基于遗传算法优化BP神经网络方法的 旋转磨料射流开窗预测
DOI:
作者:
作者单位:

胜利石油管理局 黄河钻井三公司,中国石油大学(华东)

作者简介:

通讯作者:

中图分类号:

TE257.3

基金项目:

国家重大科技专项(2011ZX05060)“山西沁水盆地南部煤层气直井开发示范工程二期”


Prediction of Casing Window With Swirling Abrasive Jet Based on genetic algorithm optimization of BP Artificial Neural Network
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    旋转磨料射流井下套管开窗施工中,开窗直径与开窗深度是最为关注的两个参数。在实际施工过程中,难以对井下套管的开窗过程进行实时监测,因此,需要进一步研究其预测技术,以保证能够开出需要的窗口。由于影响开窗直径与开窗深度的因素很多,很难用传统的数学建模方法进行预测。针对此,提出了一种利用BP神经网络预测旋转磨料射流开窗直径与开窗深度的新方法,并用遗传算法进行了优化,以得到最优的隐层学习率和输出层学习率,从而提高了BP神经网络预测磨料射流井下套管开窗直径及开窗深度的准确性。利用部分实验数据对该方法的可靠性进行了验证。通过对比预测值与实验值发现,该方法的预测精度完全满足工程要求,为现场应用提供了理论支撑。

    Abstract:

    When using swirling abrasive water jet to drill casing windows in the wellbore, the size and the depth of the casing windows are two key parameters of the operation. In practice, it is quite difficult to monitor the geometry of the casing window in real-time. Therefore, it is essential to make an accurate prediction of the quality of the penetration performance so as to optimize the process parameters. There are a lot of factors influencing the casing perforation performance, as a result, it is not realizable to make predictions with conventional mathematical model. To address this issue, in this paper a BP-neural network predictor model is proposed to investigate the quality of the casing window. In addition, the BP-neural network predictor is corrected using genetic algorithm to achieve better results and it is verified with some experimental data. The test results turned out to be satisfying with the margin of error under 10%. This technology has proven itself one of the most reliable methods and certainly meets the demands of the industry.

    参考文献
    相似文献
    引证文献
引用本文

周爱照,杨焕强. 基于遗传算法优化BP神经网络方法的 旋转磨料射流开窗预测[J]. 科学技术与工程, 2014, 14(27): .
Zhou Aizhao and. Prediction of Casing Window With Swirling Abrasive Jet Based on genetic algorithm optimization of BP Artificial Neural Network[J]. Science Technology and Engineering,2014,14(27).

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2014-04-25
  • 最后修改日期:2014-05-19
  • 录用日期:2014-05-30
  • 在线发布日期: 2014-09-24
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注