CN-115640919-B - Urban rail transit route guidance information generation method and system

CN115640919BCN 115640919 BCN115640919 BCN 115640919BCN-115640919-B

Abstract

The invention relates to a method and a system for generating guidance information of an urban rail transit path, and relates to the field of traffic passenger flow guidance, wherein the method comprises the steps of constructing a guidance information display form utility model and a guidance information content utility model based on information entropy, conditional entropy and information gain; the method comprises the steps of determining a guidance information perception coefficient of a path according to a guidance information display form utility model and a guidance information content utility model, constructing a passenger path selection model based on the guidance information perception coefficient, wherein the passenger path selection model is an improved Logit model, calibrating model parameters of the passenger path selection model according to different types of passengers, calibrating the passenger path selection model based on different model parameters, and generating guidance information with minimum distribution unbalance degree of passenger flow as a target. The invention relieves the problem of unbalanced distribution of the passenger flow of the road network.

Inventors

XU XINYUE
CAI CHANGJUN
LIU JUN
ZHANG KE

Assignees

北京交通大学
广州地铁集团有限公司

Dates

Publication Date: 20260505
Application Date: 20220620

Claims (6)

1. The method for generating the urban rail transit route guidance information is characterized by comprising the following steps of: Constructing an induced information display form utility model and an induced information content utility model based on information entropy, conditional entropy and information gain, wherein the information entropy is the information entropy of a path selection decision event of an urban rail transit passenger, the conditional entropy is the conditional entropy of the path selection decision event under the condition that one path attribute takes one display form level, one path attribute comprises a plurality of display forms, each display form comprises a plurality of display form levels, and the information gain is the information gain of the path selection decision event by the display form level of the path attribute; determining an induced information perception coefficient of a path according to the induced information display form utility model and the induced information content utility model; constructing a passenger path selection model based on the induction information perception coefficient, wherein the passenger path selection model is an improved Logit model; performing model parameter calibration on the passenger path selection model according to different types of passengers; Generating induction information by taking the minimum degree of unbalanced passenger flow distribution as a target on the basis of the passenger path selection model calibrated by different model parameters; the utility model of the induced information display form is that each path attribute corresponding to the set display form takes the sum of normalized information gains of the level of the set display form; The induced information content utility model is respectively represented by information particle number and information accuracy; The information particle number is expressed as: ; Wherein, the , , The information particle number representing the path a, Representing a non-zero positive number, The average full load rate of path a is indicated, A set of alternative paths is represented and, Represents the full rate of the interval from node u to node v, Represents the traffic passing through the interval from node u to node v, Represents the number of trains passing through the section from node u to node v, which is a point on path a, Indicating the number of passengers the train is rated to carry, Representing the total number of intervals on path a; the information accuracy is expressed as: ; Wherein, the The coefficient of variation of the travel time is represented, , Representing OD pairs Standard deviation of travel time within one day, Representing OD pairs An average value of travel time within a day of (a), The starting position is indicated as such, Indicating the end position; determining an induced information perception coefficient of a path according to the induced information display form utility model and the induced information content utility model, wherein the method specifically comprises the following steps: determining the utility of the guidance information of the path a under the display form G according to the utility model of the guidance information display form and the utility model of the guidance information content; Determining an induction information perception coefficient of the path a according to the induction information utility; The utility of the induction information is expressed as: ; Wherein, the Representing the utility model of the induction information display form; The induction information perception coefficient is expressed as: ; Wherein, the Representing OD pairs Is a path of (a) Is provided with the perceptual coefficients of the presentation form G, Is a natural constant; The improved Logit model is expressed as: ; Wherein, the Representing a path Is used to determine the selection probability of (1), , , , , In the form of a vector which is a vector, Represents the travel time attribute coefficient, Represents the coefficient of the attribute of the degree of congestion, Representing the transfer attribute coefficient, Representing a path Is used to determine the negative utility vector of (1), Is a path Is used for the travel time of the car, Is a path Is used for the degree of congestion of a vehicle, Is a path Is used for the transfer times of the number of times, To induce information perception coefficient Is a function of (2).
2. The urban rail transit route guidance information generation method according to claim 1, wherein the information entropy is expressed as: ; wherein I represents a passenger's routing decision event, I represents a routing result, , Representing a set of alternative paths; The conditional entropy is expressed as: ; Wherein, the Represents the presentation form level of the path attribute f, , Representing a presentation form level set; the information gain is expressed as: 。
3. the urban rail transit route guidance information generation method according to claim 1, wherein the route attributes include travel time attributes, crowdedness attributes, and transfer attributes; The display forms comprise a travel time display form, a crowdedness display form and a transfer display form, wherein the travel time display form is a text display form, the crowdedness display form comprises a text display form, a picture display form and a photo display property, and the transfer display form is a text display form.
4. The urban rail transit route guidance information generation method according to claim 1, wherein the model parameters include a travel time attribute coefficient, a congestion degree attribute coefficient, and a transfer attribute coefficient.
5. The urban rail transit route guidance information generation method according to claim 4, wherein the objective function targeting minimization of the degree of unbalance of the passenger flow distribution is expressed as: ; wherein E represents the road network full load rate distribution entropy of the set region, g represents the discrete value of the interval full load rate, The full load rate in the total road network interval number representing the set area is The percentage of the interval; The passenger path selection model calibrated based on different model parameters aims at minimizing the unbalanced degree of passenger flow distribution to generate induction information, and specifically comprises the following steps: For each type of passenger, combining the display form levels corresponding to the different path attribute display forms of each path and each path to determine an induction information display scheme set, wherein each display scheme in the induction information display scheme set comprises the display form level corresponding to the path attribute display form of each path; Traversing the guidance information display scheme set, calculating the selection probability of the route according to the passenger route selection model based on each guidance information display scheme, and determining the road network full load rate distribution entropy of the set area according to the selection probability of the route; and outputting the guidance information display scheme corresponding to the minimum value in the road network full load rate distribution entropy of the plurality of setting areas obtained through traversing as final guidance information.
6. An urban rail transit route guidance information generation system, comprising: The utility model construction module is used for constructing an induced information display form utility model and an induced information content utility model based on information entropy, conditional entropy and information gain, wherein the information entropy is the information entropy of a path selection decision event of an urban rail transit passenger, the conditional entropy is the conditional entropy of the path selection decision event under the condition that one path attribute takes one display form level, one path attribute comprises a plurality of display forms, each display form comprises a plurality of display form levels, and the information gain is the information gain of the path selection decision event of the display form level of the path attribute; The induction information perception coefficient determining module is used for determining induction information perception coefficients of paths according to the induction information display form utility model and the induction information content utility model; the passenger path selection model building module is used for building a passenger path selection model based on the induction information perception coefficient, and the passenger path selection model is an improved Logit model; the model parameter calibration module is used for calibrating model parameters of the passenger path selection model according to different types of passengers; The induction information generation module is used for generating induction information with the aim of minimizing the unbalanced degree of the passenger flow distribution based on the passenger path selection model calibrated by different model parameters; the utility model of the induced information display form is that each path attribute corresponding to the set display form takes the sum of normalized information gains of the level of the set display form; The induced information content utility model is respectively represented by information particle number and information accuracy; The information particle number is expressed as: ; Wherein, the , , The information particle number representing the path a, Representing a non-zero positive number, The average full load rate of path a is indicated, A set of alternative paths is represented and, Represents the full rate of the interval from node u to node v, Represents the traffic passing through the interval from node u to node v, Represents the number of trains passing through the section from node u to node v, which is a point on path a, Indicating the number of passengers the train is rated to carry, Representing the total number of intervals on path a; the information accuracy is expressed as: ; Wherein, the The coefficient of variation of the travel time is represented, , Representing OD pairs Standard deviation of travel time within one day, Representing OD pairs An average value of travel time within a day of (a), The starting position is indicated as such, Indicating the end position; determining an induced information perception coefficient of a path according to the induced information display form utility model and the induced information content utility model, wherein the method specifically comprises the following steps: determining the utility of the guidance information of the path a under the display form G according to the utility model of the guidance information display form and the utility model of the guidance information content; Determining an induction information perception coefficient of the path a according to the induction information utility; The utility of the induction information is expressed as: ; Wherein, the Representing the utility model of the induction information display form; The induction information perception coefficient is expressed as: ; Wherein, the Representing OD pairs Is a path of (a) Is provided with the perceptual coefficients of the presentation form G, Is a natural constant; The improved Logit model is expressed as: ; Wherein, the Representing a path Is used to determine the selection probability of (1), , , , , In the form of a vector which is a vector, Represents the travel time attribute coefficient, Represents the coefficient of the attribute of the degree of congestion, Representing the transfer attribute coefficient, Representing a path Is used to determine the negative utility vector of (1), Is a path Is used for the travel time of the car, Is a path Is used for the degree of congestion of a vehicle, Is a path Is used for the transfer times of the number of times, To induce information perception coefficient Is a function of (2).

Description

Urban rail transit route guidance information generation method and system Technical Field The invention relates to the technical field of traffic passenger flow induction, in particular to a method and a system for generating urban rail transit route induction information. Background Along with the continuous development of urban rail transit systems in large cities, the passenger flow is increased increasingly, and higher requirements are put forward on subway networked operation, and the existing research shows that the induction information has a great guiding effect on the path selection behavior of passengers and is an important component part of a passenger transport organization method. At present, the Chinese research is multi-oriented to information content research such as effective path search, and the like, so that the research on the passenger travel path selection behavior rule under the common influence of the guidance information content and the display form is very few, and a personalized guidance information generation method suitable for urban rail transit passenger selection preference is lacking, so that a personalized generation method for urban rail transit path guidance information content and display form is needed, and decision support is provided for subway passenger flow guidance information generation. Disclosure of Invention The invention aims to provide a method and a system for generating urban rail transit path induction information, which relieve the problem of unbalanced distribution of road network passenger flows. In order to achieve the above object, the present invention provides the following solutions: a method for generating urban rail transit route guidance information comprises the following steps: Constructing an induced information display form utility model and an induced information content utility model based on information entropy, conditional entropy and information gain, wherein the information entropy is the information entropy of a path selection decision event of an urban rail transit passenger, the conditional entropy is the conditional entropy of the path selection decision event under the condition that one path attribute takes one display form level, one path attribute comprises a plurality of display forms, each display form comprises a plurality of display form levels, and the information gain is the information gain of the path selection decision event by the display form level of the path attribute; determining an induced information perception coefficient of a path according to the induced information display form utility model and the induced information content utility model; constructing a passenger path selection model based on the induction information perception coefficient, wherein the passenger path selection model is an improved Logit model; performing model parameter calibration on the passenger path selection model according to different types of passengers; And generating induction information by taking the minimum degree of unbalanced passenger flow distribution as a target on the basis of the passenger path selection model calibrated by different model parameters. Optionally, the information entropy is expressed as: Wherein I represents a passenger's path selection decision event, I represents a path selection result, I e I R,IR＝(1,2,3,...),IR represents an alternative path set; The conditional entropy is expressed as: H(I|l)=-p(l)H(I|Lf=l); Wherein L f represents the presentation form level of the path attribute f, Representing a presentation form level set; The information gain is expressed as g (I, l) =h (I) -H (i|l). Optionally, the path attribute includes a travel time attribute, a crowdedness attribute, and a transfer attribute; The display forms comprise a travel time display form, a crowdedness display form and a transfer display form, wherein the travel time display form is a text display form, the crowdedness display form comprises a text display form, a picture display form and a photo display property, and the transfer display form is a text display form. Optionally, the utility model of the induced information display form takes the sum of normalized information gains of the level of the set display form for each path attribute corresponding to the set display form; The induced information content utility model is respectively represented by information particle number and information accuracy; The information particle count is represented as N R,a＝1/(ε'+Loadave,a), Wherein, the Load u,v＝qu,v/(Ntrain×λtrain),NR,a represents the information particle count of path a, ε' represents a non-zero positive number, load ave,a represents the average full rate of path a, I R represents the set of alternative paths, load u,v represents the full rate of the section from node u to node v, q u,v represents the volume of traffic passing through the section from node u to node v, N train represents the number of trains passing throug