Comparison of techniques for uncertainty quantification of superconducting radio frequency cavities

This study examines the influence of uncertainty in the geometry parameters on the properties of superconducting cavities considering the 1.3 GHz TESLA mid-cell cavity. Monte Carlo simulation as well as uni-variate and multi-variate generalized Polynomial Chaos expansions are applied to estimate these stochastic properties. These stochastic methods are compared based on their computational efficiency and suitability for the uncertainty quantification in the given computational model. For this purpose, crucial geometry parameters are identified and used to reduce the random model parameter space. A central result comprises that the generalised Polynomial Chaos approach is highly applicable for the quantification of uncertainty in single-cell cavities.