CN106741018A - A kind of train based on network evolution starts control method and system - Google Patents

Abstract

The present invention discloses a kind of train based on network evolution and starts control method, and methods described includes：S1：According to train passenger flow and road net data, railway network model is built, and specificity analysis is carried out to network model, export Network characteristic parameters；S2：According to the Network characteristic parameters and wagon flow net evolution rule generation wagon flow net；S3：The mapping relations of the wagon flow net and starting scheme are analyzed, initial starting scheme is generated；S4：The initial starting scheme is evaluated and optimized, obtain train running scheme, start control system the present invention discloses a kind of train based on network evolution, suitable for starting control into the train under the conditions of net on a large scale, tradition is overcome just for single line and depends on artificial establishment to start the shortcoming of control program, be that train is started there is provided more practical, efficient control device.

Description

A kind of train based on network evolution starts control method and system

Technical field

Start control field the present invention relates to train.Start control more particularly, to a kind of train based on network evolution Method and system processed.

Background technology

With the fast development of railway, especially propose to strengthen railway infrastructure construction during China 13, railway is solid Determining assets investment scale will build 30,000 kilometers of new line up to 3.5 to 3.8 trillion yuans.To the year two thousand twenty, china railway revenue kilometres will Reach 150,000 kilometers, wherein 30,000 kilometers of high-speed railway, railway network operation characteristic is obvious all the more.

Existing train starts the train running scheme that control mode depends on artificial establishment, to train running scheme Research be also limited to wall scroll rail track, the Research foundation that train network melts row control is relatively weak, and formation is different from biography The new control system of control method of uniting and method are Train Control field problem demanding prompt solutions, and to improving railway transportation The aspects such as economic benefit all have being of great significance.

Accordingly, it is desirable to provide a kind of train based on network evolution starts control method and system, solve current train and open Artificial establishment efficiency that row scheme is present is low, the problems such as be confined to single line.

The content of the invention

The invention solves the problems that a technical problem be to provide a kind of train based on network evolution and start control method, with Solve the problems, such as that conventional train starts control and relies only on artificial establishment train running scheme, while it can be considered that a plurality of circuit Obtain the optimal train of scheme and start control program, the invention solves the problems that another technical problem be to provide it is a kind of based on network The train of evolution starts control system.

In order to solve the above technical problems, the present invention uses following technical proposals：

A kind of train based on network evolution starts control method, it is characterised in that methods described includes：

S1：According to train passenger flow and road net data, railway network model is built, and specificity analysis is carried out to network model, Output Network characteristic parameters；

S2：According to the Network characteristic parameters and wagon flow net evolution rule generation wagon flow net；

S3：The mapping relations of the wagon flow net and starting scheme are analyzed, initial starting scheme is generated；

S4：The initial starting scheme is evaluated and optimized, train running scheme is obtained.

Preferably, the S1 includes：

S11：According to train passenger flow and road net data, structure includes the railway network mould of railway geographic network and railway passenger drift net Type；

S12：Specificity analysis is carried out to network model, Network characteristic parameters are exported, the Network characteristic parameters include node Group relationship metric matrix, node physical connection matrix and station significance level matrix.

Preferably, the railway geographic network is using station as network node, using connect the track between two stations as Two sides of network node of connection.

Preferably, it is side that the railway passenger drift net will have two lines at station of passenger's boarding, with described two stations Between the volume of the flow of passengers as side right.

Preferably,

The node group relationship metric matrix is

The node physical connection matrix is

The station significance level matrix is

Wherein, ω_ijIt is the parameter of the volume of the flow of passengers of interval ij in railway network model, p_ijIt is interval in railway network model The volume of the flow of passengers of ij, η is train load factor, and L is train seating capacity.

Preferably, the S2 includes：

S21：Connecting probability according to the Network characteristic parameters, between calculating any two station is

F_ij=α_ijp_ij(s_i+s_j)

Wherein, F_ijIt is the passenger flow gravitation between two stations, p_ijThe volume of the flow of passengers between two station ij, α_ijIt is two stations Group's relationship metric parameter between ij, β_ijIt is the physical connection parameters between two station ij, s_i、s_jIt is the important journey of station i, j Degree parameter；

S22：Two stations connection of the selection connection maximum probability, now the side right at two stations adds 1, while by iron Side right in road network model between two station ij subtracts 1；

S23：Repeat S21-S22 until in the railway network model all side rights be it is negative terminate to develop, after evolution Railway network model as wagon flow net.

Preferably, the S3 includes：

S31：According to the wagon flow net, all stations on same train working line are set to a group；

S32：All train dwelling schemes of each group are included, according to residue between ensureing the starting of direct train, stand Weight feedback generation through train, consideration select station and stop residual ownership between train and station and feed back again to generate done site by site and stop the rule of train All alternative starting schemes are generated, an alternative starting scheme is often generated and the weight between two stations in the group is subtracted 1；

S33：S32 is pressed into all groups and generates all alternative starting schemes, obtain the initial starting scheme of all groups.

Preferably, the S4 includes：

S41：The evaluation index value of each initial starting scheme is calculated, according to train running scheme assessment indicator system Enter row index marking to initial starting scheme each described；

S42：Each grade residing for the initial starting scheme is judged using fitness function, if the initial starting scheme Meet the requirements, then as train running scheme；If the initial starting scheme is undesirable, to the initial starting scheme Optimize, S41 is repeated, until obtaining satisfactory all initial starting schemes as train running scheme.

Preferably, the fitness function is

Wherein, y_lIt is the evaluation index value of initial starting scheme l, γ_lFor the index of initial starting scheme l is given a mark.

Start control system the present invention discloses a kind of train based on network evolution, it is characterised in that the system System includes：

Network model builds module：For according to train passenger flow and road net data, building railway network model, and to network Model carries out specificity analysis, exports Network characteristic parameters；

Network model genetic module：For according to the Network characteristic parameters and wagon flow net evolution rule generation wagon flow net；

Starting scheme generation module：Mapping relations for analyzing the wagon flow net and starting scheme, generation is initially started Scheme；

Starting scheme evaluation module：For the initial starting scheme to be evaluated and optimized, the train side of starting is obtained Case.

Beneficial effects of the present invention are as follows：

1st, overcome conventional train and start the problem that control relies on artificial establishment, can quickly realize that train starts control.

2nd, support that networking train starts control problem, improve the practicality of train running scheme.

3rd, ensure single line train start control program it is optimal while, road network integral train has been taken into account again and has started control Scheme processed it is optimal.

Brief description of the drawings

Specific embodiment of the invention is described in further detail below in conjunction with the accompanying drawings.

Fig. 1 shows that a kind of train based on network evolution starts the flow chart of control method.

Fig. 2 shows that a kind of train based on network evolution starts the schematic diagram of control system.

Specific embodiment

In order to illustrate more clearly of the present invention, the present invention is done further with reference to preferred embodiments and drawings It is bright.Similar part is indicated with identical reference in accompanying drawing.It will be appreciated by those skilled in the art that institute is specific below The content of description is illustrative and be not restrictive, and should not be limited the scope of the invention with this.

As shown in figure 1, one aspect of the present invention discloses a kind of train based on network evolution starts control method, the side Method includes：

S1：According to train passenger flow and road net data, railway network model is built, and specificity analysis is carried out to network model, Output Network characteristic parameters.

S11：According to train passenger flow and road net data, structure includes the railway network mould of railway geographic network and railway passenger drift net Type.Wherein, the railway geographic network is using station as network node, to connect the track between two stations as connection two The side of network node；The railway passenger drift net will have two lines at station of passenger's boarding for side, with described two stations it Between the volume of the flow of passengers as side right.

S12：Specificity analysis is carried out to network model, Network characteristic parameters are exported, the Network characteristic parameters include node Group relationship metric matrix, node physical connection matrix and station significance level matrix.Wherein,

The node group relationship metric matrix is

The node physical connection matrix is

The station significance level matrix is

Wherein, ω_ijIt is the parameter of the volume of the flow of passengers of interval ij in railway passenger drift net, p_ijIt is interval ij in railway network model The volume of the flow of passengers, η be train load factor, L is train seating capacity.

S2：According to the Network characteristic parameters and wagon flow net evolution rule generation wagon flow net.

S21：Connecting probability according to the Network characteristic parameters, between calculating any two station is

F_ij=α_ijp_ij(s_i+s_j)

Wherein, F_ijIt is the passenger flow gravitation between two stations, p_ijThe volume of the flow of passengers between two station ij, α_ijIt is two stations Group's relationship metric parameter between ij, belongs to Same Community and then tends to 1, otherwise takes certain numerical value for tending to 0, β_ijIt is two stations There is physical connection and take 1 in the physical connection parameters between ij, i, j, be otherwise 0, s_i、s_jIt is the significance level parameter of station i, j；

S22：Two stations connection of the selection connection maximum probability, now the side right at two stations adds 1, while by iron Side right in road network model between two station ij subtracts 1；

S23：Repeat S21-S22 until in the railway network model all side rights be it is negative terminate to develop, after evolution Railway network model as wagon flow net.

S3：The mapping relations of the wagon flow net and starting scheme are analyzed, initial starting scheme is generated.

S31：According to the wagon flow net, all stations on same train working line are set to a group；

S32：All train dwelling schemes of each group are included, according to residue between ensureing the starting of direct train, stand Weight feedback generation through train, consideration select station and stop residual ownership between train and station and feed back again to generate done site by site and stop the rule of train All alternative starting schemes are generated, an alternative starting scheme is often generated and the weight between two stations in the group is subtracted 1；

S33：S32 is pressed into all groups and generates all alternative starting schemes, obtain the initial starting scheme of all groups.

S4：The initial starting scheme is evaluated and optimized, train running scheme is obtained.

S41：The evaluation index value of each initial starting scheme is calculated, according to train running scheme assessment indicator system Enter row index marking to initial starting scheme each described.The evaluation index may include three kinds of road network, circuit and station.Its In, road network evaluation index includes road network average degree, road network average nodal betweenness, road network average aggregate coefficient, road network average service The empty gruel passenger-kilometer number of ability, road network average transfer times, road network.Circuit evaluation index includes that circuit average degree, circuit are averagely saved The empty gruel passenger-kilometer number of point betweenness, circuit average aggregate coefficient, circuit average service ability, circuit average transfer times, circuit.Car Evaluation index of standing includes station degree, station betweenness, station convergence factor, station average transfer times, station service frequency.

S42：Each grade residing for the initial starting scheme is judged using fitness function, if the initial starting scheme Meet the requirements, then as train running scheme；If the initial starting scheme is undesirable, to the initial starting scheme Optimize, S41 is repeated, until obtaining satisfactory all initial starting schemes as train running scheme.The adaptation Spending function is

Wherein, y_lIt is the evaluation index value of initial starting scheme l, γ_lFor the index of initial starting scheme l is given a mark.

As shown in Fig. 2 starting control system the present invention discloses a kind of train based on network evolution, its feature exists In the system includes：

Network model builds module：For according to train passenger flow and road net data, building railway network model, and to network Model carries out specificity analysis, exports Network characteristic parameters.According to train passenger flow and road net data, structure include railway geographic network with The railway network model of railway passenger drift net.Wherein, the railway geographic network is using station as network node, to connect two stations Between track as connection two sides of network node；The railway passenger drift net will have two lines at station of passenger's boarding It is side, using the volume of the flow of passengers between described two stations as side right.Specificity analysis, output network characteristic ginseng are carried out to network model Number, the Network characteristic parameters include node group relationship metric matrix, node physical connection matrix and station significance level square Battle array.Wherein, the node group relationship metric matrix is

The node physical connection matrix is

The station significance level matrix is

Wherein, ω_ijIt is the parameter of the volume of the flow of passengers of interval ij in railway passenger drift net, p_ijIt is interval ij in railway network model The volume of the flow of passengers, η be train load factor, L is train seating capacity.

The network model builds module and may include that network model builds rule base, network model memory cell and network spy Property analytic unit.Network model builds rule base according to network struction rule, receives passenger flow and road net data generation railway is geographical Net, passenger flow pessimistic concurrency control；Network model memory cell network model is built the network model generated in regular library unit and is stored Come, called for network characteristic analytic unit；Network characteristic analytic unit, analyze that each network model is distinctive and each network model between Dependency relation, and be responsible for output relation function and Network characteristic parameters.

Network model genetic module：For according to the Network characteristic parameters and wagon flow net evolution rule generation wagon flow net. Connecting probability according to the Network characteristic parameters, between calculating any two station is

F_ij=α_ijp_ij(s_i+s_j)

Wherein, F_ijIt is the passenger flow gravitation between two stations, p_ijThe volume of the flow of passengers between two station ij, α_ijIt is two stations Group's relationship metric parameter between ij, belongs to Same Community and then tends to 1, otherwise takes certain numerical value for tending to 0, β_ijIt is two stations There is physical connection and take 1 in the physical connection parameters between ij, i, j, be otherwise 0, s_i、s_jIt is the significance level parameter of station i, j.Choosing Two stations connection of the connection maximum probability is selected, now the side right at two stations adds 1, while by two in railway network model Side right between individual station ij subtracts 1；Repetition is calculated as above up to all side rights are negative end evolution in the railway network model, Using the railway network model after evolution as wagon flow net.

The network model genetic module may include wagon flow net evolution rule storehouse and evolution wagon flow net memory cell, wagon flow net Evolution rule storehouse receives the Network characteristic parameters that network characteristic analytic unit is transmitted, and matching adapts to the wagon flow net of existing network topology Evolution rule, and the gained wagon flow net that will develop passes to evolution wagon flow net memory cell.

Starting scheme generation module：Mapping relations for analyzing the wagon flow net and starting scheme, generation is initially started Scheme.According to the wagon flow net, all stations on same train working line are set to a group；Include each group The all train dwelling schemes existed between station, lead directly to according to Weighted residue feedback generation between ensureing the starting of direct train, stand Train, consideration select station and stop residual ownership between train and station and feed back generation done site by site again and stop the rule of train to generate and all alternatively start Scheme, often generates an alternative starting scheme and subtracts 1 by the weight between two stations in the group；By all groups based on as above The all alternative starting schemes of generation are calculated, the initial starting scheme of all groups is obtained.

The starting scheme generation module may include wagon flow net starting scheme map unit, and the mapping of wagon flow net starting scheme is single Unit is analyzed to the gained wagon flow net that develops, according to series of rules feedback generation starting scheme.

Starting scheme evaluation module：For the initial starting scheme to be evaluated and optimized, the train side of starting is obtained Case.The evaluation index value of each initial starting scheme is calculated, according to train running scheme assessment indicator system to each institute State initial starting scheme and enter row index marking.The evaluation index may include three kinds of road network, circuit and station.Wherein, road network is commented Valency index includes road network average degree, road network average nodal betweenness, road network average aggregate coefficient, road network average service ability, road network The empty gruel passenger-kilometer number of average transfer times, road network.Circuit evaluation index includes circuit average degree, circuit average nodal betweenness, line The empty gruel passenger-kilometer number of road average aggregate coefficient, circuit average service ability, circuit average transfer times, circuit.Evaluate and refer in station Mark includes station degree, station betweenness, station convergence factor, station average transfer times, station service frequency.Using fitness letter Number judges each grade residing for the initial starting scheme, if the initial starting scheme meets the requirements, starts as train Scheme；If the initial starting scheme is undesirable, the initial starting scheme is optimized, repeat S41, until To satisfactory all initial starting schemes as train running scheme.The fitness function is

Wherein, y_lIt is the evaluation index value of initial starting scheme l, γ_lFor the index of initial starting scheme l is given a mark.

The starting scheme evaluation module may include starting scheme indicator calculating unit, starting scheme index judging unit, Starting scheme adjustment unit.Starting scheme indicator calculating unit is evaluated the network index of starting scheme, and is tied evaluating Fruit is transferred to starting scheme index judging unit；Starting scheme index judging unit is opened according to railway operation administrative staff's typing The index that row scheme evaluation objective is transmitted with starting scheme indicator calculating unit is compared, and judges that next step is to enter the side of starting Case adjustment unit still directly carries out display output to starting scheme；Starting scheme adjustment unit is by regulation rule to initially opening Row scheme is optimized.

The train starts control system can also further include basic data module and human-computer interaction module.The base Plinth data module may include passenger flow data storehouse and the part of Traffic network database two.The data of passenger flow data storehouse and Traffic network database can pin Real-time update is carried out to different road networks.The human-computer interaction module is the operation end of data maintenance and displaying starting scheme result End, including data maintenance unit and the part of starting scheme display unit two.Data maintenance unit realizes passenger flow data storehouse, road network number The reading of the data such as rule base, wagon flow evolution rule storehouse, wagon flow net and starting scheme mapping ruler storehouse is built according to storehouse, network model Write, synchronization, protection；Starting scheme display unit calls the starting scheme of starting scheme generation module, and emphasis shows certain finger The starting scheme on alignment road, is equipped with explanatory note.

Obviously, the above embodiment of the present invention is only intended to clearly illustrate example of the present invention, and is not right The restriction of embodiments of the present invention, for those of ordinary skill in the field, may be used also on the basis of the above description To make other changes in different forms, all of implementation method cannot be exhaustive here, it is every to belong to this hair Obvious change that bright technical scheme is extended out changes row still in protection scope of the present invention.

Claims (10)

1. a kind of train based on network evolution starts control method, it is characterised in that methods described includes：

S1：According to train passenger flow and road net data, railway network model is built, and specificity analysis is carried out to network model, exported Network characteristic parameters；

S2：According to the Network characteristic parameters and wagon flow net evolution rule generation wagon flow net；

S3：The mapping relations of the wagon flow net and starting scheme are analyzed, initial starting scheme is generated；

S4：The initial starting scheme is evaluated and optimized, train running scheme is obtained.

2. method according to claim 1, it is characterised in that the S1 includes：

S11：According to train passenger flow and road net data, structure includes the railway network model of railway geographic network and railway passenger drift net；

S12：Specificity analysis is carried out to network model, Network characteristic parameters are exported, the Network characteristic parameters include node group Relationship metric matrix, node physical connection matrix and station significance level matrix.

3. method according to claim 2, it is characterised in that the railway geographic network using station as network node, with Track between two stations of connection is used as two sides of network node of connection.

4. method according to claim 2, it is characterised in that the railway passenger drift net will have two stations of passenger's boarding Line be side, using the volume of the flow of passengers between described two stations as side right.

5. method according to claim 2, it is characterised in that

The node group relationship metric matrix is

$\mathrm{\α}={\left\{{\mathrm{\α}}_{ij}\right\}}_{m_0\×m_0}=\left[\begin{array}{cccc}{\mathrm{\α}}_{1,1}& {\mathrm{\α}}_{1,2}& ...& {\mathrm{\α}}_{1,m_0}\\ {\mathrm{\α}}_{2,1}& {\mathrm{\α}}_{2,2}& ...& {\mathrm{\α}}_{2,m_0}\\ ...& ...& {\mathrm{\α}}_{i,i}& ...\\ {\mathrm{\α}}_{m_0,1}& {\mathrm{\α}}_{m_0,2}& ...& {\mathrm{\α}}_{m_0,m_0}\end{array}\right]$

The node physical connection matrix is

$\mathrm{\β}={\left\{{\mathrm{\β}}_{ij}\right\}}_{m_0\×m_0}=\left[\begin{array}{cccc}{\mathrm{\β}}_{1,1}& {\mathrm{\β}}_{1,2}& ...& {\mathrm{\β}}_{1,m_0}\\ {\mathrm{\β}}_{2,1}& {\mathrm{\β}}_{2,2}& ...& {\mathrm{\β}}_{2,m_0}\\ ...& ...& {\mathrm{\β}}_{i,i}& ...\\ {\mathrm{\β}}_{m_0,1}& {\mathrm{\β}}_{m_0,2}& ...& {\mathrm{\β}}_{m_0,m_0}\end{array}\right]$

The station significance level matrix is

${\mathrm{\ω}}_{ij}=\frac{p_{ij}}{L*\mathrm{\η}}$

$s_i=\underset{j\∈v\left(i\right)}{\Σ}{w}_{ij}$

$s={\left\{s_i\right\}}_{m_0\×m_0}=\[s_1,s_2,...,s_{m_0}\]$

Wherein, ω_ijIt is the parameter of the volume of the flow of passengers of interval ij in railway network model, p_ijIt is interval ij in railway network model The volume of the flow of passengers, η is train load factor, and L is train seating capacity.

6. method according to claim 1, it is characterised in that the S2 includes：

S21：Connecting probability according to the Network characteristic parameters, between calculating any two station is

$P_{ij}={\mathrm{\β}}_{ij}\·\frac{F_{ij}}{\underset{j\∈v\left(i\right)}{\Σ}{F}_{ij}}$

F_ij=α_ijp_ij(s_i+s_j)

Wherein, F_ijIt is the passenger flow gravitation between two stations, p_ijThe volume of the flow of passengers between two station ij, α_ijFor between two station ij Group's relationship metric parameter, β_ijIt is the physical connection parameters between two station ij, s_i、s_jFor the significance level of station i, j is joined Number；

S22：Two stations connection of the selection connection maximum probability, now the side right at two stations adds 1, while by the railway network Side right in network model between two station ij subtracts 1；

S23：Repeat S21-S22 until in the railway network model all side rights be it is negative terminate to develop, by the iron after evolution Road network model is used as wagon flow net.

7. method according to claim 1, it is characterised in that the S3 includes：

S31：According to the wagon flow net, all stations on same train working line are set to a group；

S32：All train dwelling schemes of each group are included, according to Weighted residue between ensureing the starting of direct train, standing Feedback generation through train, consideration select station stop train and station between residual ownership feed back again generation done site by site stop train rule generate All alternative starting schemes, often generate an alternative starting scheme and subtract 1 by the weight between two stations in the group；

S33：S32 is pressed into all groups and generates all alternative starting schemes, obtain the initial starting scheme of all groups.

8. method according to claim 1, it is characterised in that the S4 includes：

S41：The evaluation index value of each initial starting scheme is calculated, according to train running scheme assessment indicator system to every The individual initial starting scheme enters row index marking；

S42：Each grade residing for the initial starting scheme is judged using fitness function, if the initial starting scheme meets It is required that, then as train running scheme；If the initial starting scheme is undesirable, the initial starting scheme is carried out Optimization, repeats S41, until obtaining satisfactory all initial starting schemes as train running scheme.

9. method according to claim 8, it is characterised in that the fitness function is

$F=\underset{l=1}{\overset{L}{\Σ}}{\mathrm{\γ}}_l{y}_l/\underset{l=1}{\overset{L}{\Σ}}{\mathrm{\γ}}_l$

Wherein, y_lIt is the evaluation index value of initial starting scheme l, γ_lFor the index of initial starting scheme l is given a mark.

10. a kind of train based on network evolution starts control system, it is characterised in that the system includes：

Network model builds module：For according to train passenger flow and road net data, building railway network model, and to network model Specificity analysis is carried out, Network characteristic parameters are exported；

Network model genetic module：For according to the Network characteristic parameters and wagon flow net evolution rule generation wagon flow net；

Starting scheme generation module：Mapping relations for analyzing the wagon flow net and starting scheme, generate initial starting scheme；

Starting scheme evaluation module：For the initial starting scheme to be evaluated and optimized, train running scheme is obtained.