ABP法在高密度电阻率法反演中的应用
|
摘要
非线性反演方法作为地球物理反演的一个重要分支,在地球物理反演中发挥着特有的作用.近年来学者对非线性联合反演研究较多,但目前仍未有实质性的研究进展;本文尝试利用BP(Back Propagation)神经网络优化方法与蚁群算法联合演算,实现高密度电阻率法的电阻率二维非线性反演.通过两组模型的结果比较,BP与ABP法的反演较传统反演法优势较为突出,而且ABP(Ant colony optimization-Back Propagation)方法明显优于BP神经网络反演法,它可以克服BP神经网络反演方法的不足、减少迭代次数、节约计算时间,获得更好的反演结果.
As an important branch of geophysical inversion,non-linear inversion method has played a unique role in geophysical inversion.In recent years,more researchers lay emphasis on non-linear joint inversion,but has not made any substantial progresses.This paper tried to achieve high density resistivity two-dimensional non-linear inversion by using the joint calculus of BP neural network optimization method and ant colony algorithm.A comparison of the results revealed that the ABP method is much better than BP neural network inversion method,the former can overcome the deficiencies of BP neural network inversion method,reduce the number of iterations,save computing time and finally obtain a better inversion result.
As an important branch of geophysical inversion,non-linear inversion method has played a unique role in geophysical inversion.In recent years,more researchers lay emphasis on non-linear joint inversion,but has not made any substantial progresses.This paper tried to achieve high density resistivity two-dimensional non-linear inversion by using the joint calculus of BP neural network optimization method and ant colony algorithm.A comparison of the results revealed that the ABP method is much better than BP neural network inversion method,the former can overcome the deficiencies of BP neural network inversion method,reduce the number of iterations,save computing time and finally obtain a better inversion result.
引文
[1] 祁 民,张宝林,梁光河等.高分辨率预测地下复杂采空区的空间分布特征 高密度电法在山西阳泉某复杂采空区中的初步应用研究.地球物理学进展,2006,21(1) :256～262 Qi M,Zhang B L,Liang G H.High-resolution predict the ground to gather the area of space complex distribution-High density electricity in shanxi yangquan a complex pick up in the initial application.Progress in Geophysicsdn Chinese) ,2006 ,21(1) :256～262
[2] 祁民,张宝林,梁光河.高密度电法的三维数据场可视化.地球物理学进展,2006,21(3) :981～986 Qi M,Zhang B L,Liang G H.The visualization of three dimension data fields of high density electrical technique.Progress in Geophysics (in Chinese) ,2006 ,21(3) :981～986
[3] Sehlmubeger C.Etude surla prospection electique du sous-sol.Gauthie-rVillasretCie,Paris,1920
[4] 杨文采.地球物理反演的理论与方法.北京:地质出版社,1996 Yang W C.The Theory and Method of Geophysics (in Chinese).Beijing:Geological Publishing House,1996
[5] Backus G E,Gilbert J F.Numeficfl application of a formulism for geophysical problem.Geophys.J.R.astr.Soc.,1967,13:247～276
[6] Backus G E ,Gilbert J F.The resolving power of gross earth data.Geophys.J.R.astr.Soc.,1968 ,16:169～205
[7] Wiggins R A.The generalized linear inverse problem:Implication of surface waves and free oscillations for earth structure.Rev.Geophys.SpacePhys.,1972,10:251～285
[8] Jackson D D.Interpretation of inaccurate,insufficient and inconsistent data.Oeophys.J.R.astr.Soc.,1972 ,28:97～109
[9] Parker R L.Understanding inverse theory.Rev Earth Planet Sci.,1976,5:35～64
[10] 艾伯特·塔兰托拉.反演理论一数据拟合和模型参数估计.北京:学术出版社,1 989 Albert ·Talantula.The Theory of Inversion-Data and Estimates Proposed Model Parameters.Beijing:Academic Press,1989
[11] 王家映.地球物理反演理论.武汉:中国地质大学出版社,1998 Wang J Y.Geophysical Theory of Inversion (in Chinese).Wuhan:China University of Geosciences Press,1998
[12] 何继善.电法勘探的发展与展望.地球物理学报,1997,40(增):308～316 He J S.The prospecting only development and prospect.Chinese J.Geophys.(in Chinese) ,1997,40(Suppl) :308～316
[13] 李创社,张彦鹏,李实等.瞬变电磁勘探中的人工神经网络反演法.西安交通大学学报,2001,35(6) :604～615 Li C S,Zhang Y P,Li S,et al.New inversion method of artificial neural network in transient electromagnetic inversion.Journal of Xi'an Traffic University (in Chinese) ,2001,35(6) :604～615
[14] 黄科,肖慈殉,张哨楠等.基于神经网络的储层识别处理技术及在塔河油田的应用.物探化探计算技术,2006,28(1) :18～21 Huang K,Xiao C X,Zhang S N,et al.Based on the nerve network layer of recognition processing technology and applications in the field of Tahe.Computing Technology for Geophysical and Geochemical Exploration (in Chinese),2006,28(1) :18～21
[15] 丁柱,杨长春,陶宏根等.神经网络反演双侧向电阻率测井曲线的物理约束.地球物理学进展,2002,17(2) :331～336 Ding Z,Yang C C,Tao H G,et al.The nerve network the performance of the logs to the resistivity of the physical constraints.Progress in Geophysics (in Chinese),2002,17(2) :331～336
[16] 徐海浪,吴小平.电阻率二维神经网络反演.地球物理学报,2006,49(2) :584～589 Xu H L,Wu X P.2-D resistivity inversion using the neural network method.Chinese J.Geophys (in Chinese) ,2006,49 (2) :584～589
[17] 王家映.人工神经网络反演法.工程地球物理学报,2008,5(3) :255～265 Wang J Y.The artificial neural network method.Chinese Journal of Engineering Geophysics(in Chinese) ,2008,5(3) :255～265
[18] Colorni A,Dorigo M,Maniezzo V.Dist ributed optimization by ant colonies.Proc of the First European Conference on artificial Life.Paris,France:Elsevier Publishing,1991． 134～142
[19] Colorni A,Dorigo M,Maniezzo V.An investigation of some properties of an ant aigorithm .Proc of the Parallel Problem Solving from Nature Conference (PPSN' 92) .Brussels,Blegium:Elsevier Publishing,1992． 509～520
[20] Gong D X,Ruan X G.A hybrid approach of GA and ACO for TSP.Proceedings of the 5th World Congress on Intelligent Control and Automation,Hangzhou,China,2004,3:2068～2072
[21] Kumar G M.Haq A N.Hybrid genetic-ant colony algorithms for solving aggregate production plan.Journal of Advanced Manufacturing Systems ,2005 ,4(1) :103～111
[22] Pilat M L,White T.Using genetic algorithms to optimize ACS-TSP.Proceedings of Ant Algorithms;Third International Workshop,ANTS 2002 .Brussels,Belgium,2002． 282～287
[23] 丁建立,陈增强.遗传算法与蚁群算法的融合闭.计算机研究与发展,2003,40(9) :1351～1356 Ding J L,Chen Z Q.The method combinating genetic algorithm with the group of algorithms.Journal of Computer Research and Development (in Chinese) ,2003,40 (9) :1351～1356
[24] Hong B R,Jin F H,Guo Q.Hopfield neural network based on ant system.Journal of Haerbin Institue of Technology,2004,11(3) :267～269
[25] 邹政达,孙雅明,张智晟.基于蚁群优化算法递归神经网络的短期负荷预测.电网技术,2005,29(3) :59～63 Zou Z D,Sun Y M,Zhang Z S.The short-term forecast of the algorithm based on the number of recursive neural network.Power System Technology (in Chinese) ,2005 ,29(3) :59～63
[26] 张国立,王 晶,谢宏等.利用蚁群算法优化前向神经网络.电子商务,2005,18:65～67 Zhang G L,Wang J.Xie H,et al.Using of the ant colony algorithm to optimize neural network.E-business (in Chinese) .2005,18:65～67
[27] 刘彦鹏.蚁群优化算法的理论研究及其应用.杭州:浙江大学出版社,2007 Liu Y P.The ACO theory researching and application (in Chinese).Hangzhou:Zhejiang University Press,2007
[28] 王书明,刘玉兰,王家映.地球物理资料非线性反演方法讲座蚁群算法.工程地球物理学报,2009,6(2) :131～135 Wang S M,Liu Y L,Wang J Y.ant colony optimization.Chinese Journal of Engineering Geophysics (in Chinese) ,2009,6(2) :131～135
[2] 祁民,张宝林,梁光河.高密度电法的三维数据场可视化.地球物理学进展,2006,21(3) :981～986 Qi M,Zhang B L,Liang G H.The visualization of three dimension data fields of high density electrical technique.Progress in Geophysics (in Chinese) ,2006 ,21(3) :981～986
[3] Sehlmubeger C.Etude surla prospection electique du sous-sol.Gauthie-rVillasretCie,Paris,1920
[4] 杨文采.地球物理反演的理论与方法.北京:地质出版社,1996 Yang W C.The Theory and Method of Geophysics (in Chinese).Beijing:Geological Publishing House,1996
[5] Backus G E,Gilbert J F.Numeficfl application of a formulism for geophysical problem.Geophys.J.R.astr.Soc.,1967,13:247～276
[6] Backus G E ,Gilbert J F.The resolving power of gross earth data.Geophys.J.R.astr.Soc.,1968 ,16:169～205
[7] Wiggins R A.The generalized linear inverse problem:Implication of surface waves and free oscillations for earth structure.Rev.Geophys.SpacePhys.,1972,10:251～285
[8] Jackson D D.Interpretation of inaccurate,insufficient and inconsistent data.Oeophys.J.R.astr.Soc.,1972 ,28:97～109
[9] Parker R L.Understanding inverse theory.Rev Earth Planet Sci.,1976,5:35～64
[10] 艾伯特·塔兰托拉.反演理论一数据拟合和模型参数估计.北京:学术出版社,1 989 Albert ·Talantula.The Theory of Inversion-Data and Estimates Proposed Model Parameters.Beijing:Academic Press,1989
[11] 王家映.地球物理反演理论.武汉:中国地质大学出版社,1998 Wang J Y.Geophysical Theory of Inversion (in Chinese).Wuhan:China University of Geosciences Press,1998
[12] 何继善.电法勘探的发展与展望.地球物理学报,1997,40(增):308～316 He J S.The prospecting only development and prospect.Chinese J.Geophys.(in Chinese) ,1997,40(Suppl) :308～316
[13] 李创社,张彦鹏,李实等.瞬变电磁勘探中的人工神经网络反演法.西安交通大学学报,2001,35(6) :604～615 Li C S,Zhang Y P,Li S,et al.New inversion method of artificial neural network in transient electromagnetic inversion.Journal of Xi'an Traffic University (in Chinese) ,2001,35(6) :604～615
[14] 黄科,肖慈殉,张哨楠等.基于神经网络的储层识别处理技术及在塔河油田的应用.物探化探计算技术,2006,28(1) :18～21 Huang K,Xiao C X,Zhang S N,et al.Based on the nerve network layer of recognition processing technology and applications in the field of Tahe.Computing Technology for Geophysical and Geochemical Exploration (in Chinese),2006,28(1) :18～21
[15] 丁柱,杨长春,陶宏根等.神经网络反演双侧向电阻率测井曲线的物理约束.地球物理学进展,2002,17(2) :331～336 Ding Z,Yang C C,Tao H G,et al.The nerve network the performance of the logs to the resistivity of the physical constraints.Progress in Geophysics (in Chinese),2002,17(2) :331～336
[16] 徐海浪,吴小平.电阻率二维神经网络反演.地球物理学报,2006,49(2) :584～589 Xu H L,Wu X P.2-D resistivity inversion using the neural network method.Chinese J.Geophys (in Chinese) ,2006,49 (2) :584～589
[17] 王家映.人工神经网络反演法.工程地球物理学报,2008,5(3) :255～265 Wang J Y.The artificial neural network method.Chinese Journal of Engineering Geophysics(in Chinese) ,2008,5(3) :255～265
[18] Colorni A,Dorigo M,Maniezzo V.Dist ributed optimization by ant colonies.Proc of the First European Conference on artificial Life.Paris,France:Elsevier Publishing,1991． 134～142
[19] Colorni A,Dorigo M,Maniezzo V.An investigation of some properties of an ant aigorithm .Proc of the Parallel Problem Solving from Nature Conference (PPSN' 92) .Brussels,Blegium:Elsevier Publishing,1992． 509～520
[20] Gong D X,Ruan X G.A hybrid approach of GA and ACO for TSP.Proceedings of the 5th World Congress on Intelligent Control and Automation,Hangzhou,China,2004,3:2068～2072
[21] Kumar G M.Haq A N.Hybrid genetic-ant colony algorithms for solving aggregate production plan.Journal of Advanced Manufacturing Systems ,2005 ,4(1) :103～111
[22] Pilat M L,White T.Using genetic algorithms to optimize ACS-TSP.Proceedings of Ant Algorithms;Third International Workshop,ANTS 2002 .Brussels,Belgium,2002． 282～287
[23] 丁建立,陈增强.遗传算法与蚁群算法的融合闭.计算机研究与发展,2003,40(9) :1351～1356 Ding J L,Chen Z Q.The method combinating genetic algorithm with the group of algorithms.Journal of Computer Research and Development (in Chinese) ,2003,40 (9) :1351～1356
[24] Hong B R,Jin F H,Guo Q.Hopfield neural network based on ant system.Journal of Haerbin Institue of Technology,2004,11(3) :267～269
[25] 邹政达,孙雅明,张智晟.基于蚁群优化算法递归神经网络的短期负荷预测.电网技术,2005,29(3) :59～63 Zou Z D,Sun Y M,Zhang Z S.The short-term forecast of the algorithm based on the number of recursive neural network.Power System Technology (in Chinese) ,2005 ,29(3) :59～63
[26] 张国立,王 晶,谢宏等.利用蚁群算法优化前向神经网络.电子商务,2005,18:65～67 Zhang G L,Wang J.Xie H,et al.Using of the ant colony algorithm to optimize neural network.E-business (in Chinese) .2005,18:65～67
[27] 刘彦鹏.蚁群优化算法的理论研究及其应用.杭州:浙江大学出版社,2007 Liu Y P.The ACO theory researching and application (in Chinese).Hangzhou:Zhejiang University Press,2007
[28] 王书明,刘玉兰,王家映.地球物理资料非线性反演方法讲座蚁群算法.工程地球物理学报,2009,6(2) :131～135 Wang S M,Liu Y L,Wang J Y.ant colony optimization.Chinese Journal of Engineering Geophysics (in Chinese) ,2009,6(2) :131～135
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |