WO2014041439A1 - Calculation and evaluation of station capacity for high speed railway - Google Patents

Calculation and evaluation of station capacity for high speed railway Download PDF

Info

Publication number
WO2014041439A1
WO2014041439A1 PCT/IB2013/050040 IB2013050040W WO2014041439A1 WO 2014041439 A1 WO2014041439 A1 WO 2014041439A1 IB 2013050040 W IB2013050040 W IB 2013050040W WO 2014041439 A1 WO2014041439 A1 WO 2014041439A1
Authority
WO
WIPO (PCT)
Prior art keywords
station
trains
occupation
time
high speed
Prior art date
Application number
PCT/IB2013/050040
Other languages
French (fr)
Inventor
Jiamin Zhang
Original Assignee
ZHANG, Jiarui
ZHANG, Chunjiang
FU, Shuxia
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ZHANG, Jiarui, ZHANG, Chunjiang, FU, Shuxia filed Critical ZHANG, Jiarui
Publication of WO2014041439A1 publication Critical patent/WO2014041439A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q50/40—

Definitions

  • Ph.D Jiamin Zhang belongs to the technical field of the application of railway transportation , targeting at calculation & evaluation of station capacity of high speed railway and recognizing its capacity bottleneck.
  • the operations around the station region is relatively complex affected by many factors, such as the layout of the station yard, the selection of the station's facility, the unequilibrium of the trains' arrival and departure, the percentage of the types of the trains' arrival and departure, the usage scheme of the station tracks, the cooperation between the station's throat area and its arrival-departure tracks, the running lines of the multiple train unit depot, etc, according to Wen Dong (2006).
  • Hu Anzhou & Yang Hao (1994) proposed the conception of the total effective capacity of the railway system's facility group and discussed the problem of the conception, the calculation, the usage and the reinforcement of the railway capacity in the view of dynamics, which laid the foundation for the later research on the railway capacity deeply and totally.
  • This invention proposes the global — local bilevel model system & optimizing flow to calculate & evaluate the station capacity which optimizes the trains' occupation order on global level and checks the feasibility of the order on local level (Fig. 1). On global level this invention proposes the flow chart for optimization of trains occupation order of station tracks(Fig. 8) and Max-plus calculation eigenvalue & eigenvector for each train occupation of track(Fig.
  • the core element in t@s—TSDIS is the train, for which the 'service' means it designates the physical nodes for the trains entering and leaving the station as well as the platforms for the trains to dwell (or pass by) according to the station's physical connection direction & topology, the 'demand' designates the speed & class type for each train in t@s—TSDIS from the angel of the passenger demand and the connections among the trains, etc.
  • the 'service' means it designates the physical nodes for the trains entering and leaving the station as well as the platforms for the trains to dwell (or pass by) according to the station's physical connection direction & topology
  • the 'demand' designates the speed & class type for each train in t@s—TSDIS from the angel of the passenger demand and the connections among the trains, etc.
  • other elements in t@s—TSDIS are as following:
  • t@s-TSDIS determines the type and service priority for each train in t@s—TSDIS.
  • the high speed railway serves the 'passenger', so the ideas of trains routing should be changed as the comprehensive balance between the passenger demand and the capacity supply when calculating & evaluating the station capacity.
  • the maximum passenger flow in t@s-TSDIS depends on its practical demand. It presumes d i p as the passenger demand within the time period, C i as the given train capacity corresponding to its type ⁇ , as the type i train's loaded factor satisfied a certain service level, then would be the number of trains planned in t@s—TSDIS.
  • this invention discretizes the calculation time period by the equal time interval (e.g. minute), then endows a certain fuzzy probability for the trains occupation or passing by the physical station infrastructure in a certain time interval.
  • the fuzzy time function is the mapping function from the time scalar (non-negative real number) to the real number section [0,1].
  • This invention adopts the fuzzy triangle time function tackling with the uncertainty and dynamics, which lets the fuzzy triangle time slot t(a,b,c) represents the time items for the trains entering the station, leaving the station, passing by the simplified sub–region and dwelling at the station, etc.
  • h i denote the minimum time interval for the trains running on the line section connecting with the station
  • [t b ,t e ] denote within time period, according to the time interval for the trains running on the section, then is the possible arrival time for each train j in t@s-TSDIS within time period, is the departure time of the arrival–stop–departure train, where dwell(t j ) denotes the dwelling time at station, denotes the allowable possible deviation time or certain fuzzy time probability, then arr(t j ) and dep(t j )can be described as the fuzzy triangle time slot.
  • This invention constructs the high speed railway trains at station 'train service - demand intention set' (t@s—TSDIS) and regards the time needed to complete the set t@s—TSDIS as the new criteria to measure the station capacity of the high speed railway.
  • This invention proposes the global — local bilevel model system & optimizing flow to calculate & evaluate the station capacity which optimizes the trains' occupation order on global level and checks the feasibility of the order on local level (Fig. 1). On global level this invention proposes the flow chart for optimization of trains occupation order of station tracks(Fig. 8) and Max-plus calculation eigenvalue & eigenvector for each train occupation of track(Fig.
  • FTHNs is a sort of Petri nets which introduces the fuzzy set theory related to time and uses four sorts of fuzzy time function: represent the fuzzy time stamp, fuzzy enable time, fuzzy occurrence time and fuzzy delay time respectively. represents the token colored by V locating in place P, where is its fuzzy time stamp.
  • the trains operation types in station can be classified as original departure, final arrival, turning out, passing by (stopping then leaving is included).
  • the static definition for Interaction between Trains & Station t@s—FTHNs can be as following:
  • P in and P out denote the place for the trains entering and leaving the station tracks respectively
  • P free and P busy denote the station tracks are available and busy respectively.
  • T— finite transition set where T arr , T dep and T syn denote the train's arrival event, departure event and connection event.
  • the station In the global level the station is abstracted as the composition of a series of tracks and nodes, assigning the trains operation to the parallel tracks according to a certain order without considering the capacity limited. According to the machine schedule theory, it's regarded as the non-preemption job shop (train operation) problem on a certain number of the same type of machines (parallel tracks) (Fig. 2 — Fig. 7 in Drawings section ).
  • FTHNs allocates the tracks for two trains a, b and provide the information for the partial order .
  • the fuzzy triangle it takes: .
  • trains operation order on the station tracks belong to the non-preemption problem.
  • the occupation order for multi-trains share one track can be determined by introducing the corresponding strategy and turn it to the t@s—FTHNs event graph which doesn't need further decision.
  • Fig. 6 for the strategy flow chart
  • this invention constructs t@s—FTHNs event graph for the task list t@s-TSDIS and designs the optimized algorithm for the trains occupation order on station tracks (Fig.8 in Drawings section). According to Y. Zhou. T. Murata. T Defanti. (2000) this invention defines the defuzzy function deFuzzy[a, b, c]:
  • the proposed t@s—TSDIS trains' occupation order of the station tracks on the global level is on the assumption of the ignorance of the constraints of the interval time of the resource allocation and the route conflicts, it needs to check the conflicts and optimize the routes based on the logic relationship illustrated in Fig. 1: in the local level it regards the station tracks, the switches and the signals as the resources, based on the concrete resource topology using the depth-first search algorithm to enumerate the possible routes set R for the trains in task list t@s—TSDIS between the entering points of the station system and the leaving points of the station system and then optimize the trains' occupation order of the station tracks.
  • TSDIS For the t@s—TSDIS trains' occupation order of the station tracks on the global level, it determines the start–end time for the occupation of the corresponding resources, that is it proposes algorithm1 to backstep the moment that the trains passing by each subregion from its entering points of the station system according to the trains' arrival time and deduce the moment that the trains' passing by each subregion from their departure of the station tracks to the leaving points of the station system based on the trains' departure time of the station. Based on these, it proposes algorithm 2 to construct the resource tree conflict graph (RTCG) to check and optimize the trains route.
  • RTCG resource tree conflict graph
  • the improved RTCG proposed by this invention regards the station tracks and the switches as the resources, it constructs the algorithm1 to determine the start-end time of resource occupation on the trains' operation route at first and algorithm2 to find the conflict group for check the conflicts from the set of the start–end time of the resources occupation, that is it groups the conflicts for all of the possible overlap time interval of the resource occupation.
  • algorithm1 calculation of start–end time for occupation of resource i by train t on its operation route r.
  • T ts ri start time of occupation of resource i by train t in t@s—TSDIS in its route r;
  • T te ri end time of occupation of resource i by train t in t@s—TSDIS in its route r.
  • Step1 For the resource i occupied by train t in t@s—TSDIS on its route r, let distance(i, j) denote the physical distance between the resource i and j on the same route and train length(t) denote the length of the train t.
  • Step2 according to the trains occupation order in t@s—TSDIS, let
  • t ja r the moment that train t entering the point j of the station system on its route r;
  • t jd r the moment that train t leaving the point j of the station system on its route r;
  • V t r the average speed of train t on its route r
  • V t ir the approximated speed of train t going through the resource i on its route r.
  • Step2.1 if resource j the successor of the resource i on its operation direction , then
  • T te ri t ja r – distance(i,j)/V t r +trainlength(t)/V t ir
  • Step2.2 if resource j the predecessor of the resource i on its operation direction , then
  • T te ri t jd r + distance(i,j)/V t r +trainlength(t)/V t ir
  • Step3 Return the start–end time for the occupation time of resource i [T ts ri , T te ri ] .
  • Step1 Constructing the start time list L i s for occupation of resource i by trains in t@s—TSDIS by ascending order ,that is
  • Step2 Constructing the end time list L i e for occupation of resource i by trains in t@s—TSDIS by ascending order , that is
  • Step5 If list L i e isn't empty, then return to step3 for check.
  • the original trains' occupation order is effective in the local level, that is the safety interval distance for the trains' occupation of the resources in the task list t@s—TSDIS is satisfied through which the optimized train route can be found.
  • this invention 'Calculation and Evaluation of Station Capacity for High Speed Railway' dedicated by Ph.D Jiamin Zhang constructs the high speed railway trains at station 'train service - demand intention' (t@s—TSDIS) set and regards the time needed to complete the set t@s—TSDIS as the new criteria to measure the station capacity of the high speed railway.
  • This invention proposes the global - local bilevel model system & optimizing flow to calculate & evaluate the station capacity (Fig. 1)which optimizes the trains' occupation order on global level and checks the feasibility of the order on local level. On global level this invention proposes the flow chart for optimization of trains occupation order of station tracks(Fig.
  • this invention constructs the Resource Tree Conflict Graph (RTCG) and proposes algorithm1 & algorithm2 to check the conflicts and optimize the train routes by the interactive feedback between global level and local level where the composition of the switches and lines of each subregion is considered in details.
  • RTCG Resource Tree Conflict Graph
  • algorithm1 & algorithm2 proposes algorithm1 & algorithm2 to check the conflicts and optimize the train routes by the interactive feedback between global level and local level where the composition of the switches and lines of each subregion is considered in details.
  • This invention illustrates its practical application on certain high speed railway station BJSS.
  • This bilevel model system and its solution flow algorithm invented by this invention can calculate and evaluate the capacity not only of the station but also of the more complex railway junction.
  • the best mode of the invention is that it considers the dynamics of the trains operation fully and constructs the high speed railway trains at station service–demand intention set ( t@s—TSDIS ), regarding the time needed for the occupation of the infrastructure to complete the proposed set t@s—TSDIS as the standard for calculation & evaluation of the station capacity, proposing global—local bilevel model & optimizing flow chart to calculate & evaluate the station capacity and recognize bottleneck.
  • This bilevel model system and its solution flow algorithm can not only apply to the station but also the more complex railway junction for capacity calculation & evaluation (such as Fig. 1, Fig.6, Fig. 7 ,Fig. 8, Fig. 9, Fig. 10, Fig. 12, Fig. 20, algorithm1 & algorithm2 and Table 1—Table 10 ).
  • the mode for invention considers the dynamics of the trains operation fully and constructs the high speed railway trains at station service–demand intention set ( t@s—TSDIS ), regarding the time needed for the occupation of the infrastructure to complete the proposed set t@s—TSDIS as the standard for calculation & evaluation of the station capacity, proposing global—local bilevel model & optimizing flow chart to calculate & evaluate the station capacity and recognize bottleneck.
  • This bilevel model system and its solution flow algorithm can not only apply to the station but also the more complex railway junction for capacity calculation & evaluation (such as Fig. 1, Fig.6, Fig. 7 ,Fig. 8, Fig. 9, Fig. 10, Fig. 12, Fig. 20, algorithm1 & algorithm2 and Table 1—Table 10 ).
  • Certain high speed railway station BJSS is the original station , which has the average speed yard, the high speed yard, the intercity yard and 24 station tracks totally.
  • the topology of BJSS is illustrated as Fig. 11 in Drawings section.
  • the task list t@s—tsdis for station BJSS in certain period is illustrated as Table 1. list of t@s-tsdis in certain period NO.
  • the train lines occupy GS1 include gs1l6, gs1l12 and gs2l18, using the optimized calculation principle of this invention to construct the event graph and optimize the occupation order as following:
  • the event graph for trains occupation on GS1 can be referred in Fig .12 in the Drawings section.
  • This invention uses the defined defuzzy function deFuzzy[a, b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program.
  • the mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the original trains occupation order on GS1 gs1l6-gs1l12-gs2l18 is illustrated as following:
  • This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program.
  • the mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the first mutation of trains occupation order on GS1 gs1l6-gs2l18-gs1l12 is illustrated as following:
  • This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program.
  • the mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the second mutation of trains occupation order on GS1gs2l18-gs1l6-gs1l12 is illustrated as following:
  • This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program.
  • the mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the third mutation of trains occupation order on GS1gs1l6-gs1l12-gs2l18 is illustrated as following:
  • TZZ3 is smaller than TZZ2, so TZZ3 is saved.
  • This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program.
  • the mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the fourth mutation of trains occupation order on GS1gs1l6-gs2l18-gs1l12 is illustrated as following:
  • TZZ4 is smaller then TZZ3 and is saved.
  • This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program.
  • the mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the fifth mutation of trains occupation order on GS1 gs2l18-gs1l6- gs1l12 is illustrated as following:
  • This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program.
  • the mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the fifth mutation of trains occupation order on GS1 gs2l18-gs1l12-gs1l6 is illustrated as following:
  • this invention Using defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks & C# Console Program, this invention get the optimized trains occupation order on tracks of each station yard which are illustrated as Table 2, Table 3 and Table 4.
  • CST code for station track
  • ootl order for occupation of train lines
  • train lines train lines—tl
  • tos type of operation at station
  • Random defuzzy for item of operation time at station rdiots (Arrival time for entering station—ates, Departure time from station–dts)
  • wsct Startt time for occupation—sto, End time for occupatio—eto).
  • CST code for station track
  • ootl order for occupation of train lines
  • tl train lines
  • tos type of operation at station
  • AU Arriv Upstream
  • ODD Oil Downstream
  • TD Temporal Departure Downstream
  • TD Temporal Departure
  • DTD&ES Departure from Train's depot and Entering Station
  • smo start moment for occupation in certain cycle time
  • emo end moment for occupation in certain cycle time
  • rdiots Random defuzzy for item of operation time at station— (ates —Arrival time for entering station , sts —Stop time at station , dts —Departure time from station
  • ptld Pulling out time from lead track
  • ttlt Turnback time from lead track
  • wsct Within the same cycle time for the whole yard (sto — Start time for occupation, eto — End time for occupatio).
  • CST Code for station track , ootl—Order for occupation of train lines, tl— Train lines, tos—Type of operation at station(dtmd—Departure from multi-unit train depot, dtlr —Departure from lines where trains reserved, td—Turnback for departure, au—Arrival upstream), smo— start moment for occupation in certain cycle time, emo—end moment for occupation in certain cycle time, rdiots—Random defuzzy for item of operation time at station ( ates—Arrival time for entering station, dt—Dwelling time,dtls —departure time for leaving station, ptlt— Pulling out time from lead track, ttld—Turnback time from lead trackstoStart time for occupation), wsct—Within the same cycle time for the whole yard (sto—start time for occupation, eto— End time for occupation ). notion again: It doesn't exist the conflicts the
  • each train line in t@s—tsd passes by the throat area a and b in the average speed yard, it passes by the throat area c and d in the high speed yard and it passes by the throat area c,e,f and g in the intercity yard.
  • the conflicts among the trains' lines in BJSS station can only happen at the throat area c where the trains operations for pulling out of their deposit cross at the high speed yard and the intercity yard.
  • the train lines cjl29c, cjl28c, cjl30c and cjl31c in the intercity yard may conflict with the train lines gs2l18, gs1l12, and gs1l7 in the high speed yard in throat area c in station BJSS and its RTCG is illustrated as Fig. 20.
  • sn Serieal number
  • tl train linesl
  • tr Train route
  • setrotas Start-end time for resource occupation in throat region a of station
  • setrotcs Startt-end time for resource occupation in throat region c of station
  • setrost Startt-end time for resource occupation of station track
  • Start time st
  • End time et.
  • sn seriesal number
  • tl Train lines
  • tr Train route
  • tos Type of operation at station (dtmd—Departure from multi-unit train depot, dtlr —Departure from lines where trains reserved, td—Turnback for departure, au—Arrival upstream), setrotcs—Start-end time for resource occupation in throat area c of station, setrotes—Start-end time for resource occupation in throat area e of station, setrotfs—Start-end time for resource occupation in throat area f of station, setrotgs—Start-end time for resource occupation in throat area g of station, setrost —Start-end time for resource occupation of station track, st—Start time ,et—End time.
  • the pure number in the circles represents the code number of the main switches in the throat area c of station BJSS
  • the combination of alphabetic and the number represents the code number of the train line
  • the arrowed connecting line represents the route conflicts in the corresponding switch generated by two or more than two train lines for their occupation in the set of the start-end time interval.
  • the trains' occupation order of the station tracks for t@s—TSDIS in certain period for certain station BJSS on its global level is partly feasible after RTCG conflict check, namely, it exists conflict for the pulling out of the train lines cjl30c and gs1l7 from their deposit in the throat area c.
  • CST Code for station track
  • ootl Order for occupation of train lines
  • tl Train lines
  • tos Type of operation at station(DTMD—Departure from multi-unit train depot
  • DTLR Departure from lines where trains reserved
  • TD Transmissionnback for departure
  • AU–Arrival upstream smo— start moment for occupation in certain cycle time
  • emo end moment for occupation in certain cycle time
  • rdiots Random defuzzy for item of operation time at station ( ates—Arrival time for entering station, dt—Dwelling time,dtls —departure time for leaving station
  • ptlt Pulling out time from lead track
  • ttld Turnback time from lead tracks to Start time for occupation)
  • wsct Within the same cycle time for the whole yard (sto—start time for occupation, eto— End time for occupation ).
  • the bottleneck can be recognized.
  • the capacity of the station except certain subarea it calculates and evaluates the consumption of the resource occupation time for each subarea is illustrated as Table 10 and the biggest one is the bottleneck affected the capacity.
  • it Because it has the operation of the trains for pulling out-entering the deposit and the turnback for departure in the throat area c, which is relatively complex and occupy the resource for the longest time, it forms the critical capacity bottleneck for station BJSS in the throat area c.
  • time needed to complete t@s-TSDIS list in each subregion of BJSS unit minute st a B c d e f g to 22.44 22.44 189.45 74.18 8.08 27.82 107.85

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Train Traffic Observation, Control, And Security (AREA)

Abstract

Considering the dynamics of train operation based on the research by Jiamin Zhang(2011), this invention 'Calculation and Evaluation of Station Capacity for High Speed Railway' dedicated by Ph.D Jiamin Zhang constructs the high speed railway trains at station 'train service - demand intention' (t@s— TSDIS) set and regards the time needed to complete the set t@s— TSDIS as the new criteria to measure the station capacity of the high speed railway. This invention proposes the global— local bilevel model system & optimizing flow to calculate & evaluate the station capacity which optimizes the trains' occupation order on global level and checks the feasibility of the order on local level. On global level this invention proposes the flow chart for optimization of trains occupation order of station tracks (Fig. 8) and Max-plus calculation eigenvalue & eigenvector for each train occupation of track (Fig. 9) using the Fuzzy Time High Level Petri nets (FTHNs), the Max Plus algebra, the simulated annealing algorithm and Scilab ((Claude Gomez, etc), 1999) software to determine the trains' occupation order of the station tracks, where station is abstracted as topology of main nodes presuming that trains can pass each node unlimitedly targeting at minimization of time needed to complete the task list t@s— TSDIS. On local level this invention constructs the Resource Tree Conflict Graph (RTCG) and proposes algorithm 1 & algorithm2 to check the conflicts and optimize the train routes by the interactive feedback between global level and local level where the composition of the switches and lines of each subregion is considered in details. This invention illustrates its practical application on certain high speed railway station BJSS. This invention can calculate and evaluate the capacity not only of the station but also of the more complex railway junction.

Description

CALCULATION AND EVALUATION OF STATION CAPACITY FOR HIGH SPEED RAILWAY
This invention proposed by Ph.D Jiamin Zhang belongs to the technical field of the application of railway transportation , targeting at calculation & evaluation of station capacity of high speed railway and recognizing its capacity bottleneck.
The operations around the station region is relatively complex affected by many factors, such as the layout of the station yard, the selection of the station's facility, the unequilibrium of the trains' arrival and departure, the percentage of the types of the trains' arrival and departure, the usage scheme of the station tracks, the cooperation between the station's throat area and its arrival-departure tracks, the running lines of the multiple train unit depot, etc, according to Wen Dong (2006). Based on the characteristics of the capacity's systematic and effectiveness, Hu Anzhou & Yang Hao (1994) proposed the conception of the total effective capacity of the railway system's facility group and discussed the problem of the conception, the calculation, the usage and the reinforcement of the railway capacity in the view of dynamics, which laid the foundation for the later research on the railway capacity deeply and totally.
Dan Max Burkolter (2005) studied the station capacity regarding the train routes as the criteria and taking the problem of routes assignment for capacity calculation as the strategy one, which in the global level it constructs the feasible timetable using Petri net but it didn't consider the uncertainty of the trains operation at the station, and in the local level it checks the route conflicts using the conflict graph but it didn't reflect the location of the conflict as well as the selection preference for the routes. De Kort A.F., Heidergott B., Van Egmond R.J. and Hooghiemstra G. (1999) used the queueing theory to construct the different forms of the model (such as markovian and semi-markovian regarding the station as the single server or the multiple servers) based on the different service rules and they researched the maro-capacity of the station by calculating the scheduled waiting time and the delay time. Peter J. Zwaneveld, Leo G. Kroon, Stan P.M. van Hoesel (2001) & Richard Lusby, Jesper Larsen, David Ryan (2006) targeting at the maximization of the feasible train routes in the station, they abstracted the problem of the feasible train routes in the station as the weighted node packing model and solved it by branch and bound method, which can deal with plenty of train routes in large scale but once it's done it couldn't change the occupation time of the resource on the routes. Xavier Delorme, Xavier Gandibleux, Joaquin Rodriguez (2004) constructed the hybrid model and solved it by the heuristic algorithm. Jiamin Zhang (2011) compared and analyzed several strategies for the calculation and evaluation of the station capacity, she prefers the global-local bilevel model to calculate and evaluate the station capacity which is illustrated as Fig. 1 in Drawings section.
Technical Problem
All of the effects of the interactions between the fixed devices and their active counterparts are embodied in the temporal dimension, according to the insufficiency of the background art which taking the number of the trains or the train routes as the criteria for the station capacity, considering the dynamics of train operation based on the research by Jiamin Zhang(2011) , this invention 'Calculation and Evaluation of Station Capacity for High Speed Railway' dedicated by Ph.D Jiamin Zhang constructs the high speed railway trains at station 'train service - demand intention set' (t@s—TSDIS) and regards the time needed to complete the set t@s—TSDIS as the new criteria to measure the station capacity of the high speed railway. This invention proposes the global — local bilevel model system & optimizing flow to calculate & evaluate the station capacity which optimizes the trains' occupation order on global level and checks the feasibility of the order on local level (Fig. 1). On global level this invention proposes the flow chart for optimization of trains occupation order of station tracks(Fig. 8) and Max-plus calculation eigenvalue & eigenvector for each train occupation of track(Fig. 9) using the Fuzzy Time High Level Petri nets (FTHNs), the Max Plus algebra, the simulated annealing algorithm and Scilab ((Claude Gomez, etc),1999) software to determine the trains' occupation order of the station tracks, where station is abstracted as topology of main nodes presuming that trains can pass each node unlimitedly targeting at minimization of time needed to complete the task list t@s-TSDIS. On local level this invention constructs the Resource Tree Conflict Graph (RTCG) and proposes algorithm1 & algorithm2 to check the conflicts and optimize the train routes (Fig. 10) by the interactive feedback between global level and local level where the composition of the switches and lines of each sub–region is considered in details. This invention illustrates its practical application on certain high speed railway station BJSS. This invention can calculate and evaluate the capacity not only of the station but also of the more complex railway junction.
Technical Solution
1. Construction of t@s—TSDIS (the train at station 'train service - demand intention set' )
The core element in t@s—TSDIS is the train, for which the 'service' means it designates the physical nodes for the trains entering and leaving the station as well as the platforms for the trains to dwell (or pass by) according to the station's physical connection direction & topology, the 'demand' designates the speed & class type for each train in t@s—TSDIS from the angel of the passenger demand and the connections among the trains, etc. Besides the time period for calculation, other elements in t@s—TSDIS are as following:
(1)Type & Number of Trains
According to the passenger demand it determines the type and service priority for each train in t@s—TSDIS. The high speed railway serves the 'passenger', so the ideas of trains routing should be changed as the comprehensive balance between the passenger demand and the capacity supply when calculating & evaluating the station capacity. The maximum passenger flow in t@s-TSDIS depends on its practical demand. It presumes di p as the passenger demand within the time period, Ci as the given train capacity corresponding to its type ¡,
Figure zhangjiamins-appb-I000002
as the type i train's loaded factor satisfied a certain service level, then
Figure zhangjiamins-appb-I000003
would be the number of trains planned in t@s—TSDIS.
(2)Representation of Occupation Time on Infrastructure
Considering the dynamics of the capacity incurred by the uncertainty of the train's operation time and the preserved buffer time when scheduling for the railway's practical operation to ensure the service level, this invention discretizes the calculation time period by the equal time interval (e.g. minute), then endows a certain fuzzy probability for the trains occupation or passing by the physical station infrastructure in a certain time interval. The fuzzy time function is the mapping function from the time scalar (non-negative real number) to the real number section [0,1]. This invention adopts the fuzzy triangle time function tackling with the uncertainty and dynamics, which lets the fuzzy triangle time slot t(a,b,c) represents the time items for the trains entering the station, leaving the station, passing by the simplified sub–region and dwelling at the station, etc.
Let hi denote the minimum time interval for the trains running on the line section connecting with the station, [tb,te] denote within time period, according to the time interval for the trains running on the section, then
Figure zhangjiamins-appb-I000004
is the possible arrival time for each train j in t@s-TSDIS within time period,
Figure zhangjiamins-appb-I000005
is the departure time of the arrival–stop–departure train, where dwell(tj) denotes the dwelling time at station,
Figure zhangjiamins-appb-I000006
denotes the allowable possible deviation time or certain fuzzy time probability, then arr(tj) and dep(tj)can be described as the fuzzy triangle time slot.
2. Train of Thought for Calculation & Evaluation of Station Capacity Based on Bilevel Model
This invention constructs the high speed railway trains at station 'train service - demand intention set' (t@s—TSDIS) and regards the time needed to complete the set t@s—TSDIS as the new criteria to measure the station capacity of the high speed railway. This invention proposes the global — local bilevel model system & optimizing flow to calculate & evaluate the station capacity which optimizes the trains' occupation order on global level and checks the feasibility of the order on local level (Fig. 1). On global level this invention proposes the flow chart for optimization of trains occupation order of station tracks(Fig. 8) and Max-plus calculation eigenvalue & eigenvector for each train occupation of track(Fig. 9) using the Fuzzy Time High Level Petri nets (FTHNs), the Max Plus algebra, the simulated annealing algorithm and Scilab ((Claude Gomez, etc),1999) software to determine the trains' occupation order of the station tracks, where station is abstracted as topology of main nodes presuming that trains can pass each node unlimitedly targeting at minimization of time needed to complete the task list t@s—TSDIS. On local level this invention constructs the Resource Tree Conflict Graph (RTCG) and proposes algorithm1 & algorithm2 to check the conflicts and optimize the train routes (Fig. 10) by the interactive feedback between global level and local level where the composition of the switches and lines of each sub–region is considered in details.
3. Model Construction & Solution Algorithm for Global Level Based on FTHNs
(1) Fuzzy time high level Petri nets (FTHNs)
FTHNs is a sort of Petri nets which introduces the fuzzy set theory related to time and uses four sorts of fuzzy time function:
Figure zhangjiamins-appb-I000007
represent the fuzzy time stamp, fuzzy enable time, fuzzy occurrence time and fuzzy delay time respectively.
Figure zhangjiamins-appb-I000008
represents the token colored by V locating in place P, where
Figure zhangjiamins-appb-I000009
is its fuzzy time stamp.
(2) Static Defination & Dynamic Description for Interaction between Trains & Station t@s-FTHNs
The trains operation types in station can be classified as original departure, final arrival, turning out, passing by (stopping then leaving is included). According to t@s-TSDIS and FTHNs, the static definition for Interaction between Trains & Station t@s—FTHNs can be as following:
1) — finite type (color) set , including the station track strack ( where ptrack denotes the platform track), train at station service-demand set t@s—TSDIS.
2) P — finite place set, where Pin and Pout denote the place for the trains entering 　 and leaving the station tracks respectively, Pfree and Pbusy denote the station tracks are available and busy respectively.
3) T— finite transition set, where Tarr, Tdep and Tsyn denote the train's arrival event, departure event and connection event.
4) the arc set represents the flow relationship.
5) — color function.
6)D -- delay set related to 　 arcs (T x P)
7)
Figure zhangjiamins-appb-I000014
represents colored token, which represents the train j in t@s—TSDIS arrives the place strack(i) with time possibility The trains operation is a dynamic process, there are three variables changing with time such as its location, speed and accelerating & decelerating speed, which are the functions of time.
8) Let (t, bin, bout) denotes the event of a transition, where bin, bout represents the consumed token set and the produced token set respectively.
In the global level the station is abstracted as the composition of a series of tracks and nodes, assigning the trains operation to the parallel tracks according to a certain order without considering the capacity limited. According to the machine schedule theory, it's regarded as the non-preemption job shop (train operation) problem on a certain number of the same type of machines (parallel tracks) (Fig. 2 — Fig. 7 in Drawings section ).
(3) t@s—FTHNs Allocation Strategy of Station Track Resources
Regarding the minimization of the occupation time of the infrastructure for the trains occupation order on the station tracks on the global topology as the non-preemption machine job shop problem: for the situation of the multi-trains occupying the multi-tracks (Fig. 4), it proposes the set ði for the sub–region in which the elements are composed of the train set
Figure zhangjiamins-appb-I000016
passing by the sub–region and µð i the number of the parallel tracks in the sub–region; then one division of the parallel tracks by the train set consists of
Figure zhangjiamins-appb-I000017
which is a solution for the occupation order of the station tracks. The elements train set in each sub-region exist an order corresponding to parallel tracks ( where in Fig. 4 M0(Pin=n), M0(Pfree)=1) . Taking two trains occupy one station track as the example ( where in Fig. 4 M0(Pin)=2, M0(Pfree)=1) to illustrate the t@s—FTHNs allocation strategy of the tracks (where in Fig. 5 two trains a and b running on one station track strack(i)).
According to possibility calculation (using latest operator to calculate & evaluate fuzzy enable time, using min operator to calculate & evaluate fuzzy occurrence time, to determine fuzzy delay time and calculate the fuzzy time of the output place ) FTHNs allocates the tracks for two trains a, b and provide the information for the partial order . Generally, for the fuzzy triangle, it takes:
Figure zhangjiamins-appb-I000018
. However it doesn't get the total sequence, trains operation order on the station tracks belong to the non-preemption problem. The occupation order for multi-trains share one track can be determined by introducing the corresponding strategy and turn it to the t@s—FTHNs event graph which doesn't need further decision. For more details one can refer to Fig. 6 for the strategy flow chart and Fig. 7 in Drawings section for the t@s—FTHNs event graph.
4. Flow Chart for t@s—FTHNs Optimization of Trains' Occupation Order of Station Tracks on Global Level
According to t@s—FTHNs description of the interaction between trains and station and t@s—FTHNs allocation strategy for track resources, this invention constructs t@s—FTHNs event graph for the task list t@s-TSDIS and designs the optimized algorithm for the trains occupation order on station tracks (Fig.8 in Drawings section). According to Y. Zhou. T. Murata. T Defanti. (2000) this invention defines the defuzzy function deFuzzy[a, b, c]:
(1)Amplifying 100 times of the simulation time and generating a random value atime within [100a, 100c];
(2)if atime=100c, then return atime=100c;
(3)if atime locates within [100a, 100b] or (100b, 100c], then producing a random value V within [0, 1] and calculating the possibility value D(atime) ; if D(atime) is greater than or equal to V , then return atime, else return to (1).
Let ß'ji denote the holding time of the place between transition ti and tj which relating to the track resources directly, that is ß'ji=o'(tj)-o'(ti). Using deFuzzy[a, b, c] as ßji=deFuzzy(ß'ji)=deFuzzy(o'(tj)-o'(ti))=o(tj)-o(ti) to defuzzy the fuzzy time slot of the t@s-FTHNs time event graph and construct the Max Plus matrix model for each train occupation the track as following:
Defines the vector X(K)=(x1(k),x2(k),…)' as the system status, where xj(k) denotes transition tj is enabled in the kth time, then
Figure zhangjiamins-appb-I000019
where xi(k) is the directly upstream transition xj(k). Based on this, using Howard's (Farlow, Kasie Geralyn (2009)) algorithm and calling for Scilab software to calculate the Max Plus Model and making the loops mutation using the simulated annealing algorithm for the calculated eigenvalue and eigenvector. The flow chart of the eigenvalue & eigenvector calculation for each track & trains using the Max–plus & Scilab software calling for Howard's algorithm is illustrate in Fig. 9 in Drawings section .
5.Conflict Check and Optimized Route Selection on Local Level Based on RTCG on Local Level
(1)Basic Train of Thought
The proposed t@s—TSDIS trains' occupation order of the station tracks on the global level is on the assumption of the ignorance of the constraints of the interval time of the resource allocation and the route conflicts, it needs to check the conflicts and optimize the routes based on the logic relationship illustrated in Fig. 1: in the local level it regards the station tracks, the switches and the signals as the resources, based on the concrete resource topology using the depth-first search algorithm to enumerate the possible routes set R for the trains in task list t@s—TSDIS between the entering points of the station system and the leaving points of the station system and then optimize the trains' occupation order of the station tracks. For the t@s—TSDIS trains' occupation order of the station tracks on the global level, it determines the start–end time for the occupation of the corresponding resources, that is it proposes algorithm1 to backstep the moment that the trains passing by each subregion from its entering points of the station system according to the trains' arrival time and deduce the moment that the trains' passing by each subregion from their departure of the station tracks to the leaving points of the station system based on the trains' departure time of the station. Based on these, it proposes algorithm 2 to construct the resource tree conflict graph (RTCG) to check and optimize the trains route.
(2)Construction of Improved RTCG and Conflict Check of Trains' Routes
According to Martin Fuchsberger (2007), it represents the routes between the trains' entering points of the station system and the leaving points of the station system for the optimized trains occupation of the task list t@s—TSDIS in the form of the tree structure. The improved RTCG proposed by this invention regards the station tracks and the switches as the resources, it constructs the algorithm1 to determine the start-end time of resource occupation on the trains' operation route at first and algorithm2 to find the conflict group for check the conflicts from the set of the start–end time of the resources occupation, that is it groups the conflicts for all of the possible overlap time interval of the resource occupation.
algorithm1 : calculation of start–end time for occupation of resource i by train t on its operation route r.
input : trains occupation order of station tracks and the set of trains routes in t@s—TSDIS.
output: the start–end time Tts ri, Tte ri for the occupation of the resource i by train t in t@s—TSDIS on its route r. Let
Tts ri = start time of occupation of resource i by train t in t@s—TSDIS in its route r;
Tte ri = end time of occupation of resource i by train t in t@s—TSDIS in its route r.
Step1. For the resource i occupied by train t in t@s—TSDIS on its route r, let distance(i, j) denote the physical distance between the resource i and j on the same route and train length(t) denote the length of the train t.
Step2. according to the trains occupation order in t@s—TSDIS, let
tja r = the moment that train t entering the point j of the station system on its route r;
tjd r =the moment that train t leaving the point j of the station system on its route r;
Vt r = the average speed of train t on its route r;
Vt ir = the approximated speed of train t going through the resource i on its route r.
Step2.1 if resource j the successor of the resource i on its operation direction ， then
Tts ri = tja r – distance(i,j)/ Vt r ,
Tte ri = tja r – distance(i,j)/Vt r +trainlength(t)/Vt ir
Step2.2 if resource j the predecessor of the resource i on its operation direction ， then
Tts ri = tjd r + distance(i,j)/ Vt r ,
Tte ri = tjd r + distance(i,j)/Vt r +trainlength(t)/Vt ir
Step3. Return the start–end time for the occupation time of resource i [Tts ri , Tte ri ] .
algorithm2 : check the conflict group CGi of the resource occupation from the set of the start–end time..
input: the set of start–end time [Tts ri , Tte ri ] for occupation of resource i on its route r.
output: the conflict group CGi for the time occupation of resource i
Step1. Constructing the start time list Li s for occupation of resource i by trains in t@s—TSDIS by ascending order ，that is
Li s ={li,n s |li,n s =Tts ri and li,n s <li,n+1 s , n=0,1,…}
Step2. Constructing the end time list Li e for occupation of resource i by trains in t@s—TSDIS by ascending order ， that is
Li e ={li,n e |li,n e =Tte ri and li,n e <li,n+1 e , n=0,1,…}
Step3. Let ts denote the earliest time for resource occupation in the current Li s list ， te denote the end time corresponding to ts , let new_end_time=te .
Step4. For each element in the current list Li s let li,n s = Tts ri ， if li,n s < new_end_time , then add [Tts ri , Tte ri ] to the conflict group CGi ; else delete Tts ri from the list Li s , delete Tte ri from list Li e .
Step5. If list Li e isn't empty, then return to step3 for check.
It uses algorithm1 to determine the start-end time for resource occupation on the trains' operation route and algorithm2 to check the conflict group from the set of start–end time for the resource occupation. Then it checks the feasibility of the trains' occupation order in the task list t@s—TSDIS on the global level, which can be concluded as basic feasible, partly feasible and unfeasible and illustrated in Fig. 10 in Drawings sectioin :
Feasible. The original trains' occupation order is effective in the local level, that is the safety interval distance for the trains' occupation of the resources in the task list t@s—TSDIS is satisfied through which the optimized train route can be found.
Partly feasible. If the resource occupation interval for the trains in the task list t@s—TSDIS isn't satisfied which exists the conflicts, it feedbacks to the global level for the adjustment of the trains' occupation order of the station tracks for the later check.
Unfeasible. If the number of the trains in the task list t@s—TSDIS whose resource occupation safety interval isn't satisfied is greater than a certain threshold and it would consume a lot of time and energy to feedback with the global level, then a new trains' occupation order of the station tracks for t@s— TSDIS is generated and checked again.
Advantageous Effects
Considering the dynamics of train operation based on the research by Jiamin Zhang(2011) , this invention 'Calculation and Evaluation of Station Capacity for High Speed Railway' dedicated by Ph.D Jiamin Zhang constructs the high speed railway trains at station 'train service - demand intention' (t@s—TSDIS) set and regards the time needed to complete the set t@s—TSDIS as the new criteria to measure the station capacity of the high speed railway. This invention proposes the global - local bilevel model system & optimizing flow to calculate & evaluate the station capacity (Fig. 1)which optimizes the trains' occupation order on global level and checks the feasibility of the order on local level. On global level this invention proposes the flow chart for optimization of trains occupation order of station tracks(Fig. 8) and Max–plus calculation eigenvalue & eigenvector for each train occupation of track(Fig. 9) using the Fuzzy Time High Level Petri nets (FTHNs), the Max Plus algebra, the simulated annealing algorithm and Scilab ((Claude Gomez, etc),1999) software to determine the trains' occupation order of the station tracks, where station is abstracted as topology of main nodes presuming that trains can pass each node unlimitedly targeting at minimization of time needed to complete the task list t@s—TSDIS. On local level this invention constructs the Resource Tree Conflict Graph (RTCG) and proposes algorithm1 & algorithm2 to check the conflicts and optimize the train routes by the interactive feedback between global level and local level where the composition of the switches and lines of each subregion is considered in details. This invention illustrates its practical application on certain high speed railway station BJSS. This bilevel model system and its solution flow algorithm invented by this invention can calculate and evaluate the capacity not only of the station but also of the more complex railway junction.
Description of Drawings
Please refer all of the mentioned figures to the Drawings section correspondingly.
 Best Mode
The best mode of the invention is that it considers the dynamics of the trains operation fully and constructs the high speed railway trains at station service–demand intention set ( t@s—TSDIS ), regarding the time needed for the occupation of the infrastructure to complete the proposed set t@s—TSDIS as the standard for calculation & evaluation of the station capacity, proposing global—local bilevel model & optimizing flow chart to calculate & evaluate the station capacity and recognize bottleneck. This bilevel model system and its solution flow algorithm can not only apply to the station but also the more complex railway junction for capacity calculation & evaluation (such as Fig. 1, Fig.6, Fig. 7 ,Fig. 8, Fig. 9, Fig. 10, Fig. 12, Fig. 20, algorithm1 & algorithm2 and Table 1—Table 10 ).
Mode for Invention
The mode for invention considers the dynamics of the trains operation fully and constructs the high speed railway trains at station service–demand intention set ( t@s—TSDIS ), regarding the time needed for the occupation of the infrastructure to complete the proposed set t@s—TSDIS as the standard for calculation & evaluation of the station capacity, proposing global—local bilevel model & optimizing flow chart to calculate & evaluate the station capacity and recognize bottleneck. This bilevel model system and its solution flow algorithm can not only apply to the station but also the more complex railway junction for capacity calculation & evaluation (such as Fig. 1, Fig.6, Fig. 7 ,Fig. 8, Fig. 9, Fig. 10, Fig. 12, Fig. 20, algorithm1 & algorithm2 and Table 1—Table 10 ).
Industrial Applicability
Certain high speed railway station BJSS is the original station , which has the average speed yard, the high speed yard, the intercity yard and 24 station tracks totally. The topology of BJSS is illustrated as Fig. 11 in Drawings section. The task list t@s—tsdis for station BJSS in certain period is illustrated as Table 1. list of t@s-tsdis in certain period
NO. Code for trains line Occupied station track Trains route Operation type in station Station yard
0 psl0 ps1 a-ps1-b Pass by Average speed yard
1 psl1 ps2 a-ps2-b Pass by
2 psl2 ps3 a-ps3-b Pass by
3 psl3 ps4 a-ps4-b Pass by
4 psl4 ps5 a-ps5-b Pass by
5 psl5 ps1 a-ps1-b Pass by
6 gs1l6 GS8 , GS1 c-d-GS8-e-GS1-d-c Original departure downstream (consider the operation of the electronic unit trains ' out of their deposit ) High speed yard
7 gs1l7 GS9 , GS2 c-d-GS9-e-GS2-d-c High speed yard
8 gs1l8 GS10, GS3 c-d-GS10-e-GS3-d-c High speed yard
9 gs1l9 GS11 , GS4 c-d-GS11-e-GS4-d-c High speed yard
10 gs1l10 GS12 , GS5 c-d-GS12-e-GS5-d-c High speed yard
11 gs1l11 GS8, GS6 c-d-GS8-e-GS6-d-c High speed yard
12 gs1l12 GS1, GS7 c-d-GS1-e-GS7-d-c The turnback operation for departure after arrival upstream and pull out at the end of the station High speed yard
13 gs1l13 GS2, GS8 c-d-GS2-e-GS8-d-c High speed yard
14 gs2l14 GS3, GS9 c-d-GS3-e-GS9-d-c High speed yard
15 gs2l15 GS4, GS10 c-d-GS4-e-GS10-d-c High speed yard
16 gs2l16 GS5, GS11 c-d-GS5-e-GS11-d-c High speed yard
17 gs2l17 GS6, GS12 c-d-GS6-e-GS12-d-c High speed yard
18 gs2l18 GS1, GS7 c-d-GS1-e-GS7-d-c High speed yard
19 gs2l19 GS2, GS8 c-d-GS2-e-GS8-d-c High speed yard
20 gs2l20 GS3, GS9 c-d-GS3-e-GS9-d-c High speed yard
21 gs2l21 GS4, GS10 c-d-GS4-e-GS10-d-c High speed yard
22 cjl22f CJ1 f-g-CJ1-h departure from the lines reserving the trains Intercity yard
23 cjl23f CJ1 f-g-CJ1-h Intercity yard
24 cjl24f CJ1 f-g-CJ1-h Intercity yard
25 cjl25f CJ2 f-g-CJ2-h Intercity yard
26 cjl26f CJ2 f-g-CJ2-h Intercity yard
27 cjl27f CJ2 f-g-CJ2-h Intercity yard
28 cjl28c CJ1 c-d-f-CJ1-h departure from the multi-unit train deposit Intercity yard
29 cjl29c CJ1 c-d-f-CJ1-h Intercity yard
30 cjl30c CJ1 c-d-f-CJ1-h Intercity yard
31 cjl31c CJ2 c-d-f-CJ2-h Intercity yard
32 cjl32h CJ6, CJ3 h-CJ6-g-QCI-g-CJ3-h operation of turnback Intercity yard
33 cjl33h CJ6, CJ3 h-CJ6-g-QCI-g-CJ3-h Intercity yard
34 cjl34h CJ6, CJ5 h-CJ6-g-QCI-g-CJ5-h Intercity yard
35 cjl35h CJ7, CJ4 h-CJ7-g-QCI-g-CJ4-h Intercity yard
36 cjl36h CJ7, CJ4 h-CJ7-g-QCI-g-CJ4-h Intercity yard
1. Illustration of Optimized Calculation Process for Occupation Order of Trains on Tracks
Taking the station track GS1 in high speed yard as example to illustrate the optimized calculation process for the occupation order of trains on the tracks of high speed yard gsyard. The train lines occupy GS1 include gs1l6, gs1l12 and gs2l18, using the optimized calculation principle of this invention to construct the event graph and optimize the occupation order as following:
(1). event graph for train's occupation on station track
let
The event graph for trains occupation on GS1 can be referred in Fig .12 in the Drawings section.
(2) The original trains occupation order on GS1 in high speed yard (gs1l6-gs1l12-gs2l18)
This invention uses the defined defuzzy function deFuzzy[a, b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program. The random defuzzy results for the time items during the original trains occupation order on GS1can be referred in Fig.13 in the Drawings section. The mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the original trains occupation order on GS1 gs1l6-gs1l12-gs2l18 is illustrated as following:
Mathematical formulae:
By operating Scilab software can get the transposed eigenvector v'and eigenvalue l as:
v'=(117.2, 158.68, 164.68, 166.76, 127.2, 168.68, 174.68 176.76, 152.38, 179.68, 198.24, 201.94 )', l=49.56,, that is the original eigenvalue is TZZ0=49.56 and the original eigenvector is TZXL0 =(117.2, 158.68, 164.68, 166.76, 127.2, 168.68, 174.68 176.76, 152.38, 179.68, 198.24, 201.94 )' , where xgs1l6arr=158.68, xgs1l6dep=164.68, xgs1l12arr=168.68, xgs1l12dep=174.68, xgs2l18arr=179.68 , xgs2l18dep=198.24.
(3).The first mutation of trains occupation order on GS1 in high speed yard (gs1l6-gs2l18-gs1l12)
This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program. The random defuzzy results for the time items during the first mutation of trains occupation order on GS1can be referred in Fig. 14 in the Drawings section. The mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the first mutation of trains occupation order on GS1 gs1l6-gs2l18-gs1l12 is illustrated as following:
Mathematical formulae:
By operating Scilab software can get the transposed eigenvector v' and eigenvalue l as:
v'=(166.29, 208.8, 215.8, 217.49, 206.49, 249, 256, 257.69, 190.42, 220.8,238, 241.62)', l=51.2, which represent the eigenvector TZZ1 and eigenvalue TZXL1 respectively. After simulated annealing algorithm optimization, save them and do the second mutation for the trains occupation order.
(4). The second mutation of trains occupation order on GS1 in high speed yard (gs2l18-gs1l6-gs1l12)
This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program. The random defuzzy results for the time items during the second mutation of trains occupation order on GS1can be referred in Fig. 15 in the Drawings sectioin. The mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the second mutation of trains occupation order on GS1gs2l18-gs1l6-gs1l12 is illustrated as following:
Mathematical formulae:
By operating Scilab software can get the transposed eigenvector v' and eigenvalue l as:
v'=(194.17, 236.14, 243.14, 245.18, 206.17, 248.14, 155.14, 257.18, 177.82, 208.13, 225.14, 228.83)', l=51.01, which represent the eigenvector TZZ2 and eigenvalue TZXL2 respectively.TZZ1 is greater than TZZ2, so TZZ2 is saved. Do the third mutation.
(5). the third mutation of the trains occupation order on GS1 in high speed yard is back to the original one (gs1l6-gs1l12-gs2l18)
This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program. The random defuzzy results for the time items during the third mutation of trains occupation order on GS1can be referred in Fig. 16 in the Drawings section. The mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the third mutation of trains occupation order on GS1gs1l6-gs1l12-gs2l18 is illustrated as following:
Mathematical formulae:
By operating Scilab software can get the transposed eigenvector v'and eigenvalue l as:
v'=(169.15, 210.8, 217.8, 219.35, 180.15, 221.8, 238.8, 230.35, 204.36, 233.8, 251, 254.65)', l=50.2, which represent the eigenvector TZZ3 and eigenvalue TZXL3 respectively. TZZ3 is smaller than TZZ2, so TZZ3 is saved. Do the fourth mutation.
(6). lower the temperature and mutate the fourth trains occupation order on GS1 according to (gs1l6-gs2l18-gs1l12)
This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program. The random defuzzy results for the time items during the third mutation of trains occupation order on GS1can be referred in Fig. 17 in the Drawings section. The mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the fourth mutation of trains occupation order on GS1gs1l6-gs2l18-gs1l12 is illustrated as following:
Mathematical formulae:
By operating Scilab software can get the transposed eigenvector v'and eigenvalue l as:
v'=(275.91, 298.12, 305.12, 306.93, 295.93, 336.14, 343.14, 344.95, 280.76, 309.12, 326.14, 329.78)', l=49.02, which represent the eigenvector TZZ4 and eigenvalue TZXL4 respectively. TZZ4 is smaller then TZZ3 and is saved.
(7). lower temperature and mutate fifth trains occupation order on GS1 according to gs2l18-gs1l6- gs1l12
This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program. The random defuzzy results for the time items during the third mutation of trains occupation order on GS1can be referred in Fig.18 in the Drawings section. The mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the fifth mutation of trains occupation order on GS1 gs2l18-gs1l6- gs1l12 is illustrated as following:
Mathematical formulae:
By operating Scilab software can get the transposed eigenvector v'and eigenvalue l as:
v'=(143.18, 187.16, 193.16, 194,22, 154.18, 198.16, 204.16, 205.22, -inf, 158.12, 176.12, 179.68)', l=51.04, which represent the eigenvector TZZ5 and eigenvalue TZXL5 respectively. Save the fifth mutation result and continue the simulated annealing optimization.
(8). lower temperature and mutate sixth trains occupation order on GS1 according to ( gs2l18-gs1l12- gs1l6)
This invention uses the defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks to compile the C# Console Program. The random defuzzy results for the time items during the third mutation of trains occupation order on GS1can be referred in Fig. 19 in the Drawings section. The mathematical formulae and the eigenvalue & eigenvector resulted from Scilab software for the fifth mutation of trains occupation order on GS1 gs2l18-gs1l12-gs1l6 is illustrated as following:
Mathematical formulae:
By operating Scilab software can get the transposed eigenvector v' and eigenvalue l as:
v'=(52.62, 95.06, 101.06, 103.15, 41.62, 84.06,90.06, 26.16, 55.53, 73.06, 76.69)', l=50.53, hich represent the eigenvector TZZ6 and eigenvalue TZXL6 respectively. Save TZZ6 as the final result. By integration the arrival-departure time of the trains on GS1 can be optimized as :
xgs1l6arr=96, xgs1l6dep=102, xgs1l12arr=85, xgs1l12dep=91, xgs2l18arr=56, xgs2l18dep=74.
Using defined defuzzy function deFuzzy[a,b,c] and optimized model & algorithm flow for trains occupation order on station tracks & C# Console Program, this invention get the optimized trains occupation order on tracks of each station yard which are illustrated as Table 2, Table 3 and Table 4. optimized sequence schedule for occupation of tracks in average speed yard unit: minute
CST ootl tl tos rdiots wsct
ates dts sto eto
PS1 psl0-psl5 psl0 Pass by 3.74 3.74 00 00
psl5 Pass by 3.74 3.74 43 43
PS2 psl1 psl1 Pass by 3.74 3.74 09 09
PS3 psl2 psl2 Pass by 3.74 3.74 18 18
PS4 psl3 psl3 Pass by 3.74 3.74 27 27
PS5 psl4 psl4 Pass by 3.74 3.74 35 35
notion:CST—code for station track, ootl—order for occupation of train lines, train lines—tl, tos—type of operation at station, Random defuzzy for item of operation time at station—rdiots (Arrival time for entering station—ates, Departure time from station–dts), Within the same cycle time for the whole yard—wsct (Start time for occupation—sto, End time for occupatio—eto). notion again: It doesn't exist the conflicts among the trains in the same yard based on this trains' occupation order. The conflicts can only take place in the crossing area of the different yards. optimized sequence schedule for occupation of tracks in high speed yard unit: minute
CST ootl tl tos smo emo rdiots wsct
ates sts dts ptld ttlt sto eto
GS1 gs2l18-gs1l12 -gs1l6 gs2l18 AU 56 74 4.01 17.53 3.63 00 08
gs1l12 AU 85 91 2.09 6 3.53 19 25
gs1l6 ODD 96 102 6 2.09 3.53 146 152
GS2 gs1l7--gs1l13 -gs2l19 gs1l7 ODD 158 164 6 2.55 3.6 63 69
gs1l13 AU 169 175 2.55 6 3.6 74 80
gs2l19 AU 179 197 4.39 18.19 3.51 84 102
GS3 gs2l20-gs2l14- gs1l18 gs2l20 AU 367 384 4.03 17.67 3.55 142 159
gs2l14 AU 389 407 4.03 17.67 3.55 189 206
gs1l18 ODD 417 423 6 1.11 3.52 248 266
GS4 gs2l21- gs2l15- gs1l9 gs2l21 AU 683 701 5.1 18.73 3.51 281 299
gs2l15 AU 707 724 5.1 18.73 3.51 305 322
gs1l9 ODD 735 741 7 2.06 3.55 369 375
GS5 gs2l16- gs1l10 gs2l16 AU 45 63 5.23 17.66 3.52 379 397
gs1l10 ODD 74 81 7 1.03 3.55 443 451
GS6 gs2l17- gs1l11 gs2l17 AU 48 67 4.8 18.59 3.52 456 475
gs1l11 ODD 79 86 7 2.1 3.51 537 544
GS7 gs2l18- gs1l12 gs2l18 TD 166 183 17.16 3.55 3.55 16 33
gs1l12 TD 195 201 6 1.17 3.57 45 51
GS8 gs2l19- gs1l6- gs1l13- gs1l11 gs2l19 TD 344 362 18.3 3.53 3.53 110 128
gs1l6 DTD&ES 372 372 3.56 3.37 138 138
gs1l13 TD 376 383 7 2.23 3.56 270 277
gs1l11 DTD&ES 387 387 3.37 3.56 512 512
GS9 gs1l7--gs2l20- gs2l14 gs1l7 DTD&ES 66 66 3.76 3.53 55 55
gs2l20 TD 70 87 17.25 3.62 3.62 167 184
gs2l14 TD 92 109 17.25 3.62 3.62 511 528
GS10 gs1l8- gs2l15- gs2l21 gs1l8 DTD&ES 348 348 2.44 3.54 240 240
gs2l15 TD 352 370 18.05 3.51 3.51 330 348
gs2l21 TD 375 393 18.05 3.51 3.51 333 351
GS11 gs1l9- gs2l16 gs1l9 DTD&ES 44 44 3.74 3.52 361 361
gs2l16 TD 49 67 18.28 3.58 3.58 405 424
GS12 gs1l10- gs2l17 gs1l10 DTD&ES 144 144 3.6 3.53 435 435
gs2l17 TD 149 166 17.04 3.53 3.53 483 501
notion:CST—code for station track, ootl—order for occupation of train lines, tl—train lines, tos—type of operation at station (AU—Arrival Upstream, ODD—Original Departure Downstream, TD—Turnback for Departure, DTD&ES—Departure from Train's depot and Entering Station,), smo—start moment for occupation in certain cycle time, emo—end moment for occupation in certain cycle time, rdiots—Random defuzzy for item of operation time at station— (ates —Arrival time for entering station , sts —Stop time at station , dts —Departure time from station , ptld —Pulling out time from lead track , ttlt —Turnback time from lead track), wsct —Within the same cycle time for the whole yard (sto — Start time for occupation, eto — End time for occupatio). notion again: It doesn't exist the conflicts among the trains in the same yard based on this trains' occupation order. The conflicts can only take place in the crossing area of the different yards. optimized sequence schedule for occupation of the tracks in intercity yard unit: minute
CST ootl tl tos smo emo rdiots wsct
ates dt dtls ptlt ttld sto eto
CJ1 cjl29c--cjl22f- cjl23f---cjl24f -cjl28c--cjl30c cjl29c dtmd 295 303 5.19 7.52 2.17 00 08
cjl22f dltr 307 314 2.17 7.52 2.17 12 20
cjl23f dltr 318 326 2.17 7.52 2.17 24 32
cjl24f dltr 330 337 2.17 7.52 2.17 36 44
cjl28c dtmd 341 349 5.19 7.52 2.17 48 56
cjl30c dtmd 308 315 5.19 7.52 2.17 60 68
CJ2 cjl31c--cjl25f- cjl26f---cjl27f cjl31c dtmd 282 290 5.46 7.52 2.03 72 80
cjl25f dltr 294 301 2.03 7.52 2.03 84 92
cjl26f dltr 305 313 2.03 7.52 2.03 96 104
cjl27f dltr 317 325 2.03 7.52 2.03 108 116
CJ3 cjl32h--cjl33h cjl32h td 306 324 17.54 2.52 2.52 166 184
cjl33h td 328 345 17.54 2.52 2.52 188 206
CJ4 cjl35h--cjl36h cjl35h td 139 157 17.3 2.5 2.5 278 296
cjl36h td 162 179 17.3 2.5 2.5 300 318
CJ5 cjl34h cjl34h td 17.3 2.5 2.5 322 340
CJ6 cjl32h- cjl33h- cjl34h cjl32h au 342 359 2.66 17.41 1.5 121 139
cjl33h au 364 381 2.66 17.41 1.5 144 162
cjl34h au 386 404 2.66 17.41 1.5 210 228
CJ7 cjl35h- cjl36h cjl35h au 49 68 2.35 18.47 1.37 232 251
cjl36h 72 90 2.35 18.47 1.37 255 274
notion: CST—Code for station track , ootl—Order for occupation of train lines, tl— Train lines, tos—Type of operation at station(dtmd—Departure from multi-unit train depot, dtlr —Departure from lines where trains reserved, td—Turnback for departure, au—Arrival upstream), smo— start moment for occupation in certain cycle time, emo—end moment for occupation in certain cycle time, rdiots—Random defuzzy for item of operation time at station ( ates—Arrival time for entering station, dt—Dwelling time,dtls —departure time for leaving station, ptlt— Pulling out time from lead track, ttld—Turnback time from lead trackstoStart time for occupation), wsct—Within the same cycle time for the whole yard (sto—start time for occupation, eto— End time for occupation ). notion again: It doesn't exist the conflicts among the trains in the same yard based on this trains' occupation order. The conflicts can only take place in the crossing area of the different yards.
2. Interactive Feedback of Adjustment between Global and Local Level
Then calling for algorithm1 and algorithm2 proposed in this paper the start-end time of each train line's resource occupation in each station yard on the local level can be obtained. The start–end time of resource occupation for each train line in each yard in the local level is illustrated in Table 5, Table 6 and Table 7. It doesn't exist the conflicts among the trains in the same yard based on this trains' occupation order. The conflicts can only take place in the crossing area of the different yards. For the case study in this paper BJSS station, each train line in t@s—tsdis passes by the throat area a and b in the average speed yard, it passes by the throat area c and d in the high speed yard and it passes by the throat area c,e,f and g in the intercity yard. The conflicts among the trains' lines in BJSS station can only happen at the throat area c where the trains operations for pulling out of their deposit cross at the high speed yard and the intercity yard. By checking the conflicts on the local level according to the improved RTCG, it's partly feasible for the trains occupation order of the station tracks on the global level. The train lines cjl29c, cjl28c, cjl30c and cjl31c in the intercity yard may conflict with the train lines gs2l18, gs1l12, and gs1l7 in the high speed yard in throat area c in station BJSS and its RTCG is illustrated as Fig. 20. start and end moment for train line occupation in average speed yard unit: minute
sn tl tr setrotas setrotcs setrost
st et st et st et
1 psl0 a-ps1-b -3.74 00 00 3.74 00 00
2 psl1 a-ps2-b 39.26 43 43 46.74 43 43
3 psl2 a-ps3-b 5.26 09 09 12.74 09 09
4 psl3 a-ps4-b 14.26 18 18 21.74 18 18
5 psl4 a-ps5-b 23.26 27 27 30.74 27 27
6 psl5 a-ps1-b 31.26 35 35 38.74 35 35
notion: sn—Serieal number , tl—train linesl, tr—Train route, setrotas—Start-end time for resource occupation in throat region a of station, setrotcs—Start-end time for resource occupation in throat region c of station, setrost—Start-end time for resource occupation of station track , Start time —st, End time —et. start and end moment for train line occupation in high speed yard unit: minute
sn tl tr setrotcs setrotds setrost tos
st et st et st et st et
1 gs2l18 c-GS1-d-GS7- c -4.01 00 08 11.63 00 08 GS1 AU
33 36.55 12.45 16 GS7 16 33 TD
2 gs1l12 c-GS1-d-GS7- c 16.91 19 25 28.53 19 25 GS1 AU
51 52.17 41.43 45 GS7 45 51 TD
3 gs1l6 c-GS8-d-GS1- c 152 154.09 142.47 146 146 152 GS1 ODD
134.44 138 138 141.37 GS8 138 138 DTD&ES
4 gs1l7 c-GS9-d-GS2- c 69 71.55 59.4 63 63 69 GS2 ODD
51.24 55 55 58.53 GS9 55 55 DTD&ES
5 gs1l13 c-GS2-d-GS8- c 71.45 74 80 83.6 74 80 GS2 AU
277 279.23 266.44 270 GS8 270 277 TD
6 gs2l19 c-GS2-d-GS8- c 79.61 84 102 105.51 84 102 GS2 AU
128 131.53 106.47 110 GS8 110 128 TD
7 gs2l20 c-GS3-d-GS9- c 137.97 142 159 162.55 142 159 GS3 AU
184 187.62 163.38 167 GS9 167 184 TD
8 gs2l14 c-GS3-d-GS9- c 184.97 189 206 209.55 189 206 GS3 AU
528 531.62 507.38 511 GS9 511 528 TD
9 gs1l8 c--------GS10-d-GS3 -c 266 267.11 244.48 248 248 266 GS3 ODD
237.56 240 240 243.54 GS10 240 240 DTD&ES
10 gs2l21 c-GS4-d-GS10-c 275.9 281 299 302.51 281 299 GS4 AU
351 354.51 329.49 333 GS10 333 351 TD
11 gs2l15 c-GS4-d-GS10-c 299.9 305 322 308.51 305 322 GS4 AU
348 351.51 326.49 330 GS10 330 348 TD
12 gs1l9 c-------GS11-d-GS4-c 375 377.06 365.45 369 369 375 GS4 ODD
361 364.74 361 363.52 GS11 361 361 DTD&ES
13 gs2l16 c-GS5-d-GS11-c 373.77 379 397 400.52 379 397 GS5 AU
424 527.58 401.42 405 GS11 405 424 TD
14 gs1l10 c-------GS12-d-GS5-c 451 452.03 439.45 443 443 451 GS5 ODD
431.4 435 435 438.53 GS12 435 435 DTD&ES
15 gs2l17 c-GS6-d-GS12-c 451.2 456 475 478.52 456 475 GS6 AU
501 504.53 479.47 483 GS12 483 501 TD
16 gs1l11 c-GS8-d-GS6- c 544 546.1 533.49 537 537 544 GS6 ODD
512 508.63 512 515.6 GS12 512 512 DTD&ES
notion: sn—serieal number, tl—Train line, tr— Train route, setrotcs—Start-end time for resource occupation in throat area c of station, setrotds —Start-end time for resource occupation in throat area d of station , setrost —Start-end time for resource occupation of station track, tos —Type of operation(AU—Arrival Upstream, ODD—Original Departure Downstream, TD—Turnback for Departure, DTD&ES—Departure from Train's depot and Entering Station), st—Start time ,et—End time. start and end moment for train line occupation in intercity yard unit: minute
sn tl tr setrotcs setrotes setrotfs setrotgs setrost setrost tos
st et st et st et st et st et st et
1 cjl29c c-e-f---CJ1-g -5.19 -2.6 -2.6 -1.3 -1.3 00 08 10.17 00 08 CJ1 dtmd
2 cjl22f e-f------CJ1-g 　 　 9.83 10.1 10.1 12 20 22.17 12 20 CJ1 dltr
3 cjl23f e-f------CJ1-g 　 　 21.83 22.92 22.92 24 32 34.17 24 32 CJ1 dltr
4 cjl24f e-f------CJ1-g 　 　 33.83 34.92 34.92 36 44 46.17 36 44 CJ1 dltr
5 cjl28c c-e-f---CJ1-g 42.9 45.46 45.46 46.73 46.73 48 56 58.17 48 56 CJ1 dtmd
6 cjl30c c-e-f---CJ1-g 54.81 57.4 57.4 58.7 58.7 60 68 70.17 60 68 CJ1 dtmd
7 cjl31c c-e-f---CJ2-g 66.36 69.28 69.28 70.64 70.64 72 80 82.03 72 80 CJ2 dtmd
8 cjl25f e-f------CJ2-g 　 　 81.97 82.99 82.99 84 92 94.03 84 92 CJ2 dltr
9 cjl26f e-f------CJ2-g 　 　 　 93.97 94.99 94.99 96 104 106.03 96 104 CJ2 dltr
10 cjl27f 　 　 106 106.99 106.99 108 116 118.03 108 116 CJ2 dltr
11 　 cjl32h 　 g-CJ6-f-QCI-f-CJ3-g 　 　 　 　 　 163.48 166 184 186.52 166 184 CJ3 td
　 139 141 118 　 121 CJ6 121 139 au
12 　 cjl33h 　 g-CJ6-f-QCI-f-CJ3-g 　 　 　 　 　 185.48 188 206 208.52 188 206 CJ3 　 td
162 164 141 　 144 CJ6 144 162 au
13 　 cjl35h 　 g-CJ7-f-QCI-f-CJ4-g 　 　 　 　 　 275.5 278 296 298.5 278 296 CJ4 　 td
　 251 252 230 　 232 CJ7 232 251 au
14 　 cjl36h 　 g-CJ7-f-QCI-f-CJ4-g 　 　 　 　 　 297.5 300 318 320.5 300 318 CJ4 　 td
　 274 275 253 　 255 CJ7 255 274 au
15 　 cjl34h 　 gCJ6-g-QCI-f-CJ5-g 　 　 　 　 　 228 230 207 210 CJ6 　 210 228 au
　 319.5 322 340 342.5 322 　 340 CJ5 td
notion: sn—serieal number, tl— Train lines, tr— Train route, tos—Type of operation at station (dtmd—Departure from multi-unit train depot, dtlr —Departure from lines where trains reserved, td—Turnback for departure, au—Arrival upstream), setrotcs—Start-end time for resource occupation in throat area c of station, setrotes—Start-end time for resource occupation in throat area e of station, setrotfs—Start-end time for resource occupation in throat area f of station, setrotgs—Start-end time for resource occupation in throat area g of station, setrost —Start-end time for resource occupation of station track, st—Start time ,et—End time.
In Fig. 20, the pure number in the circles represents the code number of the main switches in the throat area c of station BJSS, the combination of alphabetic and the number represents the code number of the train line, while the arrowed connecting line represents the route conflicts in the corresponding switch generated by two or more than two train lines for their occupation in the set of the start-end time interval. The trains' occupation order of the station tracks for t@s—TSDIS in certain period for certain station BJSS on its global level is partly feasible after RTCG conflict check, namely, it exists conflict for the pulling out of the train lines cjl30c and gs1l7 from their deposit in the throat area c. According to the flow chart for the conflict check and route optimization, it feedbacks between the global—local level interactively. For the number of the affected trains in the high speed yard are relatively great, it doesn't change the optimized trains' occupation order of the station tracks in the high speed yard and adjusts start-end time for the resource occupation of the train line cjl30c and its affected trains in the intercity yard, which is adjusted in details as: the resource occupation time for the train line cjl30c and its affected trains in the intercity yard are all delayed by five minutes then it deletes the conflict of the pulling out operation of train line cjl30c and gs1l7 from their deposit while it doesn't generate the new conflicts. The adjusted sequence schedule for the optimized occupation of the tracks in the intercity yard is illustrated as Table 8 and the start–end time of resource occupation is illustrated as Table 9. adjusted sequence schedule for optimized occupation of tracks in intercity yard unit: minute
CST ootl tl tos smo smo rdiots wsct
ates dt dtls ptlt ttld sto eto
CJ1 cjl29c----cjl22f- cjl23f-cjl24f-cjl28c-cjl30c cjl29c DTMD 295 303 5.19 7.52 2.17 　 　 00 08
cjl22f DLTR 307 314 2.17 7.52 　 2.17 　 12 20
cjl23f DLTR 318 326 2.17 7.52 　 2.17 　 24 32
cjl24f DLTR 330 337 2.17 7.52 　 2.17 　 36 44
cjl28c DTMD 341 349 5.19 7.52 　 2.17 　 48 56
cjl30c DTMD 308 315 5.19 7.52 　 2.17 　 65 73
CJ2 cjl31c---cjl25f- cjl26f---cjl27f cjl31c DTMD 282 290 5.46 7.52 2.03 　 　 77 85
cjl25f DLTR 294 301 2.03 7.52 　 2.03 　 89 97
cjl26f DLTR 305 313 2.03 7.52 　 2.03 　 101 109
cjl27f DLTR 317 325 2.03 7.52 　 2.03 　 113 121
CJ3 cjl32h---cjl33h cjl32h TD 306 324 　 17.54 2.52 　 2.52 171 189
cjl33h TD 328 　 345 17.54 　 2.52 2.52 193 211
CJ4 cjl35h---cjl36h cjl35h TD 139 157 　 17.3 2.5 　 2.5 283 301
cjl36h TD 162 　 179 17.3 　 2.5 2.5 305 323
CJ5 cjl34h cjl34h TD 　 　 　 17.3 2.5 　 2.5 327 345
CJ6 cjl32h---cjl33h---cjl34h cjl32h AU 342 359 2.66 17.41 　 1.5 　 126 144
cjl33h AU 364 381 2.66 　 17.41 　 1.5 149 167
cjl34h AU 386 404 2.66 　 17.41 　 1.5 215 232
CJ7 cjl35h---cjl36h cjl35h AU 49 68 2.35 18.47 　 1.37 　 237 256
cjl36h AU 72 90 2.35 　 18.47 　 1.37 260 279
notion: CST—Code for station track , ootl—Order for occupation of train lines, tl— Train lines, tos—Type of operation at station(DTMD—Departure from multi-unit train depot, DTLR —Departure from lines where trains reserved, TD—Turnback for departure, AU–Arrival upstream), smo— start moment for occupation in certain cycle time, emo—end moment for occupation in certain cycle time, rdiots—Random defuzzy for item of operation time at station ( ates—Arrival time for entering station, dt—Dwelling time,dtls —departure time for leaving station, ptlt— Pulling out time from lead track, ttld—Turnback time from lead tracks to Start time for occupation), wsct—Within the same cycle time for the whole yard (sto—start time for occupation, eto— End time for occupation ). notion again: It doesn't exist the conflicts among the trains in the same yard based on this trains' occupation order. The conflicts can only take place in the crossing area of the different yards. adjusted start and end moment for train line occupation in intercity yard unit: minute
sn tl tr setrotcs setrotes setrotfs setrotgs setrost setrost tos
st et st et st et st et st et st et
1 cjl29c c-e-f-CJ1-g -5.19 -2.6 -2.6 -1.3 -1.3 00 08 10.17 00 08 CJ1 dmtd
2 cjl22f e-f-----CJ1-g 　 　 9.83 10.1 10.1 12 20 22.17 12 20 CJ1 dltr
3 cjl23f e-f-----CJ1-g 　 　 21.83 22.92 22.92 24 32 34.17 24 32 CJ1 dltr
4 cjl24f e-f-----CJ1-g 　 　 33.83 34.92 34.92 36 44 46.17 36 44 CJ1 dltr
5 cjl28c c-e-f-CJ1-g 42.9 45.46 45.46 46.73 46.73 48 56 58.17 48 56 CJ1 dmtd
6 cjl30c c-e-f-CJ1-g 59.81 62.4 62.4 63.7 63.7 65 73 75.17 65 73 CJ1 dmtd
7 cjl31c c-e-f-CJ2-g 71.36 74.28 74.28 75.64 75.64 77 85 87.03 77 85 CJ2 dmtd
8 cjl25f e-f-----CJ2-g 　 　 86.97 87.99 87.99 89 97 99.03 89 97 CJ2 dltr
9 cjl26f e-f-----CJ2-g 　 　 　 98.97 99.99 99.99 101 109 111 101 109 CJ2 dltr
10 cjl27f 　 　 111 112 112 113 121 123 113 121 CJ2 dltr
11 cjl32h 　 g-CJ6-f-----QCI-f-CJ3-g 　 　 　 　 　 168.5 171 189 191.5 171 189 CJ3 td
12 　 　 144 145.5 123 126 CJ6 126 144 au
13 cjl33h 　 g-CJ6-f-----QCI-f-CJ3-g 　 　 　 　 　 190.5 193 211 213.5 194 211 CJ3 td
14 　 168.5 　 167 146 149 CJ6 149 167 au
15 cjl35h 　 g-CJ7-f-----QCI-f-CJ4-g 　 　 　 　 　 280.5 283 301 303.5 283 301 CJ4 td
16 　 　 256 257.4 235 237 CJ7 237 256 au
17 cjl36h 　 g-CJ7-f-----QCI-f-CJ4-g 　 　 　 　 　 302.5 305 323 325.5 305 323 CJ4 td
18 　 　 279 280.4 258 260 CJ7 260 279 au
19 cjl34h 　 gCJ6-g-----QCI-f-CJ5-g 　 　 　 　 　 232 234.5 212 215 CJ6 215 233 au
20 　 　 324.5 325 345 347.5 327 345 CJ5 td
notion: sn—serieal number, tl— Train lines, tr— Train route, tos—Type of operation at station(dmtd —Departure from multi-unit train depot, dltr —Departure from lines where trains reserved, td—Turnback for departure, au—Arrival upstream), setrotcs—Start-end time for resource occupation in throat area c of station, setrotes—Start-end time for resource occupation in throat area e of station, setrotfs—Start-end time for resource occupation in throat area f of station, setrotgs—Start-end time for resource occupation in throat area g of station, setrost —Start-end time for resource occupation of station track, st—Start time ,et—End time.
According to Dan Max Burkolter (2005) the bottleneck can be recognized. In the topology of the station BJSS, assuming the capacity of the station except certain subarea, it calculates and evaluates the consumption of the resource occupation time for each subarea is illustrated as Table 10 and the biggest one is the bottleneck affected the capacity. Because it has the operation of the trains for pulling out-entering the deposit and the turnback for departure in the throat area c, which is relatively complex and occupy the resource for the longest time, it forms the critical capacity bottleneck for station BJSS in the throat area c. time needed to complete t@s-TSDIS list in each subregion of BJSS unit: minute
st a B c d e f g
to 22.44 22.44 189.45 74.18 8.08 27.82 107.85
notion: st—Subarea in throat , to —Time occupation

Claims (1)

  1. Characteristics of This Invention & Claims for Protection
    1.Characteristics of This Invention
    (1)This invention 'Calculation and Evaluation of Station Capacity for High Speed Railway' dedicated by Ph.D Jiamin Zhang constructs the high speed railway trains at station train service–demand intention set (t@s—TSDIS) regarding the time needed to complete the set t@s—TSDIS as the new criteria to measure the station capacity of the high speed railway and proposing the global—local bilevel model system to calculate and evaluate the station capacity (Fig. 1) which on the global level optimizes the trains' occupation order (Fig. 6, Fig. 8, Fig. 9) and on the local level checks the feasibility of the order (Fig. 10, algorithm1, algorithm2); this invention can calculate and evaluate the capacity not only of the station but also of the more complex railway junction.
    (2)According to what mentioned about the calculation and evaluation of the station capacity of the high speed railway above in (1), targeting at the minimization of the time needed to complete the set t@s—TSDIS, this invention proposes the t@s—FTHNs allocation strategy of the station track resources (Fig. 6, Fig. 7) as well as the t@s—FTHNs optimization model & solution algorithm flow chart for the trains occupation order on the tracks (Fig. 8, Fig. 9) and describes the static definition and dynamic process for the interaction between the station and the trains (Fig. 2,Fig. 3, Fig. 4, Fig. 5 ) using the Fuzzy time high level Petri nets (FTHNs), Max Plus algebra, Scilab software and simulated annelaing altorithm. On the global level the station is abstracted as the topology of the main nodes and presumes the trains can pass through each node unlimitedly.
    (3) According to what mentioned about the calculation and evaluation of the station capacity of the high speed railway above in (1), considering the composition of the throat nodes and switches & tracks in details in each subregion on the local level, regarding the station tracks, the switches and the signals as the resources, this invention enumerates the possible routes set for the trains in the task list t@s—TSDIS between the physical point of the trains entering the station and that of the trains leaving the station based on the concrete resources topology using the depth–first search algorithm, this invention also constructs algorithm1 and algorithm2, for the t@s—TSDIS trains' occupation order of the station tracks on the global level, it determines the start–end time for the occupation of the corresponding resources, that is it proposes algorithm1 to backstep the moment that the trains passing by each subarea from its entering points of the station system according to the trains' arrival time and deduce the moment that the trains' passing by each subregion from their departure of the station tracks to the leaving points of the station system based on the trains' departure time of the station. Based on these, it proposes algorithm 2 to construct the resource tree conflict graph (RTCG) to check and optimize the train routes by feedback between the global and local level and then calculate & evaluate the station capacity.
    (4) According to what mentioned about the calculation and evaluation of the station capacity of the high speed railway above in (1), this invention demonstrates the practice of the invented model system & solution algorithm on certain high speed railway station BJSS and proposes the task list t@s—TSDIS (table 1) for BJSS in certain time period, which obtains the optimized trains occupation order of the station tracks on the global level (table 2, table 3, table 4, table 8) and the start–end time of the resource occupation on the local level (table 5, table 6, table 7 and table 9), as well as this invention gets the station capacity and its bottleneck (table 10).
    2. Claims for Protection
    (1)Rights for Signature,
    (2)Rights for Occupancy,
    (3)Rights for Patents,
    (4)Rights for Usage and Disposition,
    (5)Rights for Indispensible Rewards,
    (6)Rights for Sale and Transfer of the Usage.
PCT/IB2013/050040 2012-09-17 2013-01-03 Calculation and evaluation of station capacity for high speed railway WO2014041439A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201220470306 2012-09-17
CN201220470306.X 2012-09-17

Publications (1)

Publication Number Publication Date
WO2014041439A1 true WO2014041439A1 (en) 2014-03-20

Family

ID=50277709

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2013/050040 WO2014041439A1 (en) 2012-09-17 2013-01-03 Calculation and evaluation of station capacity for high speed railway

Country Status (1)

Country Link
WO (1) WO2014041439A1 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105303001A (en) * 2015-11-27 2016-02-03 北京交通大学 Method for automatically drawing railway yard graph
EP3236399A1 (en) * 2016-04-21 2017-10-25 ALSTOM Transport Technologies A method for updating a time-table in order to reduce a recurrent delay
CN109814393A (en) * 2019-03-08 2019-05-28 哈尔滨理工大学 A kind of reduced dimension observer and controller design method suspending floating material servomechanism in midair
CN110769443A (en) * 2019-10-21 2020-02-07 北京交通大学 Vehicle-vehicle communication performance analysis system based on stochastic Petri network
CN111967134A (en) * 2020-07-10 2020-11-20 北京交通大学 Underground logistics system optimization control method based on subway collinear common vehicle
CN113212503A (en) * 2021-05-11 2021-08-06 卡斯柯信号(成都)有限公司 Detection method for rail transit vehicle shunting plan conflict
CN117035204A (en) * 2023-10-10 2023-11-10 中铁第四勘察设计院集团有限公司 Method and device for acquiring running buffer time between high-speed rail running chart regions

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102129522A (en) * 2011-03-17 2011-07-20 北京交通大学 Method for quickly identifying and eliminating transportation capacity bottleneck of high-speed railway

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102129522A (en) * 2011-03-17 2011-07-20 北京交通大学 Method for quickly identifying and eliminating transportation capacity bottleneck of high-speed railway

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ZHANG, JIAMIN ET AL.: "Comparison and Analysis of Calculation and Evaluation Strategy of Station Capacity for High Speed Railway.", CHINESE RAILWAYS., March 2011 (2011-03-01), pages 22 - 25 *
ZHANG, JIAMIN ET AL.: "Optimization on the Occupation Order for Trains on Tracks of the Railway Station.", JOURNAL OF TRANSPORTATION SYSTEMS ENGINEER AND INFORMATION TECHNOLOGY., vol. 11, no. 3, June 2011 (2011-06-01), pages 100 - 107 *

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105303001A (en) * 2015-11-27 2016-02-03 北京交通大学 Method for automatically drawing railway yard graph
EP3236399A1 (en) * 2016-04-21 2017-10-25 ALSTOM Transport Technologies A method for updating a time-table in order to reduce a recurrent delay
CN109814393A (en) * 2019-03-08 2019-05-28 哈尔滨理工大学 A kind of reduced dimension observer and controller design method suspending floating material servomechanism in midair
CN109814393B (en) * 2019-03-08 2021-11-02 哈尔滨理工大学 Dimension reduction observer and controller design method of suspended floating object follow-up system
CN110769443A (en) * 2019-10-21 2020-02-07 北京交通大学 Vehicle-vehicle communication performance analysis system based on stochastic Petri network
CN110769443B (en) * 2019-10-21 2020-11-10 北京交通大学 Vehicle-vehicle communication performance analysis system based on stochastic Petri network
CN111967134A (en) * 2020-07-10 2020-11-20 北京交通大学 Underground logistics system optimization control method based on subway collinear common vehicle
CN111967134B (en) * 2020-07-10 2023-12-12 北京交通大学 Underground logistics system optimization control method based on subway collinear co-vehicle
CN113212503A (en) * 2021-05-11 2021-08-06 卡斯柯信号(成都)有限公司 Detection method for rail transit vehicle shunting plan conflict
CN117035204A (en) * 2023-10-10 2023-11-10 中铁第四勘察设计院集团有限公司 Method and device for acquiring running buffer time between high-speed rail running chart regions
CN117035204B (en) * 2023-10-10 2024-01-12 中铁第四勘察设计院集团有限公司 Method and device for acquiring running buffer time between high-speed rail running chart regions

Similar Documents

Publication Publication Date Title
WO2014041439A1 (en) Calculation and evaluation of station capacity for high speed railway
Schöbel Capacity constraints in delay management
WO2019027247A1 (en) Integrated logistics management system using drone and method thereof
Wang et al. Rescheduling trains using Petri nets and heuristic search
Ladier et al. Robust cross-dock scheduling with time windows
Niu et al. Demand-driven train schedule synchronization for high-speed rail lines
Wagenaar et al. Rolling stock rescheduling in passenger railway transportation using dead-heading trips and adjusted passenger demand
Van der Hurk et al. Passenger advice and rolling stock rescheduling under uncertainty for disruption management
WO2014041438A1 (en) Calculation and estimation of line capacity for high speed railway
WO2019112379A1 (en) Method and system for designing layout of smart factory
CN112883640A (en) Digital twin station system, job scheduling method based on system and application
Li et al. Discrete-event simulations for metro train operation under emergencies: A multi-agent based model with parallel computing
Guo et al. A stochastic multiagent optimization framework for interdependent transportation and power system analyses
Liu et al. Stochastic passenger train timetabling using a branch and bound approach
Meng et al. A train dispatching model under a stochastic environment: stable train routing constraints and reformulation
Wang et al. Synchronized optimization for service scheduling, train parking and routing at high-speed rail maintenance depot
Løve et al. Using Heuristics to Solve the Dedicated Aircraft Recovery Problem.
Wang et al. Optimal parking pricing in many-to-one park-and-ride network with parking space constraints
Outafraout et al. A control approach based on colored hybrid Petri nets and (max,+) algebra: Application to multimodal transportation systems
Gaied et al. Modelling and performance evaluation of railway transport systems using P-timed petri nets
Abid et al. Sensitivity analysis of train schedule of a railway track network using an optimization modeling technique
JP2013010388A (en) Vehicle operation plan preparation device
Zu et al. Integrated Production and distribution problem with pickup and delivery and multiple trips
Kersbergen et al. Reducing the time needed to solve the global rescheduling problem for railway networks
Hartmann Tolić et al. Optimization methods in modern transportation systems

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13837323

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13837323

Country of ref document: EP

Kind code of ref document: A1