Project Details
Projekt Print View

Optimal Soundfield Generation for Sound Reinforcement in Time and Frequency Domain

Subject Area Acoustics
Fluid Mechanics
Term from 2018 to 2023
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 393106680
 
Final Report Year 2024

Final Report Abstract

Today, large-scale sound reinforcement systems for speech and music are usually realized with loudspeaker array systems, which have to be optimized regarding different degrees of freedom (distribution/geometry/housing design of the drivers) and different boundary conditions (geometry of the auditorium and the areas to be avoided, available acoustic power) concerning an optimal control of the array. This is an ill-posed inverse problem in the mathematical sense, where the number of sources is always significantly smaller than the number of receivers. In this project, a new access to the solution to this problem was developed by (a) further developing traditional approaches for the optimization of the radiation of array systems in the frequency domain by a multi-objective-goal-attainment approach and (b) implementing for the first time an optimization in the time domain with the adjoint-based method, which has proven successful in analogous problems in fluid mechanics. Thus, for the first time, the influence of a base flow on the generated sound field can be taken into account, which can be of great practical relevance - for example, when wind influences the speech intelligibility in stadiums. While a theorybased presetting of the geometric driver arrangement (Polygonal Audience Line Curving, PALC) was developed for traditional optimization in the frequency domain, the adjoint-based method has proven to be suitable for finding both suitable positions and driver functions of the loudspeakers for a target sound field. Not only the directional characteristics of the drivers, which are often complex in practice, can be modeled precisely, but also base flows can be taken into account. While the influence of the room was initially only taken into account for rigid boundary conditions, impedance-boundary conditions are also to be modeled and optimized in a follow-up project, so it will not only be possible to optimize the sound sources in the room but also the room acoustic conditions themselves.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung