Fluid identification in glutenites with the machine learning AdaBoost.M2 algorithm
CHEN Ganghua1, LIANG Shasha1,2, WANG Jun3, DI Shuhua4, ZHUGE Yueying4, LIU Youji1
1. School of Geosciences, China University of Petroleum(East China), Qingdao, Shandong 266580, China; 2. Jiaxing Jia'an Gas Technology Service Limited, Jiaxing, Zhejiang 314000, China; 3. Research Institute of Exploration and Development, Shengli Oilfield Branch Co., SINOPEC, Dongying, Shandong 257015, China; 4. North China Branch, Logging Co., Ltd., CNPC, Renqiu, Hebei 062550, China
Abstract:Glutenites are usually characterized by complex compositions,various pore structures,and strong heterogeneity,and the logging response of fluid is much smaller than that of gravel skeleton.So fluids in glutenites cannot be identified with conventional logging methods.Therefore we propose the machine learning AdaBoost.M2 algorithm for glutenite fluid identification.First combined with oil test data,the AdaBoost.M2 algorithm disassembles the K-class multi-fluid type classification problem into a two-class problem,and the sample distribution is obtained through multi-iterations.Then the decision tree algorithm called a weak learning algorithm obtains automatically a classifier ht for the discrimination.The proposed algorithm is applied to the reservoir fluid identification in A area.The sample return accuracy is 95% and the test accuracy is 91.5%,which proves the validity and applicability of proposed algorithm.
昝灵,王顺华,张枝焕,等.砂砾岩储层研究现状[J].长江大学学报(自然科学版),2011,8(3):63-66.ZAN Ling,WANG Shunhua,ZHANG Zhihuan,et al.Research status of glutenite reservoirs[J].Journal of Yangtze University(Nature Science Edition),2011,8(3):63-66.
[2]
赵军龙,李新胜,张彬.基于测井曲线小波分析与希尔伯特变换的流体识别新方法[J].石油地球物理勘探, 2010,45(2):290-294,305.ZHAO Junlong,LI Xinsheng,ZHANG Bin.New fluid detection method based on well logging curve wavelet analysis and Hilbert transform[J].Oil Geophysical Prospecting,2010,45(2):290-294,305.
[3]
孙焕泉,运华云,侯庆功,等.胜利油区油气层测井响应特征[M].北京:石油工业出版社,2008.SUN Huanquan,YUN Huayun,HOU Qinggong,et al.Logging Response Characteristics of Oil and Gas Reservoirs in Shengli Oilfield[M].Petroleum Industry Press,Beijing,2008.
[4]
谭锋奇,李洪奇,孟照旭,等.数据挖掘方法在石油勘探开发中的应用研究[J].石油地球物理勘探,2010,45(1):85-91.TAN Fengqi,LI Hongqi,MENG Zhaoxu,et al.Stu-dies on application of data mining method in oil exploration and development[J].Oil Geophysical Prospecting,2010,45(1):85-91.
[5]
宋建国,杨璐,高强山,等.强容噪性随机森林算法在地震储层预测中的应用[J].石油地球物理勘探,2018, 53(5):954-960.SONG Jianguo,YANG Lu,GAO Qiangshan,et al.Strong tolerance random forest algorithm in seismic reservoir prediction[J].Oil Geophysical Prospecting,2018,53(5):954-960.
[6]
殷文,李援,郭加树,等.基于改进蚁群追踪策略的地震层位自动识别方法[J].石油地球物理勘探,2017,52(3):553-561.YIN Wen,LI Yuan,GUO Jiashu,et al.Seismic horizon automatic identification based on ant colony tra-cking strategy[J].Oil Geophysical Prospecting,2017,52(3):553-561.
[7]
刘全稳.测井神经网络技术综述[J].石油地球物理勘探,1996,31(增刊1):64-69.LIU Quanwen.A summary of logging neural network technology[J].Oil Geophysical Prospecting,1996,31(S1):64-69.
[8]
陈雷,肖创柏,禹晶,等.基于相似性传播聚类与主成分分析的断层识别方法[J].石油地球物理勘探,2017,52(4):826-833.CHEN Lei,XIAO Chuangbai,YU Jing,et al.Fault recognition based on affinity propagation clustering and principal component analysis[J].Oil Geophysical Prospecting, 2017,52(4):826-833.
[9]
李舟波.利用测井方法识别复杂油气储层的流体性质[J].石油与天然气地质,2004,25(4):356-362.LI Zhoubo.Problem and advance on the fluild typing for complex reservoirs using well log[J].Oil & Gas Geology, 2004,25(4):356-362.
[10]
石宁,李洪奇,罗伟平.数据挖掘与测井解释:应用于砾岩储层[J].应用地球物理,2015,12(2):263-272.SHI Ning,LI Hongqi,LUO Weiping.Data mining and well logging interpretation:application to a conglo-merate reservoir[J].Applied Geophysics,2015,12(2):263-272.
[11]
刘得芳,林发武,梁忠奎,等.碳酸盐岩储层流体性质综合判别方法[J].特种油气藏,2015,22(4):104-107.LIU Defang,LIN Fawu,LIANG Zhongkui,et al.An integrated method to determine fluid properties in carbonate reservoirs[J].Special Oil & Gas Reservoirs,2015,22(4):104-107.
[12]
张银德,童凯军,郑军,等.支持向量机方法在低阻油层流体识别中的应用[J].石油物探,2008,47(3):306-310.ZHANG Yinde,TONG Kaijun,ZHENG Jun,et al.Application of support vector machine method for identifying fluid in low-resistivity oil layers[J].Geophysical Prospecting for Petroleum,2008,47(3):306-310.
[13]
王少龙,杨斌,赵倩,等.BP神经网络在复杂储层流体识别中的应用[J].石油化工应用,2018,37(7):45-48.WANG Shaolong,YANG Bin,ZHAO Qian,et al.Application of BP neural network in complex reservoir fluid identification[J].Petrochemical Application,2018,37(7):45-48.
[14]
陈钢花,张艳,毛克宇,等.声-电测井联合流体识别因子构建方法及其应用——以川东北碳酸盐岩气藏为例[J].石油物探,2017,56(2):295-301.CHEN Ganghua,ZHANG Yan,MAO Keyu,et al.Establishment of fluid identification factor by joint acoustic and resistivity logging and its application:a case study of carbonate gas reservoir in the Northeast Sichuan[J].Geophysical Prospecting for Petroleum,2017,56(2):295-301.
[15]
曹莹,苗启广,刘家辰,等.AdaBoost算法研究进展与展望[J].自动化学报,2013,39(6):745-758.CAO Ying,MIAO Qiguang,LIU Jiachen,et al.Advance and prospects of AdaBoost algorithm[J].Acta Automatica Sinica,2013,39(6):745-758.
[16]
Schaping R E,Singer Y.Improved boosting algorithms using confidence rated prediction[J].Machine Lear-ning,1999,37(3):297-336.
[17]
吴信东.数据挖掘十大算法[M].北京:清华大学出版社,2013.WU Xindong.Ten Algorithms for Data Mining[M].Tsinghua University Press,Beijing,2013.
[18]
杨新武,马壮,袁顺.基于弱分类器调整的多分类AdaBoost算法[J].电子与信息学报,2016,38(2):373-380.YANG Xinwu,MA Zhuang,YUAN Shun.Multi-class AdaBoost algorithm based on the adjusted weak classifier[J].Journal of Electronics & Information Technology,2016,38(2):373-380.
[19]
涂承胜,刁力力,鲁明羽,等.Boosting家族AdaBoost系列代表算法[J].计算机科学,2003,30(3):30-34.TU Chengsheng,DIAO Lili,LU Mingyu,et al.The typical algorithm of AdaBoost series in Boosting family[J].Computer Science,2003,30(3):30-34.
[20]
袁红旗,刘长利,肖阳,等.二连盆地乌兰花凹陷沉积特征研究[J].地球物理学进展,2016,31(5):2239-2245.YUAN Hongqi,LIU Changli,XIAO Yang,et al.Sedimentary characteristics of Wulanhua Sag in Erlian Basin[J].Progress in Geophysics,2016,31(5):2239-2245.