基于灰色-加权马尔可夫的机械钻速动态预测  

Dynamic Prediction of Drilling Rate Based on Grey-Weighted Markov

作　　者：柳军[1,2] 严顾鑫 郭晓强 万敏 罗鸣[3] 朱海燕 LIU Jun;YAN Guxin;GUO Xiaoqiang;WAN Min;LUO Ming;ZHU Haiyan(School of Mechanical and Electrical Engineering,Southwest Petroleum University,Chengdu 610500;School of Mechanical Engineering,Chengdu University,Chengdu 610106;Zhanjiang Branch,CNOOC(China)Co.,Ltd.,Zhanjiang 524057;School of Energy,Chengdu University of Technology,Chengdu 610059)

机构地区：[1]西南石油大学机电工程学院,成都610500 [2]成都大学机械工程学院,成都610106 [3]中海石油(中国)有限公司湛江分公司,湛江524057 [4]成都理工大学能源学院,成都610059

出　　处：《系统科学与数学》2022年第7期1727-1739,共13页Journal of Systems Science and Mathematical Sciences

基　　金：国家自然科学基金(面上项目)(51875489);国家建设高水平大学公派研究生项目(201908510191);四川省青年科技创新研究团队专项计划项目(2019JDTD0017)联合资助课题。

摘　　要：在钻井工程中,科学准确地预测机械钻速(Rate of Penetration,ROP)对高效钻井以及钻井智能化发展具有重要意义.针对ROP的预测实效性以及波动性大的问题,在灰色系统理论的基础上,结合马尔可夫理论,建立灰色-加权马尔可夫机械钻速动态预测模型.其方法是利用灰色预测模型对ROP进行模拟预测,根据预测值的误差范围划分几个误差分布的状态空间,以不同转移步长间的状态转移信息构建状态转移概率矩阵,再利用该矩阵和不同井段的状态信息,对下一井段的ROP值进行状态空间判断并修正,同时通过不断更新模型中的数据,实现动态实时预测.以南海莺歌海盆地M井进行实例分析.结果表明:经灰色预测模型与加权马尔可夫的有效结合,ROP预测值拟合曲线不仅更加贴近实测值,而且降低了误差的离散性,具有更高的预测可信度,进而有效验证了该模型的有效性.In drilling engineering,scientific and accurate prediction of Rate of Penetration(ROP)is of great significance to the development of efficient drilling and intelligent drilling.Aiming at the problem of ROP prediction effectiveness and volatility,a gray-weighted Markov ROP dynamic prediction model was established based on the grey system theory and Markov theory.Its method is the use of grey forecasting model for simulating ROP prediction,according to the forecast error range of several error distribution of state space,in the state transition between different shift step length information construction of state transition probability matrix,using the matrix and the state information of different interval,ROP value of the next interval state space and correction,At the same time,dynamic real-time prediction is realized by updating the data in the model.Taking M well in Yinggehai Basin,South China Sea as an example.The results show that the ROP predicted value fitting curve is not only more close to the measured values,but also reduces the discretization of the error and has higher prediction reliability by the effective combination of the grey prediction model and weighted Markov model,which verifies the effectiveness of the model.

关 键 词：钻井工程 机械钻速 灰色理论 马尔可夫理论 动态预测 

分 类 号：TE24[石油与天然气工程—油气井工程]