Options
2020
Conference Paper
Titel
Stochastic identification of parametric reduced order models of printed circuit boards
Abstract
The digital design of power electronic components is a key to accelerate the development process and to control development risks for electrically driven vehicles. Due to manufacturing, structural dynamic and vibroacoustic behavior of the same part may differ significantly. The investigations within this paper initially focus on the experimental analysis of printed circuit boards (PCBs). To this end, seven samples of a test PCB are examined. The PCBs are excited with a loudspeaker in the frequency range from 50 Hz to 8000 Hz and the modal parameters are determined by an operational modal analysis. The identified modal parameters are used to update the numerical simulation. Due to the complexity of the model and the large number of FE re-analyses, parametric model reduction is used to increase computational efficiency. After adjusting the model parameters, the simulation and measurement results present good agreement levels and thus form a basis for further investigation s.