摘要
为解决普通孕镶金刚石钻头(普通钻头)破岩效率低、钻进不稳定、不耐磨以及不能根据地层选择合适形态破岩等问题,结合仿生原理和非平面结构设计了新型聚晶金刚石复合片(polycrystalline diamond compact,PDC)齿和钻头底唇面,提出一种在钻头底唇面安装传感器进行岩性识别,可选择合适形态破岩的智能仿生PDC钻头(智能钻头)。利用机器学习,基于现场钻井数据对岩石性质进行分类。分析了智能钻头切削结构针对软中硬地层和硬地层具有两种工作形态,基于建立的切削模型和岩石失效准则,对智能钻头和普通钻头的破岩过程进行数值模拟对比,结果表明:机器学习的加权KNN(K-nearest neighbor)算法可很好避免样本重叠问题,可精确对岩石性质分类,其训练模型可提供给传感器进行岩性识别。破碎砂岩时,智能钻头经传感器调节将具有PDC齿的底唇面和仿生底唇面保持同一高度破岩,对比普通钻头,其破岩比能和切削力明显更小,表明其效率更高、切削更稳定;破碎花岗岩时,智能钻头经传感器调节将具有PDC齿的底唇面下放破岩,其破岩效率明显高于普通钻头,也更稳定。研究成果对石油钻井设备智能化发展有重要意义。
In order to solve ordinary impregnated diamond bit(normal bit),low efficiency of rock fragmentation,drilling is not stable,not wear-resisting,and not according to choose a suitable form of broken rock stratum,combined with the bionic nonplanar structural design principle and the bottom of the tooth and a new type of polycrystalline diamond compact(PDC)bit lip surface,a bit bottom lip surface sensor installed on the lithology identification was proposed.The intelligent bionic PDC bit(intelligent bit)can be selected to break rock with appropriate shape.Machine learning was used to classify rock properties based on field drilling data.Based on the established cutting model and rock failure criteria,the rock breaking process of the intelligent bit and the common bit is numerically simulated and compared.The results show that the weighted KNN(K-nearest neighbor)algorithm based on machine learning can well avoid the problem of sample overlap and accurately classify the rock properties.Its training model can be provided to the sensor for lithology identification.When sandstone is broken,the bottom lip with PDC teeth and the bionic bottom lip of the intelligent bit are adjusted by the sensor to keep the same height of rock breaking.Compared with the ordinary bit,the rock breaking specific energy and cutting force are significantly smaller,indicating that its efficiency is higher and the cutting is more stable.When granite is broken,the intelligent bit will be adjusted by the sensor to lower the bottom lip with PDC teeth to break rock,and its rock breaking efficiency is significantly higher than that of the ordinary bit,but also more stable.The research results are of great significance to the intelligent development of oil drilling equipment.
作者
吴泽兵
袁若飞
张文溪
黄俊杰
WU Ze-bing;YUAN Ruo-fei;ZHANG Wen-xi;HUANG Jun-jie(Mechanical Engineering College,Xian Shiyou University,Xian 710065,China)
出处
《科学技术与工程》
北大核心
2023年第16期6870-6880,共11页
Science Technology and Engineering
基金
陕西省重点研发计划(2022KW-10)。
