Application of Soft Computing Techniques in Prediction of Road Traffic Noise Levels

Abstract

In order to control noise pollution it is necessary to have a suitable calculation method for traffic noise prediction. Since 1950s many mathematical models for estimation of road traffic noise levels have been developed, and most of the available models are based on regression analysis of experimental data. This paper presents the application of soft computing techniques in traffic noise prediction. Two models for prediction of equivalent A-weighted level of road traffic noise are presented and their predictions are compared to experimental data collected by traffic noise monitoring in the urban environment, as well as to predictions of commonly used traffic noise models. The results obtained by statistical analysis of differences between the measured and the calculated noise levels show that the application of neural networks and optimization methods based on swarm intelligence and evolutionary algorithms may improve process of development, as well as accuracy of traffic noise prediction models.