Machine learning-based diagnosis of fracture-induced lost circulation and intelligent selection of lost circulation control fluids

Authors

• SUN Huan, Changqing Drilling General Company, CNPC Chuanqing Drilling Engineering Company Limited, Xi’an 710018, ChinaFollow
• ZHU Mingming, Changqing Drilling General Company, CNPC Chuanqing Drilling Engineering Company Limited, Xi’an 710018, China
• WANG Jinshu, Petroleum Engineering Department, Hebei Petroleum University of Technology, Chengde 067000, China;Hebei Technology Innovation Center for Big Data Application in Petroleum Enterprises, Chengde 067000, ChinaFollow
• YAO Lu, Changqing Drilling General Company, CNPC Chuanqing Drilling Engineering Company Limited, Xi’an 710018, China
• ZHOU Fangfang, Petroleum Engineering Department, Hebei Petroleum University of Technology, Chengde 067000, China
• TAN Xuebin, Changqing Drilling General Company, CNPC Chuanqing Drilling Engineering Company Limited, Xi’an 710018, China
• SUN Yan, Changqing Drilling General Company, CNPC Chuanqing Drilling Engineering Company Limited, Xi’an 710018, China
• QU Yanping, Changqing Drilling General Company, CNPC Chuanqing Drilling Engineering Company Limited, Xi’an 710018, China
• CHEN Ning, Changqing Drilling General Company, CNPC Chuanqing Drilling Engineering Company Limited, Xi’an 710018, China

Abstract

Objective and Method Drilling operations in the Changqing area face several challenges in lost circulation, including complex types, significantly different characteristics, and difficulty in designing plugging formulations. To address these issues, this study proposes a DSX-Hybrid coupled intelligent decision-making framework, which serves as an integrated scheme to identify lost circulation channels, perform intelligent selection of plugging formulations, and offer recommendations on construction process. In this framework, given the scarcity of labeled log data, the density-based spatial clustering of applications with noise (DBSCAN) is employed to mine latent geological characteristics of unlabeled data. Then, the self-attention convolutional neural network (SACNN) is constructed to integrate local features with long-range layer-wise correlations, thus enabling the semi-supervised prediction of fracture parameters. Lastly, the physics-constrained XGBoost model, combined with eight-dimensional characteristic parameters, is used to achieve the intelligent selection and performance prediction of plugging formulations. Concurrently, standardized construction procedures for lost circulation control are clarified. Results and Conclusions The diagnostic accuracy of lost circulation channels is increased from 73% to 94.6%. The semi-supervised model can make full use of massive unlabeled logging data and achieves more stable identification of fractures with various scales compared with traditional algorithms. The DSX-Hybrid model realizes accurate matching between fractures and particle sizes of plugging materials with the coefficient of determination (R²) over 0.945. The indoor plugging pressure bearing capacity of recommended formulas reaches 10.7 MPa, and the stable on-site pressure bearing capacity is 7.6 MPa, with a one-time plugging success rate of 100% in multiple blocks of Changqing Oilfield. This model provides a practical intelligent technical approach for lost circulation control in Changqing area. Further research can integrate real-time logging-while-drilling data, enrich datasets of extreme lost circulation conditions, and extend the application to other fractured oil and gas blocks.

Keywords

fracture-induced lost circulation, semi-supervised learning, diagnosis of lost circulation channels, DSX-Hybrid framework, intelligent selection of lost circulation control fluids

DOI

10.12363/issn.1001-1986.25.10.0756

Recommended Citation

SUN Huan, ZHU Mingming, WANG Jinshu, et al. (2026) "Machine learning-based diagnosis of fracture-induced lost circulation and intelligent selection of lost circulation control fluids," Coal Geology & Exploration: Vol. 54: Iss. 8, Article 24.
DOI: 10.12363/issn.1001-1986.25.10.0756
Available at: https://cge.researchcommons.org/journal/vol54/iss8/24

Reference

[1] 孙金声,刘凡,程荣超,等. 机器学习在防漏堵漏中研究进展与展望[J]. 石油学报,2022,43(1):91−100 SUN Jinsheng,LIU Fan,CHENG Rongchao,et al. Research progress and prospects of machine learning in lost circulation control[J]. Acta Petrolei Sinica,2022,43(1):91−100

[2] 孙金声,白英睿,程荣超,等. 裂缝性恶性井漏地层堵漏技术研究进展与展望[J]. 石油勘探与开发,2021,48(3):630−638 SUN Jinsheng,BAI Yingrui,CHENG Rongchao,et al. Research progress and prospect of plugging technologies for fractured formation with severe lost circulation[J]. Petroleum Exploration and Development,2021,48(3):630−638

[3] 李伟,白英睿,李雨桐,等. 钻井液堵漏材料研究及应用现状与堵漏技术对策[J]. 科学技术与工程,2021,21(12):4733−4743 LI Wei,BAI Yingrui,LI Yutong,et al. Research and application progress of drilling fluid lost circulation materials and technical countermeas-ures for lost circulation control[J]. Science Technology and Engineering,2021,21(12):4733−4743

[4] 李恒丰,李琳,赵 志 峰. 井漏预测模型的超参数优化与解释:以伊朗马伦油田为例[J]. 科学技术与工程,2024,24(35):15006−15013 LI Hengfeng,LI Lin,ZHAO Zhifeng. Drilling fluid lost-circulation prediction model hyperparameter optimization and interpretation:a casestudy of the Marun field in Iran[J]. Science Technology and Engineering,2024,24(35):15006−15013

[5] 刘峻玮,李善建,王玉功,等. 裂缝性地层高效智能堵漏技术发展综述与展望[J]. 西安石油大学学报(自然科学版),2026,41(1):118−124 LIU Junwei,LI Shanjian,WANG Yugong,et al. Review and prospect of efficient and intelligent plugging technology for fractured reservoirs[J]. Journal of Xi’an Shiyou University (Natural Science Edition),2026,41(1):118−124

[6] AHMED A,BASFAR S,ELKATATNY S. Sagging prevention for hematite–based invert emulsion mud[J]. Journal of Energy Resources Technology,2022,144(7):073003.

[7] ZHANG He,LUO Rong,YANG Hai. Calibration modeling of drilling fluid rheological parameters in variable temperature environment based on support vector machine[J]. Review of Scientific Instruments,2024,95(10):105107.

[8] ZHU Danxia. Leveraging machine–aided learning in college English education:Computational approaches for enhancing student outcomes and pedagogical efficiency[J]. International Journal of Advanced Computer Science and Applications,2025,16(2):1279−1287.

[9] ZHU Kai,DU Yonghui,WANG Qian,et al. A comparative study of five networks for reservoir classification based on geophysical logging signals[J]. IEEE Access,2020,8:197776−197786.

[10] 冯荟锦,曹渴望,王中华. 苏里格气田西区易漏地层高效治漏技术[J]. 石油工业技术监督,2025,41(12):34−37 FENG Huijin,CAO Kewang,WANG Zhonghua. Efficient leakage control technology for easily leaking strata in the west area of Sulige gas field[J]. Technology Supervision in Petroleum Industry,2025,41(12):34−37

[11] DELAVAR M R,RAMEZANZADEH A. Machine learning classification approaches to optimize ROP and TOB using drilling and geomechanical parameters in a carbonate reservoir[J]. Journal of Petroleum Exploration and Production Technology,2024,14(6):1−26.

[12] 曾锐,邵燕林,黄宇,等. 基于深度学习的测井岩性智能识别[J]. 科学技术与工程,2025,25(30):12804−12812 ZENG Rui,SHAO Yanlin,HUANG Yu,et al. Intelligent identification of logging lithology based on deep learning[J]. Science Technology andEngineering,2025,25(30):12804−12812

[13] 孙欢,朱明明,钟锋,等. 鄂尔多斯盆地浅层厘米级裂缝漏失井高效治理技术[J]. 石油化工应用,2026,45(2):27−30 SUN Huan,ZHU Mingming,ZHONG Feng,et al. Efficient treatment technology for shallow centimeter level crack leakage wells in the Ordos Basin[J]. Petrochemical Industry Application,2026,45(2):27−30

[14] 田建华,朱博华,卢志强,等. 基于井控多属性机器学习的缝洞型储层预测方法[J]. 油气地质与采收率,2023,30(1):86−92 TIAN Jianhua,ZHU Bohua,LU Zhiqiang,et al. Fracture–cavity reservoir prediction based on well–controlled multi–attribute machine learning[J]. Petroleum Geology and Recovery Efficiency,2023,30(1):86−92

[15] SABAH M,TALEBKEIKHAH M,AGIN F,et al. Application of decision tree,artificial neural networks,and adaptive neuro–fuzzy inference system on predicting lost circulation:A case study from Marun oil field[J]. Journal of Petroleum Science and Engineering,2019,177:236−249.

[16] 陈曾伟. 基于神经网络算法的井下裂缝诊断与堵漏技术[J]. 钻井液与完井液,2019,36(1):20−24 CHEN Zengwei. Downhole fracture diagnosis and mud loss control technologies bases on neural network algorithm[J]. Drilling Fluid & Completion Fluid,2019,36(1):20−24

[17] 王健,徐加放,赵密福,等. 基于神经网络的钻井液漏失裂缝宽度预测研究[J]. 煤田地质与勘探,2023,51(9):81−88 WANG Jian,XU Jiafang,ZHAO Mifu,et al. Prediction of crack width of drilling fluid leakage based on neural network[J]. Coal Geology & Exploration,2023,51(9):81−88

[18] 陈叶旺,曹海露,陈谊,等. 面向大规模数据的DBSCAN加速算法综述[J]. 计算机研究与发展,2023,60(9):2028−2047 CHEN Yewang,CAO Hailu,CHEN Yi,et al. Survey on DBSCAN acceleration algorithms for large scale data[J]. Journal of Computer Research and Development,2023,60(9):2028−2047

[19] LI Mingyang,BI Xinhua,WANG Limin,et al. A method of two–stage clustering learning based on improved DBSCAN and density peak algorithm[J]. Computer Communications,2021,167:75−84.

[20] JEBARI S，SMITI A，LOUATI A. AF–DBSCAN：An unsupervised Automatic Fuzzy Clustering method based on DBSCAN approach[C]//2019 IEEE International Work Conference on Bioinspired Intelligence (IWOBI). Budapest：IEEE，2019：000001–000006.

[21] 韩复兴,刘水源,高正辉,等. 基于机器学习的微动HVSR数据干扰信号压制方法[J]. 吉林大学学报(地球科学版),2025,55(6):2153−2163 Han Fuxing,Liu Shuiyuan,Gao Zhenghui,et al. A Transient Interference suppression for microtremor HVSR data based on machine learning[J]. Journal of Jilin University (Earth Science Edition),2025,55(6):2153−2163

[22] CHEN Tianqi，GUESTRIN C. XGBoost：A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco：ACM，2016：785–794.

[23] 赵军,汪峻宇,赖强,等. 基于XGBoost算法的走滑断裂内部特征带的精细识别[J]. 地质科技通报,2025,44(2):182−192 ZHAO Jun,WANG Junyu,LAI Qiang,et al. Fine-grained identification of intemal characteristic zones within strike-slip faults via the XGBoost algorithm[J]. Bulletin of Geological Science and Technology,2025,44(2):182−192

[24] 林朋,孙成,任珂,等. 基于PSO–XGBoost的煤层断层智能识别方法研究[J]. 矿业科学学报,2025,10(1):57−69 LIN Peng,SUN Cheng,REN Ke,et al. Research on intelligent fault identification method of coalfield based on the PSO–XGBoost algorithm[J]. Journal of Mining Science and Technology,2025,10(1):57−69

[25] ZHAO Long,WANG Yuhang,SHI Yi,et al. Selecting essential factors for predicting reference crop evapotranspiration through tree–based machine learning and Bayesian optimization[J]. Theoretical and Applied Climatology,2024,155(4):2953−2972.

[26] 彭浩,李黔,尹虎,等. Lietard天然裂缝宽度预测模型求解新方法[J]. 石油钻探技术,2016,44(3):72−76 PENG Hao,LI Qian,YIN Hu,et al. A new solution method for the Lietard natural fracture width prediction model[J]. Petroleum Drilling Techniques,2016,44(3):72−76

[27] 孙伟峰,冯剑寒,张德志,等. 结合LSTM自编码器与集成学习的井漏智能识别方法[J]. 石油钻探技术,2024,52(3):61−67 SUN Weifeng,FENG Jianhan,ZHANG Dezhi,et al. An intelligent lost circulation recognition method using LSTM–autoencoder and ensemble learning[J]. Petroleum Drilling Techniques,2024,52(3):61−67

[28] 陈卫明,王家文,凡东,等. 矿山救援钻孔中井涌井漏事故预警预测[J]. 煤田地质与勘探,2024,52(3):144−152 CHEN Weiming,WANG Jiawen,FAN Dong,et al. Early warning and prediction of kicks and lost circulation accident during rescue drilling of mine[J]. Coal Geology & Exploration,2024,52(3):144−152

[29] 张金发，冯永存，慕小龙，等. 桥接堵漏材料研究进展及发展趋势[J/OL]. 特种油气藏，2026：1–14 [2026-01-30]. https://link.cnki.net/urlid/21.1357.TE.20260129.1236.004.

[30] 李红蕾,陈石,庄建仓,等. 多源布格重力异常数据贝叶斯融合算法及其在川滇地区的应用[J]. 地球物理学报,2021,64(9):3232−3245 LI Honglei,CHEN Shi,ZHUANG Jiancang,et al. A Bayesian data fusion algorithm for multi–source Bouguer gravity anomalies and its application in the Sichuan–Yunnan region[J]. Chinese Journal of Geophysics,2021,64(9):3232−3245

[31] JAGTAP A D,EM KARNIADAKIS G. Extended physics–informed neural networks (XPINNs):A generalized space–time domain decomposition based deep learning framework for nonlinear partial differential equations[J]. Communications in Computational Physics,2020,28(5):2002−2041.