Adjoint-based optimization of sound reinforcement including non-uniform flow

Area of Science:

• Acoustics
• Computational Fluid Dynamics
• Signal Processing

Background:

• Optimizing loudspeaker arrays for sound reinforcement is a complex inverse problem.
• Traditional frequency-domain methods struggle with complex environmental conditions.
• Non-linear fluid dynamics are often simplified, limiting real-world applicability.

Purpose of the Study:

• To develop an innovative time-domain adjoint-based method for determining loudspeaker array driving functions.
• To account for complex environmental conditions like non-uniform flow, wind, and thermal stratification.
• To provide an alternative to conventional frequency-domain approaches.

Main Methods:

• An adjoint-based approach in the time domain was developed.
• The full non-linear Euler equations were used instead of the Helmholtz equation.
• Acoustic sources were optimized to generate a specified target sound field.

Main Results:

• The method successfully determined appropriate driving functions for both circular and linear monopole arrays.
• Environmental conditions including wind and thermal stratification were successfully incorporated for the linear array.
• The approach demonstrated effectiveness in generating target sound fields under complex scenarios.

Conclusions:

• The proposed time-domain adjoint method offers a robust solution for optimizing loudspeaker array drives.
• The use of non-linear Euler equations facilitates handling complex flow and boundary conditions.
• This method provides a powerful tool for advanced sound reinforcement applications.