A hybrid numerical-experimental analysis for tire air-pumping noise with application to pattern optimization

As a vehicle travels at moderate/high speeds, tire noise becomes dominant over other sources of vehicle noise. It is a challenging task to determine the spectral characteristics of the tire noise source and even much more challenging to predict the tire pattern noise. This paper proposes a hybrid experimental-numerical method to identify the tire air-pumping noise spectrum and to predict the pattern noise. A special designed test and cut tire are used to obtain the near field tire noise spectrum while the boundary element method (BEM) is used to determine the transfer characteristics between noise source and selected field points. Then, the spectral characteristics of the air pumping source can be identified by combined use of the numerical transfer characteristics and the measured noise of the cut tire by solving the Helmholtz equation. In the end, the proposed approach has been developed into an in-house software and practical tire pattern noise has been investigated successfully. The established model is able to predict the noise at any field point and the good agreement between the predicted results and tests can be found, demonstrating the reasonability of the proposed approach. In addition, an optimization method of tire tread block shift and mold shift, i.e. moving one of the two half molds and/or tread ribs along the circumferential direction to change their relative phases, are proposed based on the developed software. A practical example demonstrates that the proposed approach and in-house code can be used to predict and optimize tire tread pattern noise.