Topology Optimization of Broadband Acoustic Transition Section: A Comparison between Deterministic and Stochastic Approaches

TL;DR

This paper compares deterministic and stochastic topology optimization methods for designing broadband acoustic transition sections, demonstrating stochastic approaches' advantages in efficiency and performance in acoustic wave transmission.

Contribution

It introduces a comparative analysis of deterministic and stochastic gradient-based topology optimization methods for broadband acoustic devices.

Findings

Stochastic methods achieve better broadband transmission performance.

Stochastic approaches reduce computational effort.

Designs optimized with stochastic methods show improved practicality.

Abstract

This paper focuses on the topology optimization of a broadband acoustic transition section that connects two cylindrical waveguides with different radii. The primary objective is to design a transition section such that it maximizes the transmission of a planar acoustic wave while ensuring the planarity of the transmitted wave. Helmholtz equation is used to model linear wave propagation in the device.We utilize the finite element method to solve the state equation on a structured mesh of square elements. Subsequently, a material distribution topology optimization problem is formulated to optimize the distribution of sound-hard material in the transition section. We employ two different gradient-based approaches to solve the optimization problem: namely, a deterministic approach using the method of moving asymptotes (MMA), and a stochastic approach utilizing both stochastic gradient (SG) and continuous stochastic gradient (CSG) methods. A comparative analysis is provided among these methodologies concerning the design feasibility and the transmission performance of the optimized designs, and the computational efficiency. The outcomes highlight the effectiveness of stochastic techniques in achieving enhanced broadband acoustic performance with reduced computational demands and improved design practicality. The insights from this investigation demonstrate the potential of stochastic approaches in acoustic applications, especially when broadband acoustic performance is desired.