Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [sumo-user] Need help with graphs

1. you are simulating a high level of traffic with an (almost) constant insertion flow on a short road segment. There are actually a few datapoints in the left side of the plot from the time where the road starts to fill. Afterwards the traffic stays dense. Note, that your simulation has extreme insertion backlog (the requested flow is higher than the road capacity) and this slows down your simulation (see https://sumo.dlr.de/docs/Simulation/VehicleInsertion.html#delayed_departure)

2. vehicle insertion in SUMO generally cannot create a breakdown in flow on a single road. For this you need some kind of obstruction downstream of the insertion point (i.e. from a junction).  Hence, in your simulation the density never increases to the point where speed is lowered significantly.

3. your simulation shows saturation flows (minimum headway for the given speed) over a narrow range of speeds. In this speed range the flow is mainly determined by the (constant) time headway of the simulation and hence remains unaffected. For more math on this, see https://sumo.dlr.de/docs/Simulation/RoadCapacity.html

4. The gaps on the road are high because their speed is high. The speed remains high because vehicles can drive off without obstruction after insertion.

For a discussion on generating all traffic density states, see https://sumo.dlr.de/docs/Tutorials/FundamentalDiagram.html

regards,
Jakob


Am Di., 22. Feb. 2022 um 22:49 Uhr schrieb Sumbal Malik <maliksumbal01@xxxxxxxxx>:
I also run the simulation with a short length of edge 500m with the same repeated vehicle flows but still, the graphs are not generating correctly as they should be. I also observed that the speed is not changing and there is a lot of gaps between the vehicles on the road.

On Tue, 22 Feb 2022 at 14:29, Sumbal Malik <maliksumbal01@xxxxxxxxx> wrote:
Dear Jakob

Apologies to bother you again.

As per your suggestions, I generated the attached graphs using the below commands:

1) Occupancy vs Time Loss -- plotXMLAttributes.py edge_data.xml -i id -x occupancy -y timeLoss -s --scatterplot --xlabel Occupancy(%) --ylabel TimeLoss(s)

2) Density vs Speed -- plotXMLAttributes.py edge_data.xml -i id -x density -y speed --yfactor 3.6 -s --scatterplot --xlabel density --ylabel km/h"

3) Density vs Arrived -- plotXMLAttributes.py edge_data.xml -i id -x density -y arrived --yfactor 60 -s --scatterplot --xlabel density --ylabel VehPerHour

I would be grateful if could clear the following confusions:

1. Why no data is plotted on the left side of all graphs. 
2. In Graph_2: why the speed is high even though the density is high ... should not the speed be slow when the density is high? If we have high density, the vehicles will move slowly.
3. In Graph_3: similarly if the density is high then the number of Vehperhour leaving the edge should be less but in the attached graph it is opposite.

Please find the attached configuration files and graphs

Please guide me where I am making mistakes.

Thank you

On Mon, 21 Feb 2022 at 17:48, Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
In case that the edge in question is not a transit edge but the destination edge, use 'arrived' instead of 'left'.

Am Mo., 21. Feb. 2022 um 14:42 Uhr schrieb Sumbal Malik <maliksumbal01@xxxxxxxxx>:
Thank you very much for your reply.

I generated the edge data with <edgeData id="dump_60" freq="60" file="edge_data.xml" excludeEmpty="true"/>.
But the output of the edge data shows left vehicles "0" for each time step.

<interval begin="0.00" end="60.00" id="dump_15">
        <edge id="E1" sampledSeconds="1422.00" traveltime="80.92" overlapTraveltime="81.36"
density="23.70" laneDensity="11.85" occupancy="6.38" waitingTime="0.00" timeLoss="140.73"
speed="12.36" speedRelative="0.89" departed="48" arrived="0" entered="0" left="0" laneChangedFrom="6" laneChangedTo="6"/>
    </interval>

Please find the attached configuration files.



Thank you


On Mon, 21 Feb 2022 at 16:38, Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
1. you cannot do this with fcd-output directly since you need an intermediate processing step to compute average speeds (fcd has no aggregation). You could plot edgeData 'speed' vs 'left' for this
2. as above. You need to program your own aggregation for this.
3. edgeData 'left' vs 'density'

Am Sa., 19. Feb. 2022 um 09:45 Uhr schrieb Sumbal Malik <maliksumbal01@xxxxxxxxx>:
Dear All 

I'm struggling to plot some graphs from the attached files. 

I would be thankful if you could share the script/commands to generate the graphs:

1. How can I plot a graph of average speed versus vehicles per hour using FCD output.

2. How to plot a graph of the average leader gap versus vehicles per hour using FCD output.

3. How to calculate the number of vehicles per time that pass through a given point and plot them versus density?

Thank you in advance.








_______________________________________________
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
_______________________________________________
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