Research On Microwave Non-Destructive Testing Theory And Its Inverse Scattering Imaging Algorithm

Microwave imaging can get the information about the scattering field around imaging region, and achieve the purpose of the complex dielectric constant image of objects reconstruction. That is an inverse electromagnetic scattering of imaging technology. This article is major on how to obtain the scattering information arising from the incident plane wave in the imaging region which contains media columns and how to use such scattering information realized inverse electromagnetic scattering imaging by using genetic algorithm.Electromagnetic Scattering is on the basis of microwave imaging algorithm. By discussing the electromagnetic scattering basic theory, comparing and researching several kinds of the current widely used numerical methods of electromagnetic field, choose the finite element method which has highlighted advantages in the analysis of electromagnetic field issues to solve the positive problem of inverse scattering imaging. In accordance with the basic idea of the finite element method, combining the large-scale finite element software ANSYS's own language APDL, develops process simulation analysis of the imaging region with media column arising from the incident plane wave. Access to relevant information about the scattering phenomenon of different conditions, and provide a positive analysis tool for the later imaging algorithm. At the same time, in order to allow more users to use the ANSYS simulation result, use MATLAB's GUI tool by calling the ANSYS to achieve the 3D high-frequency electromagnetic field simulation software design with the friendly man-machine interface.Microwave imaging, or electromagnetic inverse scattering imaging is actually a very non-linear and ill-posed problem. Through studying and comparing several current broad and representative imaging algorithm, it also shows that the algorithm for solving the inverse problem directly is not possible to get more high-quality imaging algorithm. But the principle of the genetic algorithm as follows: assume the original distribution of dielectric scatters, calculate the electromagnetic scattering problem to achieve scattering data, compare with the measured data to determine whether is the end of the iteration, and find out the best conditions to meet the known scattering body structure. In this way, not only can avoid solving the ill-conditioned equations directly, and also can easily compare with a variety of prior information by adding conditions, which good for improving the imaging effects. Therefore using MATLAB develops the imaging procedure, which positive scattering numerical calculation is based on the finite element method and inverse scattering iterative part is based on the genetic algorithm. Using this method for the 8×8 unit imaging region in free space does single-column' and dual-medium columns' experiments. The imaging results error is relatively small and it verifies the effectiveness and superiority of this algorithm.