National Repository of Dissertations in Serbia
    • English
    • Српски
    • Српски (Serbia)
  • English 
    • English
    • Serbian (Cyrilic)
    • Serbian (Latin)
  • Login
View Item 
  •   NaRDuS home
  • Универзитет у Београду
  • Саобраћајни факултет
  • View Item
  •   NaRDuS home
  • Универзитет у Београду
  • Саобраћајни факултет
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Metodologija za utvrđivanje optimalnog rešenja rasputnice

Methodology to determine a railway junction optimal design

Thumbnail
2013
Disertacija.pdf (25.76Mb)
Author
Milinković, Sanjin M.
Mentor
Vukadinović, Katarina S.
Committee members
Vesković, Slavko M.
Stojić, Gordan
Metadata
Show full item record
Abstract
Simulaciono modeliranje je veoma efikasan alat za analiziranje složenih železničkih sistema kao što su sistemi saobraćaja vozova na rasputnici. Rasputnica je službeno mesto gde se sa otvorene pruge odvaja druga pruga. Model sistema rasputnice predstavlja procese kretanja vozova preko izolovanih odseka sistema rasputnica. Granice modela postavljaju se u stanicama koje okružuju rasputnicu. Model Petrijevih mreža sistema rasputnice razvijen je tako da je svaka vrsta odseka predstavljena modulom (podsistemom) u grafu Petrijevih mreža. Izrada modela sistema rasputnice vrši se povezivanjem modula odseka po planu odseka, a zatim i njihovim obeležavanjem i definisanjem. Ulazni podaci potrebni za izvršenje simulacije, definišu se u bazi podataka koja je povezana sa simulacionim programom. Ulazni podaci o primarnom kašnjenju vozova generišu se u posebnom modelu ili modulu fazi Petrijeve mreže. Modeli kašnjenja vozova zasnivaju se na tehnikama računarske inteligencije. U slučaju kada postoje stat...istički podaci o prethodnim kašnjenjima, za proračune kašnjenja koriste se modeli zasnovani na neuronskim mrežama ili adaptivni neuro-fazi modeli. Ovi modeli se obučavaju i verifikuju podacima koji dobijenim praćenjem kretanja vozova i iz dnevnika otpravnika vozova. Kada nisu dostupni podaci o kašnjenju vozova u prethodnom periodu, primenjuje se model zasnovan na fazi logici, gde se ekspertsko znanje o ponašanju sistema koristi za kreiranje modela. U toku izvršenja simulacije, podaci o kretanju voza i zauzetosti odseka snimaju se u unapred definisanu bazu podataka. Stanja odseka i kretanje vozova tokom simulacije može se pratiti pomoću animacije i grafikona saobraćaja vozova. Baza podataka u kojoj se nalaze rezultati simulacije prilagođena je za jednostavnu obradu podataka i analizu rezultata. Model je testiran na primeru rasputnice „G“ beogradskog železničkog čvora. Nakon izrade, validacije i verifikacije modela rasputnice „G“, analizirani su rezultati modela preko sekundarnih kašnjenja vozova nastalih u sistemu rasputnice. Takođe, upoređena su različita infrastrukturna rešenja rasputnice...

Simulation modelling is very efficient method for analysis of complex railway systems like railway junction system. Junction is a place on the track where another track diverges. Model of the junction system is comprised of the processes of train movement on insolated sections. Model boundaries are stations that are surrounding the junction system. Petri Net model of the junction system is defined as a model where all types of section blocks are denoted by a module or subsystem in the Petri Nets Graph. Building of the model is a process of connecting modules according to the junction section plan and then marking and defining them. Initial data for simulation is located in an external database connected to the simulation program. Input data on train primary delay is calculated either in separate model or in the fuzzy Petri net module. Models for calculating train primary delay are based on computational intelligence techniques. When there are historical data on train delays, primary de...lays are calculated by neural networks or adaptive neuro-fuzzy model. These models are trained and verified by data on previous train delays (from train detection systems or from dispatcher’s logs). When there are no data on previous delays in system, delays are calculated by a fuzzy logic model where experts’ knowledge on system behaviour is used to create a model. During simulation run data is exported to external predefined database. State of the sections and train movements are observed in animation window and in the trains’ time-distance graph. Database of simulation results is improved for data query and data analysis. Model is tested on a case study of junction “G” system located in Belgrade Railway Node. After model building, validation and verification of model, data results of simulation are analysed by secondary delays generated within the model. Also, different infrastructure solutions are compared...

Faculty:
Универзитет у Београду, Саобраћајни факултет
Date:
14-06-2013
Keywords:
železnički saobraćaj / railway traffic / junctions / simulation modelling / petri nets / fuzzy logic / artificial neural networks / adaptive network fuzzy inference system (ANFIS) / rasputnice / simulaciono modeliranje / Petrijeve mreže / fazi logika / veštačke neuronske mreže / adaptivni neuro-fazi sistemi
[ Google Scholar ]
Handle
https://hdl.handle.net/21.15107/rcub_nardus_2672
URI
https://nardus.mpn.gov.rs/handle/123456789/2672
http://eteze.bg.ac.rs/application/showtheses?thesesId=935
https://fedorabg.bg.ac.rs/fedora/get/o:7483/bdef:Content/download
http://vbs.rs/scripts/cobiss?command=DISPLAY&base=70036&RID=512326314

DSpace software copyright © 2002-2015  DuraSpace
About NaRDus | Contact us

OpenAIRERCUBRODOSTEMPUS
 

 

Browse

All of DSpaceUniversities & FacultiesAuthorsMentorCommittee membersSubjectsThis CollectionAuthorsMentorCommittee membersSubjects

DSpace software copyright © 2002-2015  DuraSpace
About NaRDus | Contact us

OpenAIRERCUBRODOSTEMPUS