但松林,刘尚奇,罗艳艳,梁光跃,杨朝蓬
DAN Songlin,LIU Shangqi,LUO Yanyan,LIANG Guangyue,YANG Zhaopeng

• 1:中国石油勘探开发研究院

摘要(Abstract)：

针对油砂储层内部存在的高含水层,运用分类分析和数值模拟等方法,系统研究了高含水层对SAGD开发效果的影响。基于数值模拟结果,采用BP神经网络方法,建立了高含水层对SAGD开发效果影响的预测模型,并讨论了该模型的参数设置。结果表明:高含水层虽然增加了热损失,但加速了蒸汽腔扩展,缩短了SAGD生产时间,其中底部高含水层对SAGD影响最大,当底部高含水层厚度为3 m,含水饱和度为80%时,采出程度降低11%;引入的4个无因次变量有效刻画了高含水层的非均质性,将无因次变量作为输入参数的BP神经网络模型,预测结果平均相对误差为1.12%,证明该模型用于预测研究区块高含水层对SAGD开发效果的影响是有效的,可以为现场提供快速准确指导。
In the light of the high-watercut one within the oil sand reservoir, with the help of the classification analysis and numerical simulation etc, the influence of the high-watercut reservoir on SAGD development effect was systematically studied. Based on the simulated results and by adopting BP neural network method, the predicting model for the effects stated above was established, and moreover the parameter setting of the model was discussed. The results show that although the lean reservoirs lead to the heat loss, accelerate the steam chamber expansion and reduce the SAGD production time, in detail, the SAGD is most influenced by the bottom high-watercut reservoir: for instance, its thickness is 3 m and the water saturation is 80%, the oil recovery factor will be reduced by 11%; four dimensionless input variables effectively characterize the heterogeneity of the water-bearing reservoir, for BP neural network model taking the dimensionless variables as the input parameters, the predicted average relative error is 1.12%, thus the model is proven to be effective in predicting the SAGD development effects for the high-watercut reservoir in the studied block, and therefore can rapidly and accurately guide the field application.

关键词(KeyWords)： BP神经网络;高含水层;SAGD;加拿大油砂
BP neural network;high-watercut reservoir;SAGD;Canadian oil sand

Abstract:

Keywords:

基金项目(Foundation): 国家油气专项课题“油砂有效开发与提高SAGD效果新技术”(2016ZX05031-002);; 国家自然科学基金项目“潮控河口湾油砂储层构型及盖层完整性评价研究”(41472119)

作者(Author): 但松林,刘尚奇,罗艳艳,梁光跃,杨朝蓬
DAN Songlin,LIU Shangqi,LUO Yanyan,LIANG Guangyue,YANG Zhaopeng

DOI: 10.19597/j.issn.1000-3754.201809002

参考文献(References)：

• [1]张方礼,张丽萍,鲍君刚,等. 蒸汽辅助重力泄油技术在超稠油开发中的应用[J].特种油气藏,2007,14(2):70-72,108-109. ZHANG Fangli,ZHANG Liping,BAO Jungang,et al. Application of SAGD in ultra heavy oil development[J].Special Oil & Gas Reservoirs,2007,14(2):70-72,108-109.
• [2]刘名,姜丹,李婷,等. 风城油田水平井辅助SAGD技术参数优化[J].特种油气藏,2017,24(4):122-126. LIU Ming,JIANG Dan,LI Ting,et al. Technical parameter optimization of SAGD technique for horizontal wells in Fengcheng Oil Field[J].Special Oil & Gas Reservoirs,2017,24(4):122-126.
• [3]魏绍蕾,程林松,张辉登,等. 夹层对加拿大麦凯河油砂区块双水平井蒸汽辅助重力泄油开发的影响[J].油气地质与采收率,2016,23(2):62-69.WEI Shaolei,CHENG Linsong,ZHANG Huideng,et al. Physical simulation of the interlayer effect on SAGD production by dual horizontal well in Mackay River oil sands block,Canada[J].Petroleum Geology and Recovery Efficiency,2016,23(2):62-69.
• [4]石兰香,李秀峦,刘荣军,等. 夹层对SAGD开发效果影响研究[J].特种油气藏,2015,22(5):133-136,157-158.SHI Lanxiang,LI Xiuluan,LIU Rongjun,et al. Research on the effect of interlayers on SAGD development[J].Special Oil & Gas Reservoirs,2015,22(5):133-136,157-158.
• [5]桑林翔,杨万立,杨浩哲,等. 重18井区J3q3层夹层分布对SAGD开发效果的影响[J].特种油气藏,2015,22(3):81-84,155.SANG Linxiang,YANG Wanli,YANG Haozhe,et al. Influence of interbedding distribution on SAGD development effect in layer J3q3,Zhong-18 Well Field[J].Special Oil & Gas Reservoirs,2015,22(3):81-84,155.
• [6]KERR R K, JONASSON H P. SAGDOX—steam assisted gravity drainage with the addition of oxygen injection[R].SPE 165509,2013.
• [7]LAW D H S,NASR T N,GOOD W K. Field-scale numerical simulation of SAGD process with top-water thief zone[J].Journal of Canadian Petroleum Technology,2000,42(8):32-38.
• [8]BAO Xia,CHEN Zhangxin,WEI Yizheng,et al. Geostatistical modeling and numerical simulation of the SAGD process:Case study of an Athabasca reservoir with top water and gas thief zones[R].SPE 137435,2010.
• [9]XU Jinze,CHEN Zhangxin. Numerical simulation and optimization of steam-assisted gravity drainage in Long Lake Field with lean zone and shale layer[R].WHOC14-161,2014.
• [10]XU Jinze,PAN Yi,CHEN Zhangxin. Understanding impacts of lean zones on thermal recovery in view of mobile water[R].SPE 180712,2016.
• [11]皮建,朱磊,武静,等. 不同类型水层对油砂SAGD开发的影响分析[J].长江大学学报(自然科学版),2015,12(20):69-72. PI Jian,ZHU Lei,WU Jing,et al. The influence of different water layers on oil sand SAGD development[J].Journal of Yangtze University (Natural Science Edition),2015,12(20):69-72.
• [12]SHEIKHA H. SAGD application in reservoirs with overlying lean bitumen[R].WHOC14-147,2014.
• [13]JOHNSON M D,HANSEN L,LAU T,et al. Production optimization at Connacher[R].SPE 145091,2011.
• [14]AUSTIN-Adigio M E,WANG J,GATES I D,et al. On the performance of SAGD in Athabasca point bar deposit reservoir with top water[R].SPE 180735,2016.
• [15]MASIH S,SANCHEZ J,PATINO F,et al. The effect of bottom water coning and its monitoring for optimization in SAGD[R]. SPE 157797,2012.
• [16]MUNOZ R. Simulation sensitivity study and design parameters optimization of SAGD process[R].SPE 165387,2013.
• [17]FAIRBRIDGE J K,CEY E,GATES I D. Impact of intraformational water zones on SAGD performance[J].Journal of Petroleum Science & Engineering,2012,82-83(2):187-197.
• [18]POOLADI D M,MATTAR L. SAGD operations in the presence of overlying gas cap and water layer-effect of shale layers[J].Journal of Canadian Petroleum Technology,2002,41(6):40-51.
• [19]NASR T N,LAW D,BEAULIEU G,et al. SAGD application in gas cap and top water oil reservoirs[J].Journal of Canadian Petroleum Technology,2003,42(1):32-38.
• [20]LAW D,NASR T,GOOD W. Lab-scale numerical simulation of SAGD process in the presence of top thief zones:A mechanistic study[J].Journal of Canadian Petroleum Technology,2003,42(3):29-35.
• [21]POPA A,RAMOS R,COVER A B,et al. Integration of artificial intelligence and lean sigma for large field production optimization:Application to Kern River Field[R].SPE 97247,2005.
• [22]RAGHAVENDA C S,LIU Y,WU A,et al. Global model for failure prediction for rod pump artificial lift systems[R].SPE 165374,2013.
• [23]SHAHKARAMI A,MOHAGHEGH S D,GHOLAMI V,et al. Artificial intelligence (AI) assisted history matching[R].SPE 169507,2014.
• [24]陈志礼,宁正福,杜华明,等. 基于改进BP神经网络的页岩吸附量预测模型[J].断块油气田,2018,25(2):208-212. CHEN Zhili,NING Zhengfu,DU Huaming,et al. Prediction model of shale adsorption based on improved BP neural network[J].Fault-Block Oil & Gas Field,2018,25(2):208-212.
• [25]陈磊,李长俊.基于BP神经网络预测硫在高含硫气体中溶解度[J].石油与天然气化工,2015,44(3):1-5. CHEN Lei,LI Changjun. Prediction of sulfur solubility in high sulfur gas based on BP neural network[J].Chemical Engineering of Oil & Gas,2015,44(3):1-5.
• [26]袁兆祺,李长俊,杜强,等. 基于BP神经网络的天然气集输管道结垢预测[J].西安石油大学学报(自然科学版),2017,32(3):78-82. YUAN Zhaoqi,LI Changjun,DU Qiang,et al. Scaling prediction of natural gas gathering pipeline based on BP neural network[J].Journal of Xi'an Shiyou University( Natural Science Edition),2017,32(3):78-82.
• [27]杨宇,康毅力,郭春华,等. GA-BP神经网络模型在洛带气田产量预测中的应用[J].油气地质与采收率,2006,13(6):84-85.YANG Yu,KANG Yili,GUO Chunhua,et al. Application of GA-BP neural network in prediction of oil production of Luodai Gasfield[J].Petroleum Geology and Recovery Efficiency,2006,13(6):84-85.
• [28]李虎,蒲春生,吴飞鹏. 基于广义回归神经网络的CO2驱最小混相压力预测[J].岩性油气藏,2012,24(1):108-111.LI Hu,PU Chunsheng,WU Feipeng. Prediction of minimum miscibility pressure in CO2 flooding based on general regression neural network[J].Lithologic Reservoirs,2012,24(1):108-111.
• [29]杨建,杨程博,张岩,等. 基于改进神经网络的渗透率预测方法[J].岩性油气藏,2011,23(1):98-102.YANG Jian,YANG Chengbo,ZHANG Yan,et al. Permeability prediction method based on improved BP neural network[J].Lithologic Reservoirs,2011,23(1):98-102.
• [30]符翔,韩国庆,安永生. 应用人工神经网络优化五点法水平井井网[J].油气地质与采收率,2007,14(2):93-95.FU Xiang,HAN Guoqing,AN Yongsheng. Application of artificial neural network to optimizing five-spot horizontal well patterns[J].Petroleum Geology and Recovery Efficiency,2007,14(2):93-95.
• [31]MA Z,LEUNG J Y W,ZANON S D J,et al. Practical implementation of knowledge-based approaches for SAGD production analysis[R].SPE 170144,2014.
• [32]ZHENG Jingwen,LEUNG J Y,SAWATZKY R P,et al. A proxy model for predicting SAGD production from reservoirs containing shale barriers[R].SPE 180715,2016.