The Use of Machine Learning Algorithms in the Simulation of Multi-Layer Acoustic Palliatives

2024-01-2928

06/12/2024

Event
13th International Styrian Noise, Vibration & Harshness Congress: The European Automotive Noise Conference
Authors Abstract
Content
As palliative acoustic material mixtures and compositions become more complex, the ability to accurately simulate their acoustic performance within an installed NVH component is becoming increasingly difficult. Historically, Biot parameters and their associated TMM models have been used to simulate the acoustic performance of multi-layered material compositions. However, these simulations are not able to account for real-world complexities such as manufacturing imperfections or inter-layer gluing effects. The assumptions made by simulation models, such as the perfectly diffuse field, are rarely true in actual measurements, let alone in the vehicle, further increasing the uncertainty when comparing measurement versus simulation.
There already exists widely accepted methods for obtaining Biot parameters for single-layer materials. Typically, a multi-layer simulation considers each individual layer in isolation rather than its interactions with the rest of the composition after heating, compression, or gluing. The current trend towards sustainability is also adding restrictions to the types of materials that can be used. Target compliance for NVH components includes acoustic parameters and environmental impact, increasing the effort required for component quotation.
This paper examines four possible approaches used to satisfy an OEM’s quotation request which range from flat samples to fully built vehicle systems. It successfully examines the suitability of bespoke machine learning algorithms combined with large measurement and simulation databases.
Meta TagsDetails
DOI
https://doi.org/10.4271/2024-01-2928
Pages
6
Citation
Harry, E., Morris-Kirby, R., Caponio, E., and Hoang, M., "The Use of Machine Learning Algorithms in the Simulation of Multi-Layer Acoustic Palliatives," SAE Technical Paper 2024-01-2928, 2024, https://doi.org/10.4271/2024-01-2928.
Additional Details
Publisher
Published
Jun 12
Product Code
2024-01-2928
Content Type
Technical Paper
Language
English