Early warning and prediction of kicks and lost circulation accident during rescue drilling of mine

Authors

CHEN Weiming, Faculty of Engineering, China University of Geosciences, Wuhan 430074, China; Institute for Natural Disaster Risk Prevention and Emergency Management, China University of Geosciences, Wuhan 430074, China; Shaanxi Key Laboratory of Prevention and Control Technology for Coal Mine Water Hazard, Xi’an 710077, ChinaFollow
WANG Jiawen, Faculty of Engineering, China University of Geosciences, Wuhan 430074, China
FAN Dong, CCTEG Xi’an Research Institute (Group) Co., Ltd., Xi’an 710077, China
HAO Shijun, CCTEG Xi’an Research Institute (Group) Co., Ltd., Xi’an 710077, China
ZHAO Jiangpeng, CCTEG Xi’an Research Institute (Group) Co., Ltd., Xi’an 710077, China
QIU Yu, Faculty of Engineering, China University of Geosciences, Wuhan 430074, China

Abstract

In order to solve the problems such as the difficulty in early warning and prediction of kicks and lost circulation accidents during emergency rescue drilling of mine, a machine learning-based early for warning and prediction model of drilling process was established. Firstly, the accident characterization parameters of the drilling parameters in the early stage of kicks and lost circulation accidents were analyzed. Secondly, the accident characterization parameters were cleaned and processed. On this basis, XGBoost and early warning model was used to carry out the early diagnosis and identification of kicks and lost circulation accidents. Then, the PSO-LSTM accident development prediction model was established to predict the development trend of downhole pressure parameters after the accident, so as to understand the development status of drilling accidents in advance. Finally, the effectiveness of the early warning prediction model was verified by the actual drilling data. The results show that XGBoost and early warning model could quickly and accurately diagnose the kick and loss circulation accidents in the drilling process according to the abnormal changes of four drilling parameters, including the total tank volume, riser pressure, inflow-outflow differential, and power head load. The PSO-LSTM accident development status prediction model could fully learn the development law of downhole pressure parameters. With full consideration to the four error evaluation indicators, such as E_MAP, E_MA, E_RMS and R², the prediction performance of the PSO-LSTM models is the best compared with BP, RNN and SVM, capable of accurately predicting the development trend of the downhole pressure after the accident, thereby knowing about the severity and development situation of kick and lost circulation accidents. Generally, the research results enrich the early warning and prediction methods of kicks and lost circulation accidents in the drilling process, improve the reliability of surface rescue in mine accident, and have a reference and guiding effect on accident control during the emergency rescue drilling of mine.

Keywords

mine accident rescue, drilling process, early warning and prediction, XGBoost, PSO-LSTM, machine learning, kick and lost circulation

DOI

10.12363/issn.1001-1986.23.07.0449

Recommended Citation

CHEN Weiming, WANG Jiawen, FAN Dong, et al. (2024) "Early warning and prediction of kicks and lost circulation accident during rescue drilling of mine," Coal Geology & Exploration: Vol. 52: Iss. 3, Article 15.
DOI: 10.12363/issn.1001-1986.23.07.0449
Available at: https://cge.researchcommons.org/journal/vol52/iss3/15

Reference

[1] 赵江鹏，郝世俊. 气动潜孔锤技术在矿山事故钻孔救援中的应用[J]. 煤田地质与勘探，2022，50(11)：24−34.

ZHAO Jiangpeng，HAO Shijun. Application of pneumatic DTH hammer in drilling rescue of mine accidents[J]. Coal Geology & Exploration，2022，50(11)：24−34.

[2] 郑学召，孙梓峪，郭军，等. 矿山钻孔救援多源信息探测技术研究与应用[J]. 煤田地质与勘探，2022，50(11)：94−102.

ZHENG Xuezhao，SUN Ziyu，GUO Jun，et al. Research and application of multi–source information detection technology for drilling rescue of mine[J]. Coal Geology & Exploration，2022，50(11)：94−102.

[3] 王雷，朱玉芹，张维娜，等. 煤矿大直径救援钻孔顶管钻进装备关键技术研究[J]. 煤田地质与勘探，2022，50(11)：58−66.

WANG Lei，ZHU Yuqin，ZHANG Weina，et al. Research on key technology of pipe jacking drilling equipment for large–diameter rescue borehole of coal mine[J]. Coal Geology & Exploration，2022，50(11)：58−66.

[4] 田宏亮，邹祖杰，郝世俊，等. 矿山灾害生命保障救援通道快速安全构建关键技术与装备[J]. 煤田地质与勘探，2022，50(11)：1−13.

TIAN Hongliang，ZOU Zujie，HAO Shijun，et al. Key technologies and equipment of quickly and safely building life support and rescue channel in mine disaster[J]. Coal Geology & Exploration，2022，50(11)：1−13.

[5] 张会利. 钻井现场井涌井漏事故原因分析及对策研究[J]. 西部探矿工程，2022，34(9)：107−108.

ZHANG Huili. Research on cause analysis and countermeasures of welling–loss accident in drilling site[J]. West–China Exploration Engineering，2022，34(9)：107−108.

[6] 陈凯枫，杨学文，宋先知，等. 基于工程录井数据的井漏智能诊断方法[J]. 石油机械，2022，50(11)：16−22.

CHEN Kaifeng，YANG Xuewen，SONG Xianzhi，et al. An intelligent diagnosis method for lost circulation based on engineering logging data[J]. China Petroleum Machinery，2022，50(11)：16−22.

[7] 万康，马志超，郭青松，等. 人工智能技术在石油钻井工程事故预警中的应用[J]. 录井工程，2022，33(2)：24−29.

WAN Kang，MA Zhichao，GUO Qingsong，et al. Application of artificial intelligence on petroleum drilling engineering accident early warning[J]. Mud Logging Engineering，2022，33(2)：24−29.

[8] GALDINO J F，OLIVEIRA G M，FRANCO A T，et al. Gas kick detection and pressure transmission in thixotropic，compressible drilling fluids[J]. Journal of Petroleum Science and Engineering，2019，180：138−149.

[9] SUN Wenfeng，LI Maoran，LIU Chunyang，et al. Experimental research on lost circulation and overflow monitoring system for drilling engineering based on the requirements of safe and efficient oil and gas drilling[J]. Fresenius Environmental Bulletin，2021，30(2)：1511−1517.

[10] KANG Yili，MA Chenglin，XU Chengyuan，et al. Prediction of drilling fluid lost–circulation zone based on deep learning[J]. Energy，2023，276：127495.

[11] 朱晓东，袁坤杰，王艳玲. 基于隐马尔科夫模型的石油钻井事故预警方法[J]. 郑州大学学报(工学版)，2018，39(4)：51−57.

ZHU Xiaodong，YUAN Kunjie，WANG Yanling. An accident prewarning method of oil drilling based on hidden Markov model[J]. Journal of Zhengzhou University (Engineering Science)，2018，39(4)：51−57.

[12] 王海彪. 井漏智能识别及处理决策研究[D]. 成都：西南石油大学，2017.

WANG Haibiao. Study of intelligent recognition and management decisions of lost circulation[D]. Chengdu：Southwest Petroleum University，2017.

[13] ADEDIGBA S A，OLORUNTOBI O，KHAN F，et al. Data–driven dynamic risk analysis of offshore drilling operations[J]. Journal of Petroleum Science and Engineering，2018，165：444−452.

[14] PANG Huiwen，MENG Han，WANG Hanqing，et al. Lost circulation prediction based on machine learning[J]. Journal of Petroleum Science and Engineering，2022，208：109364.

[15] 张正，赖旭芝，陆承达，等. 基于贝叶斯网络的钻进过程井漏井涌事故预警[J]. 探矿工程(岩土钻掘工程)，2020，47(4)：114−121.

ZHANG Zheng，LAI Xuzhi，LU Chengda，et al. Lost circulation and kick accidents warning based on Bayesian network for the drilling process[J]. Exploration Engineering (Rock & Soil Drilling and Tunneling)，2020，47(4)：114−121.

[16] 王钰豪. 钻井过程中的溢流复杂预警算法研究与实现[D]. 成都：电子科技大学，2022.

WANG Yuhao. Research and implementation of early kick detection algorithm in drilling process[D]. Chengdu：University of Electronic Science and Technology of China，2022.

[17] 孟翰. 基于机器学习的井漏风险评估及优化设计[D]. 北京：中国石油大学(北京)，2020.

MENG Han. Fluid loss risk prediction and optimization design based on machine learning[D]. Beijing：China University of Petroleum(Beijing)，2020.

[18] 王茜，张菲菲，李紫璇，等. 基于钻井模型与人工智能相耦合的实时智能钻井监测技术[J]. 石油钻采工艺，2020，42(1)：6−15.

WANG Xi，ZHANG Feifei，LI Zixuan，et al. Real–time intelligent drilling monitoring technique based on the coupling of drilling model and artificial intelligence[J]. Oil Drilling & Production Technology，2020，42(1)：6−15.

[19] 张欣. 基于案例推理的井漏诊断与决策系统研究[D]. 北京：中国石油大学(北京)，2019.

ZHANG Xin. Study on lost circulation diagnosis and decision–making system based on case–based reasoning[D]. Beijing：China University of Petroleum (Beijing)，2019.

[20] 涂曦予. 工业数据挖掘在井漏事故上的应用与分析[D]. 上海：上海交通大学，2019.

TU Xiyu. Analysis and application of industrial data mining of well leakage[D]. Shanghai：Shanghai Jiao Tong University，2019.

[21] MA Meihong，ZHAO Gang，HE Bingshun，et al. XGBoost–based method for flash flood risk assessment[J]. Journal of Hydrology，2021，598：126382.

[22] 许飞青，李潇，李凯，等. 随机森林回归模型在地下水水质评价的新应用[J]. 地质与勘探，2023，59(2)：408−417.

XU Feiqing，LI Xiao，LI Kai，et al. New application of random forest regression model on groundwater quality assessment[J]. Geology and Exploration，2023，59(2)：408−417.

[23] 郑卓，宋峙潮，陈波，等. 基于XGBoost算法的井漏预警模型研究[J]. 石油化工应用，2023，42(1)：112−115.

ZHENG Zhuo，SONG Zhichao，CHEN Bo，et al. Research on leakage warning model based on XGBoost algorithm[J]. Petrochemical Industry Application，2023，42(1)：112−115.

[24] 董丽丽，费城，张翔，等. 基于LSTM神经网络的煤矿突水预测[J]. 煤田地质与勘探，2019，47(2)：137−143.

DONG Lili，FEI Cheng，ZHANG Xiang，et al. Coal mine water inrush prediction based on LSTM neural network[J]. Coal Geology & Exploration，2019，47(2)：137−143.

[25] 刘慧，刘桂芹，宁殿艳，等. 基于 VMD–DBN 的矿井涌水量预测方法[J]. 煤田地质与勘探，2023，51(6)：13−21.

LIU Hui，LIU Guiqin，NING Dianyan，et al. Mine water inrush prediction method based on VMD–DBN model[J]. Coal Geology & Exploration，2023，51(6)：13−21.

[26] 李占利，邢金莎，靳红梅，等. 基于EMD和时序注意力机制的明渠流量预测模型[J]. 高技术通讯，2022，32(2)：122−130.

LI Zhanli，XING Jinsha，JIN Hongmei，et al. Open channel flow prediction model based on EMD and temporal attention mechanism[J]. Chinese High Technology Letters，2022，32(2)：122−130.

[27] 唐晓灵，刘嘉敏. 基于PSO–LSTM网络模型的建筑碳排放峰值预测[J]. 科技管理研究，2023，43(1)：191−198.

TANG Xiaoling，LIU Jiamin. Forecast of peak carbon emissions of buildings based on PSO–LSTM model[J]. Science and Technology Management Research，2023，43(1)：191−198.

[28] XU Yuanhao，HU Caihong，WU Qiang，et al. Research on particle swarm optimization in LSTM neural networks for rainfall–runoff simulation[J]. Journal of Hydrology，2022，608：127553.