Global Sensitivity Analysis and Uncertainty Quantification of Radiated Susceptibility in PCB Using Nonintrusive Polynomial Chaos Expansions
Tarek Bdour, Alain Reineix
Abstract
Tarek Bdour, Alain Reineix
Abstract
In this paper, we introduce a workflow that uses uncertainty quantification (UQ) and sensitivity analysis (SA) in conjunction with a metamodeling technique called polynomial chaos expansion (PCE) to investigate the stochastic terminal response of a printed circuit board (PCB) due to an external plane wave excitation. The objective of PCE use is to reduce the computational burden of commonly used Monte Carlo (MC) approach for calculating statistical moments and Sobol' sensitivity indices in the presence of uncertain geometrical and electrical parameters of the investigated PCB. The statistical results computed by PCE has been shown to be very accurate compared to MC simulation results.
OpenAlex reports 23 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper, we introduce a workflow that uses uncertainty quantification (UQ) and sensitivity analysis (SA) in conjunction with a metamodeling technique called polynomial chaos expansion (PCE) to investigate the stochastic terminal response of a printed circuit board (PCB) due to an external plane wave excitation. The objective of PCE use is to reduce the computational burden of commonly used Monte Carlo (MC) approach for calculating statistical moments and Sobol' sensitivity indices in the presence of uncertain geometrical and electrical parameters of the investigated PCB. The statistical results computed by PCE has been shown to be very accurate compared to MC simulation results.
Key concepts: Sobol sequence, Polynomial chaos, Uncertainty quantification, Sensitivity (control systems), Monte Carlo method, Printed circuit board, Polynomial, Metamodeling