Predicting Reservoir Petrophysical Geobodies from Seismic Data Using Enhanced Extended Elastic Impedance Inversion

The study aims to implement a high-resolution Extended Elastic Impedance (EEI) inversion estimate the petrophysical properties (e.g., porosity, saturation and volume of shale) from seismic well log data. resolves pitfall basic EEI in inverting below-tuning resolution, dimensionality absolute value are improved by employing stochastic perturbation constrained integrated energy spectra attribute Bayesian Markov Chain Monte Carlo framework. A general regression neural network (GRNN) is trained learn memorize relationship between stochastically perturbed associated GRNN then used predict any given processed EEI. proposed was successfully conducted invert shale, porosity water 4.0 m thick gas sand reservoir Sarawak Basin, Malaysia. three geobodies were built using discovery wells cut-off values, showing that inverted satisfactorily reconstruct logs with sufficient resolution an accurate at site laterally conformable Inversion provides reliable prediction potentially helps further development for field.