井控压井节流管汇自适应控制方法与试验

Adaptive control method and experiment of throttling manifolds for well killing

摘要

针对现有压井节流管汇手动控制压力波动大、滞后性强等问题,考虑压井过程中压力响应、含气率变化、设备状态等对控制参数的影响,设计耦合 BP 神经网络的压井节流管汇自适应 PID 控制器(PIDNN),研制压井节流管汇自适应控制系统,开展室内物理模拟试验和现场测试试验。结果表明:不同恒定和连续变化目标压力情况下,PIDNN自适应控制系统完全追踪至目标压力的时间小于 30 s,实际压力波动范围不超过系统本身;相对于手动控制和常规PID 自动控制,响应速度提高 30%以上,平均绝对误差降幅超 40%;该系统能够适用于不同类型的节流阀,在全开度范围内可以实现对目标压力的准确、稳定、快速跟踪,且对含气率变化引起的压力突变具有较强的抗干扰能力。

Abstract

The existing manually controlled throttle manifold has problems of large pressure fluctuation and high hysteresis. In this study, considering the influence of pressure response, gas fraction and equipment status on the control parameters during well killing, an adaptive PID controller (PIDNN) for throttling manifold coupled with BP neural network was designed, and an adaptive control method for the throttle manifold was developed. Indoor physical simulation experiments and field testing were carried out. The results show that the time for the PIDNN adaptive control system to fully track to the target pressure can be less than 30 s at different constant and continuously variable target pressure, and the actual pressure fluctuation range is no more than that of the system pressure itself. The response speed can be increased by more than 30% and the mean absolute error (MAE) is reduced by more than 40% compared to the manual control and conventional PID automatic control. The new method can be applied to different types of throttle valves, and it can achieve accurate, stable and fast tracking of the target pressure within the full opening range, and has strong anti-interference ability to the sudden change of pressure caused by the variation of gas fraction.

关键词

压井节流管汇 / PID 自动控制 / BP 神经网络 / 自适应控制 / 压力波动

Key words

throttle manifold for well killing / PID automatic control / BP neural network / adaptive control / pressure fluctuation

参考文献

[1] 高德利, 黄文君. 深层、超深层定向钻井中若干基础研究进展与展望[J]. 天然气工业, 2024, 44(1): 1-12.

[2] GAO Deli, HUANG Wenjun. Basic research progress and prospect in deep and ultra-deep directional drilling[J]. Natural Gas Industry, 2024, 44(1): 1-12.

[3] 陈国明, 朱高庚, 朱渊. 深水油气开采安全风险评估与管控研究进展[J]. 中国石油大学学报(自然科学版), 2019, 43(5): 136-145.

[4] CHEN Guoming, ZHU Gaogeng, ZHU Yuan. Advances in safety assessment and risk management for deepwater oil and gas exploitation[J]. Journal of China University of Petroleum (Edition of Natural Science), 2019, 43(5): 136-145.

[5] 蒋振新, 李军, 郭勇, 等 . 井下双梯度控压钻井井筒多相流动规律[J]. 断块油气田, 2024, 31(5): 936-944.

[6] JIANG Zhenxin, LI Jun, GUO Yong, et al. Multiphase flow law in wellbore during downhole dual-gradient controlled pressure drilling[J]. Fault-Block Oil and Gas Field, 2024, 31(5): 936-944.

[7] 高永海, 于鑫, 李昊, 等 . 深水钻井溢漏同存条件下温压分析及溢漏层位判别[J]. 中国石油大学学报(自然科学版), 2025, 49(5): 93-101.

[8] GAO Yonghai, YU Xin, LI Hao, et al. Temperature and pressure analysis under coexistence of overflow and drilling fluid loss conditions and determination of overflow and loss locations in deep water drilling[J]. Journal of China University of Petroleum (Edition of Natural Science), 2025, 49(5): 93-101.

[9] AL SALMI A H, AL MAKHMARI N I, AL NAAMANI N H . Deployment of automated dynamic kill system to manage well control activities in high risk shallow gas cap wells[R]. SPE 216312-MS, 2023.

[10] 李辉, 李军, 连威, 等 . 页岩油气井多级压裂温压耦合套管应力演化规律[J]. 断块油气田, 2025, 32(5): 781-789.

[11] LI Hui, LI Jun, LIAN Wei, et al. Evolution laws of casing stress under thermo-pressure coupling during multi-stage fracturing in shale oil and gas wells[J]. Fault-Block Oil & Gas Field, 2025, 32(5): 781-789.

[12] 胡文锋. 复杂井况下关井立管压力求取的探索及实例分析[J]. 天然气勘探与开发, 2025, 48(3): 113-123.

[13] HU Wenfeng. Discussion and case study on SIDPP determination under complex well conditions[J]. Natural Gas Exploration and Development, 2025, 48(3): 113-123.

[14] NANDAN A, IMTIAZ S . Nonlinear model predictive control of managed pressure drilling[J]. ISA Transactions, 2017, 69: 307-314.

[15] CHEN X, HE M, XU M, et al. Early gas kick detection-inversion-control integrated system: the significance of applications of managed pressure drilling: a review[J]. Geoenergy Science and Engineering, 2023, 229: 212134.

[16] CARLSEN L A, NYGAARD G, NIKOLAOU M . Evaluation of control methods for drilling operations with unexpected gas influx[J]. Journal of Process Control, 2013, 23(3): 306-316.

[17] SHEIKHI M A, NIKOOFARD A, KHAKI-SEDIGH A . Control of managed pressure drilling systems using nonlinear predictive generalized minimum variance approach based on a Volterra model[J]. ISA Transactions, 2022, 128: 380-390.

[18] 刘绘新, 孟英峰, 唐继平, 等 . 油气井多级节流压井系统研究[J]. 天然气工业, 2007, 27(8): 63-65.

[19] LIU Huixin, MENG Yingfeng, TANG Jiping, et al. Research on multi-stage choking & killing system of oil and gas wells[J]. Natural Gas Industry, 2007, 27(8): 63-65.

[20] 王德玉, 刘绘新, 孟英峰, 等 . 多级节流智能控制装置的研制[J]. 西南石油学院学报, 2005(5): 82-84, 87.

[21] WANG Deyu, LIU Huixin, MENG Yingfeng, et al. Strength calculation of ring angle welding line subjected to bending moment or torque[J]. Journal of Southwest Petroleum Institute, 2005(5): 82-84, 87.

[22] 张涛, 柳贡慧, 李军, 等 . 精细控压多级并联节流管汇系统研究[J]. 石油钻探技术, 2012, 40(2): 98-103.

[23] ZHANG Tao, LIU Gonghui, LI Jun, et al. Research on multi-level parallel choke manifold system[J]. Petroleum Drilling Techniques, 2012, 40(2): 98-103.

[24] 任伟. 大排量工况下控压钻井节流阀节流特性研究[J]. 中国设备工程, 2023(19): 119-121.

[25] REN Wei. Study on throttling characteristics of throttle valve in controlled pressure drilling under large displacement condition[J]. China Plant Engineering, 2023(19): 119-121.

[26] 张涛, 柳贡慧, 孟振期, 等 . 控压钻井复合型节流管汇压力控制机制研究[J]. 钻采工艺, 2014, 37(4): 36-38.

[27] ZHANG Tao, LIU Gonghui, MENG Zhenqi, et al. Pressure control mechanism of composite choke manifold in managed pressure drilling[J]. Drilling & Production Technology, 2014, 37(4): 36-38.

[28] SIAHAAN H B, JIN H, SAFONOV M G . An adaptive PID switching controller for pressure regulation in drilling[J]. IFAC Proceedings Volumes, 2012, 45(8): 90-94.

[29] 张项飞, 李敬兆, 刘泽朝. 基于 BSO 改进模糊神经网络 PID 的管道压力控制策略研究[J]. 煤矿机械, 2024, 45(9): 167-170.

[30] ZHANG Xiangfei, LI Jingzhao, LIU Zechao. Research on pipeline pressure control strategy based on BSO improving fuzzy neural network PID[J]. Coal Mine Machinery, 2024, 45(9): 167-170.

[31] LI J, CHENG J H, SHI J Y, et al. Brief introduction of back propagation (BP) neural network algorithm and its improvement[C]// Advances in Computer Science and Information Engineering. Berlin, Heidelberg: Springer, 2012: 553-558.

[32] DING S, SU C, YU J . An optimizing BP neural network algorithm based on genetic algorithm[J]. Artificial Intelligence Review, 2011, 36(2): 153-162.

[33] ZHANG Z, SUN B, WANG Z, et al. Intelligent well killing control method driven by coupling multiphase flow simulation and real-time data[J]. Journal of Petroleum Science and Engineering, 2022, 213: 110337.

[34] LIANG H, WEI Q, LU D, et al. Application of GA-BP neural network algorithm in killing well control system[J]. Neural Computing and Applications, 2021, 33(3): 949-960.

[35] YANG H W, LI J, JIANG J W, et al. A dynamic managed pressure well-control method for rapid treatment of gas kick in deepwater managed pressure drilling[J]. Petroleum Science, 2022, 19(5): 2297-2313.

[36] 孙宝江, 宋荣荣, 王志远. 高含硫化氢天然气气侵时的溢流特性[J]. 中国石油大学学报(自然科学版), 2012, 36(1): 73-79.

[37] SUN Baojiang, SONG Rongrong, WANG Zhiyuan. Overflow behaviors of natural gas kick well with high content of H₂S gas [J]. Journal of China University of Petroleum (Edition of Natural Science), 2012, 36(1): 73-79.

[38] 耿亚楠, 任美鹏, 刘书杰, 等 . 海上非常规压井井筒多相流动规律实验[J]. 石油学报, 2019, 40(增2): 123-130.

[39] GENG Yanan, REN Meipeng, LIU Shujie, et al. An experiment on law of multiphase flow in wellbore during offshore unconventional well killing[J]. Acta Petrolei Sinica, 2019, 40(sup2): 123-130.

[40] 王志远, 孙宝江, 高永海, 等 . 深水司钻法压井模拟计算[J]. 石油学报, 2008, 29(5): 786-790.

[41] WANG Zhiyuan, SUN Baojiang, GAO Yonghai, et al. Simulation computation of well killing with deepwater driller̍s method[J]. Acta Petrolei Sinica, 2008, 29(5): 786-790.

基金资助

国家自然科学基金项目(52474018)

国家自然科学基金项目(52227804)

国家自然科学基金项目(U22B2072)

国家重点研发计划项目(2023YFC3009200)