Low-Frequency Sound Absorption Mechanism and Bidirectional Prediction of a Viscoelastic Rubber-Based Underwater...

Area of Science:

• Acoustics
• Materials Science
• Computational Modeling

Background:

• Underwater acoustic coatings are crucial for noise reduction and sonar stealth in marine applications.
• Existing models often lack accuracy in predicting low-frequency sound absorption in complex multilayered structures.

Purpose of the Study:

• To develop and validate a novel Micro-Perforated Panel Acoustic Coating Layer (MPPACL) model for underwater applications.
• To investigate the low-frequency sound absorption mechanisms of viscoelastic rubber and micro-perforated panel structures.
• To establish an efficient predictive modeling framework for acoustic coating design.

Main Methods:

• Development of the MPPACL model based on an improved transfer function method.
• Investigation of viscoelastic material properties (damping, complex modulus).
• Construction of an ensemble learning-based deep neural network (ELDNN) for forward and inverse predictions.

Main Results:

• The MPPACL model accurately describes coupled acoustic behavior in multilayer coatings.
• Low-frequency sound absorption is enhanced by viscoelastic properties and MPP resonance.
• ELDNN achieved 3.7 times higher frequency prediction accuracy than DNN.
• Significant sound absorption enhancement observed between 50-2000 Hz, experimentally verified.

Conclusions:

• The proposed MPPACL model and ELDNN framework offer an efficient and reliable approach for designing and optimizing underwater acoustic coatings.
• This research advances the understanding and application of acoustic metamaterials in marine environments.