Microwave Nondestructive Defect Identification Using Sensitivity Analysis

We propose a novel near-field approach to microwave non-destructive detection and evaluation of defects. The problem is expressed as an inverse least square problem solved using a gradient-based optimizer. The required response gradient is obtained with an adjoint-based sensitivity analysis technique developed here to handle materials with complex permittivity. It allows the recovery of the shape and the material parameters of the defect. The proposed semi-analytical sensitivity expression features improved accuracy and simplified implementation over the preceding approaches. It is verified with a practical example where we identify the parameters of a defect. The electromagnetic analysis is done with a frequency-domain solver based on the transmission line method. Its initial discretization grid is preserved throughout the optimization iterations.