Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Try #2813

Closed
wants to merge 16 commits into from
Closed

Try #2813

wants to merge 16 commits into from

Conversation

synas34
Copy link

@synas34 synas34 commented Oct 8, 2023

No description provided.

…ne class for the handler and another as the main method to run the analysis
…tersection densities. Created the contrib file SyntheticModel for this and synthetic population scripter. Created a class to generate nodes and another to generate links from the nodes created from IntersectionCoordinateCalculator. Extended it using MultiSquareCoordinateCalculator, which takes multiple slice numbers and generates a line of dense meshes. Writes to xml file that will be used for synthetic model.
…t uses xml files of household and commerical locations to generate house to work to house trips for specified people. RandomCoordinatesGenerator.java script is used to generate the semi-random coordinates, with a density and number in core(250 m around centre) parameter to control for spatial variables. Plans file needs to be extended with ability to make shopping trips, and household and commerical location files can be updated with more realistic locations
…adjusted for each new module attached to the simulation. Trouble shooted and added necessary changes so that drt could work. Will optimise vehicle fleet and plans.
…adjusted for each new module attached to the simulation. Trouble shooted and added necessary changes so that drt could work. Will optimise vehicle fleet and plans.
…changed to allow for multiple initial modes. RunCreateDrtMesh.java changed so that the fleet is around 1/40 of pop size. mode change is working. NEXT need to incorporate public transit through synthetic transit vehicle and schedule files. Also need to incorporate mode choice model utility values.
…distances between stations. 6 urban stations, 4 suburban. Used RailLinkCreator.java to create them, it needs number of links and distances and origin defined
…distances between stations. 6 urban stations, 4 suburban. Used RailLinkCreator.java to create them, it needs number of links and distances and origin defined
…h contains the route of the train route and its reverse. The departure times are gained from an assumed trend. with rush houe being 15 mins apart, going into 20 min headways and finaly an hour in the afternoon and night. Can improve them. Use of only two trains currently, as route should take less than 15 mins. Will test and see if the transit file works or not.
…at vehicles run properly. Successfully ran a 500 pop scenario with car,walk,bike,drt,pt. config5 is saved. Will see if pt runs as expected without drt module.
… transitschedule and transitvehicle files. Successfully ran pt-car simulation
…uccessfully. Incorporated basic mode choice parameters in modeparams and changeMode. Use of subtourModeChoice and multiModeDrt to see if simulations are correct. However, parameters when inputed seem to favour bike mode very heavily. MATSIM simulation for 50 iterations is still unstable but seems to converge to around 50% bike modes. May be due to the inherent small size of network05.xml of only 8 by 1 km. Irregardless will update mode choice modules with subtourModeChoice, extend the multiModeDrt module, and intermodalAccessEgress module to incorporate FMLM modes.

		<parameterset type="modeParams">
			<param name="mode" value="car"/>
			<param name="constant" value="-1.8" />
			<param name="marginalUtilityOfTraveling_util_hr" value="-5.7" />
			<param name="monetaryDistanceRate" value="-0.01" />
		</parameterset>
		<parameterset type="modeParams">
			<param name="mode" value="drt"/>
			<param name="constant" value="-3.3" />
			<param name="marginalUtilityOfTraveling_util_hr" value="-6.2" />
			<param name="monetaryDistanceRate" value="-0.05" />
			<!-- [utils] daily utility constant. default=0 to be backwards compatible -->
			<param name="dailyUtilityConstant" value="-0.0" />
		</parameterset>
		<parameterset type="modeParams">
			<param name="mode" value="pt"/>
			<param name="constant" value="-0.7" />
			<param name="marginalUtilityOfTraveling_util_hr" value="-0.0" />
			<param name="monetaryDistanceRate" value="-0.03" />
			<!-- [utils] daily utility constant. default=0 to be backwards compatible -->
			<param name="dailyUtilityConstant" value="0.0" />
		</parameterset>
		<parameterset type="modeParams">
			<param name="constant" value="0.0" />
			<param name="marginalUtilityOfTraveling_util_hr" value="-7.5" />
			<param name="mode" value="walk" />
			<param name="monetaryDistanceRate" value="0.0" />
		</parameterset>
		<parameterset type="modeParams">
			<param name="constant" value="-1.5" />
			<param name="marginalUtilityOfTraveling_util_hr" value="-5.8" />
			<param name="mode" value="bike" />
			<param name="monetaryDistanceRate" value="0.0" />
		</parameterset>
…ress module. Also extended multiModeDRT to manually write drt positioning through zonalsystem parameterset and rebalancing through rebalancing parameterset, and drt pricing through "drtfare" parameterset. Successful run, however a large discrepancy in mode share,

 walk = 13.6
 bike = 23.6
 car = 37.8
 drt = 12.4
 pt = 12.6

 Most likely due to network size differences. toy network only 8km long when odakyu line alomst 80 km, scale factor of 10. width is 500m when it actual simulation should be 3km, or drt access/egress radius.
…ated Scenario 40x1km to replicate distance of Suburban rail built environment, especially the distance of trips and reliance on rail. Updated IntersectionCoordinateCalculator, MultiSquareCoordinateCalculator, LinksCalculator to use Coord, Node and Links classes from Matsim rather than to generate strings that would be used to write an XML file. Updated the Network generator main method to SyntheticNetworkCreator and updated it to create 40 squares. Need to make it more three dimensional, may simply make it wider, 3 x 1 km. In lieu of new scenario 40x1km, created estimated household and commercial location xml files. Updated PlansXMLSynthesizer to use Coord class, and generated 5000 plans. Updated RailLinkCreator by putting all the logic of creatign rail links into RailLinkCreator.generateLinks so that RailScheduleCreator.createSchedule can call it and not need to manually list the links that the rail route and rail stops will be made on. Updated RailScheduleCreator with 40 stations and departure times more similar to actual, with high frequency rail as in every 3-5 minutes. Need to update use of train for that one. Irregardless, bigger network is causing now noticeable time and computational cost(more than 10 mintues). Need to save, run a long simulation through scripting or through Lab computers. Also updated RunMatsimUnique to printstats of transit boardings as well as automatically open bar graph of mode share. Next steps will be to run large sim on Super Godzilla(Lab Server), figure out if it really is just a measure of network complexity, and expand event handlers to gather necessary data. Need to understand how to generate effective policy tests within synthetic simulation.
…ogether. Moved editable part of SyntheticNetworkCreator PlansXMLSynthesizer, RandomCoordinatesGenerator, RailScheduleCreator and VehicleCreator to create all parts of a synthetic scenario:

		// 1. Create network.xml
		// 2. Create households.xml and commercial.xml
		// 3. Create plans.xml
		// 4. Create transitschedule.xml
		// 5. Create transitVehicles.xml
Need to run DRTMesh for drtvehicle file*
Ran and troubleshooted a lot of simulations, ultimately decided that a general rough of 2500 nodes and 5000 link, or 30, 30, and 30 slices. Otherwise starts a memoryleak that will not clear up. Need to generate simulations that wont have that many links.
@synas34 synas34 closed this Oct 8, 2023
@synas34 synas34 deleted the try branch October 8, 2023 14:04
@synas34 synas34 restored the try branch October 8, 2023 14:06
@synas34 synas34 deleted the try branch October 8, 2023 14:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant