EMC Simulation Based on FDTD Analysis Considering Uncertain Inputs with Arbitrary Probability Density

Stochastic Galerkin Method, a prevailing uncertainty analysis method, has been successfully used in today’s EMC simulation, in order to consider nonideality and unpredictability in actual circumstance. In this case, the inputs of the simulation are no longer certain values, but random variables with corresponding probability density distribution. This paper focuses on the arbitrary probability density cases at inputs. Two constructing orthogonal basis methods, the Wiener Haar expansion and the Stieltjes procedure, are generalized into the Stochastic Galerkin Method which is combined with the Finite Difference Time Domain analysis. With the help of the Feature Selective Validation, the quantitative precision comparison of the proposed methods in different cases (the probability density function is continuous or discontinuous) can be presented in detail.