西南石油大学学报(自然科学版) ›› 2023, Vol. 45 ›› Issue (2): 126-134.DOI: 10.11885/j.issn.1674-5086.2021.03.12.03

• 石油与天然气工程 • 上一篇    下一篇

基于井下环空参数的溢流智能预警技术研究

葛亮1,2, 滕怡1, 肖国清3, 肖小汀4, 邓红霞1,2   

  1. 1. 西南石油大学机电工程学院, 四川 成都 610500;
    2. 西南石油大学人工智能研究院, 四川 成都 610500;
    3. 西南石油大学化学化工学院, 四川 成都 610500;
    4. 西南石油大学电气信息学院, 四川 成都 610500
  • 收稿日期:2021-03-12 发布日期:2023-05-05
  • 通讯作者: 葛亮,E-mail:cgroad@swpu.edu.cn
  • 作者简介:葛亮,1985年生,男,汉族,湖北咸宁人,教授,博士,主要从事石油仪器和井下测试技术方面的工作。E-mail:cgroad@swpu.edu.cn
    滕怡,1997年生,女,汉族,宁夏吴忠人,硕士研究生,主要从事仪器仪表技术和信号处理方面的研究。E-mail:Tengyi42396@163.com
    肖国清,1968年生,男,汉族,湖南湘潭人,教授,博士研究生导师,主要从事安全工程研究、化工安全、油气安全和矿山安全等方面的教学科研工作。E-mail:gqxiao68@sina.com
    肖小汀,1988年生,女,汉族,四川什邡人,讲师,博士后,主要从事探地雷达和建筑材料的无损检测方面的研究工作。E-mail:xt_xiao@foxmail.com
    邓红霞,1994年生,女,汉族,四川绵阳人,硕士研究生,主要从事电磁流量计及相关信号处理方面的研究工作。E-mail:1677978305@qq.com
  • 基金资助:
    国家自然科学基金面上项目(51974273);西南石油大学井下智能测控青年科研创新团队项目(2018CXTD04);四川省国际科技合作与交流研究项目(18GJHZ0195)

Research on Overflow Intelligent Warning Technology Based on Downhole Annulus Parameters

GE Liang1,2, TENG Yi1, XIAO Guoqing3, XIAO Xiaoting4, DENG Hongxia1,2   

  1. 1. School of Mechatronic and Electrical Engineering, Southwest Petroleum University, Chengdu, Sichuan 610500, China;
    2. Institute of Artificial Intelligence, Southwest Petroleum University, Chengdu, Sichuan 610500, China;
    3. College of Chemistry and Chemical Engineering, Southwest Petroleum University, Chengdu, Sichuan 610500, China;
    4. School of Electrical Engineering and Information Science, Southwest Petroleum University, Chengdu, Sichuan 610500, China
  • Received:2021-03-12 Published:2023-05-05

摘要: 随着油气勘探开发向复杂地层发展,钻井过程中发生井喷的风险增加,而溢流是井喷的前兆,所以早期溢流预警是实现安全井控预防的关键方向。针对传统预测算法在基于地面参数进行溢流预警时未分析溢流严重程度以及预测准确度不高的问题,通过对溢流征兆及溢流发生机理的研究,在利用环空电磁流量系统及其他系统直接测量井下近钻头处的环空流量和其他环空参数的基础上,建立了一种基于人工智能算法——随机森林的溢流智能预警模型来对溢流严重程度进行分类预测。为了验证该预警模型的可行性,通过搭建模拟实验平台进行测试,并与常规的BP神经网络相比较,结果显示该方法正确率高达92.68%,其分类预测的准确性明显高于BP神经网络。研究结果验证了随机森林模型进行井下溢流预警的可靠性,很好地实现了溢流的早期预警,为钻井提供了安全技术保证,具有较好的应用前景。

关键词: 溢流智能预警, 溢流征兆, 环空参数, 随机森林, 人工智能

Abstract: With the development of oil and gas exploration and development toward complex formations, the risk of blowout during drilling has increased, and overflow is the precursor of blowout, so the early warning of overflow becomes a key direction to well control and safety prevention. Aiming at the problem that the traditional prediction algorithm fails to analyze the severity of overflow and the prediction accuracy is not high when performing overflow warning based on ground parameters, through the study of overflow symptoms and the mechanism of overflow, the annulus electromagnetic flow system and other systems are used to directly measure the underground near the bit, and an overflow intelligent early warning model was established based on artificial intelligence algorithm—Random Forest to classify and predict the severity of overflow. In order to verify the feasibility of the early warning model, a simulation experimental platform was built for testing, and compared with the conventional BP neural network. The results show that the accuracy of this method is as high as 92.68%, and the accuracy of classification prediction is significantly higher than that of the BP neural network. The research results verify the reliability of the random forest model for downhole overflow early warning, which well realizes the early warning of overflow, and provides a safety technical guarantee for drilling, and has good application prospects.

Key words: overflow intelligent warning, overflow symptoms, annulus parameters, random forest, artificial intelligence

中图分类号: