Type: E13 Position
Focus: Theoretical/Numerical
Timber construction presents challenges in sound insulation, particularly in low-frequency impact noise. Traditionally, the acoustic properties of wooden floors are determined through complex and costly measurements on prototypes. This project aims to develop planning software that uses neural networks to accurately predict the airborne and impact sound levels of wooden floors based on various construction variables. The project seeks to significantly improve the prediction accuracy of planning models, especially in the low-frequency range. Ultimately, the models will be implemented in a user-friendly software application, allowing planners to conduct acoustic simulations and develop optimized floor structures. The goal is to enhance acoustic conditions, increase the acceptance of timber construction, optimize the planning process, and reduce costs.
Contact: Marcus Mäder
Room: MW1530
Phone: +49 89 289 55123
Email: Marcus.maeder@tum.de