Direct, nonlinear inversion algorithm for hyperbolic problems via projection-based model reduction

TL;DR

This paper introduces a nonlinear inversion algorithm for hyperbolic problems that uses projection-based model reduction and snapshot orthogonalization to estimate wave speed from boundary data.

Contribution

The method employs a data-driven Gram-Schmidt process on snapshots to suppress reflections and reconstruct wave speed with a novel grid approximation approach.

Findings

Effective in 1D and 2D synthetic models

Suppresses multiple reflections via orthogonalization

Provides accurate wave speed estimates

Abstract

We estimate the wave speed in the acoustic wave equation from boundary measurements by constructing a reduced-order model (ROM) matching discrete time-domain data. The state-variable representation of the ROM can be equivalently viewed as a Galerkin projection onto the Krylov subspace spanned by the snapshots of the time-domain solution. The success of our algorithm hinges on the data-driven Gram--Schmidt orthogonalization of the snapshots that suppresses multiple reflections and can be viewed as a discrete form of the Marchenko--Gel'fand--Levitan--Krein algorithm. In particular, the orthogonalized snapshots are localized functions, the (squared) norms of which are essentially weighted averages of the wave speed. The centers of mass of the squared orthogonalized snapshots provide us with the grid on which we reconstruct the velocity. This grid is weakly dependent on the wave speed in traveltime coordinates, so the grid points may be approximated by the centers of mass of the analogous set of squared orthogonalized snapshots generated by a known reference velocity. We present results of inversion experiments for one- and two-dimensional synthetic models.