A primary concern for NASA engineers currently is to improve the effectiveness of terahertz frequency electromagnetic inspection of the Sprayed on Foam Insulation (SOFI) of the space shuttle external tank. Current efforts have been limited to signal processing techniques such as peak detection. Therefore, much of the information contained in the return signal is neglected. An inverse problem formulation utilizing physics-based modeling to generate forward solutions enables improved characterization of flaws in the SOFI.
I will discuss our mathematical approaches to the damage detection problem using modeling techniques involving Maxwell's equations coupled with polarization. First, I will present numerical results for a gap detection inverse problem that is designed to estimate the depth and size of delamination voids. Second, I will present our efforts to account for the disruptive presence of "knit lines".