Nonlinear Processes in Geophysics www.nonlin-processes-geophys.net 1023-5809 1607-7946 17 1 2010 10.5194/npg-17-65-2010 http://www.nonlin-processes-geophys.net/17/65/2010/ http://www.nonlin-processes-geophys.net/17/65/2010/npg-17-65-2010.html http://www.nonlin-processes-geophys.net/17/65/2010/npg-17-65-2010.pdf 65 76 2010-02-03 Inversion of 2-D DC resistivity data using rapid optimization and minimal complexity neural network U. K. Singh upendra{_}bhu1@rediffmail.com R. K. Tiwari S. B. Singh Department of Applied Geophysics, Indian School of Mines, Dhanbad-826 004, India National Geophysical Research Institute, Hyderabad-500 007, India The backpropagation (BP) artificial neural network (ANN) technique of optimization based on steepest descent algorithm is known to be inept for its poor performance and does not ensure global convergence. Nonlinear and complex DC resistivity data require efficient ANN model and more intensive optimization procedures for better results and interpretations. Improvements in the computational ANN modeling process are described with the goals of enhancing the optimization process and reducing ANN model complexity. Well-established optimization methods, such as Radial basis algorithm (RBA) and Levenberg-Marquardt algorithms (LMA) have frequently been used to deal with complexity and nonlinearity in such complex geophysical records. We examined here the efficiency of trained LMA and RB networks by using 2-D synthetic resistivity data and then finally applied to the actual field vertical electrical resistivity sounding (VES) data collected from the Puga Valley, Jammu and Kashmir, India. The resulting ANN reconstruction resistivity results are compared with the result of existing inversion approaches, which are in good agreement. The depths and resistivity structures obtained by the ANN methods also correlate well with the known drilling results and geologic boundaries. The application of the above ANN algorithms proves to be robust and could be used for fast estimation of resistive structures for other complex earth model also. Battiti, R.: First and second order methods for learning between steepest descent and Newton's methods, Neural Comput., 4, 141–166, 1992. Baum, E. and Haussler, D.: What size net gives valid generalization?, Neural Comput., 1, 151–160, 1989. Brown, M. and Poulton, M.: Locating buried objects for environmental site investigations using neural networks, J. Environ. Eng. Geoph., 1, 179–188, 1996. Constable, S., Parker, R., and Constable, C.: Occam's inversion: a practical algorithm for generating smooth models from electromagnetic sounding data, Geophysics, 52, 4919–4930, 1987. Dahlin, T.: 2-D resistivity surveying for environmental and engineering applications, First Break, 14, 275–284, 1996. Dahlin, T., Bernstone, C., and Loke, M. H.: A 3-D resistivity investigation of a contaminated site at Lernacken, Sweden, Geophysics, 67, 1692–1700, 2002. Dey, A. and Morrison, H. F.: Resistivity modeling for arbitrarily shaped two-dimensional structures, Geophys. Prospect., 27, 1020–1036, 1979. Edwards, L.: SA modified pseudosection for resistivity and induced polarization, Geophysics, 42, 1020–1036, 1977. Ellis, R. G. and Oldenburg, D. W.: Applied geophysical inversion, Geophys. J. Int., 116, 5–11, 1994. El-Qady, G. and Ushijima, K.: Inversion of DC resistivity data using neural networks, Geophys. Prospect., 49, 417–430, 2001. Griffiths, D. and Barker, R.: Two-dimensional resistivity imaging and modeling in areas of complex geology, J. Appl. Geophys., 29, 211–226, 1993. Hagan, M. T. and Menhaj, M.: Training feed forward networks with the Marquardt algorithms, IEEE T. Neural Networ., 5, 4~pp., 1994. Haykin, S.: Neural networks: A comprehensive foundation, Macmillan Publishing Co., 1994. Hornik, K.: Approximation capabilities of multilayer feed forward network, Neural Networks, 4, 251–257, 1991. Keller, G. V. and Frishchknecht, F. C.: Electrical methods in geophysical prospecting, Pergamon press, 519~pp., 1996. Levenberg, K.: A Method for the Solution of Certain Non-Linear Problems in Least Squares, Q. Appl. Math., II(2), 164–168, 1994 Loke, M. and Barker, R.: Rapid least squares inversion of apparent resistivity pseudosections using a quasi Newton's method, Geophys. Prospect., 44, 131–152, 1996. Marquardt, D. W.: An algorithm for least squares estimation of non-linear parameters, J. Soc. Ind. Appl. Math., 11, 431–41, 1963 Masters, T.: Advanced Algorithm for Neural Networks, Wiley, New York, 1995. McCormack, M., Zaucha, D., and Dushek, D.: First break refraction picking and seismic data trace editing using neural networks, Geophysics, 58, 67–78, 1993. Meju, M.: An effective ridge regression procedure for resistivity data inversion, Computer and Geosciences, 18, 99–118, 1992. Oldenburg, D. W. and Li, Y.: Estimating the depth of investigation in DC resistivity and IP surveys, Geophysics, 64, 403–416, 1999. Padhi, Ravishanker, Arora, R. N., Gyan Prakash, C. L., Thussa, J. L. and Dua, K. J. S.: Geothermal exploration of Puga and Chumatang geothermal fields, Ladakh, India, in: Proc. UN symposium, Development use geothermal resources, 2nd edn., San Francisco, Calif., 245–258, 1992. Pelton, W., Rijo, L., and Swift, C.: Inversion of two-dimensional resistivity and induced polarization data, Geophysics, 43, 788–803, 1978. Poulton, M., Sternberg, K., and Glass, C.: Neural network pattern recognition of subsurface EM images, J. Appl. Geophys., 29, 21–36, 1992. Raiche, A.: A pattern recognition approach to geophysical inversion using neural networks, Geophys. J. Int., 105, 629–648, 1991. Raina, B. N., Nanda, M., Bhat, M. L., Mehrotra, P., and Dhalli, B. N.: Report on the investigations of coal, limestone, borax, and sulphur deposits of Ladakh: Geological survey of India, unpublished report, 1963. Roth, G. and Tarantola, A.: Neural networks and inversion of seismic data, J. Geophys. Res., 99, 6753–6768, 1994. Rummelhart, D. E., Hinton, G. E., and Williams, J.: Learning representations by back-propagating errors, Nature, 323(9), 533–535, 1986. Singh, S. B.: Application of Geoelectrical and Geothermal methods for the exploration of geothermal resources in India, Banaras Hindu University, Varanasi, India, 1985. Singh, U. K., Somvanshi, V. K., Tiwari, R. K., and Singh, S. B.: Inversion of DC resistivity data using neural network approach, IGC-2002, 57–64, 2002. Singh, U. K., Tiwari, R. K., and Singh, S. B.: One-dimensional inversion of geo-electrical resistivity sounding data using Artificial Neural Networks-a case study, Computers and Geosciences, 31, 99–108, 2006. Singh, U. K., Tiwari, R. K., Singh, S. B., and Rajan, S.: Prediction of electrical resistivity structures using artificial neural networks, J. Geol. Soc India, 67, 234–242, 2005. Sasaki, Y.: Two dimensional joint inversion of magneto-telluric and dipole-dipole resistivity data, Geophysics, 54, 254–262, 1989. Smith, N. and Vozoff, K.: Two-dimensional DC resistivity inversion for dipole-dipole data, IEEE T. Geosci. Remote, 22, 21–28, 1984. Tripp, A., Hohmann, G., and Swift, C.: Two-dimensional resistivity inversion, Geophysics, 49, 1708–1717, 1984. Uchida, T.: Two-dimensional resistivity inversion for Schlumberger sounding, Geophysical Exploration of Japan, 44, 1–17, 1991. Zhang, Z., Mackie, R. L., and Madden, T. R.: 3-D resistivity forward modeling and inversion using conjugate gradients, Geophysics, 60, 1313–1325, 1995. Zhang, Z. and Zhou, Z.: Real time quasi 2-D inversion or array resistivity logging data using neural networks, Geophysics, 67, 517–524, 2002. Zohdy, A. R.: A new method for automatic interpretation of Schlumberger and Wener sounding curves, Geophysics, 54, 245–253, 1989.