Estimation of Parameter Variability for High Dimensional Microwave Problems via Partial Least Squares

This paper deals with the uncertainty quantification applied to the analysis of Wireless Power Transfer (WPT) efficiency. It considers the development of a surrogate model based on Partial Least Squares (PLS) Regression, in order to estimate statistical quantities of an output with relatively low computational cost in comparison with Monte Carlo (MC) simulation. The method is extendable to other microwave problems as well.