Second order of reflection or higher

[eliserd]

Hello,
I seem to be unable to get a second (or higher) order of reflection when using the Noise_level_from_source function. The change from 0 to 1 reflection works however. I would appreciate any advice on what I'm doing wrong !

Thanks in advance,

BUILDINGS.txt

test_reflection.txt
Point_Source.txt

[nicolas-f]

Hi,

How do you conclude that the second (or higher) order of reflection doesn't work ?

[eliserd]

I tested this kind of configuration (here only the rays reaching the top left receiver are plotted). If the wall absoprtion is fixed at 0.1, the 2nd order of reflection should have some impact ?

[nicolas-f]

You should have a reflection on the right building if the distance between the rays and the wall is inferior than maxRefDist

[eliserd]

Yes sorry, it was not the best illustration (here is something more accurate).

[nicolas-f]

Hi,

As there is a direct field, second order reflected rays may not be significant enough to be kept. Maybe try with receiver not in direct view with source

[nicolas-f]

You can change the parameter here (to 0), but the computation time will be greatly increased without any change in results value:

https://github.com/Universite-Gustave-Eiffel/NoiseModelling/blob/v4.0.5/wps_scripts/src/main/groovy/org/noise_planet/noisemodelling/wps/NoiseModelling/Noise_level_from_source.groovy#L693

[eliserd]

Hello,
Thanks for the advices !
I did some additional tests but I still don’t find the 2nd order reflections.
Initially I noticed my problem while doing a sensitivity analysis where the reflection order {1;2;3;4} did not have any influence on the results. The analysis has been done for 3 different source positions and 3 different heights (3, 25 and 70m). The source is omnidirectional and its sound power level is over 70 dB.
Here is an example of 6 sources (tested one by one) and some receivers (small brown spheres).

I obtain this kind of results for the first (light pink) and third order (dark pink) of specular reflection:

The maximum distances confMaxReflDist and confMaxSrcDist are set to 1000 m.
Two types of grid have been used for the receivers (Delauney and regular).
For most of my scenarios, I don’t enable the diffraction parameters because I want to focus first on the understanding of the specular reflections. The presence of diffraction doesn’t seem to solve my problem.

I also used your advice to look at a receiver outside of a direct path and a MaximumError set at 0.01d and 0.0, but I did not see any change. For example, here I used a high reflection order (7) and I was expected some rays to reach the receivers circled in blue:

[eliserd]

I have also some strange vertical rays for one of my tests :

and for all of the tests where I enabled both the 2nd order of reflection and the horizontal diffraction :

I mention this in case it is related to my initial issue.