Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [sumo-user] How to get high densities?

Hi Jakob,

Just to confirm that this fix does seem to have fixed the problem. Many thanks.

On Thu, 10 Feb 2022 at 13:24, Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
Thank you for the example. I did uncover a bug in the insertion code that has now been fixed (https://github.com/eclipse/sumo/issues/10137).
You can download an updated version of sumo tomorrow at https://sumo.dlr.de/docs/Downloads.php#sumo_-_latest_development_version

Am Mi., 9. Feb. 2022 um 16:18 Uhr schrieb Ste Millington <stemillington@xxxxxxxxx>:
Please find attached my config. I've tidied things up a bit and do appear to be now getting slightly higher densities but there still seems to be a bottleneck at densities lower than I would expect.

I have also attached a few graph showing how the average gap, effective flow and average speed alter as i increase the vehsPerHour parameter. The all seem to flatline at around 6000.

Many thanks for taking a look at this.


On Wed, 9 Feb 2022 at 09:54, Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
The spacing of 90m is unexpected for your setup, With the above configuration, you should see an average spacing of roadSpeed + vehLength + minGap (i.e. ~20m for 13.89/ms = 50km/h road and 7.5m length+minGap).
Depending on road speed you should be seeing 1800 - 2000 v/h/lane.
If you use departSpeed="avg" instead of "max", capacity goes up to 2400 v/h/lane

Please post your (zipped) scenario files to figure out why you are not seeing this.

However, insertion flow alone will never lead to a breakdown of the flow. For that you need some sort of speed disturbance downstream.
If you set sigma="1" you will see occasional breakdowns but they will be short-lived (and due to a known issue with the Krauss model, this only works at 1s step length).
An alternative for inducing break-down would be a variable-speed-sign

regards,
Jakob

Am Di., 8. Feb. 2022 um 17:23 Uhr schrieb Ste Millington <stemillington@xxxxxxxxx>:
I'm struggling to get high flow rates in a simple model, just a straight 5km stretch of 4 lane highway.

I'm using multiple flow elements with probability set close to 1. However i can't seem to get the flow rate to ever go above 4000 vehicles per hour. This only gives me 1000 vehicles per hour per lane and an average vehicle spacing of 90 metres. I was hoping to be able to achieve a breakdown of the flow.

I've tried tried setting step length to 0.1 and using multiple copies of

<flow id="flow_0" type="DEFAULT_VEHTYPE" begin="0.00" departSpeed="max" departPos="last" departLane="best" route="route_0" end="3600.00" probability="0.9"/>

But after a certain point nothing seems to make much difference. I'm sure I'm doing something wrong or not perhaps just misunderstanding the traffic dynamics somehow.

Should I be able to ramp up the flow rate to a point where the traffic flow breaks down?
_______________________________________________
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
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user

Back to the top