Abstract:The difference in logging responses of reservoirs with different fluid properties in Dongfang X gas field is not obvious, and it is especially difficult to determine the bottom limit of the resistivity of different fluids. In addition, it is susceptible to physical factors when using porosity logging curves to identify fluids, making different fluid properties of sample data overlap. A Stacking model fusion method is proposed. It includes various machine learning algorithms (i.e. decision tree, support vector machine, random forest and extreme gradient boosting) and has better effects on fluid identification at high temperature and high pressure. Through 10 fold and cross validations, the algorithms are iteratively optimized to achieve a final optimal model. Compared with a single machine learning algorithm, the Stacking model fusion algorithm can take into account the differences in data observation and training principles of different algorithms, and give full play to the advantages of each model. Tests on real data indicate that the Stacking model can improve prediction accuracy from 87.08% to 92%, compared with the best-performing single model XGBoost. It has a stronger learning ability, and more suitable for fluid identification in high-temperature and high-pressure reservoir. It provides a new idea for building models of logging interpretation.
吴磊,徐怀民,季汉成.基于交会图和多元统计法的神经网络技术在火山岩识别中的应用[J].石油地球物理勘探,2006,41(1):81-86.WU Lei,XU Huaimin and JI Hancheng.Application of neural networks technique based on crossplot and multielement statistics to recognition of volcanic rocks[J].Oil Geophysical Prospecting,2006,41(1):81-86.
[2]
梁丽梅,喻高明,黎明,等.神经网络模拟交会图在H油田低阻油层流体识别中的应用[J].石油地质与工程,2011,25(1):39-40,43.LIANG Limei,YU Gaoming,LI ming,et al.Application of neural network simulated rendezvous graph in fluid identification of low-resistance reservoir in H oilfield[J].Petroleum Geology and Engineering,2011,25(1):39-40,43.
[3]
何旭,李忠伟,刘昕,等.应用卷积神经网络识别测井相[J].石油地球物理勘探,2019,54(5):1159-1165.HE Xu,LI Zhongwei,LIU Xin,et al.Log facies re-cognition based on convolutional neural network[J].Oil Geophysical Prospecting,2019,54(5):1159-1165.
[4]
王礼常,王志章,陶果.样本分解交汇图法与决策树方法识别苏丹某油田流体性质[J].科技导报,2012,30(2):22-27.WANG Lichang,WANG Zhizhang,TAO Guo.Fluid nature identification of an oil-gas field in Sultan using stylebook hierarchical decomposition cross-plot and decision tree methods[J].Science & Technology Review,2012,30(2):22-27.
[5]
赵倩,杨斌,李星,等.基于图版法决策树在流体识别中的应用[J].测井技术,2018,42(6):641-646.ZHAO Qian,YANG Bin,LI Xing,et al.Application of cross-plot-based decision tree template method in fluid identification[J].Well Logging Technology,2018,42(6):641-646.
[6]
于代国,孙建孟,张振城,等.支持向量机方法识别储集层流体性质[J].新疆石油地质,2005,26(6):675-677.YU Daiguo,SUN Jianmeng,ZHANG Zhencheng,et al.Reservoir fluid property identification with support vector machine method[J].Xinjiang Petroleum Geology,2005,26(6):675-677.
[7]
赵学松,高强山,唐传章,等.基于支持向量回归机与井导向的三角洲岩性油气藏储层参数预测[J].石油地球物理勘探,2016,51(5):976-982.ZHAO Xuesong,GAO Qiangshan,TANG Chuan-zhang,et al.Delta stratigraphic reservoir parameter estimation based on support vector regression machine and well logging data[J].Oil Geophysical Prospecting,2016,51(5):976-982.
[8]
王丽,袁伟,丁磊,等.基于常规测井资料的储层流体识别方法[J].地质科技情报,2018,37(2):241-245.WANG Li,YUAN Wei,DING Lei,et al.Reservoir fluid identification based on normal logging data[J].Geological Science and Technology Information,2018,37(2):241-245.
[9]
田时芸,刘子云.利用测井资料判别油水层时几种判别分析方法的判别效果比较[J].江汉石油学院学报,1983,5(1):37-46.TIAN Shiyun,LIU Ziyun.Comparison of discriminant effects of several discriminant analysis methods when using logging data to discriminate oil and water layers[J].Journal of Jianghan Petroleum Institute,1983,5(1):37-46.
[10]
陈军,范晓敏,莫修文.火山碎屑岩岩性的测井识别方法[J].吉林大学学报(地球科学版),2007,37(增刊):99-101,113.CHEN Jun,FAN Xiaomin,MO Xiuwen.The resrarch of volcaniclastic rock lithologic identification based logging[J].Journal of Jilin University(Earth Science Edition),2007,37(S):99-101,113.
[11]
袁少阳,张占松,李权,等.利用贝叶斯判别法识别岩性基础上的孔隙度评价[J].测井技术,2016,40(3):281-285.YUAN Shaoyang,ZHANG Zhansong,LI Quan,et al.Porosity evaluation based on lithology indentification using Bayesian classifier[J].Well Logging Technology,2016,40(3):281-285.
[12]
何健,文晓涛,聂文亮,等.利用随机森林算法预测裂缝发育带[J].石油地球物理勘探,2020,55(1):161-166.HE Jian,WEN Xiaotao,NIE Wenliang,et al.Fracture zone prediction based on random forest algorithm[J].Oil Geophysical Prospecting,2020,55(1):161-166.
[13]
苏瑞.基于随机森林的F区块水淹层测井定性识别[D].黑龙江大庆:东北石油大学,2019.SU Rui.Qualitative Logging Identification of Water- Flooded Zones in F Block Based on Random Forest[D].Northeast Prtroleum University,Daqing,Heilongjiang,2019.
[14]
陈钢花,梁莎莎,王军,等.机器学习AdaBoost.M2算法在砂砾岩流体识别中的应用[J].石油地球物理勘探,2019,54(6):1357-1362.CHEN Ganghua,LIANG Shasha,WANG Jun,et al.Fluid identification in glutenites with the machine learning AdaBoost.M2 algorthm[J].Oil Geophysical Prospecting,2019,54(6):1357-1362.
[15]
周志华.机器学习[M].北京:清华大学出版社,2016.ZHOU Zhihua.Machine Learning[M].Tsinghua University Press,Beijing,2016.
[16]
Huang X,Zhang L.An SVM ensemble approach combining spectral,structural,and semantic features for the classification of high-resolution remotely sensed imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2013,51(1):257-272.
[17]
De-la-Torre M,Granger E,Sabourin R,et a1.An adaptive ensemble-based system for face recognition in person re-identification[J].Machine Vision and Applications,2015,26(6):741-773.
[18]
Zhang L,Zhang L,Zhang D,et al.Ensemble of local and global information for finger-knuckle-print recognition[J].Pattern Recognition,2011,44(9):1990-1998.
[19]
Kavitha B,Karthi K S,Maybell P S.An ensemble design of intrusion detection system for handling uncertainty using neutrosophic logic classifier[J].Know-ledge-Based Systems,2012,28(1):88-96.
[20]
Cai Z,Xu D,Zhang Q,et al.Classification of lung cancer using ensemble-based feature selection and machine learning methods[J].Molecular BioSystems,2015,11(3):791-800.
[21]
方育柯.集成学习理论研究及其在个性化推荐中的应用[D].四川成都:电子科技大学,2011.FANG Yuke.Research of Ensemble Learning Theory and Its Application in Personalized Recommendation[D].University of Electronic Science and Technology of China,Chengdu,Sichuan,2011.
[22]
Chen T,Guestrin C.XGBoost:A scalable tree boosting system[C].ACM SIGKDD International Confe-rence on Knowledge Discovery and Data Mining,2016,785-794.