Comparing railway noise prediction results for passenger trains using various models

Abstract

Railway noise modeling is a requirement for all European countries. Under the European Directive on the Assessment and Management of Environmental Noise, 2002/49/EC all member states were obliged to create strategic noise maps of the major railways by June 30th, 2007. Some of the participating states have their own national schemes dedicated to noise prediction. For those not having their own noise prediction model, the recommendation is to use the Dutch SRM II Model. The most important assessment criterion is an inaccuracy value, which is defined as a difference between the result calculated based on a model and the actual measurements under the same atmospheric conditions simulated. Therefore, it is essential to check out which of the main calculation schemes produces most accurate results. Therefore, all above models were implemented numerically and calculations of the noise maps were performed with the use of: Schall 03 developed in Germany, Dutch SRM II and the Nordic model engineered in Norway. Those models were tested employing sample passenger trains data exploited in Gdansk, Poland. Noise maps were drawn in CadnaA software and then were compared to the field noise measurement test results.