Inversion of nuclear well‐logging data using neural networks

This work looks at the application of neural networks in geophysical well-logging problems and specifically their utilization for inversion of nuclear downhole data. Simulated neutron and γ-ray fluxes at a given detector location within a neutron logging tool were inverted to obtain formation properties such as porosity, salinity and oil/water saturation. To achieve this, the forward particle-radiation transport problem was first solved for different energy groups (47 neutron groups and 20 γ-ray groups) using the multigroup code EVENT. A neural network for each of the neutron and γ-ray energy groups was trained to re-produce the detector fluxes using the forward modelling results from 504 scenarios. The networks were subsequently tested on unseen data sets and the unseen input parameters (formation properties) were then predicted using a global search procedure. The results obtained are very encouraging with formation properties being predicted to within 10% average relative error. The examples presented show that neural networks can be applied successfully to nuclear well-logging problems. This enables the implementation of a fast inversion procedure, yielding quick and reliable values for unknown subsurface properties such as porosity, salinity and oil saturation.

[1]  Bjørn Ursin,et al.  Porosity and permeability prediction from wireline logs using artificial neural networks: a North Sea case study , 2001 .

[2]  T. Dahlin,et al.  A comparison of the Gauss-Newton and quasi-Newton methods in resistivity imaging inversion , 2002 .

[3]  Albert Tarantola,et al.  Neural networks and inversion of seismic data , 1994 .

[4]  Douglas W. Oldenburg,et al.  Inversion of 3-D DC resistivity data using an approximate inverse mapping , 1994 .

[5]  Evaluation of small-loop transient electromagnetic soundings to locate the Sherwood Sandstone aquifer and confining formations at well sites in the Vale of York, England , 2000 .

[6]  Mrinal K. Sen,et al.  Artificial neural networks for parameter estimation in geophysics , 2000 .

[7]  Mansor Nakhkash,et al.  Automatic detection of buried utilities and solid objects with GPR using neural networks and pattern recognition , 2000 .

[8]  Y. Sasaki.  Full 3-D inversion of electromagnetic data on PC , 2001 .

[9]  C.R.E. de Oliveira,et al.  An arbitrary geometry finite element method for multigroup neutron transport with anisotropic scattering , 1986 .

[10]  Mary M. Poulton,et al.  Location of subsurface targets in geophysical data using neural networks , 1992 .

[11]  Joseph Robert Inman,et al.  RESISTIVITY INVERSION WITH RIDGE REGRESSION , 1975 .

[12]  Mrinal K. Sen,et al.  Automatic NMO correction and velocity estimation by a feedforward neural network , 1998 .

[13]  Christopher C. Pain,et al.  Anisotropic resistivity inversion , 2003 .

[14]  D. Oldenburg,et al.  Inversion of induced polarization data , 1994 .

[15]  G. El-Qady,et al.  Inversion of DC resistivity data using neural networks , 2001 .

[16]  T. Mogi,et al.  ANN reconstruction of geoelectrical parameters of the Minou fault zone by scalar CSAMT data , 2002 .

[17]  W. A. Sandham,et al.  3D horizon tracking using artificial neural networks , 1996 .

[18]  S. Constable,et al.  Occam's inversion to generate smooth, two-dimensional models from magnetotelluric data , 1990 .

[19]  David Beamish,et al.  Quantitative 2D VLF data interpretation , 2000 .

[20]  Mirko van der Baan,et al.  Neural networks in geophysical applications , 2000 .

[21]  R. Gerhard Pratt,et al.  Seismic Amplitude Inversion for Interface Geometry of Multi-layered Structures , 2000 .

[22]  Michael E. Murat,et al.  AUTOMATED FIRST ARRIVAL PICKING: A NEURAL NETWORK APPROACH1 , 1992 .

[23]  G. Wadge,et al.  Inferring the lithology of borehole rocks by applying neural network classifiers to downhole logs: an example from the Ocean Drilling Program , 1999 .

[24]  K. V. Paulson,et al.  Magnetotelluric inversion using regularized Hopfield neural networks , 1997 .

[25]  Subrata Chakraborty,et al.  Velocity inversion in cross-hole seismic tomography bycounter-propagation neural network, genetic algorithmand evolutionary programming techniques , 1999 .