Fuel Cell Stack Magnetic Tomography with Adjoint Method

To diagnose Fuel Cell stacks (FC-stacks) and aug-ment their durability, this paper presents a novel approach with an inverse algorithm based on magnetic tomography. The goal is to determine internal local resistivities of the stack from external magnetic measurements in order to provide a diagnosis. The inverse problem is solved by minimizing the difference between simulated and measured magnetic field. Sensitivities are computed with the adjoint state method and first numerical results are presented.