BibTex RIS Cite

Güncel Optimizasyon Yöntemleri Kullanılarak Rezidüel Gravite Anomalilerinden Parametre Kestirimi

Year 2015, Volume: 36 Issue: 1, 31 - 44, 23.12.2015
https://doi.org/10.17824/yrb.71895

Abstract

Bu çalışmada, jeofizik modellemede yaygın olarak kullanılan global ve geleneksel yöntemlere ek olarak, yapay sinir ağları yöntemleri yeraltı boşluklarına ait rezidüel gravite anomalisinden parametre kestirimi amacıyla kullanılmıştır. İleri Beslemeli Geri Yayılımlı sinir ağı günümüzde ters çözüm problemlerinde sıklıkla kullanılan bir yöntemdir. Bu yönteme ek olarak bu çalışmada İleri Kademeli Geri Yayılımlı ve Doğrusal Olmayan Otoregresif sinir ağı, parametre kestirimi için denenmiş ve sonuçlar karşılaştırılmıştır. Ayrıca global bir yöntem olan Genetik Algoritma ve geleneksel bir yöntem olan Levenberg-Marquardt algoritması ile rezidüel anomaliden derinlik ve yarıçap parametreleri hesaplanmış ve sonuçlar karşılaştırılmıştır. Hem teorik hem arazi verisi üzerinde bu yöntemler denenmiştir. Kuramsal çalışmalarda, yeraltı boşluklarını temsil eden yatay silindir modeli kullanılmıştır. Yöntemlerin etkinliği yatay silindir gravite anomalisine gürültü eklenerek sınanmıştır. Hata değerleri incelendiğinde Levenberg-Marquardt algoritması ve doğrusal olmayan otoregresif sinir ağının gürültüden en az etkilenen yöntemler olduğu görülmektedir. Arazi verisi olarak Medford (ABD) alanındaki yeraltı boşluğuna ait rezidüel gravite anomalisi kullanılmıştır. Sonuçlar incelendiğinde ileri beslemeli geri yayılımlı ve doğrusal olmayan otoregresif sinir ağının sondajdan bilinen derinlik değerine en yakın sonucu verdiği görülmektedir. Levenberg-Marquardt algoritması kullanılarak arazi verisinin ters çözümü ile en düşük ortalama karekök hata değeri hesaplanmasına rağmen, hesaplanan derinlik sondajdan bilinen derinlik değerine en uzaktır.


Anahtar Kelimeler: Yapay sinir ağları, genetik algoritma, doğrusal olmayan otoregresif sinir ağı, ileri kademeli geri yayılımlı sinir ağı.

References

  • Abdelrahman, E.M., Bayoumi, A.I., ve El-Araby, H.M., 1991. A least-squares minimizati- on approach to invert gravity data, Ge- ophysics, 56, 1, 115-118.
  • Basu, A., ve Frazer, L.N., 1990. Rapid determi- nation of critical temperature in simula- ted annealing inversion, Science, 249, 1409-1412.
  • Başokur, A.T., 2002. Doğrusal ve Doğrusal Ol- mayan Problemlerin Ters-Çözümü. Je- ofizik Mühendisleri Odası Eğitim Yayın- ları, Ankara.
  • Beiki, M., ve Pedersen, L.B., 2010. Eigenvector analysis of gravity gradient tensor to lo- cate geologic bodies, Geophysics, 75, 6, I37-I49.
  • Butler, D.K., 1984. Microgravimetric and gravity gradient techniques for detection of subsurface cavities, Geophysics, 49, 7, 1084-1096.
  • Calderón-Macías, C., Sen, M.K., ve Stoffa, P.L.,
  • Hopfield neural networks, and
  • mean field annealing for seismic de
  • convolution and multiple attenuation,
  • Geophysics, 62, 3, 992-1002.
  • Calderón-Macías, C., Sen, M.K. ve Stoffa, P.L.,
  • Automatic NMO correction and
  • velocity estimation by a feedforward
  • neural network, Geophysics, 63, 5, 1696-1707.
  • Chamoli, A., Srivastava, R.P. ve Dimri, V.P., 2006. Source depth characterization of poten- tial field data of Bay of Bengal by conti- nuous wavelet transform, Indian Journal of Marine Sciences, 35, 3, 195-204.
  • Dosso, S.E., ve Oldenburg, D.W., 1991. Mag- netotelluric appraisal using simulated annealing, Geophysical Journal Inter- national, 106, 370-85.
  • Elawadi, E., Salem, A., ve Ushijima, K., 2001. Detection of cavities and tunnels from gravity data using a neural network, Exploration Geophysics, 32, 4, 204-208.
  • Gupta, O.P., 1983. A least-squares approach to depth determination from gravity data, Geophysics, 48, 3, 357-360.
  • Günaydın, K., ve Günaydın, A., 2008. Peak gro- und acceleration prediction by artificial neural networks for Northwestern Tur- key, Mathematical Problems in Engine- ering, 2008.
  • Hajian, A., 2004. Depth Estimation of Gravity Anomalies by Neural Network, M.Sc.
  • Thesis, Tehran University, Iran (in Per- sian).
  • Hajian, A., Styles, P. ve Zomorrodian, H., 2011. Depth estimation of cavities from mic- rogravity data through multi adaptive neuro fuzzy interference system. In:
  • “Near Surface” 2011, Proc. 17th Euro
  • pean Meeting of Environmental and En
  • gineering Geophysics, 12-14 Septem
  • ber 2011, Leicester, UK.
  • Hajian, A., Zomorrodian, H., ve Styles, P., 2012. Simultaneous Estimation of Shape Fac- tor and Depth of Subsurface Cavities from Residual Gravity Anomalies using Feed-Forward Back-Propagation Ne- ural Networks, Acta Geophysica, 60, 1043-1075.
  • Haykin, S., 1999. Neural Networks: A Compre- hensive Foundation, 2nd ed., Prentice- Hall Inc., Englewood Cliffs.
  • Ho, T. L., 2009. 3-D inversion of borehole-to- surface electrical data using a back- propagation neural network, Journal of Applied Geophysics, 68, 4, 489-499.
  • Holland, J., 1975. Adaptation in Natural and Ar- tificial Systems, University of Michigan Press.
  • Huang, Z., Shimeld, J., Williamson, M., ve Kat- sube, J., 1996. Permeability prediction with artificial neural network modeling in the Ventura gas field, offshore eas- tern Canada, Geophysics, 61, 2, 422- 436.
  • Landa, E., Beydoun, W., ve Tarantola, A., 1989. Reference velocity model estimation from prestack waveforms: coherency optimization by simulated annealing, Geophysics, 54, 984-990.
  • Langer, H., Nunnari G., ve Occhipinti L., 1996. Estimation of seismic waveform gover- ning parameters with neural networks, Journal of Geophysical Research, 101, B9, 20109-20118.
  • Lee, K. Y., ve Mohamed, P. S., 2002. A real-co- ded genetic algorithm involving a hybrid crossover method for power plant control system design, in Proceedings of the 2002 Congress on Evolutionary Computation, pp. 1069-1074.
  • Levenberg, K., 1944. A Method for the solution of certain nonlinear problems in least squares. The Quarterly of Applied Mat- hematics, 2, 164-168.
  • Lines, L.R. ve Treitel, S., 1984. A review of least- squares inversion and its application to geophysical problems, Geophysical Prospecting, 32, 2, 159-186.
  • Luke, S., 2009. Essentials of metaheuristics. Lulu, Retrieved January 20th, 2012, from http://cs.gmu.edu/~sean/book/ metaheuristics/.
  • Marquardt, D.W., 1963. An algorithm for le- ast-squares estimation of nonlinear parameters, Journal of the Society for Industrial and Applied Mathematics, 11(2): 431-441.
  • McCulloch, W.S., ve Pitts, W., 1943. A logical calculus of the ideas immanent in ner- vous activity, Bulletin of Mathematical Biology, 5, 4, 115-133.
  • Oruç, B., 2010. Depth estimation of simple ca- usative sources from gravity gradient tensor invariants and vertical compo- nent, Pure and Applied Geophysics, 167, 10, 1259-1272.
  • Oruç, B., 2012. Teori ve Örneklerle Jeofizikte Modelleme. Umuttepe Yayınları, Koca- eli.
  • Osman, O., Albora, A.M., ve Ucan O.N., 2007. Forward modeling with forced neural networks for gravity anomaly profile, Mathematical Geology, 39, 6, 593-605.
  • Poulton, M.M., Sternberg, B.K., ve Glass, C.E.,
  • Location of subsurface targets
  • in geophysical data using neural net
  • works, Geophysics, 57, 12, 1534- 1544.
  • Reid, A.B., Allsop, J.M., Granser, H., Millett, A.J., ve Somerton, I.W., 1990. Magnetic in- terpretation in three dimensions using Euler deconvolution, Geophysics, 55, 1, 80-91.
  • Roest, W. R., Verhoef, J., ve Pilkington, M.,
  • Magnetic interpretation using 3-D
  • analytic signal: Geophysics, 57, 116– 125.
  • Roth, G., ve Tarantola, A., 1994. Neural net- works and inversion of seismic data,
  • Journal of Geophysical Research, 99, B4, 6753-6768.
  • Roy, A., 1962. Ambiguity in geophysical interp- retation, Geophysics, 27, 1, 90-99.
  • Roy, L., Agarwal, N.P., ve Shaw, R.K., 2000. A new concept in Euler deconvolution of isolated gravity anomalies, Geophysi- cal Prospecting, 48, 3, 559-575.
  • Rumelhart, D.E., Hinton, G.E., ve Williams, R.J.,
  • Learning internal representati
  • on by error back propagation. In: D.E.
  • Rumelhart and J.L. Mc Clelland (eds.),
  • Parallel Distributed Processing: Exp
  • lorations in the Microstructure of Cog
  • nition. Vol. 1. Foundations, MIT Press,
  • Cambridge, USA, 318-362.
  • Salem, A., Elawadi, E., ve Ushijima, K., 2003. Short note: Depth determination from residual gravity anomaly data using a simple formula, Computer and Geosci- ence, 29, 6, 801-804.
  • Sen, M. K., ve Stoffa, P. L., 1991. Nonlinear mul- tiparameter optimization using genetic algorithms: Inversion of plane wave seismograms, Geophysics, 56, 1794- 1810.
  • Sen, M.K., ve Stoffa, P. L., 1992. Seismic wa- veform inversion using global optimiza- tion. Journal of Seismic Exploration, 1, 9-27.
  • Szu, H., ve Hartley, R., 1987. Fast Simulated Annealing, Physical Letters A, 122, No. 3, 157-162.
  • Thompson, D.T., 1982. EULDPH: A new tech- nique for making computer-assisted depth estimates from magnetic data, Geophysics, 47, 1, 31-37.
  • Vestergaard, P. D., ve Mosegaard, K. 1991. Inversion of post-stack seismic data using simulated annealing, Geophysi- cal Prospecting, 39, 613-624.
  • Wang, L.X., ve Mendel, J.M., 1992. Adaptive minimum prediction-error deconvoluti- on and source wavelet estimation using Hopfield neural networks, Geophysics, 57, 5, 670-679.
  • Zhang, Y., ve Paulson K.V., 1997. Magnetotellu- ric inversion using regularized Hopfield neural networks, Geophysical Prospec- ting, 45, 5, 725-743.
Year 2015, Volume: 36 Issue: 1, 31 - 44, 23.12.2015
https://doi.org/10.17824/yrb.71895

Abstract

References

  • Abdelrahman, E.M., Bayoumi, A.I., ve El-Araby, H.M., 1991. A least-squares minimizati- on approach to invert gravity data, Ge- ophysics, 56, 1, 115-118.
  • Basu, A., ve Frazer, L.N., 1990. Rapid determi- nation of critical temperature in simula- ted annealing inversion, Science, 249, 1409-1412.
  • Başokur, A.T., 2002. Doğrusal ve Doğrusal Ol- mayan Problemlerin Ters-Çözümü. Je- ofizik Mühendisleri Odası Eğitim Yayın- ları, Ankara.
  • Beiki, M., ve Pedersen, L.B., 2010. Eigenvector analysis of gravity gradient tensor to lo- cate geologic bodies, Geophysics, 75, 6, I37-I49.
  • Butler, D.K., 1984. Microgravimetric and gravity gradient techniques for detection of subsurface cavities, Geophysics, 49, 7, 1084-1096.
  • Calderón-Macías, C., Sen, M.K., ve Stoffa, P.L.,
  • Hopfield neural networks, and
  • mean field annealing for seismic de
  • convolution and multiple attenuation,
  • Geophysics, 62, 3, 992-1002.
  • Calderón-Macías, C., Sen, M.K. ve Stoffa, P.L.,
  • Automatic NMO correction and
  • velocity estimation by a feedforward
  • neural network, Geophysics, 63, 5, 1696-1707.
  • Chamoli, A., Srivastava, R.P. ve Dimri, V.P., 2006. Source depth characterization of poten- tial field data of Bay of Bengal by conti- nuous wavelet transform, Indian Journal of Marine Sciences, 35, 3, 195-204.
  • Dosso, S.E., ve Oldenburg, D.W., 1991. Mag- netotelluric appraisal using simulated annealing, Geophysical Journal Inter- national, 106, 370-85.
  • Elawadi, E., Salem, A., ve Ushijima, K., 2001. Detection of cavities and tunnels from gravity data using a neural network, Exploration Geophysics, 32, 4, 204-208.
  • Gupta, O.P., 1983. A least-squares approach to depth determination from gravity data, Geophysics, 48, 3, 357-360.
  • Günaydın, K., ve Günaydın, A., 2008. Peak gro- und acceleration prediction by artificial neural networks for Northwestern Tur- key, Mathematical Problems in Engine- ering, 2008.
  • Hajian, A., 2004. Depth Estimation of Gravity Anomalies by Neural Network, M.Sc.
  • Thesis, Tehran University, Iran (in Per- sian).
  • Hajian, A., Styles, P. ve Zomorrodian, H., 2011. Depth estimation of cavities from mic- rogravity data through multi adaptive neuro fuzzy interference system. In:
  • “Near Surface” 2011, Proc. 17th Euro
  • pean Meeting of Environmental and En
  • gineering Geophysics, 12-14 Septem
  • ber 2011, Leicester, UK.
  • Hajian, A., Zomorrodian, H., ve Styles, P., 2012. Simultaneous Estimation of Shape Fac- tor and Depth of Subsurface Cavities from Residual Gravity Anomalies using Feed-Forward Back-Propagation Ne- ural Networks, Acta Geophysica, 60, 1043-1075.
  • Haykin, S., 1999. Neural Networks: A Compre- hensive Foundation, 2nd ed., Prentice- Hall Inc., Englewood Cliffs.
  • Ho, T. L., 2009. 3-D inversion of borehole-to- surface electrical data using a back- propagation neural network, Journal of Applied Geophysics, 68, 4, 489-499.
  • Holland, J., 1975. Adaptation in Natural and Ar- tificial Systems, University of Michigan Press.
  • Huang, Z., Shimeld, J., Williamson, M., ve Kat- sube, J., 1996. Permeability prediction with artificial neural network modeling in the Ventura gas field, offshore eas- tern Canada, Geophysics, 61, 2, 422- 436.
  • Landa, E., Beydoun, W., ve Tarantola, A., 1989. Reference velocity model estimation from prestack waveforms: coherency optimization by simulated annealing, Geophysics, 54, 984-990.
  • Langer, H., Nunnari G., ve Occhipinti L., 1996. Estimation of seismic waveform gover- ning parameters with neural networks, Journal of Geophysical Research, 101, B9, 20109-20118.
  • Lee, K. Y., ve Mohamed, P. S., 2002. A real-co- ded genetic algorithm involving a hybrid crossover method for power plant control system design, in Proceedings of the 2002 Congress on Evolutionary Computation, pp. 1069-1074.
  • Levenberg, K., 1944. A Method for the solution of certain nonlinear problems in least squares. The Quarterly of Applied Mat- hematics, 2, 164-168.
  • Lines, L.R. ve Treitel, S., 1984. A review of least- squares inversion and its application to geophysical problems, Geophysical Prospecting, 32, 2, 159-186.
  • Luke, S., 2009. Essentials of metaheuristics. Lulu, Retrieved January 20th, 2012, from http://cs.gmu.edu/~sean/book/ metaheuristics/.
  • Marquardt, D.W., 1963. An algorithm for le- ast-squares estimation of nonlinear parameters, Journal of the Society for Industrial and Applied Mathematics, 11(2): 431-441.
  • McCulloch, W.S., ve Pitts, W., 1943. A logical calculus of the ideas immanent in ner- vous activity, Bulletin of Mathematical Biology, 5, 4, 115-133.
  • Oruç, B., 2010. Depth estimation of simple ca- usative sources from gravity gradient tensor invariants and vertical compo- nent, Pure and Applied Geophysics, 167, 10, 1259-1272.
  • Oruç, B., 2012. Teori ve Örneklerle Jeofizikte Modelleme. Umuttepe Yayınları, Koca- eli.
  • Osman, O., Albora, A.M., ve Ucan O.N., 2007. Forward modeling with forced neural networks for gravity anomaly profile, Mathematical Geology, 39, 6, 593-605.
  • Poulton, M.M., Sternberg, B.K., ve Glass, C.E.,
  • Location of subsurface targets
  • in geophysical data using neural net
  • works, Geophysics, 57, 12, 1534- 1544.
  • Reid, A.B., Allsop, J.M., Granser, H., Millett, A.J., ve Somerton, I.W., 1990. Magnetic in- terpretation in three dimensions using Euler deconvolution, Geophysics, 55, 1, 80-91.
  • Roest, W. R., Verhoef, J., ve Pilkington, M.,
  • Magnetic interpretation using 3-D
  • analytic signal: Geophysics, 57, 116– 125.
  • Roth, G., ve Tarantola, A., 1994. Neural net- works and inversion of seismic data,
  • Journal of Geophysical Research, 99, B4, 6753-6768.
  • Roy, A., 1962. Ambiguity in geophysical interp- retation, Geophysics, 27, 1, 90-99.
  • Roy, L., Agarwal, N.P., ve Shaw, R.K., 2000. A new concept in Euler deconvolution of isolated gravity anomalies, Geophysi- cal Prospecting, 48, 3, 559-575.
  • Rumelhart, D.E., Hinton, G.E., ve Williams, R.J.,
  • Learning internal representati
  • on by error back propagation. In: D.E.
  • Rumelhart and J.L. Mc Clelland (eds.),
  • Parallel Distributed Processing: Exp
  • lorations in the Microstructure of Cog
  • nition. Vol. 1. Foundations, MIT Press,
  • Cambridge, USA, 318-362.
  • Salem, A., Elawadi, E., ve Ushijima, K., 2003. Short note: Depth determination from residual gravity anomaly data using a simple formula, Computer and Geosci- ence, 29, 6, 801-804.
  • Sen, M. K., ve Stoffa, P. L., 1991. Nonlinear mul- tiparameter optimization using genetic algorithms: Inversion of plane wave seismograms, Geophysics, 56, 1794- 1810.
  • Sen, M.K., ve Stoffa, P. L., 1992. Seismic wa- veform inversion using global optimiza- tion. Journal of Seismic Exploration, 1, 9-27.
  • Szu, H., ve Hartley, R., 1987. Fast Simulated Annealing, Physical Letters A, 122, No. 3, 157-162.
  • Thompson, D.T., 1982. EULDPH: A new tech- nique for making computer-assisted depth estimates from magnetic data, Geophysics, 47, 1, 31-37.
  • Vestergaard, P. D., ve Mosegaard, K. 1991. Inversion of post-stack seismic data using simulated annealing, Geophysi- cal Prospecting, 39, 613-624.
  • Wang, L.X., ve Mendel, J.M., 1992. Adaptive minimum prediction-error deconvoluti- on and source wavelet estimation using Hopfield neural networks, Geophysics, 57, 5, 670-679.
  • Zhang, Y., ve Paulson K.V., 1997. Magnetotellu- ric inversion using regularized Hopfield neural networks, Geophysical Prospec- ting, 45, 5, 725-743.
There are 70 citations in total.

Details

Primary Language Turkish
Journal Section Articles
Authors

Fikret Doğru

Publication Date December 23, 2015
Submission Date December 23, 2015
Published in Issue Year 2015 Volume: 36 Issue: 1

Cite

EndNote Doğru F (December 1, 2015) Güncel Optimizasyon Yöntemleri Kullanılarak Rezidüel Gravite Anomalilerinden Parametre Kestirimi. Yerbilimleri 36 1 31–44.