Correlation matrices in kriging-assisted optimisation of electromagnetic devices

Kriging surrogate modelling offers efficient decision making assistance on where to place the next point for evaluation during optimisation. This is particularly helpful in the design of electromagnetic devices where computationally expensive finite element modelling needs to be used. The disadvantage, however, is that correlation matrices are required which, for problems with many design variables and multiple objectives, may grow in size leading to the need for page swapping and slowing down of what in principle should be a very fast process. This presentation addresses the issue of how to exploit the structure of these matrices to reduce computational burden.