Uncertainty quantification of cable inductances and capacitances via mixed-fidelity models

In this paper, we investigate a mixed-fidelity approach for the uncertainty quantification of the per-unit-length (p.u.l.) capacitance and inductance of cables with random geometrical and material parameters. Polynomial chaos expansion is used to model uncertainty, whereas a numerical discretization technique is used to calculate p.u.l. inductances and capacitances. However, instead of using a model with high fidelity in both features, the results are obtained as a combination of two complementary models with mixed fidelity in each feature. Numerical examples concerning the statistical assessment of the p.u.l. inductance and capacitance matrices of two shielded cables show that similar accuracy is attained at a fraction of the computational cost compared to conventional approaches.