An Intelligent Prediction Method for Drilling Stuck Risk Based on Mechanism Data Fusion

doi:10.12388/j.issn.1673-2677.2025.02.004

Abstract

Abstract:

Stuck pipe incidents significantly disrupt drilling operations and cause major economic losses. Traditional physics-based models for stuck pipe prediction and analysis are subjective with large errors，while intelligent models suffer from high false alarm rates and low interpretability. To address these issues，an unsupervised stuck pipe risk evaluation method based on fuzzy mathematics was proposed. This approach involves：1） establishing a tubular mechanics model to quantify wellbore friction characteristics；2） constructing a deep autoencoder to detect abnormal parameters through reconstruction error analysis；and 3） developing a dual-factor membership function for a comprehensive fuzzy evaluation of friction coefficient trends and reconstruction errors. This method avoids the dependence of the conventional supervised learning on labeled data，integrates the interpretability of mechanistic models with the generalization ability of data-driven models，and creates a physics-constrained intelligent risk assessment framework. Tests using real drilling data show that this model effectively identifies early signs of stuck pipe. It improves warning accuracy by 7.1%，compared to the single-parameter methods，reduces false alarms，and delivers a 30-minute-earlier alert. The proposed method provides a promising new technique for predicting downhole complex issues and possesses significant application potential.

Key words:

"> stuck pipe prediction, fuzzy mathematics, soft-string model, friction calculation, autoencoder

摘要：

卡钻事故严重制约着钻井过程的正常进行，并造成大量的经济损失。现有卡钻预测分析的物理模型方法存在主观性强、误差大等缺陷，智能模型方法存在虚警高、解释性差等问题。针对上述问题，研究提出基于模糊数学的卡钻风险无监督融合评价方法。首先，建立管柱力学软杆模型反演摩阻系数，量化井筒摩擦特征；其次，构建深度自编码器模型，通过重构误差的分析实现阻卡风险的异常参数检测；最后，建立双因素隶属度函数体系，将摩阻系数趋势特征与重构误差统计特征进行模糊数学综合评判。该方法突破传统监督学习对标注数据的依赖，通过机理模型的可解释性与数据模型的泛化能力互补，形成具有物理约束的智能风险评估框架。研究成果通过某油田区块的实钻数据验证，该模型能有效识别卡钻早期征兆，相较于单一参数预警方法，综合预警准确率提升7.1%，虚警显著减少，并实现了提前30 min报警，为井下复杂工况的智能预判提供了新的技术路径，具有较大的工程应用价值和前景。

关键词:

卡钻预测, 模糊数学, 软杆模型, 摩阻反演, 自编码器

CLC Number:

TE28 ','1');return false;" target="_blank">
TE28

ZHANG Yuqiang , YANG Yanlong , ZHANG Wenping , LIU Muchen , ZHU Zhaopeng , WANG Yiwei.

An Intelligent Prediction Method for Drilling Stuck Risk Based on Mechanism Data Fusion

[J]. Xinjiang Oil & Gas, 2025, 21(2): 35-44.

张玉强 , 杨彦龙 , 张文平 , 刘慕臣 , 祝兆鹏 , 王易玮.

机理数据融合的钻井阻卡风险智能预测方法

[J]. 新疆石油天然气, 2025, 21(2): 35-44.

References

［1］赵超. 影响卡钻事故因素的系统分析. 新疆石油科技, 1998, (2, ):1–-6.

Z~~hao~~HAO Chao. Systematic Analysis of Factors Affecting Stuck Pipe Accidents. Xinjiang Petroleum Science and Technology, 1998,(2):1-6~~1998, 2, 1–6~~.

［2］ JOHANCSIK C A， FRIESEN D B， DAWSON R. Torque and drag in directional wells–prediction and measurement[J]. Journal of Petroleum Technology，1984，36(6)：987−-992.

［3］ LESAGE M，FALCONER I G，WICK C J. Evaluating drilling practice in deviated wells with torque and weight data[J]. SPE Drilling Engineering，1992，3(3)：248−-252.

［4］ BRETT J F，BECKETT A D，HOLT C A，et al. Uses and limitations of drillstring tension and torque models for monitoring hole conditions[J]. SPE Drilling Engineering，1992，4(3)：223−-229.

［5］ M~~enand~~ENAND, S~~. et al~~. Advancements in 3D Ddrillstring mechanics: From the Bbit to the Ttopdrive[C]. in IADC/SPE drilling conference (,OnePetro, 2006).

［6］ H~~are~~ARE~~land~~LAND, G., WuU, A~~. &~~ ,L~~ei,~~EI L. The field tests for measurement of downholeweight on bit (DWOB) and the calibration of a real-time DWOB Model[C].. ~~in IPTC 2014:~~ International Petroleum Technology Conference cp-395 (,European Association of Geoscientists & Engineers, 2014).

［7］黄文君，高德利. 受井眼约束带接头管柱的纵横弯曲分析[J]. 西南石油大学学报(:自然科学版)，2015，37(5)：152−-158.

HUANG Wenjun, GAODeli, et al.Longitudinal and transverse bending analysis of pipe strings with joints constrained by wellbore~~Joints Under Wellbore Constraints~~ [J]. Journal of Southwest Petroleum University (Natural Science Edition), 2015, 37(5): 152−-158.

［8］张建群, 孙学增赵俊平. 定向井中摩擦阻力模式及其应用的初步研究[J]..大庆石油学院学报, ~~13, 23– 28 (1989)~~1989,(4):23-28.

Z~~hang~~HANG Jianqun, SunUN Xuezeng & Z~~hao~~HAO Junping. A Preliminary Study on Friction Resistance Models and Their Applications in Directional Wells. Journal of Daqing Petroleum Institute, 1989,(4):23-28.~~1989, 13, 23–28.~~

［9］秦永和, 付胜利高德利. 大位移井摩阻扭矩力学分析新模型[J].. 天然气工业 ,2006,(11):77-79.2~~6, 77– 79 (2006~~).

QinIN Yonghe, FuU Shengli & ,GaoAO Deli. A Nnew Mmodel for Mmechanical Aanalysis of Ffriction and Ttorque in Eextended Rreach Wwells[J]. Natural Gas Industry,2006,(11):77-79.~~, 2006, 26, 77–79.~~

［10］高德利,刘希圣,徐秉业.井眼轨迹控制[M].石油工业出版社,1994.~~高德利, 刘希圣, 许秉业. 井眼轨迹控制. 东营石油大学出版社, 1994.~~

GaoAO Deli, LiuIU Xisheng, XuU Bingye. Wellbore Ttrajectory Ccontrol[M]. Petroleum Industry Press~~Dongying: Petroleum University Press~~, 1994.

［11］朱硕，宋先知，李根生，等. 钻柱摩阻扭矩智能实时分析与卡钻趋势预测[J]. 石油钻采工艺，2021，43(4)：428−-435.

ZhuHU Shuo, S~~ong~~ONG Xianzhi, LiI Gensheng, et al. Intelligent Rreal-time Aanalysis of Ddrillstring Ffriction and Ttorque and Pprediction of Sstuck Ppipe Ttrends [J]. Oil Drilling & Production Technology, 2021, 43(4): 428−-435.

［12］刘慕臣,宋先知,李大钰,等.钻柱摩阻扭矩智能预测模型与解释[J].煤田地质与勘探,2023,51(09):89-99.

LiuIU Muchen, S~~ong~~ONG Xianzhi, LiI Dayu, et al. Intelligent prediction model and interpretation of drill string friction torque [J]. Coal Geology & Exploration, 2023, 51(9): 89-99.

［13］ Z~~hang~~HANG X ,D~~ong~~ONG P ,Y~~ang~~ANG Y , et al.Identification Mmethod of Sstuck Ppipe Bbased on Ddata Aaugmentation and ATT-LSTM[J].Processes,2024,12(7):1296-1296.

［14］ M~~agana~~AGANA-M~~ora~~ORA, A.; ,G~~har~~HARbiBI, S.; ,AlsLS~~hai~~HAIkhKH, A., et al.~~; Al-Yami, A. AccuPipePred:~~ A framework for the accurate and early detection of stuck pipe for real-time drilling operations[C]. In Proceedings of the SPE Middle East Oil and Gas Show and Conference. OnePetro, 2019.

［15］刘光星, 陶宇龙朱丹. ARMA 建模在神经网络卡钻预测方法中的应用研究. 现代电子技术 36, 17–-19 (2013).

LiuIU Guangxing, ,TaoAO Yulong &, ZhuHU Dan. Research on the Aapplication of ARMA Mmodeling in Nneural Nnetwork-Bbased Sstuck Ppipe Pprediction Mmethods[J].Modern Electronic Technology, 2013, 36(22),: 17–-19.

［16］ WuU, J.; ,Z~~ang~~ANG, Y.; ,C~~hen~~HEN, X. Stuck drill pipe type identification analysis based on pattern recognition theory[J].. Mining Engineering (Rock and Soil Boring Engineering), 2015, (42,): 31–-34.

［17］ M, R. K.;, B, H.,; T, N.;, et al. Early sign detection for the stuck pipe scenarios using unsupervised deep learning[J].. Journal of Petroleum Science and Engineering, 2022.

［18］ NygYG~~aard~~AARD R, H~~are~~ARE~~land~~LAND G, BudUD~~ining~~INING~~sih~~SIH Y, et al. Eight years experience with adrilling optimization simulator in the North Sea[C]//. SPE/IADC Asian Conference, Jakarta, Indonesia. 2002: 9-11.

［19］ N~~aka~~AKA~~gawa~~GAWA Y,~~inoue~~INOUE T,B~~ilen~~ILEN H,et al.An unsupervised learning model for pipe stuck predictions using a long short-termmemory autoencoder architecturel[C]//.SPE Asia Pacific Oil and Gas Conference and Exhibition.Adelaide:SPE,2021:D01 1S009R006.

［20］ E, A.; ,T, G. Holistic autonomous model for early detection of downhole drilling problems in real-time[J]. Process Safety and Environmental Protection, 2022, ~~pp.~~ 418–-434.

［21］刘建明,李玉梅,张涛,等.一种基于PCA-RF的卡钻预测方法[J].北京信息科技大学学报(:自然科学版),2021,36(01):18-22.

LiuIU Jianming, LiI Yumei, Z~~hang~~HANG Tao, et al. A Sstuck Ppipe Pprediction Mmethod Bbased on PCA-RF [J]. Journal of Beijing Information Science & Technology University (Natural Science Edition), 2021, 36(01): 18-22.

［22］富浩,张涛,李玉梅,等.基于井下参数的PCA-SVM卡钻预测研究[J].计算机仿真,2021,38(12):386-390.

FuU Hao, Z~~hang~~HANG Tao, LiI Yumei, et al. Research on PCA-SVM Sstuck Ppipe Pprediction Bbased on Ddownhole Pparameters [J]. Computer Simulation, 2021, 38(12): 386-390.

［23］邱少林, 张来斌, 梁伟,等. ~~栾国华.~~ 钻井井下事故风险模糊评估技术研究与应用[J]. 钻采工艺 ,2017,40(4), :17– -20 ~~(2017)~~.

QiuIU Shaolin, Z~~hang~~HANG Laibin, L~~iang~~IANG Wei , et al.~~& Luan Guohua.~~ Research and Aapplication of Ffuzzy Eevaluation Ttechnology for Ddownhole Aaccident Rrisks in Ddrilling[J].. Drilling & Production Technology, 2017,40(4):17-20.~~2017, 40, 17–20.~~

［24］ B~~ran~~RAN~~kovic~~KOVIC A,M~~atte~~ATTE~~ucci~~UCCI M, R~~este~~ESTE~~lli~~LLI M,et al.A data-based approach for the prediction of stuck-pipe events in oil drillinooperations[C]//.Abu Dhabilnternational Petroleum Exhibition and Conference.Abu Dhabi:SPE,2020:D041S11-1R003.

[1]	LIU Wei , XIE Fengmeng , LI Jianchao , LIU Xifeng , HU Bin , ZHANG Yu , GAO Deli. Intelligent Recognition Method for PDC Bit Damage [J]. Xinjiang Oil & Gas, 2025, 21(2): 45-.
[2]	WANG Jinlong , ALENA Pevcheva. Design of New Electric Control Downhole Flow Control Valve for Intelligent Wells [J]. Xinjiang Oil & Gas, 2025, 21(2): 73-.
[3]	LIU Mingyan , ZHANG Zhunxi , WANG Xintong , ZHAO Haiyan , WU Yuanyue , LIU Yongjie. Research on Optimization of Drilling Investment Estimation Based on Parameter Extraction and Simulation [J]. Xinjiang Oil & Gas, 2025, 21(2): 82-.
[4]	DING Jianxin , WANG Junshan , WANG Xiao , WANG Haitao , Qin Ji , LI Hui , LU Gang. Optimization Algorithm for Trajectory Design of Double-Step Horizontal Wellbore [J]. Xinjiang Oil & Gas, 2025, 21(2): 91-.
[5]	XU Wenrui , WANG Liye , WEI Chencheng , QIN Xin , LIU Yiyan , PENG Xiao. Application and Prospect of LiDAR Technology in Seismic Acquisition of Complex Mountainous Areas#br# [J]. Xinjiang Oil & Gas, 2025, 21(1): 1-9.
[6]	ZHANG Li , WANG Bo , LYU Zhenhu , LYU Bei , LI Lizhe , ZHOU Hang. Intelligent Optimization of Integral Fracturing in Unconventional Reservoirs [J]. Xinjiang Oil & Gas, 2024, 20(4): 36-43.
[7]	XU Jiangwen, WANG Mingxing, WANG Junchao, SUN Haoran, WANG Liang. Practice and Understanding of Volume Fracturing Technology for Horizontal Wells in Jimsar Shale Oil Reservoir [J]. Xinjiang Oil & Gas, 2024, 20(3): 23-30.
[8]	SHENG Mao , DENG Chao , LI Jie , GU Mingzhe , WANG Tianyu , TIAN Shouceng, . Non-Uniform Perforation to Balance Multi-Cluster Fractures Propagation and Parameter Optimization [J]. Xinjiang Oil & Gas, 2024, 20(3): 54-63.
[9]	MOU Jianye , YAN Xiaolun , ZHANG Shicheng , SHI Lei , LI Dong , MA Xinfang. Research on CO2 Matrix Penetration Distance in Tight Reservoirs [J]. Xinjiang Oil & Gas, 2024, 20(3): 64-71.
[10]	JIANG Beibei, ZHANG Guoqiang, WANG Dong, LIU Jiabo, YANG Qianlong, XU Xiaohui. Prediction of Critical Drawdown Pressure of Sand Production in Ultra-Deep High Pressure Sandstone Gas Reservoirs [J]. Xinjiang Oil & Gas, 2024, 20(2): 11-.
[11]	LIU Fan, LIU Yushuang, ZHANG Zhen, LI Yongjian, LIU Ce, MA Zhihu, . A Machine Learning-Based Bridging Particle Size Recommendation Method for Lost Circulation Control [J]. Xinjiang Oil & Gas, 2024, 20(1): 13-20.
[12]	LIU Fengbao, SUN Jinsheng, WANG Jianhua. A Global Review of Technical Status and Development Trend of Drilling Fluids for Deep and Ultra-Deep Wells [J]. Xinjiang Oil & Gas, 2023, 19(2): 34-39.
[13]	CUI Xiaobo, ZHANG Jiaqi, YAN Lili, ZOU Jun, YANG Haijun, LIU Zheng. Application of High-Temperature Resistant and High-Density Oil-Based Drilling Fluid in Well DS1 [J]. Xinjiang Oil & Gas, 2023, 19(2): 56-61.
[14]	YUAN Yuehui, QU Yuanzhi, GAO Shifeng, RONG Kesheng, YE Cheng, LIU Kecheng. Advances in Study on Temperature-Resistant and Salt-Tolerant Fluid Loss Reducers for Water-Based Drilling Fluids [J]. Xinjiang Oil & Gas, 2023, 19(2): 62-68.
[15]	LIU Lili, XIE Yingming, LI Yuzhan, Cao Yi, Zeng Shunpeng. The Law of Gas-Liquid Two-Phase Seepage for Workover Gas Injection Huff and Puff Production Recovery in Low-Pressure Gas Wells [J]. Xinjiang Oil & Gas, 2023, 19(2): 75-81.