改进麻雀搜索算法优化支持向量机的井漏预测
DOI:
作者:
作者单位:

西安石油大学电子工程学院

作者简介:

通讯作者:

中图分类号:

TE21

基金项目:

陕西省科学技术重点研发计划项目(2017ZDXM-GY-097)


Improved Sparrow Search Algorithm to Optimize Lost Circulation Prediction of Support Vector Machine
Author:
Affiliation:

School of Electronic Engineering, Xi’an Shiyou University

Fund Project:

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

    在钻井过程中,受地质环境,钻井技术等多种因素的影响,容易发生井漏事故。为预防井漏事故,减少因钻井事故带来的损失,本文提出了一种改进麻雀搜索算法(ISSA)优化支持向量机的井漏预测方法。首先,在发现者位置更新公式中引入了一种改进的自适应非线性惯性递减权重,提高算法全局搜索能力; 其次,在警戒者位置更新公式中引入莱维(Levy)飞行策略,减少算法陷入局部最优的风险。为验证改进算法的寻优能力,将麻雀搜索算法(SSA)、遗传算法(GA)、灰狼算法(GWO)以及改进的麻雀搜索算法(ISSA)在8个基准测试函数上做了对比实验。实验结果表明,改进的麻雀搜索算法(ISSA)在寻优精度,稳定性等方面都较其它算法更为优异。最后,将改进的麻雀搜索算法用于优化支持向量机(ISSA-SVM)的惩罚参数 和核参数 ,进行井漏事故的预测。结果表明,ISSA-SVM预测准确率为97.7654 ,相比于麻雀算法(SSA)-SVM、遗传算法(GA)-SVM以及灰狼算法(GWO)-SVM预测准确率都高,且收敛速度快,迭代次数少,能够高效、快速预测井漏事故,提高钻井效率和可靠性。

    Abstract:

    Lost circulation accidents are common during the drilling process due to the influence of geological environment, drilling technology, and other factors. This paper proposes an improved sparrow search algorithm (ISSA) optimized support vector machine lost circulation prediction method to prevent lost circulation accidents and reduce the losses caused by drilling accidents. Firstly, an improved adaptive nonlinear inertia decreasing weight is introduced to improve the global search ability of the Algorithm; Secondly, Levy flight strategy is introduced into the vigilance position update formula to reduce the risk of falling into local optimization.The Sparrow Algorithm (SSA), Genetic Algorithm (GA), Grey Wolf Algorithm (GWO), and improved Sparrow Search Algorithm (ISSA) were compared on 8 benchmark functions to verify the improved algorithm's optimization ability. The results show that the improved sparrow search algorithm (ISSA) is superior to other algorithms in terms of optimization accuracy and stability. Finally, the improved sparrow search algorithm is used to optimize the penalty and kernel parameters of the support vector machine (ISSA-SVM) in order to predict lost circulation accidents. The results show that the prediction accuracy of ISSA-SVM is 97.7654, which is higher than that of Sparrow Search Algorithm (SSA)-SVM, Genetic Algorithm (GA)-SVM and Grey Wolf Algorithm (GWO)-SVM, it can effectively and rapidly predict lost circulation accidents and improve drilling efficiency and reliability.

    参考文献
    相似文献
    引证文献
引用本文

王鑫,张奇志. 改进麻雀搜索算法优化支持向量机的井漏预测[J]. 科学技术与工程, , ():

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:2021-12-23
  • 最后修改日期:2022-03-25
  • 录用日期:2022-04-30
  • 在线发布日期:
  • 出版日期:
×
关于近期《科学技术与工程》编辑部居家办公的说明
亟待确认的版面费信息