Regularized Inverse Methods for Tire Noise Source Identification and Pass-by Noise Estimation in Electric Vehicles Autoneum

Format:

Book

Conference/Event

Author/Creator:

Morin, Benjamin, author.

Di Marco, Federico, author.

Horák, Jan, author.

Lafont, Thibault, author.

Kim, Minkyu, author.

Kang, Min Kyoo, author.

Yoo, Ji Woo, author.

Conference Name:

14th International Styrian Noise, Vibration & Harshness Congress: The European Automotive Noise Conference (2026-06-17 : Graz, Austria)

Language:

English

Subjects (All):

Exterior noise.

Noise measurement.

Electric vehicles.

Battery packs.

Noise, vibration, and harshness standards and regulations.

Local Subjects:

Exterior noise.

Noise measurement.

Electric vehicles.

Battery packs.

Noise, vibration, and harshness standards and regulations.

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 2026

Summary:

Tire exterior noise has become increasingly critical in vehicle acoustics due to two key developments: updated pass-by noise regulations, which amplify the relative contribution of tire noise, and the rise of Battery Electric Vehicles (BEVs), which lack traditional powertrain noise. Design trends in BEVssuch as increased vehicle mass from battery packs and the widespread use of large-diameter, wide, low-profile tiresfurther intensify tire noise due to stiffer constructions and altered contact dynamics. A common method for predicting tire noise is the source-transfer-receiver model, where the tire is represented by a set of monopoles with volume velocity Q derived from near-field measurements. Acoustic propagation is modeled via p/Q transfer functions. Despite its simplifications, this approach is practical for vehicle development, enabling clear separation between source and transfer mechanisms and facilitating targeted noise control strategies. In previous work, we proposed a rigorous framework to optimize both the spatial distribution and strength of the monopole sources. Positions were identified using an L1-norm regularization via the Lasso algorithm, promoting sparsity and physical interpretability. Strengths were estimated using an L2-norm Tikhonov regularization, which stabilizes the solution against measurement noise. While the Tikhonov regularization parameter was previously tuned manually through trial and error, we now enhance predictive accuracy by selecting it via a cross-validation technique, ensuring a more robust and data-driven optimization. Besides this, compared to the previous work the approach here is validated for the prediction of both indoor and outdoor pass-by noise, as well as for multiple tire types providing different noise levels. Results demonstrate the method's robustness, accuracy, and applicability for acoustic development in modern vehicle platforms

Notes:

Vendor supplied data

Access Restriction:

Restricted for use by site license