluation and productivity optimization experiments of three fractured shale gas wells with different geological conditions were carried, and the maximum average daily gas production rate in the first year (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({Q}_{g}\)</EquationSource> </InlineEquation>) of three fractured shale gas wells is increased by 15.9%, 10.9% and 16.9% respectively. The results indicate that the DQN model has the advantages of variable step size search and multivariate synchronous optimization, which can optimize the fracturing operation parameters of shale gas wells with different geological properties and evaluate the fracturing production potential. This research result helps to optimize fracturing operation parameters more quickly and provides guidance for on-site engineers to design fracturing plans.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Design of the optimal fracturing operating parameters of shale gas wells based on deep Q-network

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月31日

查看全文
  • Chaodong Tan,
  • Hanwen Deng,
  • Chunqiu Wang,
  • Huizhao Niu,
  • Jian Song,
  • Zhaobin Li

摘要

The design of fracturing operating parameters of shale gas wells plays an important role in improving the fracturing productivity. This paper studies a deep Q-network-based optimization method for shale gas well operation parameters. In this paper, an optimization model for operating parameters of shale gas wells based on deep Q-network (DQN) is established by utilizing the sample data of 282 fractured shale gas wells in Weiyuan (WY) block in Sichuan Basin, and a proxy model based on light gradient boosting machine is used to simulate the real environment and give corresponding feedback to the DQN agent. The model optimizes the amounts of proppant and fracturing fluid per perforated meter in the way of unfixed step-size, and obtains globally optimal design strategy through training, so as to realize self-optimization of operating parameters. Fracturing effect evaluation and productivity optimization experiments of three fractured shale gas wells with different geological conditions were carried, and the maximum average daily gas production rate in the first year ( \({Q}_{g}\) ) of three fractured shale gas wells is increased by 15.9%, 10.9% and 16.9% respectively. The results indicate that the DQN model has the advantages of variable step size search and multivariate synchronous optimization, which can optimize the fracturing operation parameters of shale gas wells with different geological properties and evaluate the fracturing production potential. This research result helps to optimize fracturing operation parameters more quickly and provides guidance for on-site engineers to design fracturing plans.