Level set method for the inverse elliptic problem in nonlinear electromagnetism

An inverse problem of inhomogeneity identification inside a nonlinear magnetic material from the local measurements of the magnetic induction is investigated. The representation of the shape of the inhomogeneity and its evolution during an iterative reconstruction process is achieved by the level set method. The reconstruction is realized by the minimization of a cost function using the steepest descent method. The gradient directions are evaluated using the sensitivity equation and the adjoint variable method. Simulations has been performed showing the robustness of the algorithm and its ability to reconstruct single inhomogeneities, convex and non-convex, as well as multiple inhomogeneities.