Solving the acoustic VTI wave equation using physics-informed neural network (PINN) (Song et al., 2020)

  1. Applying a target-oriented strategy through seismic redatuming to reduce the computational cost by focusing our inversion on a target zone.

  2. Incorporating well information to the time-lapse inversion by regularization. The required prior model is predicted from a trained deep neural network (DNN), which identifies the statistical connection between an initial inverted model and facies interpreted from well logs.


References

Song C., Alkhalifah T., and Waheed Ub., (2020), "Solving the acoustic VTI wave equation using physics-informed neural networks", Geophysical Journal International: submitted.