论文详情

基于神经网络的深层机械钻速预测方法

石油钻采工艺

2018年 40卷第S1期

阅读：83

查看详情

Title

A ROP prediction method based on neutral network for the deep layers

作者

王文刘小刚窦蓬林海陶林

Authors

WANG Wen LIU Xiaogang DOU Peng LIN Hai TAO Lin

单位

中海石油(中国)有限公司天津分公司

Organization

Tianjin Branch of China National Offshore Oil Co., Ltd., Tianjin 300459, China

摘要

随着渤海油田开发步伐向中深层迈进，需要对深层机械钻速进行准确预测。首先对影响深层机械钻速的各项因素进行分析并建立了渤中区域深层机械钻速预测神经网络模型，然后结合BZ19-6-X探井实例对钻速预测模型进行了验证，实践证明神经网络预测的结果和实际机械钻速吻合较好，最后结合提出的模型对深层提速工具进行了优选。研究成果对于渤中区域开发工期成本估算和钻井提速增效有指导意义。

Abstract

As Bohai Oilfield steps to the development of medium and deep layers, it is necessary to predict the rate of penetration(ROP) in deep layers accurately.We firstly analyzed all factors which influence the ROP in deep layers and established a neural network model for predicting the ROP in deep layers in Bozhong area.Then, the ROP prediction model was verified based on the case of exploratory well BZ19-6-X.It is practically proved that the prediction result of the neural network is better accordant with the actual ROP.Finally, the tools for improving the ROP in deep layers were optimized based on the neural network model.The research results can be used as the guidance for development cost estimate and ROP and efficiency improvement of Bozhong area.

关键词：

机械钻速; 神经网络; 深层钻井; 渤海油田;

Keywords:

rate of penetration (ROP); neural network; deep drilling; Bohai Oilfield;

基金项目

DOI

10.13639/j.odpt.2018.S0.034