充西气田须四段气水分布特征研究
|
摘要
充西气藏是一个低压、低渗透、低丰度、大面积分布的岩性气藏,勘探开发资料表明储层段岩性复杂,非均质性强,因此准确识别气水层、正确认识气水分布规律成为提高气田采收率的关键技术问题。
1、孔隙度分布区间主要为2～15%,平均7.48%;渗透率分布区间主要为0.01～10×10~(-3)μm~2,平均0.601×10~(-3)μm~2,储层为低孔、低渗储层;岩性以中粒岩屑砂岩、长石岩屑砂岩为主夹薄层泥岩;储集类型为裂缝-孔隙型。
2、在储层特征研究的基础上,以岩心分析数据为标准,测井数据为基本思路,对本地区物性参数进行了常规公式计算,基本上可以实现储层物性参数的精确预测。同时使用神经网络方法建立模型,结果显示取心井预测孔隙度的平均绝对误差为0.807%、相对误差为12.56%,渗透率的相对误差为41.34%,可见孔隙度解释模型预测的物性参数精度较高,有着较高的可靠性,而渗透率模型预测的物性参数精度达到误差要求。通过对比常规测井解释物性公式计算结果与人工神经网络结算结果,神经网络模型精度更高,具有较高的可靠性。
3、分析了砂体的剖面、平面展布规律,加深了对储集层认识。对各井储层级别进行了分类。充西气田属低孔低渗的储层,非均质性较强,储层主要以Ⅱ类、Ⅲ类储层为主。利用测井处理成果做出储层参数等值线图,认识储层在纵横向上分布规律。须四段下亚段的储集物性要好于上亚段。
4、采用多种方法研究不同产层流体测井响应特征并对全区单井储层流体进行了有效识别,筛选出了流体判别的一些较好方法;即以孔隙度-电阻率交会图为主,地层水孔隙度法、多元判别分析和视地层水电阻率法为辅助进行充西气藏须四段流体识别。通过建立SPSS判别模型,正确率为88.09%,达到了建模精度要求。
5、绘制流体类型平面分布图与纵、横向剖面图,研究气水平面分布特征及进行综合评价。因此认为须四段气水层集中发育于中下部,气水分布态势受岩性控制,构造起伏对气水分布起辅助作用。气水剖面剖析了研究区气水空间分布特征,指出气藏不存在统一的气水界面。
Chongxi gas reservoir is low-pressure, low permeability, low abundance, large size distribution of lithologic gas reservoir.Reservoir. Exploration and development data indicate that lithology is complex and heterogeneous so accurate identification of gas and water layer, the correct understanding of gas and water distribution are the key technical issues for restricting the development of gas field.
On the basis of the previous research results about Chongxi gas reservoir, with the drilling, logging and water quality analysis data of study area, Artificial neural network learning log interpretation and qualitative and quantitative analysis as the theoretical basis, it is comprehensive studied on the reservoir physical property, reservoir characteristic, logging interpretation, reservoir evaluation, gas/water pattern and so on taken the means of multidisciplinary analysis with geological, logging, core analysis tools and other subjects There are following outcomes and understanding:
1、Porosity distribution range is mainly 2～15%,average is 7.48%。Permeability distribution range is mainly 0.01~10×10~(-3)μm~2,Reservoir is low porosity and low permeability. Lithology is thin mudstone main folder mainly, medium-grained lithic sandstone and Feldspathic lithic sandstone. Reservoir type is fractured– porous.
2、In the study, based on reservoir characteristics, core analysis data, logging data, Physical parameters of the region for the conventional formula.Basically, the reservoir can accurately predict the physical parameters. At the same time, using the neural network to model, The results show porosity of cored wells, the average absolute forecast error is 0.807%、Relative error is 12.56%.The relative error of permeability is 41.34%, Porosity predicted by the model to explain physical parameters of high precision, the model has a higher reliability.The permeability predicted by the model error of physical parameters of accuracy required. By comparing the properties of conventional log interpretation results and calculated results of artificial neural network settlement, Neural network model has more accurate and High reliability.
3、After analysis of the sand profile, distribution, it is deepen understanding of reservoir. Through natural gamma, porosity, permeability values, it is identified the reservoir, established logging evaluation interpretation and classified the reservoir. Chongxi gas field is the reservoir with low porosity, low permeability and strongly heterogeneity. The reservoir is mainlyⅡ,Ⅲtypes . it is using the contour map of reservoir parameters through logging processing results to understand the reservoir distribution patters on the horizontal and vertical directions. The reservoir physical property of Sub-section of low Xu IV is better than the sub-section of up of it.
4、The research used various methods to study fluid log response characteristics of different producing formation and make effective recognition about reservoir fluids of single well from the whole region, and finally selected some better methods to discriminate fluids. It is recommended to use the porosity/resistivity cross plot as the main method, and the formation water porosity, multivariate discriminant analysis and apparent water resiscivity technique as auxiliary method to recognize the fluids from Xu 4 member of western Sichuan gas field. Through the established SPSS discrimination model, it is verified the accuracy is 88.09%, which meets the requirement for modeling.
5、Drawing plane distribution of fluid type and longitudinal and transverse profile is to study the plane distribution of gas and water layer. It is considered that the gas and water layer is developed in the middle-low section of Xu IV member of Xujiahe formation, gas and water distribution is controlled by lithology and assisting controlled by structure. After the analysis of gas-water spatial distribution through gas/water profile of the study area, it is fingered t that there is no uniform gas-water contact in this gas reservoir.
1、孔隙度分布区间主要为2～15%,平均7.48%;渗透率分布区间主要为0.01～10×10~(-3)μm~2,平均0.601×10~(-3)μm~2,储层为低孔、低渗储层;岩性以中粒岩屑砂岩、长石岩屑砂岩为主夹薄层泥岩;储集类型为裂缝-孔隙型。
2、在储层特征研究的基础上,以岩心分析数据为标准,测井数据为基本思路,对本地区物性参数进行了常规公式计算,基本上可以实现储层物性参数的精确预测。同时使用神经网络方法建立模型,结果显示取心井预测孔隙度的平均绝对误差为0.807%、相对误差为12.56%,渗透率的相对误差为41.34%,可见孔隙度解释模型预测的物性参数精度较高,有着较高的可靠性,而渗透率模型预测的物性参数精度达到误差要求。通过对比常规测井解释物性公式计算结果与人工神经网络结算结果,神经网络模型精度更高,具有较高的可靠性。
3、分析了砂体的剖面、平面展布规律,加深了对储集层认识。对各井储层级别进行了分类。充西气田属低孔低渗的储层,非均质性较强,储层主要以Ⅱ类、Ⅲ类储层为主。利用测井处理成果做出储层参数等值线图,认识储层在纵横向上分布规律。须四段下亚段的储集物性要好于上亚段。
4、采用多种方法研究不同产层流体测井响应特征并对全区单井储层流体进行了有效识别,筛选出了流体判别的一些较好方法;即以孔隙度-电阻率交会图为主,地层水孔隙度法、多元判别分析和视地层水电阻率法为辅助进行充西气藏须四段流体识别。通过建立SPSS判别模型,正确率为88.09%,达到了建模精度要求。
5、绘制流体类型平面分布图与纵、横向剖面图,研究气水平面分布特征及进行综合评价。因此认为须四段气水层集中发育于中下部,气水分布态势受岩性控制,构造起伏对气水分布起辅助作用。气水剖面剖析了研究区气水空间分布特征,指出气藏不存在统一的气水界面。
Chongxi gas reservoir is low-pressure, low permeability, low abundance, large size distribution of lithologic gas reservoir.Reservoir. Exploration and development data indicate that lithology is complex and heterogeneous so accurate identification of gas and water layer, the correct understanding of gas and water distribution are the key technical issues for restricting the development of gas field.
On the basis of the previous research results about Chongxi gas reservoir, with the drilling, logging and water quality analysis data of study area, Artificial neural network learning log interpretation and qualitative and quantitative analysis as the theoretical basis, it is comprehensive studied on the reservoir physical property, reservoir characteristic, logging interpretation, reservoir evaluation, gas/water pattern and so on taken the means of multidisciplinary analysis with geological, logging, core analysis tools and other subjects There are following outcomes and understanding:
1、Porosity distribution range is mainly 2～15%,average is 7.48%。Permeability distribution range is mainly 0.01~10×10~(-3)μm~2,Reservoir is low porosity and low permeability. Lithology is thin mudstone main folder mainly, medium-grained lithic sandstone and Feldspathic lithic sandstone. Reservoir type is fractured– porous.
2、In the study, based on reservoir characteristics, core analysis data, logging data, Physical parameters of the region for the conventional formula.Basically, the reservoir can accurately predict the physical parameters. At the same time, using the neural network to model, The results show porosity of cored wells, the average absolute forecast error is 0.807%、Relative error is 12.56%.The relative error of permeability is 41.34%, Porosity predicted by the model to explain physical parameters of high precision, the model has a higher reliability.The permeability predicted by the model error of physical parameters of accuracy required. By comparing the properties of conventional log interpretation results and calculated results of artificial neural network settlement, Neural network model has more accurate and High reliability.
3、After analysis of the sand profile, distribution, it is deepen understanding of reservoir. Through natural gamma, porosity, permeability values, it is identified the reservoir, established logging evaluation interpretation and classified the reservoir. Chongxi gas field is the reservoir with low porosity, low permeability and strongly heterogeneity. The reservoir is mainlyⅡ,Ⅲtypes . it is using the contour map of reservoir parameters through logging processing results to understand the reservoir distribution patters on the horizontal and vertical directions. The reservoir physical property of Sub-section of low Xu IV is better than the sub-section of up of it.
4、The research used various methods to study fluid log response characteristics of different producing formation and make effective recognition about reservoir fluids of single well from the whole region, and finally selected some better methods to discriminate fluids. It is recommended to use the porosity/resistivity cross plot as the main method, and the formation water porosity, multivariate discriminant analysis and apparent water resiscivity technique as auxiliary method to recognize the fluids from Xu 4 member of western Sichuan gas field. Through the established SPSS discrimination model, it is verified the accuracy is 88.09%, which meets the requirement for modeling.
5、Drawing plane distribution of fluid type and longitudinal and transverse profile is to study the plane distribution of gas and water layer. It is considered that the gas and water layer is developed in the middle-low section of Xu IV member of Xujiahe formation, gas and water distribution is controlled by lithology and assisting controlled by structure. After the analysis of gas-water spatial distribution through gas/water profile of the study area, it is fingered t that there is no uniform gas-water contact in this gas reservoir.
引文
[1]杨斌,匡立春,孙中春,施泽进.神经网络在石油测井中的应用[M].北京:石油工业出版社, 2005.
[2]唐大海等,(国家储委)充西气田须四段气藏新增天然气探明储量报告(科研报告),2005
[3]张松扬.致密砂岩气层测井特征参数提取方法研究及识别评价技术[D].北京:中国石油大学工学博士学位论文,2007.
[4]杨克明,徐进编.川西坳陷致密碎屑岩领域天然气成藏理论与勘探开发方法技术[M].北京:地质出版社,2004.
[5]罗启后,王世谦.四川盆地中西部二叠系重点含气层系天然气富集条件研究[J].天然气工业,1996,16(增刊1) : 40- 54.
[6]董贞环.四川盆地西部须家河组致密砂岩成岩作用与孔隙演化模式[J].岩相古地理, 1994, 14( 2) : 33-41.
[7]刘金华,张世奇,孙耀庭,等.川西前陆盆地层序地层模式与油气成藏[J].石油学报,2006, 27( 5) : 35-40.
[8]张金亮,工宝清.四川盆地中西部上二叠统沉积相[J].西安石油学院学报:自然科学版,2000, 15( 2) : 1-6
[9]库丽曼,刘树根,张长俊,等.川中地区上二叠统须家河组一段和四段沉积相[J].油气地质与采收率,2007, 14( 2) : 35-37.
[10]卢颖忠,李保华等.测井综合特征参数在碳酸盐岩储层识别中的应用[J].中国X部油气地质,2006,2(1):109-113.
[11]丁纯军等.志丹地区低孔、低渗储层测井解释评价[J].特种油气藏.2002.9(5):26-29.
[12]司马立强.测井地质应用技术.[M]北京:石油工业出版社,2002.10
[13]王贵文、郭荣坤.测井地质学.[M]北京:石油工业出版社,2000.11
[14]国景星,王纪祥,张立强等.油气田开发地质学[M].山东东营:中国石油大学出版社, 2008. 236-239.
[15]雍世和主编.测井数据处理与综合解释[M],北京:石油大学出版社.2007 .
[16]周文,童孝华.地球物理测井原理及应用[M].成都理工大学石油工程内部教材.
[17]刘瑞林.用神经网络建立孔隙度预测模型[J].江汉石油学院学报,1993,15(1):28-32.
[18]于建华.应用人工神经网络自动识别岩性[J].石油地球物理勘探,1993,28(1):59-66.
[19]单敬福.改进人工神经网络原理对储层渗透率的预测—以北部湾盆地涠西南凹陷为例[J].石油与天然气地质,2007,28(1):106-109.
[20]戴胜群.应用神经网络方法求解储层参数[J].石油天然气学.2005.27(5):583-584.
[21]徐守余.油藏描述方法原理[M].北京:石油工业出版社,2005.
[22]王树寅等.复杂储层测井评价原理[M].北京:石油工业出版社.2006.
[23]黄果骞,顾红英.测井曲线标准化方法及改进[C],2003年中国地球物理学会年刊.2003.
[24]戴胜群.应用神经网络方法求解储层参数[J].石油天然气学.2005.27(5):583-584.
[25]张金亮,王宝清.四川盆地中西部上三叠统沉积相[J].西安石油学院学报,2000,15(2):1-7.
[26]赵翰卿,付志国,吕晓光.大型河流—三角洲沉积储精细描述方法[J].石油学报,2000, 21(4): 109-113.
[27]郭秀蓉,程守田,刘星.油藏描述中的小层划分与对比[J].地质科技情报,2001,20(2):55-58.
[28]董立全,王信.川西坳中段上三叠统须家河组二段层序地层初析[J].沉积与特提斯地质.2004, 24(4): 66-75.
[29]何鲤,段勇,罗潇,等.川西坳陷上三叠统层序地层划分新方案[J].天然气工业.2007, 27(2): 6-11.
[30]叶泰然,郑荣才,刘兴艳.高分辨率层序地层学在川西须家河组砂体分布研究中的应用[J].天然气工业, 2008,28(2):38-41,57.
[31]杨光荣.四川盆地上三叠统组和段的划分与对比[J].川煤地勘,1996,(12):14.
[32]张健,李国辉,谢继荣,等.四川盆地上三叠统划分与对比研究[J].天然气工业, 2006,26(1): 12- 15.
[33]杨晓萍,赵文智,邹才能,等.低渗透储层成因机理及优质储层形成与分布[J].石油学报,2007, 28( 4) : 57-61
[34]朱彤,叶军.川西坳陷致密碎屑岩气藏类型划分及特征[J].石油实验地质,2004,26(6):537-541.
[35]陈恭洋.油气田地下地质学[M].北京:石油工业出版社,2007.
[36]汪中浩,章成广.低渗透砂岩储层测井评价方法[M].北京:石油工业出版社,2004.
[37]王树寅等.复杂储层测井评价原理[M].北京:石油工业出版社.2006.
[38]杨克明,徐进.川西致密碎屑岩领域天然气成藏理论与勘探开发方法技术[M].北京:地质出版社,2003.
[39]李明,侯连华,邹才能.岩性地层油气藏地球物理勘探技术与应用[M].北京:石油工业出版社.2005.
[40]丁纯军等.志丹地区低孔、低渗储层测井解释评价[J].特种油气藏.2002.9(5):26-29.
[41]孟凡顺,郝晓玉,冯庆付,孙铁军.于测井资料的BP神经网络模型在储层渗透率定量预测中的应用[J].内蒙古石油化工,2008, 9:198-201.
[42]闫家宁.葵花岛油田储层流体识别方法研究[J].石油地质与工程,2007,27(2):24-26.
[43]王洪辉等.地质数据处理技术及方法[M].成都:校内印刷版.2005.
[44]张宇晓.磁共振测井在低孔低渗油气层识别中的应用[J].新疆地质.2004. 22(3):315-318.
[45]张金亮,王宝清.四川盆地中西部上三叠统沉积相[J].西安石油学院学报,2000,15(2):1-7.
[46]蒋裕强,郭贵安,陈辉,等.川中地区上三叠统须家河组二段和四段砂岩优质储层成因探讨[J].油气地质与采收率,2007,14(1):18-21.
[47]尹晓贺.川中地区上三叠统须家河组埋藏史恢复及重点构造气水分布控制因素探讨[D].成都:西南石油大学硕士学位论文,2006.
[48]肖亮等.测井资料识别气层方法研究[J].工程地球物理学报,2006,3(6),471-472.
[49]李贤庆等著.鄂尔多斯盆地中部气田地层流体特征与天然气成藏[M].北京:地质出版社,1997.
[50]谢武仁,李熙哲,张满郎等.川中地区上三叠统须四段厚层砂体成因及油气运移通道分析[J].石油学报,2008,6(4):504-509
[51]张宇晓.磁共振测井在低孔低渗油气层识别中的应用[J].新疆地质.2004. 22(3):315-318.
[52]胡明毅,李士祥,魏国齐等.川西前陆盆地上三叠统须家河组致密砂岩储层评价[J] .天然气地球科学,2006, 17 (4):456-458.
[53]蒋裕强,陈义才,董兆雄等.川中地区上二叠统香溪群油气富集规律研究及勘探目标选择[R].遂宁:川中矿区油气勘探开发研究所,2005:140- 186
[54]闫家宁.葵花岛油田储层流体识别方法研究[J].石油地质与工程,2007,27(2):24-26.
[55]张宇晓.磁共振测井在低孔低渗油气层识别中的应用[J].新疆地质.2004. 22(3):315-318.
[56]卢颖忠,李保华等.测井综合特征参数在碳酸盐岩储层识别中的应用[J].中国西部油气地质,2006,2(1):109-113.
[57]欧阳健,工贵文,吴继余等.测井地质分析与油气层定量评价[M].北京:石油工业出版社.1999: 131- 139.
[58]杨克明,叶军,吕正祥等.川西坳陷上二叠统须家河组天然气分布及成藏特征[J].石油与天然气地质,2004, 25(5) :501-505.
[59] D.J.Soeder and P.L.R an dolph,Porocity,Permerbility,And Pore Structure of the Tight Mesaverde Sandstone, Piceance Basin,Colorado,SPE Formation Evaluation, June. 1987, P129-137.
[60] LI K. Characterization of rock heterogeneity using fractal geometry[C]. SPE86975,2004.
[61] A liM M, Kaelo P. Improved particle swarm algorithms for global optimization[J]. Applied Mathematics and Computation, 2008,1(2): 578-593.
[62]Nestor A Rivera, Shubhankar Ray.Well-log Feature Extraction Using Wavelets and Genetic Algorithms[J].AAPG Annual Meeting,2002.
[63] Ertekin and Silpngarmlers N.Optimization of formation analysis and evaluation protocols using neuro simulation[J]. Journal of Petroleum Science and Engineering,2005,49:97-109.
[64] Alimonti C and Falcone G. Knowledge Discovery in Databases and Multiphase-Flow Metering: The Integration of Statistics, Data Mining, Neural Networks, Fuzzy Logic, and Ad Hoc Flow Measurements Toward Well Monitoring and Diagnosis[A]. paper SPE 77407 presented at the 2002 SPE Annual Technical Conference and Exhibition,San Antonio,29 September-2 October, 2002.
[65] Abdelkader Kouider El Ouaheda, et al .Application of Artificial Intelligence Tools to Characterize Naturally Fractured Reservoirs in Hassi Messaoud Oil Field, Algeria: a case study[J]. Journal of Petroleum Science and Engineering, 2005,49:122-141.
[66] Balch R S,et al. Regional Data Analysis To Better Predict Drilling Success: Brushy Canyon Formation, Delaware Basin, New Mexico[A]. paper SPE 75145 presented at the 2002 SPE/DOE Improved Oil Recovery Symposium, Tulsa, 13-17 April, 2002.
[67] Weiss W W, Balch R S and Stubbs B A. How Artificial Intelligence Methods Can Forecast Oil Production[A]. paper SPE 75143 presented at the 2002 SPE/DOE Improved Oil Recovery Symposium[C], Tulsa,13-17 April.
[68] D.J.Soeder and Prasan Chowdlah,Pore Geometry in High-and Low-Permerbility Sandstones, Traviis Peak Formation, EastTexas, SPE Formation Evaluation, Dec.1990,P420-431.
[69]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering . 2007, 97-105.
[70]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering. 2007, 59:97-105.
[71] F.Aminzadeh.Reservoir parameter estimation using a hybrid neural network[J]. Computer&Geosciences.26(2000)869-875.
[72]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering . 2007, 97-105.
[73]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering. 2007, 59:97-105.
[74] F.Aminzadeh.Reservoir parameter estimation using a hybrid neural network[J]. Computer&Geosciences.26(2000)869-875.
[75] Mostafa H. Kamel.Estimation of shale volume using a combination of the three porosity logs[J]. Journal of Petroleum Science and Engineering, 2003,40:145-157.
[76]Deuth C V.Geostatistical Reservoir Modeling[M].New York:Oxford University Press, 2002.
[77]Ama Ilcea Fischetti,Andre Andrade.“Porosity image from well log”[J].“Journal of Petroleum Science and Engineering”.2002.36:149-158
[78]Liaqat Ali,Sandip Bordoloi“.Modeling Permeability in Tight Gas Sands Using Intellingent and Innovative Data Mining Techniques”[J].“Society of Petroleum”.2008.21-24
[79]R.B.Ferreira,V.M.Vieira,Iram Gleria,M.L.Lyra.“Correlation and complexity analysis of well logs via Lyapunov,Hurst Lempel-Ziv and neural network algorithms”[J].“Physical A”.2009.747-754 [1]Alabert F,Modot V.Stochastic models of reservoir heterogeneity:Impact on connectivity and average rermeabilities[J].SPE 24893,1992,355-370.
[2]唐大海等,(国家储委)充西气田须四段气藏新增天然气探明储量报告(科研报告),2005
[3]张松扬.致密砂岩气层测井特征参数提取方法研究及识别评价技术[D].北京:中国石油大学工学博士学位论文,2007.
[4]杨克明,徐进编.川西坳陷致密碎屑岩领域天然气成藏理论与勘探开发方法技术[M].北京:地质出版社,2004.
[5]罗启后,王世谦.四川盆地中西部二叠系重点含气层系天然气富集条件研究[J].天然气工业,1996,16(增刊1) : 40- 54.
[6]董贞环.四川盆地西部须家河组致密砂岩成岩作用与孔隙演化模式[J].岩相古地理, 1994, 14( 2) : 33-41.
[7]刘金华,张世奇,孙耀庭,等.川西前陆盆地层序地层模式与油气成藏[J].石油学报,2006, 27( 5) : 35-40.
[8]张金亮,工宝清.四川盆地中西部上二叠统沉积相[J].西安石油学院学报:自然科学版,2000, 15( 2) : 1-6
[9]库丽曼,刘树根,张长俊,等.川中地区上二叠统须家河组一段和四段沉积相[J].油气地质与采收率,2007, 14( 2) : 35-37.
[10]卢颖忠,李保华等.测井综合特征参数在碳酸盐岩储层识别中的应用[J].中国X部油气地质,2006,2(1):109-113.
[11]丁纯军等.志丹地区低孔、低渗储层测井解释评价[J].特种油气藏.2002.9(5):26-29.
[12]司马立强.测井地质应用技术.[M]北京:石油工业出版社,2002.10
[13]王贵文、郭荣坤.测井地质学.[M]北京:石油工业出版社,2000.11
[14]国景星,王纪祥,张立强等.油气田开发地质学[M].山东东营:中国石油大学出版社, 2008. 236-239.
[15]雍世和主编.测井数据处理与综合解释[M],北京:石油大学出版社.2007 .
[16]周文,童孝华.地球物理测井原理及应用[M].成都理工大学石油工程内部教材.
[17]刘瑞林.用神经网络建立孔隙度预测模型[J].江汉石油学院学报,1993,15(1):28-32.
[18]于建华.应用人工神经网络自动识别岩性[J].石油地球物理勘探,1993,28(1):59-66.
[19]单敬福.改进人工神经网络原理对储层渗透率的预测—以北部湾盆地涠西南凹陷为例[J].石油与天然气地质,2007,28(1):106-109.
[20]戴胜群.应用神经网络方法求解储层参数[J].石油天然气学.2005.27(5):583-584.
[21]徐守余.油藏描述方法原理[M].北京:石油工业出版社,2005.
[22]王树寅等.复杂储层测井评价原理[M].北京:石油工业出版社.2006.
[23]黄果骞,顾红英.测井曲线标准化方法及改进[C],2003年中国地球物理学会年刊.2003.
[24]戴胜群.应用神经网络方法求解储层参数[J].石油天然气学.2005.27(5):583-584.
[25]张金亮,王宝清.四川盆地中西部上三叠统沉积相[J].西安石油学院学报,2000,15(2):1-7.
[26]赵翰卿,付志国,吕晓光.大型河流—三角洲沉积储精细描述方法[J].石油学报,2000, 21(4): 109-113.
[27]郭秀蓉,程守田,刘星.油藏描述中的小层划分与对比[J].地质科技情报,2001,20(2):55-58.
[28]董立全,王信.川西坳中段上三叠统须家河组二段层序地层初析[J].沉积与特提斯地质.2004, 24(4): 66-75.
[29]何鲤,段勇,罗潇,等.川西坳陷上三叠统层序地层划分新方案[J].天然气工业.2007, 27(2): 6-11.
[30]叶泰然,郑荣才,刘兴艳.高分辨率层序地层学在川西须家河组砂体分布研究中的应用[J].天然气工业, 2008,28(2):38-41,57.
[31]杨光荣.四川盆地上三叠统组和段的划分与对比[J].川煤地勘,1996,(12):14.
[32]张健,李国辉,谢继荣,等.四川盆地上三叠统划分与对比研究[J].天然气工业, 2006,26(1): 12- 15.
[33]杨晓萍,赵文智,邹才能,等.低渗透储层成因机理及优质储层形成与分布[J].石油学报,2007, 28( 4) : 57-61
[34]朱彤,叶军.川西坳陷致密碎屑岩气藏类型划分及特征[J].石油实验地质,2004,26(6):537-541.
[35]陈恭洋.油气田地下地质学[M].北京:石油工业出版社,2007.
[36]汪中浩,章成广.低渗透砂岩储层测井评价方法[M].北京:石油工业出版社,2004.
[37]王树寅等.复杂储层测井评价原理[M].北京:石油工业出版社.2006.
[38]杨克明,徐进.川西致密碎屑岩领域天然气成藏理论与勘探开发方法技术[M].北京:地质出版社,2003.
[39]李明,侯连华,邹才能.岩性地层油气藏地球物理勘探技术与应用[M].北京:石油工业出版社.2005.
[40]丁纯军等.志丹地区低孔、低渗储层测井解释评价[J].特种油气藏.2002.9(5):26-29.
[41]孟凡顺,郝晓玉,冯庆付,孙铁军.于测井资料的BP神经网络模型在储层渗透率定量预测中的应用[J].内蒙古石油化工,2008, 9:198-201.
[42]闫家宁.葵花岛油田储层流体识别方法研究[J].石油地质与工程,2007,27(2):24-26.
[43]王洪辉等.地质数据处理技术及方法[M].成都:校内印刷版.2005.
[44]张宇晓.磁共振测井在低孔低渗油气层识别中的应用[J].新疆地质.2004. 22(3):315-318.
[45]张金亮,王宝清.四川盆地中西部上三叠统沉积相[J].西安石油学院学报,2000,15(2):1-7.
[46]蒋裕强,郭贵安,陈辉,等.川中地区上三叠统须家河组二段和四段砂岩优质储层成因探讨[J].油气地质与采收率,2007,14(1):18-21.
[47]尹晓贺.川中地区上三叠统须家河组埋藏史恢复及重点构造气水分布控制因素探讨[D].成都:西南石油大学硕士学位论文,2006.
[48]肖亮等.测井资料识别气层方法研究[J].工程地球物理学报,2006,3(6),471-472.
[49]李贤庆等著.鄂尔多斯盆地中部气田地层流体特征与天然气成藏[M].北京:地质出版社,1997.
[50]谢武仁,李熙哲,张满郎等.川中地区上三叠统须四段厚层砂体成因及油气运移通道分析[J].石油学报,2008,6(4):504-509
[51]张宇晓.磁共振测井在低孔低渗油气层识别中的应用[J].新疆地质.2004. 22(3):315-318.
[52]胡明毅,李士祥,魏国齐等.川西前陆盆地上三叠统须家河组致密砂岩储层评价[J] .天然气地球科学,2006, 17 (4):456-458.
[53]蒋裕强,陈义才,董兆雄等.川中地区上二叠统香溪群油气富集规律研究及勘探目标选择[R].遂宁:川中矿区油气勘探开发研究所,2005:140- 186
[54]闫家宁.葵花岛油田储层流体识别方法研究[J].石油地质与工程,2007,27(2):24-26.
[55]张宇晓.磁共振测井在低孔低渗油气层识别中的应用[J].新疆地质.2004. 22(3):315-318.
[56]卢颖忠,李保华等.测井综合特征参数在碳酸盐岩储层识别中的应用[J].中国西部油气地质,2006,2(1):109-113.
[57]欧阳健,工贵文,吴继余等.测井地质分析与油气层定量评价[M].北京:石油工业出版社.1999: 131- 139.
[58]杨克明,叶军,吕正祥等.川西坳陷上二叠统须家河组天然气分布及成藏特征[J].石油与天然气地质,2004, 25(5) :501-505.
[59] D.J.Soeder and P.L.R an dolph,Porocity,Permerbility,And Pore Structure of the Tight Mesaverde Sandstone, Piceance Basin,Colorado,SPE Formation Evaluation, June. 1987, P129-137.
[60] LI K. Characterization of rock heterogeneity using fractal geometry[C]. SPE86975,2004.
[61] A liM M, Kaelo P. Improved particle swarm algorithms for global optimization[J]. Applied Mathematics and Computation, 2008,1(2): 578-593.
[62]Nestor A Rivera, Shubhankar Ray.Well-log Feature Extraction Using Wavelets and Genetic Algorithms[J].AAPG Annual Meeting,2002.
[63] Ertekin and Silpngarmlers N.Optimization of formation analysis and evaluation protocols using neuro simulation[J]. Journal of Petroleum Science and Engineering,2005,49:97-109.
[64] Alimonti C and Falcone G. Knowledge Discovery in Databases and Multiphase-Flow Metering: The Integration of Statistics, Data Mining, Neural Networks, Fuzzy Logic, and Ad Hoc Flow Measurements Toward Well Monitoring and Diagnosis[A]. paper SPE 77407 presented at the 2002 SPE Annual Technical Conference and Exhibition,San Antonio,29 September-2 October, 2002.
[65] Abdelkader Kouider El Ouaheda, et al .Application of Artificial Intelligence Tools to Characterize Naturally Fractured Reservoirs in Hassi Messaoud Oil Field, Algeria: a case study[J]. Journal of Petroleum Science and Engineering, 2005,49:122-141.
[66] Balch R S,et al. Regional Data Analysis To Better Predict Drilling Success: Brushy Canyon Formation, Delaware Basin, New Mexico[A]. paper SPE 75145 presented at the 2002 SPE/DOE Improved Oil Recovery Symposium, Tulsa, 13-17 April, 2002.
[67] Weiss W W, Balch R S and Stubbs B A. How Artificial Intelligence Methods Can Forecast Oil Production[A]. paper SPE 75143 presented at the 2002 SPE/DOE Improved Oil Recovery Symposium[C], Tulsa,13-17 April.
[68] D.J.Soeder and Prasan Chowdlah,Pore Geometry in High-and Low-Permerbility Sandstones, Traviis Peak Formation, EastTexas, SPE Formation Evaluation, Dec.1990,P420-431.
[69]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering . 2007, 97-105.
[70]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering. 2007, 59:97-105.
[71] F.Aminzadeh.Reservoir parameter estimation using a hybrid neural network[J]. Computer&Geosciences.26(2000)869-875.
[72]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering . 2007, 97-105.
[73]Mohsen Saemi. Design of neural networks using genetic algorithm for the reservoir[J].Journal of Petroleum Science and Engineering. 2007, 59:97-105.
[74] F.Aminzadeh.Reservoir parameter estimation using a hybrid neural network[J]. Computer&Geosciences.26(2000)869-875.
[75] Mostafa H. Kamel.Estimation of shale volume using a combination of the three porosity logs[J]. Journal of Petroleum Science and Engineering, 2003,40:145-157.
[76]Deuth C V.Geostatistical Reservoir Modeling[M].New York:Oxford University Press, 2002.
[77]Ama Ilcea Fischetti,Andre Andrade.“Porosity image from well log”[J].“Journal of Petroleum Science and Engineering”.2002.36:149-158
[78]Liaqat Ali,Sandip Bordoloi“.Modeling Permeability in Tight Gas Sands Using Intellingent and Innovative Data Mining Techniques”[J].“Society of Petroleum”.2008.21-24
[79]R.B.Ferreira,V.M.Vieira,Iram Gleria,M.L.Lyra.“Correlation and complexity analysis of well logs via Lyapunov,Hurst Lempel-Ziv and neural network algorithms”[J].“Physical A”.2009.747-754 [1]Alabert F,Modot V.Stochastic models of reservoir heterogeneity:Impact on connectivity and average rermeabilities[J].SPE 24893,1992,355-370.