Project Tyre Road Noise
Data-based study of effects on controlled and real drive noise emission
Data-based study of effects on controlled and real drive noise emissionCopyright: © KIT, Institute of Vehicle System Technology
Road traffic noise impairs the quality of life of many citizens. With electric vehicles, the noise is mainly caused by the contact of the tires with the road surface due to the low engine noise. For the approval of vehicles, the noise level is measured between microphones while driving on a specially asphalted road surface. The transferability to real driving situations is limited, as real roads do not correspond to the test environment. To protect the population, the limit values for noise emissions are being gradually reduced. Effective measures must therefore be developed in order to comply with the new limits. This requires large amounts of measurement data on noise emissions on real roads, which can only be collected at great expense in the conventional test environment.
The aim of the project is to generate a large database from extensive pass-by noise measurements to enable the analysis of influencing factors (e.g. road surface geometry, tire profile, environmental conditions, ...) and the derivation of noise reduction measures. In addition, the conversion of noise emissions from measurement results to other test conditions is to be improved. As a result, more operating conditions can be investigated and optimized with less measurement effort.
A cross-manufacturer vehicle fleet records the pass-by noise in different environments using the simplest possible measurement methods and data on influencing parameters using a large number of other sensors. Artificial intelligence (AI) methods are used to create a model for predicting the pass-by noise, which can be used to determine the relevant influencing variables on the noise emission of the tire-road noise. The necessary measurement technology will be developed together with the required AI models in the project and validated by measurements in controlled environments. The developed models and measurement data will be published in a public database after the end of the project in a quality-assured manner.