| Re: [sumo-user] Calculation of HBEFA4 emission model |
_______________________________________________Dear Jakob,
I just wanted to kindly report that the emission calculation is now considering the road slope. The results look like what we would expect, for instance a higher fuel consumption on an uphill road and vice versa. However, I discovered that the road slope given by the FullOutput seems to be incorrect. In the excerpt below the slope remains constant at 1,15 although the z-coordinates change. I have set up a slope of 0,02 (20 meters increase over a 1000 m stretch).
timestep
id
eclass
CO2
CO
HC
NOx
PMx
fuel
electricity
noise
route
type
waiting
lane
pos
speed
angle
x
y
z
slope
0
Petrol_Euro_4_3.0
HBEFA4/PC_petrol_Euro-4
1657,3
0
0
0,61
0
537,28
0
55,94
uphill
PC_P_4
0
3_0
5,1
0
90
5,1
198,4
0,1
1,15
1
Petrol_Euro_4_3.0
HBEFA4/PC_petrol_Euro-4
3213,67
0,88
0,01
0,75
0,04
1041,78
0
63,68
uphill
PC_P_4
0
3_0
6,72
1,62
90
6,72
198,4
0,13
1,15
2
Petrol_Euro_4_3.0
HBEFA4/PC_petrol_Euro-4
3815,57
3,98
0,03
0,95
0,1
1236,92
0
64,78
uphill
PC_P_4
0
3_0
10,05
3,33
90
10,05
198,4
0,2
1,15
3
Petrol_Euro_4_3.0
HBEFA4/PC_petrol_Euro-4
4713,77
10,88
0,07
1,25
0,16
1528,11
0
66,4
uphill
PC_P_4
0
3_0
15,27
5,22
90
15,27
198,4
0,31
1,15
4
Petrol_Euro_4_3.0
HBEFA4/PC_petrol_Euro-4
4383,13
13,55
0,09
1,27
0,2
1420,94
0
64,89
uphill
PC_P_4
0
3_0
21,88
6,61
90
21,87
198,4
0,44
1,15
5
Petrol_Euro_4_3.0
HBEFA4/PC_petrol_Euro-4
4854,68
20,27
0,13
1,47
0,26
1573,81
0
65,63
uphill
PC_P_4
0
3_0
29,89
8,01
90
29,88
198,4
0,6
1,15
I tried doing a backwards calculation (0,02 slope) of the fuel using the updated formula (https://github.com/eclipse-sumo/sumo/issues/16307) and it lead to the same results except for the „0“ fuel values. This means that the emissions seem to be correct and only slope is the issue here.
The emissions in FullOutput give out „0“ at certain deceleration value range – I believe it depends on whether the engine is off or the vehicle is currently coasting. Is it really based on the deceleration value? If yes, is it prescribed as a singular value (emissions are set at 0 if deceleration is i.e. below -0,15 mm/s2) or calculated based on other parameters? Thank you in advance.
Warmest regards,
Addina
Von: sumo-user <sumo-user-bounces@xxxxxxxxxxx> Im Auftrag von Addina Shafiyya Ediansjah via sumo-user
Gesendet: Freitag, 21. März 2025 11:24
An: Sumo project User discussions <sumo-user@xxxxxxxxxxx>
Cc: Ediansjah, Addina Shafiyya <addina_shafiyya.ediansjah@xxxxxxxxxxxxxxxxxxxxx>
Betreff: Re: [sumo-user] Calculation of HBEFA4 emission model
Dear Jakob,
this is well noted, I will until tomorrow. Thanks!
Warmest regards,
Addina
Von: sumo-user <sumo-user-bounces@xxxxxxxxxxx> Im Auftrag von Jakob Erdmann via sumo-user
Gesendet: Freitag, 21. März 2025 11:04
An: Sumo project User discussions <sumo-user@xxxxxxxxxxx>
Cc: Jakob Erdmann <namdre.sumo@xxxxxxxxx>
Betreff: Re: [sumo-user] Calculation of HBEFA4 emission model
Yes. But point 3 requires version v1_22_0-946-gd227df44261 or later whereas the nightly version available as of now is still on v1_22_0+0935-7ffb9238b00 so you have to wait until tomorrow.
Am Fr., 21. März 2025 um 10:39 Uhr schrieb Addina Shafiyya Ediansjah via sumo-user <sumo-user@xxxxxxxxxxx>:
Dear Jakob,
thank you for the answers. Just one more thing – so both point 2 & 3 would be incorporated if I install the nightly version?
Warmest regards,
Addina
Von: sumo-user <sumo-user-bounces@xxxxxxxxxxx> Im Auftrag von Jakob Erdmann via sumo-user
Gesendet: Freitag, 21. März 2025 10:14
An: Sumo project User discussions <sumo-user@xxxxxxxxxxx>
Cc: Jakob Erdmann <namdre.sumo@xxxxxxxxx>
Betreff: Re: [sumo-user] Calculation of HBEFA4 emission model
1. The coefficients are obtained by computing a fit that optimizes over all the colored curves at once.
2. The code at https://github.com/eclipse-sumo/sumo/blob/7308a1b9569b4fc8edc99c891063b4366a31d0a8/src/utils/emissions/HelpersHBEFA4.h#L114 is from an older version where road slope was not take into account. This was changed only recently: https://github.com/eclipse-sumo/sumo/issues/16307
3. The angle in full output is the bearing of the vehicle. To get the slope you may download the nightly version starting tomorrow (https://github.com/eclipse-sumo/sumo/issues/16389)
regards,
Jakob
Am Fr., 21. März 2025 um 09:24 Uhr schrieb Addina Shafiyya Ediansjah via sumo-user <sumo-user@xxxxxxxxxxx>:
Dear all,
I am trying to understand the calculation of the HBEFA4 emission module and came across a few questions:
- According to a previous discussion (https://github.com/eclipse-sumo/sumo/issues/15206), SUMO takes emissions for a particular vehicle class and emission type from the database and fits it over a range of speed and acceleration. The different colored lines in the attached graphs correspond to seven discrete acceleration values. SUMO then takes f[0] to f[6] to calculate the emissions for a particular vehicle class. Are the coefficients f[0] to f[6] thus the average value of the seven different colored lines?
- HBEFA which is not dependant on acceleration is compensated by taking its dependency on road slope. As far as I understand, SUMO calculates the added acceleration it needs to exert to overcome the road slope. One documentation (https://sumo.dlr.de/doxygen/d2/db0/_helpers_h_b_e_f_a4_8h_source.html) (line 113) states the calculation of a2, which takes the vehicle acceleration and then adds it with (gravity * sin θ). I could not find this calculation step on another documentation page (https://github.com/eclipse-sumo/sumo/blob/7308a1b9569b4fc8edc99c891063b4366a31d0a8/src/utils/emissions/HelpersHBEFA4.h#L114). Is it already embedded in another function for the second documentation or is there another reason behind it?
- Not directly related to the HBEFA4 emission module – but does the angle in the FullOutput (https://sumo.dlr.de/docs/Simulation/Output/FullOutput.html) refer to the vehicle angle with respect to the road slope? I tried to set up a road stretch with a 0.02 (2%) & -0.02 (-2%) road slope by patching the network with z-coordinates. Since not all models incorporate elevation data (https://sumo.dlr.de/docs/Networks/Elevation.html), I have also selected the carFollowModel=”KraussPS”. All the angles in the FullOutput remained constant at 90 degrees. Is there an explanation for this result?
Any pointers are appreciated. Thank you in advance!
Warmest regards,
Addina
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user