Environment
Reducing the noise from steel bridges
Innovative solutions for noise reduction
The issues
There are 900 steel railway bridges in France, and every one of them is a source of noise. This is especially a problem in urban areas, where the noise is an annoyance to people living near them.
Réseau Ferré de France (RFF), the owner of the railway infrastructures, wants to have a methodology that will identify the causes of the noise and indicate the action that should be taken. It therefore commissioned a research project on reducing noise from steel bridges to find out what phenomena are responsible for the higher noise levels in the environment when a train goes over a steel bridge.
Methodology
The project has four phases: selection of a reference bridge; measurements and development of vibroacoustical models for forecasting noise in the environment; simulations to assess the effectiveness of noise reduction solutions; and validation of these solutions on the reference bridge.
Bridge selection: There are six types of steel bridges. A cross-braced truss bridge, one of the nosiest of such structures, was chosen for this study (specifically, the Gavignot Bridge at Enghien les Bains, near Paris).
Measurements: Measurements were made to determine the respective contributions of the bridge's components. Two models were then constructed, one a finite elements analysis for low frequencies, and the other a statistical energy analysis (SEA) for high frequencies, in order to study the full spectrum of audible frequencies.
Simulations to assess the solutions' effectiveness: The objective of these simulations was to assess how effective each of the noise reduction solutions was and to determine the right combination of low- and high-frequency solutions.
Validation of the solutions on the reference bridge: The solutions for reducing noise will be tested separately and in combination on the Gavignot Bridge in order to validate the forecasts made in the preceding phase. These tests will also allow precise quantification not only of the solutions' performance, but also of the cost and constraints of using them.
Results
The results for the reference bridge are consistent with those in the literature concerning the problem of noise from steel railway bridges. The measurements and modelling showed that the principal source of noise around 40 Hz is the bridge deck, while at higher frequencies it is the rail attached to the bridge. An optimized attach system, dynamic absorbers on the rail and/or the deck, absorbing panels were both tested by modelling and by being installed on the reference bridge. In combining several solutions, the improvement could amount 11db(A). The passing train on the bridge could be just perceptible (~3 dB(A)).
This project allowed to validate innovative solutions and define a methodology to reduce noise from steel bridges. With specific measures and some calculations; it is possible to define the optimal solutions combination to achieve a fixed noise reduction.
Contact : Franck Poisson