CN103208034A - Rail transit passenger flow distribution predicting model building method and predicting method - Google Patents

Abstract

The invention discloses a rail transit passenger flow distribution predicting model building method and a predicting method. The model building method includes the following steps: introducing the discrete variable for describing site property and scale, building a utility function of a passenger flow distribution predicting model by combining rail site generation attraction quantity, a network topological structure and relevant operation parameters, utilizing an individual representing method to convert the site generation attraction quantity belonging to aggregate data into disaggregate data and utilizing the maximum likelihood estimation to demarcate a passenger flow distribution predicting model. A rail transit passenger flow distribution predicting method is further disclosed. The two methods have the advantages of being low in data acquisition difficulty, high in practicability, accurate in prediction and the like.

Description

A kind of track traffic for passenger flow forecast of distribution model is set up and Forecasting Methodology

Technical field

The present invention relates to the track traffic technical field, relate in particular to a kind of track traffic for passenger flow forecast of distribution model and set up and Forecasting Methodology.

Background technology

In recent years, the continuous expansion of track traffic network, influencing each other between circuit increases gradually.Simultaneously, along with the track traffic ew line inserts existing road network and comes into operation, network topology structure and the soil utilization of website periphery etc. will change, the passenger flow spatial and temporal distributions is also incited somebody to action layout again, in order to assess the ew line access to the influence of existing line netter distributions, and being make rational planning for train running scheme and formulate passenger flow and induce strategy that the data support is provided of urban track traffic operating administration, the passenger flow forecast of distribution under the ew line access conditions is essential.But, owing to there is not the passenger flow historical data under the ew line access network condition, can not follow and traditional come passenger flow forecast changes in distribution trend based on the historical data statistical law, particularly under city rail traffic route networking operation condition, can not estimate the ew line access to the influence of existing gauze by the passenger flow changes in distribution situation of analyzing uniline simply.Therefore, make up that a kind of the urban track traffic ew line inserts the passenger flow distribution forecasting method under the net condition at becoming, insert influencing evaluation, hold passenger flow Changing Pattern, planning train running scheme under the new situation and formulate passenger flow and induce operation management link such as strategy to provide comprehensively and accurate data support has great importance existing gauze to implementing ew line.

At present, existing trip track traffic for passenger flow distribution forecasting method great majority are statistical model, be to carry out the passenger flow forecast of distribution at the distribution form of supposing following each website passenger flow with the identical basis of present existing distribution form, as growth factor method, Gravity Models etc.Obviously, said method is difficult to use between the change, each website of following website periphery land use pattern and intensity for the passenger flow forecast of distribution under the situations such as road network changes under influencing each other of attracting of passenger flow and the ew line access conditions.

Summary of the invention

In order to overcome the deficiency of prior art structure, the invention provides a kind of track traffic for passenger flow forecast of distribution model and set up and Forecasting Methodology.

The embodiment of the invention discloses a kind of track traffic for passenger flow forecast of distribution method for establishing model, may further comprise the steps:

1.1, the discrete variable of website character and scale introduce to be described, in conjunction with track website generation traffic attraction, topology of networks and relevant operation parameter, make up the utility function of passenger flow forecast of distribution model;

1.2, the website generation traffic attraction that uses the individual method of representative will belong to collection counting certificate is converted into non-collection counting certificate;

1.3, use maximum likelihood to estimate to demarcate passenger flow forecast of distribution model.

Further, as preferably, comprise further that in described step 1.1 introducing OD(starts and the destination) the accessibility index, characterize the influence that network topology structure and traffic convenience distribute for track traffic for passenger flow.

Further, as preferably, the utility function that makes up passenger flow forecast of distribution model in the described step 1.1 is specially: V _Ij=f (C _Ij, D _j, AOD ^Ij, XZ, GM, k)

Wherein, V _IjFor starting point is that selection terminal point in i station is the utility function at j station; C _IjBe the required admission fee in j station of standing from i; D _jTraffic attraction for the j station; AOD _IjFor website i to the OD accessibility index between j; XZ is for characterizing the 0-1 variable of website character, and is relevant with the website character at j station with the i station; GM is for characterizing the 0-1 variable of website scale, and is relevant with the scale at j station with the i station; K is the 0-1 variable, and when i station or j station belong to ew line, constant term is 1, otherwise is 0.

The invention also discloses a kind of track traffic for passenger flow distribution forecasting method, may further comprise the steps:

4.1, the classification according to the following track website generation traffic attraction that obtains in advance, website, topology of networks and relevant operation parameter, extract and obtain corresponding forecast model basic data;

4.2, in conjunction with the forecast model basic data, any described passenger flow forecast of distribution model of substitution claim 1 to 3;

4.3, calculate the passenger flow distributed data of corresponding time period.

Further, as preferably, further comprise the OD accessibility index that obtains in advance in the described step 4.1.

The present invention considers the relation between indexs such as track website attribute, topology of networks and relevant operation parameter and the passenger flow distribution, and different OD between influence each other, choose the discrete variable of describing website character and scale etc. as the influence factor of model, reduce the difficulty of obtaining of data, strengthened the practicality of model; Introduce OD accessibility index, described the influence that road network structure distributes for passenger flow more accurately; Setting up track traffic network passenger flow forecast of distribution based on non-collection meter model influences model, has not only improved the precision of prediction of model, has also strengthened the applicability of model; Simultaneously, proposed a kind of passenger flow distribution forecasting method, the passenger flow that can accurately predict under the net condition distributes.

Description of drawings

When considered in conjunction with the accompanying drawings, by the reference following detailed, can more completely understand the present invention better and learn wherein many attendant advantages easily, but accompanying drawing described herein is used to provide further understanding of the present invention, constitute a part of the present invention, illustrative examples of the present invention and explanation thereof are used for explaining the present invention, do not constitute to improper restriction of the present invention, wherein:

The process flow diagram of the structure track traffic for passenger flow forecast of distribution model that Fig. 1 provides for present embodiment.

The process flow diagram of the track traffic for passenger flow distribution forecasting method that Fig. 2 provides for present embodiment.

Embodiment

Describe with reference to the embodiments of the invention of Fig. 1-2.

For above-mentioned purpose, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.

As shown in Figure 1, a kind of track traffic for passenger flow forecast of distribution method for establishing model may further comprise the steps:

Select corresponding influence factor the indexs such as S1, the classification from track website generation traffic attraction, website, topology of networks and relevant operation parameter.

The utility function of S2, structure model, structure is shown below:

V _ij＝f(C _ij,D _j,AOD ^ij,XZ,GM,k)

Wherein, V _IjFor starting point is that selection terminal point in i station is the utility function at j station; C _IjBe the required admission fee in j station of standing from i; D _jFor _jThe traffic attraction of standing; XZ is for characterizing the 0-1 variable of website character.The division of website classification is as follows among the present invention: the inhabitation class is 1; The office class is 2; Other classes are 3; The office (inhabitation is dominant) of living is 4; The office (office is dominant) of living is 5; Hinge or commercial center are 6.The value condition of corresponding XZ is as shown in the table:

GM is for characterizing the 0-1 variable of website scale, if the whole day generating capacity at the station of setting out then is labeled as 1, otherwise is 0 greater than 30000 people; Same, the traffic attraction of terminal station is designated as 1 during greater than 30000 people, otherwise is 0.When the mark of set out station and terminal station was 1, the scale variable was 1, otherwise is 0.K is the 0-1 variable, and when i station or j station belong to ew line, constant term is 1, otherwise is 0; OD accessibility index can calculate by following formula.

${\mathrm{AOD}}^{\mathrm{rs}}=\ln \underset{k=1}{\overset{m}{\mathrm{\Σ}}}\exp \left(E_k^{\mathrm{rs}}\right)$

$E_k^{\mathrm{rs}}=-0.19\mathrm{IO}{t}_k^{\mathrm{rs}}-2.9\mathrm{lIV}{t}_k^{\mathrm{rs}}-0.1W{t}_k^{\mathrm{rs}}-7.5{\mathrm{VP}}_k^{\mathrm{rs}}-0.52T{\mathrm{rt}}_k^{\mathrm{rs}}-3.9T{\mathrm{rs}}_k^{\mathrm{rs}}-3.03{\mathrm{AC}}_k^{\mathrm{rs}}$

In the formula, AOD _RsFor the OD from r to s to OD accessibility index; E _k ^RsFor the OD from r to s to the k paths select effectiveness; M be from r to s OD to active path bar number.IOt _k ^RsThe turnover station time of representing the k paths; IVt _k ^RsThe riding time of representing the k paths; Wt _k ^RsThe platform Waiting time of representing the k paths; VP _k ^RsThe compartment load factor of representing the k paths; Trt _k ^RsRepresent that the transfer of k paths walks line time; Trs _k ^RsThe number of transfer of representing the k paths; AC _k ^RsThe angle expense of representing the k paths.

S3, the website generation traffic attraction that uses the individual method of representative will belong to collection counting certificate are converted into non-collection counting certificate, and concrete grammar is as follows:

At first, the present invention is by introducing weights omega _Ij, will collect counting according to being converted into non-collection counting certificate, corresponding computing formula is shown below.

$w_{\mathrm{ij}}=\frac{Q_{\mathrm{ij}}}{H_{\mathrm{ij}}}={\mathrm{rows}}_{\mathrm{ij}}\·\frac{{\mathrm{trip}}_{\mathrm{ij}}}{{\mathrm{trips}}_i}$

$P_{\mathrm{ij}}=\frac{\exp \left(V_{\mathrm{ij}}\right)}{\underset{j\∈C_j}{\mathrm{\Σ}}\exp \left(V_{\mathrm{ij}}\right)}$

$L\left(\mathrm{\θ}\right)=\underset{j}{\mathrm{\Σ}}\underset{i}{\mathrm{\Σ}}{w}_{\mathrm{ij}}\ln P_{\mathrm{ij}}$

Wherein, P _IjFor starting point is that the passenger flow selection j station at i station is as the probability of terminal point; C _jBe the set of the selectable terminus of passenger flow at i station for starting point; θ represents utility function V _IjIn all undetermined parameters; L (θ) is undetermined parameter maximum likelihood value; Q _IjFor starting point is that the passenger flow selection terminal point that i stands is the ratio at j station; H _Ij=1/rows _Ij, rows _IjFor being that i station terminal point is the representative quantity of choosing the passenger flow at j station from starting point; Trip _IjFor starting point is that i station terminal point is the volume of the flow of passengers at j station; Trips _iFor starting point is the passenger flow total amount that i stands.

S4, the undetermined parameter that uses the maximum likelihood estimation to demarcate among the L (θ) obtain forecast model, and the result is shown below.

V _ij＝0.1984·D _j/10 ⁴+0.1066·AOD ^ij+0.0809·XZ+0.2346·GM-0.1165·k

S5, set up track traffic for passenger flow forecast of distribution model.

As shown in Figure 2, a kind of track traffic for passenger flow distribution forecasting method may further comprise the steps:

Indexs such as S6, the classification according to the following track website generation traffic attraction that obtains in advance, website, OD accessibility index, topology of networks and relevant operation parameter, index basis of formation databases such as S7, extraction website traffic attraction, working time, website scale, website character, OD website accessibility;

Wherein, following track website generation traffic attraction can obtain by methods such as growth factor method, fuzzy matching methods; The topological structure of future network then can obtain according to the future plan that rail transportation operation department proposes with relevant operation parameter.The website classification can obtain by the Changing Pattern of analyzing whole day turnover station amount, thereby can determine website character variable.

S8, referring to Fig. 1 method, obtain the track passenger flow forecast of distribution model demarcated.

S9, in conjunction with basic data,, substitution passenger flow forecast of distribution model; Calculate the passenger flow distributed data of corresponding time period.

Though more than described the specific embodiment of the present invention, but those skilled in the art is to be understood that, these embodiments only illustrate, those skilled in the art can carry out various omissions, replacement and change to the details of said method and system under the situation that does not break away from principle of the present invention and essence.For example, merge the said method step, then belong to scope of the present invention thereby carry out the essence identical functions according to the identical method of essence to realize the identical result of essence.Therefore, scope of the present invention is only limited by appended claims.

Claims (5)

1. a track traffic for passenger flow forecast of distribution method for establishing model is characterized in that, may further comprise the steps:

1.1, the discrete variable of website character and scale introduce to be described, in conjunction with track website generation traffic attraction, topology of networks and relevant operation parameter, make up the utility function of passenger flow forecast of distribution model;

1.2, the website generation traffic attraction that uses the individual method of representative will belong to collection counting certificate is converted into non-collection counting certificate;

1.3, use maximum likelihood to estimate to demarcate passenger flow forecast of distribution model.

2. according to the described track traffic for passenger flow forecast of distribution of claim 1 method for establishing model, it is characterized in that, in described step 1.1, further comprise and introduce OD accessibility index, characterize the influence that network topology structure and traffic convenience distribute for track traffic for passenger flow.

3. according to the described track traffic for passenger flow forecast of distribution of claim 1 method for establishing model, it is characterized in that the utility function that makes up passenger flow forecast of distribution model in the described step 1.1 is specially: V _Ij=f (C _Ij, D _j, AOD ^Ij, XZ, GM, k)

Wherein, V _IjFor starting point is that selection terminal point in i station is the utility function at j station; C _IjBe the required admission fee in j station of standing from i; D _jTraffic attraction for the j station; XZ is for characterizing the 0-1 variable of website character, and is relevant with the website character at j station with the i station; GM is for characterizing the 0-1 variable of website scale, and is relevant with the scale at j station with the i station; K is the 0-1 variable, and when i station or j station belong to ew line, constant term is 1, otherwise is 0.

4. a track traffic for passenger flow distribution forecasting method is characterized in that, may further comprise the steps:

4.1, the classification according to the following track website generation traffic attraction that obtains in advance, website, topology of networks and relevant operation parameter, extract and obtain corresponding forecast model basic data;

4.2, in conjunction with the forecast model basic data, any described passenger flow forecast of distribution model of substitution claim 1 to 3;

4.3, calculate the passenger flow distributed data of corresponding time period.

5. according to the described track traffic for passenger flow distribution forecasting method of claim 4, it is characterized in that, further comprise the OD accessibility index that obtains in advance in the described step 4.1.

Images