Electromagnetic characterisation of structures using Krylov subspace model order reduction method

The analysis of electromagnetic (EM) emissions from Information Technology Equipment (ITE) using numerical modelling techniques is presented. The modelling has been developed to predict the characteristics of greatest relevance to the IT industry; particularly to identify elements in equipment design that might lead to non-compliance with the prevailing EM standards. Electromagnetic emissions and susceptibility are an operational by-product of IT equipment. Typical high performance electronic equipment is modelled in a mathematical domain designed to be representative of the measurement environment. The mathematical modelling presented within this thesis extends beyond the current requirements for EM measurement and compliance, anticipating future changes to EM standards and predicting performance for future changes in technology. Comparisons between different numerical modelling approaches are made. This is further assessed in the context of the environment to be modelled. Microwave structure characterisation is achieved by application of the Lanczos model order reduction (MOR) technique to the Finite Difference Time Domain (FDTD) method. The eigenvalues of a linear model are computed recursively such that the model output converges to the transfer function of the original system. The extraction of eigenvalues is rapid and provides significant savings in computational resource. The modelling improves the simulation performance and provides a scalable platform for future developments. The Lanczos method for model reduction offers significant performance benefits for the algorithms chosen. This offers several orders of magnitude reduction in processing time when compared to more conventional methods of numerical modelling. In addition, weaknesses in the FDTD method are overcome. The FDTD and model reduction methods are combined in their use as a tool to investigate EM performance. Conclusions are drawn regarding the effectiveness of the approach taken.