Macromodels for large scale electromagnetic simulation in the time-domain

Research Staff: Vladimir Podlozny, Christos Christopoulos and John Paul

This work is supported by The Engineering and Physical Sciences Research Council (EPSRC), UK

We have developed a universal and effective technique for fine features characterization in 3D field modelling methods.

Fine features characterization is a permanent problem in 3D field modelling. Thin panels, perforated screens, composite plates and perfectly conductive boards require a very fine mesh for their description. Frequently, fine spatial discretization is very expensive or practically impossible. We offer an effective alternative way for the large-scale description of fine features.

The technique is based on the characterization of scattering data for fine features obtained in the frequency-domain by analytical, experimental or numerical results. The choice of the data extraction method depends on availability and convenience. · Applying the Prony method results in scattering functions, which in turn are transformed into the discrete time domain using the z-transform. The resulting filters are implemented by state–space methods. Filter parameters can be stored in a library and inserted into the 3-D mesh as needed. The approach has been used by us in electromagnetic compatibility studies to predict shielding effectiveness of the enclosure with perforated wall and enclosure populated with perfectly conductive boards (PCBs). A number of tests showed close agreement between experimental and model predictions. The numerical results have been obtained at a fraction of the computational costs inherent in conventional methods. It is estimated that in 3-D problems, the developed method results in a reduction of memory and run-time requirements by factors of 10 and 100, respectively.

We have used the approach in connection with the TLM field modeling method. However, our technique can be adapted to work with other techniques such as FDTD.

The advantages of the approach are:

   Accuracy small number of parameters after pre-processing

   Possibility of extracting parameters from raw frequency-domain data

   Low computational costs compared to the standard models

   Mesh resolution determined by global considerations and not by the fine features

Publications

[1] J. Paul, V. Podlozny, D.W.P. Tomas, C. Christopoulos, Time-Domain Simulation of Thin Material Boundaries and Thin Panels Using Digital Filters in TLM, Elektrik (Turkish Journal of Electrical Engineering and Computer Science), Vol. 10, N.2, 2002., pp. 185-198.
[2] V. Podlozny, J. Paul, and C. Christopoulos, Efficient Description of Fine Features Using Digital Filters in Time-Domain Computational Electromagnetics, Science, Measurement and Technology, IEE Proceedings, Volume: 149 Issue: 5 , Sep 2002 Pages: 254 –257
[3] J. Paul, V. Podlozny and C. Christopoulos, The Use of Digital Filtering Techniques for the Simulation of Fine Features in EMC Problems Solved in the Time­Domain, IEEE Transactions on Electromagnetic Compatibility, Volume: 45 Issue: 2 , May 2003 Pages: 238 -244 .
[4] V. Podlozny, J. Paul, and C. Christopoulos, Efficient Description of Fine Features Using Digital Filters in Time-Domain Computational Electromagnetics, Proc. The Fourth International Conference on Computation in Electromagnetics – CEM 2002, 8-11 April 2002, Bournemouth, UK, IEE Conf. Digest 02/063, Section 5.
[5] V. Podlozny, J. Paul, and C. Christopoulos, Efficient Calculation of the Shielding Effectiveness of Equipment Cabinets in Full-Field Numerical Models , International Symposium on Electromagnetic Compatibility – EMC Europe 2002,Vol.2, pp.853-857.