• • 上一篇    

基于灰色-加权马尔可夫的机械钻速动态预测

柳军1,2, 严顾鑫1, 郭晓强1,2, 万敏1, 罗鸣3, 朱海燕4   

  1. 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
  • 收稿日期:2021-05-19 修回日期:2021-11-22 发布日期:2022-08-31
  • 基金资助:
    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.

柳军, 严顾鑫, 郭晓强, 万敏, 罗鸣, 朱海燕. 基于灰色-加权马尔可夫的机械钻速动态预测[J]. 系统科学与数学, 2022, 42(7): 1727-1739.

LIU Jun, YAN Guxin, GUO Xiaoqiang, WAN Min, LUO Ming, ZHU Haiyan. Dynamic Prediction of Drilling Rate Based on Grey-Weighted Markov[J]. Journal of Systems Science and Mathematical Sciences, 2022, 42(7): 1727-1739.

Dynamic Prediction of Drilling Rate Based on Grey-Weighted Markov

LIU Jun1,2, YAN Guxin1, GUO Xiaoqiang1,2, WAN Min1, LUO Ming3, ZHU Haiyan4   

  1. 1. School of Mechanical and Electrical Engineering, Southwest Petroleum University, Chengdu 610500;
    2. School of Mechanical Engineering, Chengdu University, Chengdu 610106;
    3. Zhanjiang Branch, CNOOC (China) Co., Ltd., Zhanjiang 524057;
    4. School of Energy, Chengdu University of Technology, Chengdu 610059
  • Received:2021-05-19 Revised:2021-11-22 Published:2022-08-31
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.

MR(2010)主题分类: 

()
[1] 肖鹏.论石油勘探技术发展状况和发展趋势研究.石化技术, 2020, 27(9):285-286.(Xiao P.Research on development status and trend of petroleum exploration technology.Petrochemical Technology, 2020, 27(9):285-286.)
[2] 李根生,宋先知,田守嶒.智能钻井技术研究现状及发展趋势.石油钻探技术, 2020, 48(1):1-8.(Li G S, Song X Z, Tian S C.Intelligent drilling technology research status and development trends.Petroleum Drilling Techniques, 2020, 48(1):1-8.)
[3] 王敏生,光新军.智能钻井技术现状与发展方向.石油学报, 2020, 41(4):505-512.(Wang M S, Guang X J.Status and development trends of intelligent drilling technology.Acta Petrolei Sinica, 2020, 41(4):505-512.)
[4] 路保平.中国石化石油工程技术新进展与发展建议.石油钻探技术, 2021, 49(1):1-10.(Lu B P.New progress and development proposals of sinopec's petroleum engineering technologies.Petroleum Drilling Techniques, 2021, 49(1):1-10.)
[5] 管志川,陈庭根,等.钻井工程理论与技术.东营:中国石油大学出版社, 2017.(Guan Z C, Chen T G, et al.Drilling Engineering Theory and Technology.Dongying:China University of Petroleum Press, 2017.)
[6] 苗健.石油钻井工程技术的优化措施.化工管理, 2021,(16):80-81.(Miao J.Optimization measures of oil drilling engineering technology.Chemical Enterprise Management, 2021,(16):80-81.)
[7] 王迪.石油钻井工程技术的优化措施.中国石油和化工标准与质量, 2020, 40(11):219-220.(Wang D.Optimization measures of oil drilling engineering technology.China Petroleum and Chemical Industry Standards and Quality, 2020, 40(11):219-220.)
[8] 李昌盛.基于多元回归分析的钻速预测方法研究.科学技术与工程, 2013, 13(7):1740-1744.(Li C S.Study of method for predict rate of penetration based of multiple regression analysis.Science Technology and Engineering, 2013, 13(7):1740-1744.)
[9] Young F S.Computerized drilling control.Journal of Petroleum Technology, 1969, 21(4):493-496.
[10] 许明泽,韦明辉,邓霜,等.多模型集成学习在机械钻速预测中的新应用.计算机科学, 2021, 48(S1):619-622, 657.(Xu M Z, Wei M H, Deng S, et al.Application of multi-model ensemble learning in prediction of mechanical drilling rate.Computer Science, 2021, 48(S1):619-622, 657.)
[11] 李谦,曹彦伟,朱海燕.基于人工智能的钻速预测模型数据有效性下限分析.钻探工程, 2021, 48(3):21-30.(Li Q, Cao Y W, Zhu H Y.Discussion on the lower limit of data validity for ROP prediction based on artificial intelligence.Drilling Engineering, 2021, 48(3):21-30.)
[12] Gan C, Cao W H, Wu M, et al.Prediction of drilling rate of penetration (ROP) using hybrid support vector regression:A case study on the Shennongjia area, Central China.Journal of Petroleum Science and Engineering, 2019, 181:106200.
[13] Hamid R A, Mohammad J, Masoud A.Drilling rate of penetration prediction through committee support vector regression based on imperialist competitive algorithm.Carbonates and Evaporites, 2017, 32(2):205-213.
[14] 赵颖,孙挺,杨进,等.基于极限学习机的海上钻井机械钻速监测及实时优化.中国海上油气, 2019, 31(6):138-142.(Zhao Y, Sun T, Yang J, et al.Extreme learning machine-based offshore drilling ROP monitoring and real-time optimization.China Offshore Oil and Gas, 2019, 31(6):138-142.)
[15] 刘思峰,谢乃明,等.灰色系统理论及其应用.北京:科学出版社, 2017.(Liu S F, Xie N M, et al.Grey System Theory and Its Application.Beijing:Science Press, 2017.)
[16] 肖金山,何涛.基于改进灰色GM (1, 1)模型的铁路货运量预测.兰州交通大学学报, 2021, 40(3):40-45.(Xiao J S, He T.Forecast of railway freight volume based on improved gray GM (1, 1) Model.Journal of Lanzhou Jiaotong University, 2021, 40(3):40-45.)
[17] 杜文然,陆航,杨涛存,等.基于灰色GM (1, 1)残差修正模型的动车组故障率预测.铁道机车车辆, 2021, 41(2):17-21.(Du W R, Lu H, Yang T C, et al.Fault rate prediction of EMU Based on grey GM (1, 1) residual correction model.Railway Locomotive&Car, 2021, 41(2):17-21.)
[18] Du H J, Liu W.Forecast of container demand based on grey Markov model.Scientific Journal of Intelligent Systems Research, 2020, 2(10):14-27.
[19] Balochian S, Baloochian H.Improving grey prediction model and its application in predicting the number of users of a public road transportation system.Journal of Intelligent Systems, 2021, 30(1):104-114.
[20] 张文杰,袁红平.基于灰色马尔可夫模型的节能设备故障预测研究.系统科学与数学, 2019, 39(1):65-75.(Zhang W J, Yuan H P.Forecasting energy-saving equipment failure based on Grey-Markov model.Journal of Systems Science and Mathematical Sciences, 2019, 39(1):65-75.)
[21] Luo D, Zhang G Z.A multiperiod grey prediction model and its application.Journal of Intelligent&Fuzzy Systems, 2021(Preprint):1-10.
[22] 李昊燃,郑度奎,程远鹏,等.优化的灰色马尔科夫模型在油气集输管道腐蚀预测中的应用.腐蚀与防护, 2021, 42(5):42-46.(Li H R, Zheng D K, Cheng Y P, et al.Application of optimized grey Markov model to corrosion prediction of oil and gas gathering and transmission pipeline.Corrosion and Protection, 2021, 42(5):42-46.)
[23] 刘克,曹平,等.马尔可夫决策过程理论与应用.北京:科学出版社, 2015.(Liu K, Cao P, et al.Markov Decision Process Theory and Application.Beijing:Science Press, 2015.)
[24] 白天晟,陈永刚.高铁中基于改进GM (1, 1)马尔可夫模型的频谱预测策略.重庆邮电大学学报(自然科学版), 2021, 33(1):34-43.(Bai T S, Chen Y G.Spectrum prediction strategy based on improved GM (1, 1) Markov model in high-speed railway.Journal of Chongqing University of Posts and Telecommunications (Natural Science), 2021, 33(1):34-43.)
[25] 郑卓,曹辉,高鹤元,等.基于加权马尔可夫链修正的ARIMA预测模型的研究.计算机应用与软件, 2020, 37(12):63-67.(Zheng Z, Cao H, Gao H Y, et al.Research on ARIMA prediction model based on weighted Markov chain modification.Computer Applications and Software, 2020, 37(12):63-67.)
[26] 陈雯,吕王勇,代娟,等.改进的加权马尔可夫链预测方法及应用.统计与决策, 2021, 37(10):179-183.(Chen W, Lü W Y, Dai J, et al.An improved weighted Markov chain prediction method and its application.Statistics and Decision, 2021, 37(10):179-183.)
[27] 罗亚琼.基于灰色马尔科夫模型的基坑开挖变形预测.昆明理工大学, 2021.(Luo Y Q.Prediction of excavation deformation based on Grey Markov Model.Kunming University of Science and Technology, 2021.)
[28] 李绪深,杨计海,范彩伟,等.南海北部海域高温超压天然气勘探新进展与关键技术——以莺歌海盆地乐东斜坡带为例.中国海上油气, 2020, 32(1):23-31.(Li X S, Yang J H, Fan C W, et al.New progress and key technologies for high temperature and overpressure natural gas exploration in the northern part of South China Sea:Taking the Ledong slope belt of Yinggehai Basin as an example.China Offshore Oil and Gas, 2020, 32(1):23-31.)
[1] 王勇,董恒新. 大数据背景下中国季度失业率的预测研究------基于网络搜索数据的分析[J]. 系统科学与数学, 2017, 37(2): 460-472.
[2] 臧人霖,李树荣,葛玉磊,常鹏. 一种求解化工动态优化问题的改进磷虾觅食优化算法[J]. 系统科学与数学, 2016, 36(7): 961-972.
[3] 段渊. 一种新的多目标人工蜂群算法[J]. 系统科学与数学, 2016, 36(1): 93-106.
[4] 马晓君,金爽,杨淑田. 中国电子商务平台产品评论意见挖掘------基于条件随机场模型的实证研究[J]. 系统科学与数学, 2015, 35(11): 1327-1346.
[5] 高玉兰,于俊燕,禹梅. 带有时滞的离散多智能体系统的$H_{\infty}$ 一致性问题[J]. 系统科学与数学, 2015, 35(3): 317-326.
[6] 王泰,高闯,胡祥恩. 突发重大新闻事件中基于兴趣的网民活跃度模型[J]. 系统科学与数学, 2014, 34(3): 294-308.
[7] 王俊杰,王连堂. 一类广义Camassa-Holm方程的多辛Preissmann格式[J]. 系统科学与数学, 2013, 33(11): 1321-1331.
[8] 陶桂平,韩立岩,宋捷. 基于矩的概率分布差异度量[J]. 系统科学与数学, 2013, 33(9): 1071-1082.
阅读次数
全文


摘要