石油学报 ›› 2023, Vol. 44 ›› Issue (11): 1949-1958.DOI: 10.7623/syxb202311015

• 油田开发 • 上一篇    下一篇

基于大数据分析算法的深部煤层气地质—工程一体化智能决策技术

龚斌1,2, 王虹雅3, 王红娜4, 宋伟5, 孙雄伟6, 杨京华1   

  1. 1. 中国地质大学(武汉)资源学院 湖北武汉 430074;
    2. 特雷西能源科技股份有限公司 浙江杭州 310000;
    3. 中石油煤层气有限责任公司 北京 100028;
    4. 中油油气勘探软件国家工程研究中心有限公司 北京 100080;
    5. 中国石油青海油田公司 甘肃敦煌 736202;
    6. 中联煤层气国家工程研究中心有限责任公司 北京 100095
  • 收稿日期:2023-07-01 修回日期:2023-10-06 出版日期:2023-11-25 发布日期:2023-12-08
  • 通讯作者: 王虹雅,女,1993年1月生,2017年获中国地质大学(北京)硕士学位,现为中石油煤层气有限责任公司工程师,主要从事勘探开发管理及数智化建设工作。Email:1012400030@qq.com;杨京华,男,1994年3月生,2021年获西南石油大学硕士学位,现为中国地质大学(武汉)资源学院博士研究生,主要从事人工智能辅助下的地质—工程一体化研究。Email:yangjinghuaswpu@sina.com
  • 作者简介:龚斌,男,1977年12月生,2007年获美国斯坦福大学博士学位,现为中国地质大学(武汉)资源学院教授,主要从事油气藏勘探开发过程中地质—工程一体化研究。Email:gongbin@tracyenergy.cn
  • 基金资助:
    国家重点研发计划项目"地质资源精准开发风险预测的大数据智能分析技术及平台建设"(2022YFF0801200)资助。

Integrated intelligent decision-making technology for deep coalbed methane geology and engineering based on big data analysis algorithms

Gong Bin1,2, Wang Hongya3, Wang Hongna4, Song Wei5, Sun Xiongwei6, Yang Jinghua1   

  1. 1. School of Earth Resources, China University of Geosciences (Wuhan), Hubei Wuhan 430074, China;
    2. Tracy Energy Technologies Company Limited, Zhejiang Hangzhou 310000, China;
    3. PetroChina Coalbed Methane Company Limited, Beijing 100028, China;
    4. CNPC Exploration Software Co., Ltd., Beijing 100080, China;
    5. PetroChina Qinghai Oilfield Company, Gansu Dunhuang 736202, China;
    6. China United Coalbed Methane National Engineering Research Center Co., Ltd., Beijing 100095, China
  • Received:2023-07-01 Revised:2023-10-06 Online:2023-11-25 Published:2023-12-08

摘要: 煤层气勘探开发形势日趋复杂、资源品质劣化程度加剧。明确煤层气高产主控因素与作用机理、建立地质—工程综合甜点评价方法、提高煤层气动态预测精确度和工程决策有效性,成为亟待解决的技术挑战。以区域数据湖及气藏精细描述研究成果为基础,运用人工智能技术对煤层气藏综合数据进行深度挖掘,构建了基于大数据分析算法的煤层气地质—工程一体化智能决策系统,实现了煤层气地质—气藏—工程一体化数据的集成和管理、大数据驱动下的煤层气单井产量快速预测及主控因素分析、融合地质及工程因素各参数的煤层气储层综合甜点分析、基于煤层气井压后产量主控因素分析的压裂参数优化等关键技术。该系统在鄂尔多斯东缘深部煤层气大宁—吉县区块进行了试点,推广应用结果显示,系统显著提升了气藏研究的效率与工程决策的有效性,为全面了解和掌握气藏的资源潜力、全力推进智能化在科研生产中的深度应用提供了有力的支撑。

关键词: 深部煤层气, 地质—工程一体化, 大数据分析, 智能决策系统, 地质—工程评价方法

Abstract: The exploration and development of coalbed methane are facing increasingly complex challenges, accompanied by a worsening of the quality of resources. It is imperative to address the technical challenges of identifying the primary control factors and mechanisms for high coalbed methane production, establishing a reliable integrated geological-engineering evaluation method, enhancing the accuracy of dynamic coalbed methane prediction, and improving the effectiveness of engineering decisions. Building upon research achievements related to regional data repositories and detailed descriptions of gas reservoirs, and leveraging artificial intelligence technology, a comprehensive data mining effort was conducted on coalbed methane reservoirs. This innovative approach led to the development of an integrated intelligent decision-making system for coalbed methane geological and engineering activities, based on big data analysis algorithms. This system integrates and manages data related to coalbed methane geological, gas reservoir, and engineering aspects, offers rapid predictions of single-well gas production under big data-driven conditions, analyzes the primary control factors, conducts comprehensive analyses of coalbed methane reservoir parameters by incorporating geological and engineering factors, and optimizes fracturing parameters based on post-fracturing production analysis. The system was piloted in the Daning-Ji xian block in the eastern margin of Hubei, and its subsequent widespread application significantly improved the efficiency of gas reservoir research and the effectiveness of engineering decisions. It provided strong support for gaining a comprehensive understanding of the resource potential of gas reservoirs and advancing the deep application of intelligent technology in research and production.

Key words: deep coalbed methane, geological and engineering integration, big data analysis, intelligent decision-making system, geological-engineering evaluation method

中图分类号: