Physics-Consistent Data-Driven Waveform Inversion With Adaptive Data Augmentation

Seismic full-waveform inversion (FWI) is a nonlinear computational imaging technique that can provide detailed estimates of subsurface geophysical properties. Solving the FWI problem be challenging due to its ill-posedness and high cost. In this work, we develop new hybrid approach solve combines physics-based models with data-driven methodologies. particular, data augmentation strategy not only improve representativity training set but also incorporate important governing physics into process therefore accuracy. To validate performance, apply our method synthetic elastic seismic waveform generated from geologic model built on carbon sequestration site at Kimberlina, California. We compare physics-consistent both purely approaches observe yields higher accuracy greater generalization ability.