Version HTML de base
12
Journée SFA / Renault / SNCF
Acoustique
&
Techniques n° 44
surfaces only and, besides geometrical calculation, uses
statistical prolongation for the higher order of reflections
and also takes into account diffusion.
ICARE, a more recent code, is based on an adaptative
beam-tracing, each beam being constituted by three rays,
takes into account reflection on both plane and curved
surfaces, and implements diffraction by edges using
the Uniformed Theory of Diffraction (UTD). The solver
of \textsc{icare} is purely deterministic (geometrical
calculation), which implies to check the convergence
and to determine if the absorption is sufficient to avoid
statistical prolongations: it seems to be always the case
in automotive applications.
The Ray-Tracing code ICARE has been preferred to
EBINAUR in this study for numerical flexibility reasons.
The first reason is the necessity to be able to separate
the geometrical computation which is highly CPU time
demanding from the acoustical computation (changing the
material properties for example) which is much quicker.
The second reason is the limitation of EBINAUR interfaced
with OPTIMA (internal PSA- Peugeot Citroën pre-post) to
compute more than 32 receivers at a time, the transfer
functions simulation being done using the reciprocity
principle (like the measurements).
Indeed, the 3D digitized mesh of the car has been imported
to build the model and therefore requires to compute 654
transfer functions (as many as the number of facets). The
convergence of the simulation has been reached at a depth
of 5 reflections (the depth 0 meaning the direct field).
The Ray-Tracing model has been used in combination with
the poro-elastic simulation in order to find the optimized
trade-off between good insulation and broadband
absorption, either implemented on the same part
leading to use Light-Weight Concept four or three layers
constructions (patented) or on other parts leading to use
bi-permeable concepts.
By implementing these solutions, we have reached the
weight target which was a weight reduction of 10\% (while
maintaining the acoustic performance).
Vehicle Validation: Final Acoustic Results
The figure 18 and table 1 show that not only we have
reduced the weight of the insulators by 10\% but also we
have slightly improved the acoustic performance in the
middle and high frequency range, particularly in terms of
Articulation Index, which leads to a more pleasant sound.
4th gear front
-1,8dB(A)
+ 5,1%IA
4th gear rear
- 2,4dB(A)
+ 3,5%IA
2nd gear front
- 0,8dB(A)
+ 3,1%IA
2nd gear rear
+ 0,1dB(A)
- 0,1%IA
Table 1 : Final results: improvement of the Articulation
Index and of the overall SPL dB(A)(500-5000Hz)
Fig. 17 : Icare Ray-Tracing Model
Fig. 18 : SPL [dB(A)] measurement comparison serial optimized, 4th gear 100 km/h
Vehicle Acoustic Synthesis Method 2nd Generation: an effective hybrid simulation tool to implement acoustic lightweight strategies