A FDTD Study For The Typical Scene Of The Coupling Between Human And Electromagnetic Wave From Quasi-Static To S-Band

The Finite Difference Time Domain (FDTD) method is an iterative algorithm for discreting the time-space domain, and has become a significant tool for electromagnetic dosimetry evaluation. However, due to the restriction of the stable condition, it has some limitation on dealing with the qusi-static electromagneticfield (EMF), wide band (WB) and multi-reflection non-uniform EMF. How to improve FDTD’s efficiency while ensuring its accuracy is one of the research focuses in applying numerical methods to investigate the interaction between EMF and biological tissues.In order to solve these problems, this thesis proposed and validated the approximate method for qusi-static EMF, the time and frequency domain solve method for WB electromagnetic energy absorption, the reconstruction algorithm for multi-reflection electromagnetic environment using Huygens box, and the uncertainty analysis for EMF numerical computation based on polynomial chaos expansion. The main research contents and innovations are as follows:(1) To address the time consuming problem during FDTD compuatation for the quasi-static EMF, we proposed an approximate method which converted the Maxwell equations to the solve equations of quasi-static EMF, based on the approximate condition of quasi-static EMF in free space and conducting medium. Compared to the conventional FDTD method, this method was confirmed to save about360times computational time with lower than5%differences on the obtained electric field intensity. Lastly, the method was used to investigate the effects of various postures and physical contact on the electric field distribution in the infant models while exposured to the ultra-low frequency uniform magnetic field. The results indicated that the variation of the induced electric field strength caused by the different postures was lower than5%, while the significant increase of peak electric field strength appeared around the contact area between two human models. The results have been considered as a priority research issue by IEEE TC95SC6working group.(2) The time and frequency domain methods were proposed in consideration of the complicated process during evaluating the WB electromagnetic energy absorption for bio-tissues with dispersive dielectric property. These methods were used to evaluate the dosimetries for rats’ exposure to electromagnetic pulse. The results demonstrated that the approximate results with lower than5%differences were obtained by the time domain and frequency domain methods. However, the time domain method saves much computational time due to avoid the Fourier transform for the electric field when compare with the frequency domain method. We also discussed lots of technical details and parameters selection of this method which would help to formulate a standardized evaluation protocol.(3) A reverberation chamber (RC) for rodents’ exposure with a wide cover band (800MHz-3GHz), vaste work space (volume is1.25x10s cm3), multi-small-fading scenes was designed. Electric field distribution in RC was measured to validate the fading type. And the electromagnetic environment in the RC was reconstructed by the FDTD method based on Huygens box. Lastly, the rats’ exposure to the RC was realized for the dosimetry evaluation, and the Absorption Cross-Section method was utilized to validate the accuracy of numerical results. The results indicated that the difference was less than30%between the numerical and experimental results. This part of work, for the first time, provided a research tool on fading factors for EMF bio-effects, and expanded the dose-effect relationship field while only the field strength and frequency were focused in the past research. The developed equipment has been used by many high-level research institutions.(4) The organisms have many variability parameters, and it is difficult to reflect the actual scenario results through the limited numerical simulations. The polynomial chaos expansion method was introduced into the numerical computation for evaluating its uncertainty and sensitivity. We selected a cylinder with uncertain geometry and dielectric parameters as an example, and it was exposured under a plane wave EMF. The results of uncertainty indicated that the influence of specific absorption rate (SAR)’s uncertainties vary with the frequency due to the input parameters’ uncertainties. The results of sensitivity showed that the impact of SAR due to the dielectric parameters is more than4times as much as that of SAR due to the geometry parameters. This was a first study that the geometry parameters were also selected to analyze uncertainties in dosimetry. It is significant for futher research to analyze EM dosimetry uncertainties due to various human bodies.