| 基于ε-SVR的钻柱延伸能力预测技术研究 |
 点此下载全文 |
| 引用本文：李井辉，孙丽娜?覮，申静波，邹龙朱.基于ε-SVR的钻柱延伸能力预测技术研究[J].计算技术与自动化,2018,(3):56-60 |
| 摘要点击次数: 1120 |
| 全文下载次数: 0 |
|
|
| 中文摘要:钻柱的极限延伸能力是钻井设计和施工中易于忽视的关键参数，影响钻柱延伸能力的因素较多且相互关系复杂，而众多的影响因素与钻柱的延伸极限之间存在着某种非线性联系。以指定工况下钻柱能够继续钻进的约束条件及其对下入极限深度的影响分析为基础，充分利用ε-SVR在解决小样本、非线性及高维模式识别问题中的特有优势挖掘钻柱延伸能力与其不易觉察的影响变量之间的相互关系，通过精确确定回归参数获得了钻柱延伸能力及其影响因素之间的隐含关系预测模型。实验结果表明可利用该预测模型实现钻柱极限延伸长度的快速预测，同时为钻柱延伸能力的预测提供了一种新的解决方案。 |
| 中文关键词:钻柱力学 延伸能力 支持向量机 ε-SVR |
| |
| Research on Drillstring Extended Capability Prediction Based on ε-SVR |
|
|
| Abstract:The drillstring extension capability is the key parameter that is easy to ignore in the drilling design and construction，there are many factors influencing the drillstring extension capability and the relationship between the influence factors，furthermore the relationship between the extension capability limit and its influence factors is nonlinear.Based on the analysis on the constraints condition to drillstring drilling and its influence on the running depth，the implied prediction model existing between the drillstring extension capability and its influencing factors is obtained by determining the regression parameters accurately，which makes full use of ε-SVR advantages in solving the small sample，nonlinear and high dimensional pattern recognition problems to dig hard for the interplay between the extension capability and its imperceptible influence variables.The experimental results show that the prediction model can be used to realize the fast prediction of the drill string extension length，and provides a new solution for the drillstirng extension capability prediction. |
| keywords:drillstring mechanics extension capability support vector machine ε-SVR |
| 查看全文 查看/发表评论 下载pdf阅读器 |
位访问者,今日访问