摘要
最小二乘支持向量机是在统计学习理论基础上发展起来的模式识别方法。与传统统计学相比,它能有效解决有限样本、非线性、高维数模型的建立问题,而且建立的模型具有很好的预测性能。岩性识别本质是解决分类问题,本文基于最小二乘支持向量机解决分类问题的优势,首先用GR、CNL、DEN、AC、RLLD等常规测井曲线数据建立样本空间;然后通过耦合模拟退火和交叉验证的方法寻找最佳参数,优化最小二乘支持向量机分类器;最后建立了最小二乘支持向量机岩性识别模型。通过取心段岩心描述和岩心/岩屑薄片鉴定,确定辽河盆地40口井315m井段2 520个岩性样品作为训练样本,建立岩性识别标准。对8口井13 866m井段110 928个火山岩数据采样点进行测井识别,可识别致密玄武岩、气孔玄武岩、粗面岩等8种主要火山岩类型。识别结果与8口测试井中316个有取心段岩心描述和岩心/岩屑薄片的精确岩矿定名对比,符合率达到75.2%,与以往测井识别复杂火山岩岩性相比,在识别准确率和效率上都有明显提高。
Least squares support vector machine(LS-SVM)is a pattern recognition method developed from the statistical learning theory.Compared with the traditional statistics,LS-SVM caneffectively resolve the problems of the finite of samples,non-linearity and high-dimension,and it can achieve accurate prediction.The essence of lithology identification is for classification.We take the advantage of the classification by LS-SVM:at first,the sample space is established by using the conventional logging curves of GR、CNL、DEN、AC、RLLD;and then,the classifier of LS-SVM is optimized by searching optimal parameters using the simulated annealing algorithm and cross validation method;finally,the model of LS-SVM lithology identification is determined.Through the description of core section and the analysis of core/cuttings,2 520 samples from the 315 msection of the 40 wells are taken as training samples in Liaohe basin,for establishing the standard of lithology identification.110 928 logging data from 13 866 msection of the 8wells are taken as the predicting samples.8types of volcanic rocks have been identified such as vesicular basalt,compact basalt,trachyte,et al.In comparison with the 316 samples from 8 wells,the corresponding identification rate is 75.2%.The accuracy and velocity of the LS-SVM lithology identification is improved distinctly compared with other well logging methods.
引文
[1]Vapnik V N.An Overview of Statistical Learning Theory[J].IEEE Trans Neural Network,1999,10(5):988-999.
[2]Suykens J A K,Vandewalle J.Least Squares Support Vector Machine Classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[3]Suykens J A K,De Brabanter J,Lukas L,et al.Weighted Least Squares Support Vector Machines:Robustness and Sparse Approximation[J].Neurocomputing,2002,48:85-105.
[4]李锋,汤宝平,章国稳.基于舍一交叉验证优化最小二乘支持向量机的故障诊断模型[J].振动与冲击,2010,29(9):170-174.Li Feng,Tang Baoping,Zhang Guowen.Fault Diagnosis Model Based on Least Square Support Vector Machine Optimized by Leave-One-Out Cross-Validation[J].Journal of Vibration and Shock,2010,29(9):170-174.
[5]陈刚,王宏琦,孙显.基于核函数原型和自适应遗传算法的SVM模型选择方法[J].中国科学院研究生院学报,2012,29(1):62-69.Chen Gang,Wang Hongqi,Sun Xian.Model Selection for SVM Classification Based on Kernel Prototype and Adaptive Genetic Algorithm[J].Journal of Graduate University of Chinese Academy of Sciences,2012,29(1):62-69.
[6]刘鲭洁,陈桂明,刘小方,等.基于遗传算法的SVM参数组合优化[J].计算机应用与软件,2012,29(4):94-100.Liu Qingjie,Chen Guiming,Liu Xiaofang,et al.Genetic Algorithm Based on SVM Parameter Composition Optimization[J].Computer Applications and Software,2012,29(4):94-100.
[7]王璞珺,衣健,陈崇阳,等.火山地层学与火山架构:以长白山火山为例[J].吉林大学学报:地球科学版,2013,43(2):319-336.Wang Pujun,Yi Jian,Chen Chongyang,et al.Volcanostratigraphy and Volcanic Architecture of the Changbaishan Volcanos,NE,China[J].Journal of Jilin University:Earth Science Edition,2013,43(2):319-336.
[8]王淑盛,徐正光,刘黄伟,等.改进的K近邻方法在岩性识别中的应用[J].地球物理学进展,2004,19(2):478-480.Wang Shusheng,Xu Zhengguang,Liu Huangwei,et al.The Advanced K-Nearest Neighborhood Method Used in the Recognition of Lithology[J].Progress in Geophysics,2004,19(2):478-480.
[9]周波,李舟波,潘保芝,等.火山岩岩性识别方法研究[J].吉林大学学报:地球科学版,2005,35(3):394-397.Zhou Bo,Li Zhoubo,Pan Baozhi,et al.A Study on Lithology Identification Methods for Volcanic Rocks[J].Journal of Jilin University:Earth Science Edition,2005,35(3):394-397.
[10]王军,杨长春,许大华,等.微电阻率扫描成像测井方法应用及发展前景[J].地球物理学进展,2005,20(2):357-364.Wang Jun,Yang Changchun,Xu Dahua,et al.Application and Prospect of the Formation Micro Resistivity Image Well Logging[J].Progress in Geophysics,2005,20(2):357-364.
[11]李舟波,王祝文.科学钻探中的元素测井技术[J].地学前缘,1998,5(1/2):119-130.Li Zhoubo,Wang Zhuwen.Technique of Elemental Logging in Scientific Drilling Project[J].Earth Science Frontiers,1998,5(1/2):119-130.
[12]程希,任战利.利用地球化学元素与矿物关系识别GS油田下干柴沟组地层岩性[J].地球物理学进展,2008,23(6):1903-1908.Cheng Xi,Ren Zhanli,Identification of Ganchaigou Group Rock Lithology in the GS Oilfield by Relationship Between Elements and Minerals Using Geochemical Logging Technology[J].Progress in Geophys,2008,23(6):1903-1908.
[13]袁祖贵,成晓宁,孙娟.地层元素测井(ECS):一种全面评价储层的测井新技术[J].原子能科学技术,2004,38(增刊):208-213.Yuan Zugui,Cheng Xiaoning,Sun Juan.ECS:New Well-Logging Technique for Completely Evaluating the Reservior[J].Atomic Energy Science and Technology,2004,38(Sup.):208-213.
[14]张治国,杨毅恒,夏立显.自组织特征映射神经网络在测井岩性识别中的应用[J].地球物理学进展,2005,20(2):332-336.Zhang Zhiguo,Yang Yiheng,Xia Lixian.The Application of Self-Organizing Feature Map Neural Network to Logging Lithological Identification[J].Progress in Geophysics,2005,20(2):332-336.
[15]于代国,孙建孟,王焕增,等.测井识别岩性新方法:支持向量机方法[J].大庆石油地质与开发,2005,24(5):93-96.Yu Daiguo,Sun Jianmeng,Wang Huanzeng,et al.A New Method for Logging Lithology Identification:SVM[J].Petroleum Geology&Oilfield Development in Daqing,2005,24(5):93-96.
[16]朱怡翔,石广仁.火山岩岩性的支持向量机识别[J].石油学报,2013,34(2):312-322.Zhu Yixiang,Shi Guangren.Identification of Lithologic Characteristics of Volcanic Rocks by Support Vector Machine[J].Acta Petrolei Sinica,2013,34(2):312-322.
[17]张莹,潘保芝.支持向量机与微电阻率成像测井识别火山岩岩性[J].物探与化探,2011,35(5):634-642.Zhang Ying,Pan Baozhi.The Application of SVM and FMI to the Lithologic Identification of Volcanic Rocks[J].Geophysical&Geochemical Exploration,2011,35(5):634-642.
[18]韩学辉,卢时林,支乐菲,等.应用最小二乘支持向量机识别J13井区杜家台油层岩性[J].特种油气藏,2011,18(6):18-21.Han Xuehui,Lu Shilin,Zhi Lefei,et al.Identification of the Lithology of Dujiatai Formation in J13Well Block with Least Squares Support Vector Machine[J].Special Oil and Gas Reservoirs,2011,(18)6:18-21.
[19]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.Zhang Xuegong.Introduction to Statistical Learning Theory and Support Vector Machines[J].Acta Automatica Sinica,2000,26(1):32-42.
[20]阎威武,邵惠鹤.支持向量机和最小二乘支持向量机的比较及应用研究[J].控制与决策,2003,18(3):358-360.Yan Weiwu,Shao Huihe.Application of Support Vector Machines and Least Squares Support Vector Machines to Heart Disease Diagnoses[J].Control and Decision,2003,18(3):358-360.
[21]阎辉,张学工,李衍达.支持向量机与最小二乘法的关系研究[J].清华大学学报:自然科学版,2001,41(9):77-80.Yan Hui,Zhang Xuegong,Li Yanda.Relation Between a Support Vector Machine and the Least Square Method[J].Journal of Tsinghua Univetsity:Science&Technology,2001,41(9):77-80.
[22]顾燕萍,赵文杰,吴占松.最小二乘支持向量机的算法研究[J].清华大学学报:自然科学版,2010,50(7):1063-1066.Gu Yanping,Zhao Wenjie,Wu Zhansong.Least Squares Support Vector Machine Algorithm[J].Journal of Tsinghua University:Science&Technology,2010,50(7):1063-1066.
[23]Maysam A,Gholam H N,Abbas B.Support Vector Machine for Multi-Classification of Mineral Prospectivity Areas[J].Computers&Geosciences,2012,46:272-283.
[24]郭辉,刘贺平,王玲.最小二乘支持向量机参数选择方法及其应用研究[J].系统仿真学报,2006,18(7):2033-2036.Guo Hui,Liu Heping,Wang Ling.Method for Selecting Parameters of Least Squares Support Vector Machines and Application[J].Journal of System Simulation,2006,18(7):2033-2036.
[25]王健峰,张磊,陈国兴,等.基于改进的网格搜索法的SVM参数优化[J].应用科技,2012,39(3):28-31.Wang Jianfeng,Zhang Lei,Chen Guoxing,et al.A Parameter Optimization Method for an SVM Based on Improved Grid Search Algorithm[J].Applied Science and Technology,2012,39(3):28-31.
[26]黄玉龙,王璞珺,舒萍,等.松辽盆地营城组中基性火山岩储层特征及成储机理[J].岩石学报,2010,26(1):82-92.Huang Yulong,Wang Pujun,Shu Ping,et al.Characteristics and Formation Mechanism of the Cretaceous Intermediate and Mafic Volcanic Reservoirs in Songliao Basin[J].Acta Petrologica Sinica,2010,26(1):82-92.
[27]王璞珺,郑常青,舒萍,等.松辽盆地深层火山岩岩性分类方案[J].大庆石油地质与开发,2007,26(4):17-22.Wang Pujun,Zheng Changqing,Shu Ping,et al.Classification of Deep Volcanic Rocks in Songliao Basin[J].Petroleum Geology&Oilfield Development in Daqing,2007,26(4):17-22.