Uncertainty Quantification of the Crosstalk in Multiconductor Transmission Lines via Degree Adaptive Stochastic Response Surface Method

The degree adaptive stochastic response surface method is applied to analyze statistically the crosstalk in multiconductor transmission lines (MTLs). The coefficient of polynomial chaos expansion (PCE) is obtained based on the least angle regression. The truncation degree of PCE is iterated using the degree adaptive truncation algorithm, and the optimal proxy model of the crosstalk of the original MTLs that satisfies the actual error requirements is calculated. The statistical properties of crosstalk in MTLs (such as mean, standard deviation, skewness, kurtosis, and probability density distribution) are obtained. The failure probability of the electromagnetic compatibility in the MTLs system is considered. The global sensitivity indices of crosstalk-related factors are analyzed. Finally, the proposed method is proved to be effective compared with the conventional Monte Carlo method. The uncertainty quantification of crosstalk in MTLs can be calculated efficiently and accurately.

Radiated Susceptibility Analysis of Multiconductor Transmission Lines Based on Polynomial Chaos

To address the uncertainties of the radiated susceptibility of multiconductor transmission lines (MTLs), a surrogate model of the MTLs radiated susceptibility is established based on generalized polynomial chaos (gPC), and the gPC is made sparser by combining the adaptive hyperbolic truncation (AHT) scheme and the least angle regression (LAR) method. The uncertainties of the radiated susceptibility of transmission lines are calculated using the adaptive-sparse polynomial chaos (AS-PC) scheme. The parameters related to the incident field, such as elevation angle theta, azimuth angle psi, polarization angle eta, and field amplitude E, are inevitably random. Therefore, these four variables are taken as random input variables, and each of them is subject to different variable distributions. The MTLs model with infinite ground as the reference conductor is adopted, different impedances are used and the AS-PC scheme is combined with transmission line theory to calculate the average, standard deviation and probability distribution of the radiated susceptibility of MTLs. Sobol global sensitivity analysis based on variance decomposition is adopted to calculate the influence of random input variables on the MTLs radiated susceptibility model. The calculation results are compared with the results of the Monte Carlo (MC) method, proving that the proposed method is correct and feasible.