Abstract

The precise litho-facies change from well log records is a complex non-linear problem in geophysical data processing. Recorded well log signals are a complex superposition of non-stationary/non-linear signals of varying wavelengths and frequencies, shaped by the heterogeneous composition and structural variation of rock types in the earth. This weakens our ability to use traditional statistical techniques, which either fail in most cases to discriminate and/or, at best, do not precisely extract facies changes from such a complex well log signals. We propose here a new method, set in a Bayesian neural network (BNN) framework and employing a powerful Hybrid Monte Carlo (HMC)/Markov Chain Monte Carlo (MCMC) simulation scheme to identify facies changes, from complex well log data.