Industrial Methodologies for the Prediction of Interior Noise Inside Railway Vehicles: Airborne and Structure Borne Transmission

This paper presents computational methods that are used by rolling stock manufacturers to predict noise inside vehicles. For airborne transmission, which dominates in the medium-high frequency range, a four-step procedure is applied: source description by their emitted sound power level, propagation of noise to the train’s exterior surface, panel transmission loss and acoustic response of the interior cavity. Reasonable agreement between computations and measurements is usually obtained, and the method makes it possible to rank the different source contributions and airborne transmission paths. Structure borne noise dominates in low frequencies. Finite Element models are used to improve car body design (dynamic stiffness at input points and carbody vibroacoustic transfers), but they do not cover the whole problem since the modelling of excitation from the bogie is not included. Recent research allowing the computation of blocked forces at car body input points and starting with wheel/rail interaction is briefly presented. Concerning source modelling, a focus is made on traction noise, including electromagnetic excitations in electric motors and mechanical excitations due to the meshing process inside gearboxes. Efficient computational methods and validation examples are presented. The coupling of these methods with optimization methods has great potential for improvement of motor noise and vibration design.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01874963
  • Record Type: Publication
  • ISBN: 9783030702885
  • Files: TRIS
  • Created Date: Feb 27 2023 9:33AM