基于神经网络方法的井下机械钻速研究  

A Research on Downhole Mechanical Drilling Rate based on Artificial Intelligence Neural Network

作　　者：刘维凯[1] 徐文 LIU Weikai;XU Wen(School of Petroleum Engineering,Northeast Petroleum University,Daqing,Heilongjiang 163000,China)

机构地区：[1]东北石油大学石油工程学院

出　　处：《中国锰业》2019年第4期90-94,共5页China Manganese Industry

基　　金：中国石油科技创新基金项目(2017D-5007-0309)

摘　　要：为了实现在钻井过程中预测实时机械钻速,提出使用神经网络BP算法,利用BP算法全局搜索问题解的特性,对BP神经网络的权值进行优化,约束BP神经网络训练学习过程。通过BP算法对样本空间进行训练测试,将井下钻速与钻井液密度、钻压、转速、水力参数、地层可钻形相结合,实现井下机械钻速的预测。In order to realize the prediction of real-time ROP in the drilling process,this paper proposes to use BP neural network algorithm to optimize the weight of BP neural network by BP algorithm.It constrains BP neural network training learning process.Through the BP algorithm,the sample space is trained and tested.The downhole drilling speed is combined with the drilling fluid density,incluidng the drilling pressure,the rotational speed,the hydraulic parameters.The formation-drillable shape will realize the prediction of the downhole mechanical drilling rate.

关 键 词：钻速方程 神经网络BP算法 机械钻速 

分 类 号：TE242[石油与天然气工程—油气井工程]