Re: [sumo-user] HBEFA Emission Model

[Mirko Barthauer <m.barthauer@xxxxxxxxxxx>]

Hi Florian,

 

I have two points before we start to investigate in detail:

• Which car-following model did you use? Unfortunately, elevation data is not used universally in sumo yet but only by certain components (see docs)
• Even if the elevation is not used, it increases the edge lengths a bit (Pythagoras ...). Nonetheless, it seems strange the consumption is lower elevation...

Best regards

Mirko

 

 

 

 

 

-----Original-Nachricht-----

Betreff: Re: [sumo-user] HBEFA Emission Model

Datum: 2023-08-21T09:26:48+0200

Von: "Florian Schnepf via sumo-user" <sumo-user@xxxxxxxxxxx>

An: "Daniel.Krajzewicz@xxxxxx" <Daniel.Krajzewicz@xxxxxx>, "sumo-user@xxxxxxxxxxx" <sumo-user@xxxxxxxxxxx>

 

 

 

Hello Daniel,

 

I've got it now. Thank you for taking the effort to give me this detailed answer.

 

It would be great if someone could explain the consumption behaviour of my experiment.

 

 

Best regards
Florian

 

 

Gesendet: Montag, 21. August 2023 um 09:17 Uhr
Von: Daniel.Krajzewicz@xxxxxx
An: florian.schnepf97@xxxxxx
Cc: sumo-user@xxxxxxxxxxx
Betreff: AW: AW: [sumo-user] HBEFA Emission Model

Hello Florian,

 

Well, obviously, I did not explain it properly.

Vehicular emissions mainly depend on the acceleration and the speed, with acceleration being the major influence.

As SUMO is a microscopic model where the vehicles’ speed and acceleration are known, this information is used in all of SUMO’s emission models.

But within the original HBEFA, no dependency between accelerations and emissions exists. HBEFA is an inventory models which only gets vehicle types (the vehicle fleet in fact), the mileage, and the traffic state as inputs – and the slope.

Now slope can be transformed into acceleration – the power needed to climb is physically of the same type as the power needed to accelerate. And that has been done when transferring HBEFA-values into SUMO’s HBEFA-based emission model.

This means SUMO uses the acceleration and the speed though acceleration was not given in the original HBEFA.

 

Now I must admit that I do not know why your values differ only by such small values. I have never worked with elevations in SUMO.

 

Maybe someone else can answer this.

 

Sincerely,

Daniel

 

Von: Florian Schnepf <florian.schnepf97@xxxxxx>
Gesendet: Samstag, 19. August 2023 15:40
An: Krajzewicz, Daniel <Daniel.Krajzewicz@xxxxxx>; sumo-user@xxxxxxxxxxx
Betreff: Aw: AW: [sumo-user] HBEFA Emission Model

 

Hi Daniel,

 

What happens when there is no information about the slope?

 

I have made some tests with one network with increases and one without, as well as two HBEFA emission models.
The results show that there is very little difference on average.

• Diesel vehicle: 

With elevation: 22.319 g
without elevation: 22.495 g

• Fuel cell vehicle: 

With elevation: 6.374 g
without elevation: 6.522 g

 

I would expect consumption to be higher with elevation than without.

Does this mean that the information about the slope does not contribute to the calculation of the consumption, since it replaces the influence of acceleration?

 

Best regards
Florian

 

Gesendet: Montag, 14. August 2023 um 10:54 Uhr
Von: Daniel.Krajzewicz@xxxxxx
An: sumo-user@xxxxxxxxxxx
Cc: florian.schnepf97@xxxxxx
Betreff: AW: [sumo-user] HBEFA Emission Model

Hello Florian,

 

well, no.

HBEFA has the information about the influence of slope on emissions. On the other hand, as HBEFA uses (macroscopic) traffic states, it does not use (microscopic) acceleration.

So, when deriving emissions from HBEFA, we used the slope information for replacing the missing influence of accelerations.

 

Sincerely,

Daniel

 

 

Von: sumo-user <sumo-user-bounces@xxxxxxxxxxx> Im Auftrag von Florian Schnepf via sumo-user
Gesendet: Montag, 14. August 2023 10:51
An: sumo-user@xxxxxxxxxxx
Cc: Florian Schnepf <florian.schnepf97@xxxxxx>
Betreff: [sumo-user] HBEFA Emission Model

 

Hello,

 

 

I read in a paper about the emission models that the HBEFA model does not depend on acceleration and therefore the slope is used for compensation. 
 

I do not understand how this is meant. Can someone explain to me how this was implemented?

 

 

Thanks a lot

 

Best regards
Florian