PREDICTING MISSING SONIC LOGS WITH SEISMIC CONSTRAINT

Abstract

Compressional and shear sonic transit-time logs (DTC and DTS, respectively) provide important petrophysical and geomechanical information for subsurface characterization. However, they are often not acquired in all wells because of cost limitations or borehole problems. We propose a method to estimate DTC and DTS simultaneously, from other commonly acquired well logs like gamma-ray, density, and neutron porosity. Our method consists of two consecutive models to predict the sonic logs and predict the seismic traces at well locations. The model predicting the seismic traces adds a spatial constraint to the model predicting sonic logs. Our method also quantifies uncertainties of the prediction, which come from uncertainties of neural network parameters and input data. We train the network on four wells from the Poseidon dataset located on the Australian shelf, in the Browse basin. We test the network on other two wells from Browse basin. The test results show better predictions of sonic logs when we add the seismic constraint.