天然气地球科学 ›› 2020, Vol. 31 ›› Issue (4): 552–558.doi: 10.11764/j.issn.1672-1926.2020.01.004

• 天然气开发 • 上一篇    下一篇

一种基于BP神经网络的气井重复压裂井优选方法

霍雅迪(),江厚顺()   

  1. 长江大学石油工程学院,湖北 武汉 430100
  • 收稿日期:2019-11-06 修回日期:2020-01-18 出版日期:2020-04-10 发布日期:2020-04-26
  • 通讯作者: 江厚顺 E-mail:469156935@qq.com;jhs_hust@sina.com
  • 作者简介:霍雅迪(1995-),女,天津人,硕士研究生,主要从事油气田开发研究.E-mail: 469156935@qq.com.

A preferred method for gas well re-fracturing well based on BP neural network

Ya-di HUO(),Hou-shun JIANG()   

  1. Yangtze University Petroleum Engineering,Wuhan 430100, China
  • Received:2019-11-06 Revised:2020-01-18 Online:2020-04-10 Published:2020-04-26
  • Contact: Hou-shun JIANG E-mail:469156935@qq.com;jhs_hust@sina.com

摘要:

目前我国一些气田经过初次压裂后,增产效果不明显,或生产一段时间后,产量明显下降。为了提高气井的产量,可对气井进行重复压裂。进行重复压裂优选,可使气井得到较好的压裂效果。分析了影响气井重复压裂效果的因素,可分为地质因素和工程因素2部分。地质因素包括孔隙度、渗透率、表皮系数、产层厚度、含气饱和度、地层压力系数及剩余可采储量,工程因素包括前一次压裂是否成功、前一次压裂液用量、前一次压裂加砂量。基于BP神经网络理论,结合气井重复压裂效果影响因素分析,建立了重复压裂井优选模型。使用粒子群算法对其进行了优化,提高收敛速度的同时有效防止了局部最优解情况的发生,预测重复压裂井的日产气量,以此为依据优选重复压裂井。通过对C区重复压裂效果预测表明,基于BP神经网络优选重复压裂井可以提高选井的准确性。

关键词: BP神经网络, 气井, 重复压裂, 优选方法

Abstract:

At present, after initial fracturing in some gas fields, the effect of increasing production is not obvious, or after a period of production, the output is significantly reduced. In order to increase the production of gas wells, the gas wells can be repeatedly fractured. It is preferred to carry out repeated fracturing to obtain a better fracturing effect of the gas well. The factors affecting the effect of repeated fracturing of gas wells are analyzed. The influencing factors are divided into two parts: geological factors and engineering factors. Geological factors include porosity, permeability, skin coefficient, production layer thickness, gas saturation, formation pressure coefficient, remaining recoverable reserves and engineering factors include the success of the previous fracturing, the amount of fracturing fluid used in the previous fracturing, and the amount of sand added during the previous fracturing. Based on BP neural network theory, combined with the analysis of influencing factors of gas well repeated fracturing effect, the optimal model of re-fracturing well was established. The particle swarm optimization algorithm was used to optimize it. While increasing the convergence rate, it effectively prevented the occurrence of local optimal solutions, and predicted the daily gas production rate of the repeated fracturing wells. Based on this, the repeated fracturing wells were optimized. By predicting the effect of repeated fracturing in Zone C, it is better to repeat the fracturing well based on BP neural network to improve the accuracy of well selection.

Key words: BP neural network, Gas well, Repeated fracturing, Preferred method

中图分类号: 

  • TE242.9

表1

影响因素权重"

影响因素权重
渗透率0.12
孔隙度0.06
表皮系数0.18
产层厚度0.06
地层压力系数0.12
含气饱和度0.06
剩余可采储量0.25
前一次压裂是否成功0.06
前一次压裂加砂量0.045
前一次压裂液用量0.045

图1

气井重复压裂影响因素"

图2

BP神经网络法层次图"

图3

BP神经网络法优选井步骤"

图4

PSO-BP神经网络法优选井步骤"

表2

学习样本"

井号渗透率/(10-3 μm2孔隙度/%表皮因子产层厚度/m地层压力系数含气饱和度/%剩余可采储量/(104 m3前一次压裂是否成功前一次压裂加砂量/t前一次压裂液用量/m3日产气量/(104m3/d)
A10.367.811.59.40.8550.81 926129.6218.60.78
A20.518.523.93.60.8749.61 514019.1142.30.25
A30.386.583.18.20.8464.12 102140.1248.20.91
A40.698.910.33.10.8546.31 507016.9141.10.22
A50.528.512.38.60.8762.61 635122.6168.60.85
A60.557.891.88.10.8651.51 712123.1172.50.78
A70.498.323.62.90.8548.21 593019.8146.40.17
A80.419.130.93.50.8760.31 779125.4185.40.96
A90.286.72.74.80.8656.81 987127.3196.60.99
A100.629.220.95.60.8249.21 395014.2136.20.23
A110.598.840.63.90.8447.91 429017.5158.30.31
A120.386.942.66.70.8757.32 592134.8226.51.05
A130.549.120.53.20.8246.51 457020.3152.90.23
A140.416.432.86.10.8753.82 004136.6236.61.06
A150.297.152.57.30.8759.31 896132.8212.20.98
A160.588.870.85.20.8348.11 416018.4138.40.19
A170.346.723.38.40.8656.21 938140.2269.41.18
A180.628.370.64.90.8245.71 509020.6152.60.32
A190.456.881.96.50.8658.51 793133.7208.90.98
A200.367.162.77.60.8763.22 398141.6273.21.16

表3

检测样本"

井号渗透率/(10-3 μm2孔隙度/%表皮因子产层厚度/m地层压力系数含气饱和度/%剩余可采储量/(104 m3前一次压裂是否成功前一次压裂加砂量/t前一次压裂液用量/m3日产气量/(104 m3/d)
B10.398.262.54.50.8753.31 778121.9144.20.76
B20.576.290.72.40.8247.61 429022.5185.80.23
B30.439.822.45.60.8665.22 354138.2243.61.12
B40.697.161.13.10.8151.51 532023.4181.70.34
B50.626.780.12.80.8344.71 569018.2143.80.39
B60.377.372.88.90.8763.42 013137.2230.61.15
B70.458.583.23.50.8652.92 516119.5146.50.78
B80.587.342.67.40.8754.91 496129.7219.90.87

表4

BP神经网络计算结果"

井号渗透率/(10-3 μm2孔隙度/%表皮因子产层厚度/m地层压力系数含气饱和度/%剩余可采储量/(104 m3)前一次压裂是否成功前一次压裂加砂量/t前一次压裂液用量/m3预测日产气量/(104 m3/d)
C10.428.372.12.50.8145.31 752017.9114.20.39
C20.516.120.72.30.8257.11 829022.5185.70.56
C30.639.820.55.60.8765.22 306140.3261.20.68
C40.7411.161.13.10.8155.31 637045.4251.70.51
C50.678.782.12.80.8544.71 589137.2226.80.92
C60.376.372.88.90.8163.42 213139.3240.20.35
C70.498.423.23.40.8651.52 812018.5138.40.98
C80.387.561.810.10.8452.91 993028.7212.30.75
1 胡永全,林辉,赵金洲,等.重复压裂技术研究[J].天然气工业,2004,(3):72-75,147.
HU Y Q, LIN H, WANG S Z, et al. Study on repetitive fracturing technology[J]. Natural Gas Industry,2004,(3):72-75,147.
2 赵金洲,尹庆,李勇明,等.重复压裂气井热—固耦合应力场拟三维模型[J].天然气地球科学,2015,26(11):2131-2136.
Zhao J Z, Yin Q, Li Y M, et al. A pseudo-three-dimensional model for thermoelastic stress oiled around a re-fractured gas well [J].Natural Gas Geoscience,2015 26(11):2131-2136.
3 邵先杰,董新秀,汤达祯,等.韩城矿区煤层气中低产井治理技术与方法[J].天然气地球科学,2014,25(3):435-443.
Shao X J, Dong X X, Tang D Z, et al. Control technology and method for low and medium production wells in coalbed methane in Hancheng mining area [J].Natural Gas Geoscience,2014,25(3):435-443.
4 MAUNISH S, SUBHASH S, ANIRBID S. A comprehensive overview on recent developments in refracturing technique for shale gas reservoirs[J]. Journal of Natural Gas Science and Engineering,2017,46:350-364.
5 YANCHAO L, YUNBIN H, JIANFENG X, et al. Optimal selection and effect evaluation of re-fracturing intervals in shale gas horizontal wells[J]. Natural Gas Industry B,2018,38(7):59-64.
6 马晓彬,周文,钱春华.模糊数学和灰色关联相结合的水力压裂选井选层评判方法研究[J].物探化探计算技术,2013,35(3):369-372, 252.
MA X B,ZHOU W,QIAN C H,et al. Study on judgment method of hydraulic fracturing and well selection by combining fuzzy mathematics and grey relation[J]. Computing Techniques for Geophysical and Geochemical Exploration,2013,35(3):369-372, 252.
7 杨小松,孙良田,孙雷,等.模糊数学在气井压裂酸化选井选层因素分析中的应用[J].钻采工艺,2005, 28(4):39-43,16-17.
YANG X S, SUN L T, LIU Y B, et al. Application of fuzzy mathematics for factors analysis of wells and layers choice in fracturing and acidizing gas wells[J]. Drilling & Production Technology,2005,28(4):39-43,16-17.
8 闫文华,魏勋,潘洪灏,等.应用模糊综合评判方法优选水力压裂井[J].数学的实践与认识,2011,41(19):69-74.
YAN W H, WEI X, PAN H H, et al. Optimizing hydraulic fracturing wells with the method of fuzzy comprehensive evaluation[J].Mathematics in Practice and Theory,2011,41(19):69-74.
9 李春月,张士诚,马新仿.多元回归与模糊识别相结合优选重复压裂井[J].内蒙古石油化工,2009,35(3):132-133.
LI C Y, ZHANG S C, MA X F. Repetitive fracturing wells by combining multiple regression with fuzzy recognition[J].Inner Mongolia Petrochemical Industry,2009,35(3):132-133.
10 SOPHIE S Y,MUKUL M S. A new method to calculate slurry distribution among multiple fractures during fracturing and refracturing[J]. Journal of Petroleum Science and Engineering,2018,170:304-314.
11 龚杨,张秦汶,王楠.模糊识别方法在压裂选井选层中的应用[J].重庆科技学院学报:自然科学版,2013,15(6):59-61,74.
GONG Y, ZHANG Q W, WANG N. Application of well selection for fracturing by fuzzy recognition method[J].Journal of Chongqing University of Science and Technology:Social Sciences Edition,2013,15(6):59-61,74.
12 姜梅枝.低渗透重复压裂井选层研究[J].油气井测试,2012,21(2):37-40,43,76-77.
JIANG M Z. Research on layer selection in low permeability and re-fracturing well[J]. Well Testing,2012,21(2):37-40,43,76-77.
13 位云生,胡永全,赵金洲,等.人工神经网络方法在水力压裂选井评层中的应用[J].断块油气田,2005,12(4):42-44,91-92.
WEI Y S, HU Y Q, ZHAO J Z, et al. Application of artificial nerve network method to evaluating & choosing target layer well in hydraulic fracturing treatment[J]. Fault-Block Oil and Gas Field, 2005,12(4):42-44,91-92.
14 张洲,鲜保安,连小华,等.低渗煤储层背景下高渗带主控地质因素及模式[J].天然气地球科学,2018,29(11):1656-1663.
ZHANG Z, XIAN B A, LIAN X H, et al. Main geological factors and models of high permeability zone under the background of low permeability coal reservoir [J].Natural Gas Geoscience,2018,29(11):1656-1663.
15 杨晓盈,李永臣,朱文涛,等.贵州煤层气高产主控因素及甜点区综合评价模型[J].天然气地球科学,2018,29(11):1664-1671,1678.
YANG X Y, LI Y C, ZHU W T, et al. Main control factors of guizhou coalbed methane high yield and comprehensive evaluation model for dessert areas[J].Natural Gas Geoscience,2018,29(11):1664-1671, 1678.
16 吕志凯,贾爱林,唐海发,等.大型致密砂岩气藏水平井产能评价与新认识[J].天然气地球科学,2018,29(6):873-879.
LÜ Z K, JIA A L, TANG H F, et al. Productivity evaluation and new understanding of horizontal wells in large tight sandstone gas reservoirs[J]. Natural Gas Geoscience, 2018,29(6):873-879.
[1] 张辉, 尹国庆, 王海应. 塔里木盆地库车坳陷天然裂缝地质力学响应对气井产能的影响[J]. 天然气地球科学, 2019, 30(3): 379-388.
[2] 杨浩珑,向祖平,袁迎中,李龙. 低渗气藏压裂气井稳态产能计算新方法[J]. 天然气地球科学, 2018, 29(1): 151-157.
[3] 武俊文,贾文峰,雷群,熊春明,曹光强,张建军,李隽,廖东,胡志国. 纳米粒子增强泡排剂性能及影响因素研究[J]. 天然气地球科学, 2017, 28(8): 1274-1279.
[4] 乔霞,罗明高, 王洪峰,肖香姣,阳建平. #br# 一种利用生产动态历史评价异常高压气井产能的新方法[J]. 天然气地球科学, 2017, 28(12): 1908-1913.
[5] 顾岱鸿, 崔国峰, 刘广峰, 丁道权. 多层合采气井产量劈分新方法[J]. 天然气地球科学, 2016, 27(7): 1346-1351.
[6] 徐兵祥. 页岩凝析气井生产数据分析新方法[J]. 天然气地球科学, 2016, 27(5): 905-909.
[7] 刘永良,袁迎中,邓力菁,隆正峰,戚志林,徐艳霞,张诗通,刘彬. 一种确定水平气井二项式产能方程的新方法[J]. 天然气地球科学, 2016, 27(2): 371-376.
[8] 汪吉林,朱炎铭,宫云鹏,方辉煌. 重庆南川地区龙马溪组页岩微裂缝发育影响因素及程度预测[J]. 天然气地球科学, 2015, 26(8): 1579-1586.
[9] 赵金洲,尹庆,李勇明,唐志娟. 重复压裂气井热—固耦合应力场拟三维模型[J]. 天然气地球科学, 2015, 26(11): 2131-2136.
[10] 邵先杰,董新秀,汤达祯,武泽,李士才,孙玉波,胥昊. 韩城矿区煤层气中低产井治理技术与方法[J]. 天然气地球科学, 2014, 25(3): 435-443.
[11] 赵俊芳,王生维,秦义,赵文秀,陈文文,杨健,李瑞. 煤层气井煤粉特征及成因研究[J]. 天然气地球科学, 2013, 24(6): 1316-1320.
[12] 胡俊坤,李晓平,肖 强,敬 伟. 利用生产动态资料确定气井产能方程新方法[J]. 天然气地球科学, 2013, 24(5): 1027-1031.
[13] 张培河, 刘钰辉, 王正喜, 刘娜娜. 基于生产数据分析的沁水盆地南部煤层气井产能控制地质因素研究[J]. 天然气地球科学, 2011, 22(5): 909-914.
[14] 熊友明, 徐家年, 冯胜利, 张启汉, 刘理明. 气田防砂效果评价方法与标准探索[J]. 天然气地球科学, 2011, 22(2): 331-334.
[15] 张典坤, 王生维, 李梦溪, 张建国, 杨青雄, 唐江林. 晋城煤层气井产能的地质控制因素分析[J]. 天然气地球科学, 2010, 21(3): 513-517.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!