Michael V. Klibanov,Jingzhi Li,Vladimir G. Romanov,Zhipeng Yang
Abstract:The travel time tomography problem is a coefficient inverse problem for the eikonal equation. This problem has well known applications in seismic. The eikonal equation is considered here in the circular cylinder, where point sources run along its axis and measurements of travel times are conductes on the whole surface of this cylinder. A new version of the globally convergent convexification numerical method for this problem is developed. Results of numerical studies are presented.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the three - dimensional travel time tomography problem (3D Travel Time Tomography Problem, TTTP). Specifically, the author focuses on the inverse problem of solving the coefficients of the eikonal equation in the cylindrical coordinate system. This problem has important applications in seismology, especially in down - hole seismic imaging.
### Specific description of the problem
1. **Background and motivation**:
- The travel time tomography problem (TTTP) is a long - standing problem and is widely used in the seismic study of the earth.
- The eikonal equation describes the propagation time of waves in an inhomogeneous medium, where the wave velocity \( c(x)=\frac{1}{n(x)} \), and \( n(x) \) is the refractive index.
- In traditional research, it is usually assumed that the observation area is a vertically elongated rectangular prism, and point sources are distributed along a straight line below the horizontal plane, which limits the imaging of only horizontal - direction anomalies.
2. **Objectives of the new method**:
- This paper proposes a new globally convergent convexification numerical method for solving 3D TTTP.
- The new method considers a vertical cylinder, with point sources distributed along its axis, and the measurement data are obtained on the side surface, top, and bottom boundaries of the cylinder.
- This configuration can better image anomalies in different directions, not just horizontal - direction anomalies.
3. **Mathematical model and theoretical contributions**:
- The author first proves the monotonicity theorem of the solution of the eikonal equation with respect to the radius of the cylindrical coordinate system.
- Then, based on this theorem, a convexification method adapted to the new source/detector configuration is proposed.
- Next, the author derives an integro - differential equation without unknown coefficients and introduces some important theoretical results, such as Carleman estimates.
4. **Numerical experiments**:
- Finally, the author conducts numerical experiments to verify the effectiveness and global convergence of the new method.
### Formula representation
- Eikonal equation:
\[
(\nabla_x\tau)^2 = n^2(x)
\]
where \( \tau(x, x_0) \) is the propagation time from the source point \( x_0 \) to the observation point \( x \).
- Eikonal equation in the cylindrical coordinate system:
\[
(\tau_r)^2+\frac{1}{r^2}(\tau_\phi)^2+(\tau_z)^2 = n^2(r,\phi,z)
\]
- Convexification functional:
\[
J_\lambda(V_h)=\left\|A_N V_h + S(V_{h\phi}, V_{hz}, r)\right\|^2_{L^2_h,N(D_h)}
\]
where \( e^{2\lambda r} \) is the Carleman weight function.
Through these formulas and theoretical analysis, the author shows how to effectively solve 3D TTTP and provides numerical experiment results to verify the effectiveness of the method.