Design Optimization of Acoustic Structures Using Geometry Parameterization and Particle Swam Optimization Algorithm

Li C, Cao H. Design Optimization of Acoustic Structures Using Geometry Parameterization and Particle Swam Optimization Algorithm. In: AES Convention 153. Audio Engineering Society; 2022. Paper 3. Available from: https://aes.org/publications/elibrary-page/?id=21911

@inproceedings{Li2022_21911,
  author = {Li, Chaopen and Cao, Huixain},
  title = {{Design Optimization of Acoustic Structures Using Geometry Parameterization and Particle Swam Optimization Algorithm}},
  booktitle = {AES Convention 153},
  note = {Paper 3},
  year = {2022},
  month = oct,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=21911}
}

TY  - CPAPER
TI  - Design Optimization of Acoustic Structures Using Geometry Parameterization and Particle Swam Optimization Algorithm
AU  - Li, Chaopen
AU  - Cao, Huixain
T2  - AES Convention 153
M1  - Paper 3
PY  - 2022
DA  - 2022/10/06
UR  - https://aes.org/publications/elibrary-page/?id=21911
PB  - Audio Engineering Society
LA  - en
AB  - The geometric design of acoustic components, such as horns and surroundings, is time-consuming and complicated. A design process is introduced in this study to shorten the design-test-optimization cycle. The design process consists of three parts. The first part is the geometry parameterization. Formulas to build the shape of the geometry are provided, from a 2D profile to a 3D geometry. In the second part, the acoustic properties of the acoustic components are simulated; for example, the sound pressure level (SPL), which is advised in most acoustic component designs. Optimization is performed in the third part. Using the particle swarm optimization (PSO) algorithm, the geometric parameters are modified to improve the acoustic properties. We consider the horn as an example in this study. However, the parametric geometry and optimization method above are highly flexible and can be performed in diverse structures, in which the derivative of the surface is expected to be continuous. The entire process is assembled in one software, which makes the design program easy to handle and highly efficient.
ER  -