Rail transit management method, system, equipment and medium

Abstract

The invention discloses a rail transit management method, system and device and a medium, and belongs to the technical field of rail transit, and the method comprises the steps: obtaining the current activity information, and preliminarily predicting the passenger flow during the activity through a first passenger flow prediction model built based on historical activity information; designing a target function by taking maximization of the rail transit income as a target, and generating an initial rail transit strategy during an activity period; based on the preliminarily predicted passenger flow volume and historical data of roads around the area where the activity is located, the congestion degree of self-driving routes of the roads around the activity area during the activity is predicted, and a detouring suggested route is generated based on the predicted congestion degree; and based on the actual activity data during the activity period, accurately predicting the passenger flow volume through the constructed second passenger flow volume prediction model, and adjusting and publishing the initial rail transit strategy based on the accurately predicted passenger flow volume. When large-scale activities are held, sharp increase of passenger flow volume can be effectively coped with, and the transportation efficiency and the passenger satisfaction degree are guaranteed.

Description

Technical Field

[0001] The present invention belongs to the technical field of rail transit, and specifically relates to a rail transit management method, system, equipment and medium. Background Art

[0002] When a large-scale event is held in a city, it usually leads to a significant increase in passenger flow in the city, especially in the areas surrounding the event. In particular, urban rail transit systems may experience full trains and long queues when faced with a surge in passenger flow in a short period of time.

[0003] In order to effectively cope with the surge in passenger traffic, urban rail transit systems generally increase the frequency or interval between various rail transit systems. However, if the frequency and interval between trains are increased without reasonable scheduling planning, the punctuality and stability of trains may be affected, which may lead to more congestion and passenger dissatisfaction. In addition, the capacity of the rail transit system is not unlimited. Especially during peak hours, when the density of trains reaches a certain level, it is difficult to significantly improve transportation efficiency even if the frequency is increased. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a rail transit management method, system, equipment and medium, which can effectively cope with the surge in passenger flow when holding large-scale events and ensure transportation efficiency and passenger satisfaction.

[0005] The present invention provides the following technical solutions:

[0006] In a first aspect, a rail transit management method is provided, comprising:

[0007] Step S1: obtaining current activity information, and using a first passenger flow prediction model established based on historical activity information to preliminarily predict passenger flow during the activity;

[0008] Step S2: Based on the preliminary predicted passenger flow, the objective function is designed to maximize the rail transit revenue, and the initial rail transit strategy during the activity period is generated and announced;

[0009] Step S3: based on the preliminary predicted passenger flow and historical data of roads around the activity area, predict the congestion level of self-driving routes around the activity area during the activity period, and generate and publish detour recommended routes based on the predicted congestion level;

[0010] Step S4: Based on the actual activity data during the activity, the passenger flow is accurately predicted by the constructed second passenger flow prediction model, and the initial rail transit strategy is adjusted and announced based on the accurately predicted passenger flow.

[0011] Optionally, in step S1, the mathematical expression of the first passenger flow prediction model is:

[0012] P(t)=F S (S) F T (t,t0,T day )·F D (t,t1,D)·F C (C)

[0013] F S (S) = k S ·S+P0

[0014]

[0015] F D (t,t1,D)=exp(-k D (t-t1))

[0016] F C (C) = 1-k C ·C

[0017] Among them, P(t) is the passenger flow initially predicted at time t, F s (S) is the activity scale impact function, F T (t,t0,T day ) is the activity time influence function, F D (t, t1, D) is the activity cycle influence function, F C (C) is the construction condition influence function, k S is the sensitivity coefficient of the activity scale, S is the activity scale, P0 is the basic passenger flow when there is no activity, A is the maximum amplitude of passenger flow fluctuation, t0 is the start time of the activity, T day is the length of a day, P avg is the average passenger flow, k D is the decay rate, t1 is the start time of the activity, D is the duration of the activity, k C is the construction impact coefficient, C is a binary variable of the construction situation. When there is no construction, C takes the value of 0, and when there is construction, C takes the value of 1.

[0018] Optionally, the step S2 specifically includes:

[0019] Based on the preliminary predicted passenger flow and historical data, obtain the rail transit lines that need to be adjusted;

[0020] Set the objective function to maximize the total revenue of rail transit operation and establish constraints;

[0021] Solve the objective function and output the optimal solution to realize the generation of rail transit adjustment strategy.

[0022] Optionally, the objective function is:

[0023]

[0024] Among them, J is the total operating revenue of rail transit, N is the total number of rail transit lines that need to be adjusted, T is the number of time periods in a day, and R ikt is the revenue of the kth route at time t, and is obtained based on the preliminary predicted passenger flow, O k is the operating cost of the kth line, f ikt is the departure frequency of the kth line at time t;

[0025] Constraints include: passenger flow carrying capacity constraints, passenger flow and demand matching constraints, and operating time constraints;

[0026] The passenger flow carrying capacity constraint is:

[0027]

[0028] in, is the predicted passenger flow from station i to station j at time t in the nth candidate rail transit strategy, C k is the vehicle capacity of the kth route; is the departure frequency of the kth line in the nth candidate rail transit strategy at time t;

[0029] The constraints for matching passenger flow with demand are:

[0030]

[0031] Among them, α ij is the adjustment factor, The passenger flow initially predicted from station i to station j at time t is obtained based on the passenger flow initially predicted by the first passenger flow prediction model;

[0032] The operating time constraints are:

[0033]

[0034] Among them, t start and t end are the operation start time and operation end time of line k respectively.

[0035] Optionally, the specific process of step S3 is as follows:

[0036] The following formula is used to predict the congestion level of the road corresponding to the rail transit platform i to platform j at time t;

[0037] C ijt =λ(1-Vt V max )F ijt +μA t +νS t

[0038] Among them, C ijt is the congestion index of the road from rail transit platform i to platform j at time t, λ is the congestion sensitivity coefficient, V t is the average speed of the self-driving vehicle at time t, V max is the maximum average speed of vehicles in the historical data without congestion, μ is the accident impact coefficient, A t is the number of accident reports at time t in the historical data, ν is the construction impact coefficient, S t is the number of construction information during the activity period, F ijt It is the self-driving traffic flow from rail transit platform i to platform j on the corresponding road at time t, and is obtained based on the initial predicted passenger flow;

[0039] Based on the congestion level of the self-driving route, the objective function is to minimize the self-driving time. The Dijkstra algorithm is used to solve the objective function and output the optimal detour route.

[0040] T ijt =D ij V t +ΔT(C ijt )

[0041] Among them, T ijt is the driving time from rail transit station i to station j at time t, D ij is the distance from the rail transit platform i to the corresponding road of platform j, ΔT(C ijt ) is based on C ijt Extended self-driving time.

[0042] Optionally, in step S4, the mathematical expression of the constructed second passenger flow prediction model is:

[0043] N(t)=αf a (t)+βf w (t)+γf h (t)+∈

[0044] f a (t) = W A A(t)+W S S(t)+W T T(t)+W D D(t)+W L L(t)

[0045] f w (t) = Wt t(t)+W P P(t)

[0046] f h (t)=K+W1H(t)+W2Th(t)+W3W(t)+W4SW(t)+W5Dh(t)

[0047] Among them, N(t) is the accurate predicted passenger flow at time t, f a (t) is the activity characteristic function, f w (t) is the weather condition function, f h (t) is the holiday effect function, ∈ is the random error term, α, β and γ are f a (t), f w (t) and f h (t) corresponding weight coefficient; A(t) is the actual activity attention at time t, S(t) is the actual activity scale at time t, T(t) is a categorical variable representing the activity type, D(t) is the activity duration at time t, L(t) is the location accessibility, W A , W S , W T , W D and W L are the weights corresponding to A(t), S(t), T(t), D(t) and L(t), t(t) is the temperature at time t, W t and W P are the weights corresponding to f(t) and P(t), P(t) is the precipitation at time t, K is the average passenger flow under daily conditions, H(t) is a binary variable indicating whether it is a holiday, Th(t) is a categorical variable about the type of holiday, W(t) is a binary variable indicating whether it is a weekday, SW(t) is a categorical variable indicating whether it is a special weekday, Dh(t) is the duration of the holiday, W1, W2, W3, W4 and W5 are the weights corresponding to H(t), Th(t), W(t), SW(t) and Dh(t), respectively.

[0048] Optionally, the initial rail transit strategy is adjusted and announced based on the accurately predicted passenger flow, including adjustment of the total frequency of rail transit and adjustment of rail transit operation hours;

[0049] The adjustment to the total frequency of rail transit is as follows:

[0050] B(t)=B0+ΔB=B0+δN(t)

[0051] Wherein, B(t) is the total frequency of rail transit after adjustment, B0 is the original total frequency of rail transit in the initial rail transit strategy, δ is the frequency adjustment coefficient, and N(t) is the passenger flow accurately predicted based on the second passenger flow prediction model;

[0052] The adjustments to rail transit operating hours are as follows:

[0053]

[0054] Among them, ΔT is the extension value of rail transit operation time, N threshold In order to accurately predict the threshold of passenger flow, τ is the operating time adjustment coefficient.

[0055] In a second aspect, a rail transit management system is provided, comprising:

[0056] A preliminary prediction module, used to obtain current activity information and use a first passenger flow prediction model established based on historical activity information to preliminarily predict passenger flow during the activity;

[0057] The rail transit strategy generation module is used to design the objective function based on the preliminary predicted passenger flow to maximize the rail transit revenue, generate the initial rail transit strategy during the activity period, and publish it;

[0058] A detour route generation module is used to predict the congestion level of self-driving routes around the activity area during the activity period based on the preliminary predicted passenger flow and historical data of roads around the activity area, and to generate and publish detour recommended routes based on the predicted congestion level;

[0059] The rail transit strategy optimization module is used to accurately predict the passenger flow based on the actual activity data during the activity by means of the constructed second passenger flow prediction model, and to adjust and announce the initial rail transit strategy based on the accurately predicted passenger flow.

[0060] In a third aspect, a computer device is provided, comprising a processor and a memory; wherein, when the processor executes a computer program stored in the memory, the steps of the rail transit management method described in any one of the first aspects are implemented.

[0061] In a fourth aspect, a computer-readable storage medium is provided for storing a computer program; when the computer program is executed by a processor, the steps of the rail transit management method described in any one of the first aspects are implemented.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] The present invention establishes a first passenger flow prediction model by combining current activity information with historical data, which can make a relatively accurate preliminary prediction of passenger flow during the activity, and provide a data basis for rail transit operators to formulate strategies. Secondly, the objective function is designed with the goal of maximizing rail transit revenue, and the initial rail transit strategy is generated accordingly, which helps to improve the overall operational efficiency and economic benefits of rail transit. At the same time, the timely publication of these strategies can also allow passengers to understand and reasonably arrange travel plans in advance; the present invention constructs a second passenger flow prediction model, which can accurately predict passenger flow based on actual activity data during the activity, and adjust the initial rail transit strategy accordingly. This dynamically adjusted strategy is not only more in line with the actual situation, but also can better meet the travel needs of passengers, and improve the service quality and passenger satisfaction of rail transit. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a flow chart of the steps of the rail transit management method of the present invention;

[0065] Figure 2 It is a structural schematic diagram of the rail transit management system of the present invention. DETAILED DESCRIPTION

[0066] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0067] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the term "comprising" and any variation thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0068] Example 1

[0069] like Figure 1 As shown, a rail transit management method comprises the following steps:

[0070] Step S1: obtaining current activity information, and using a first passenger flow prediction model established based on historical activity information to preliminarily predict the passenger flow during the activity.

[0071] Current event information and historical event information include location, scale, type, date and duration, etc.

[0072] The first passenger flow prediction model can use time series analysis, regression analysis or machine learning algorithms (such as random forest, neural network).

[0073] In this embodiment, the expression of the first passenger flow prediction model is:

[0074] P(t)=F S (S)·F T (t,t0,T day )·F D (t,t1,D)·F C (C)

[0075] F S (S) = k S ·S+P0

[0076]

[0077] F D (t,t1,D)=exp(-k D (t-t1))

[0078] F C (C) = 1-k C ·C

[0079] Among them, P(t) is the passenger flow initially predicted at time t, F s (S) is the activity scale impact function, F T (t,t0,T day ) is the activity time influence function, F D (t, t1, D) is the activity cycle influence function, F C (C) is the construction condition influence function, k S is the sensitivity coefficient of the activity scale, S is the activity scale, P0 is the basic passenger flow when there is no activity, A is the maximum amplitude of passenger flow fluctuation, t0 is the start time of the activity, T day is the length of a day, usually 24 hours, P avg is the average passenger flow, indicating the average passenger flow level without considering time fluctuations, k D is the decay rate, which indicates the speed at which the passenger flow decreases over time during the activity period, t1 is the start time of the activity, D is the duration of the activity, and k C is the construction impact coefficient, C is a binary variable of the construction situation. When there is no construction, C takes the value of 0, and when there is construction, C takes the value of 1.

[0080] The various coefficients of the first passenger flow prediction model need to be determined by fitting historical data.

[0081] Step S2: Based on the preliminary predicted passenger flow, the objective function is designed to maximize the rail transit revenue, and the initial rail transit strategy during the activity period is generated and announced.

[0082] Step S2 specifically includes the following sub-steps:

[0083] Step S21: Based on the preliminary predicted passenger flow and historical data, obtain the rail transit lines that need to be adjusted.

[0084] Rail transit routes that need to be adjusted include those that need to increase frequency or adjust routes to better serve the activity area. Historical data includes passenger flow data, historical traffic patterns (which routes and stations are prone to congestion), historical operation data, historical activity data, weather and holiday data, and data on various modes of transportation.

[0085] Step S22: Setting the objective function to maximize the total revenue of rail transit operation and establishing constraints.

[0086] The objective function is:

[0087]

[0088] Among them, J is the total operating revenue of rail transit, N is the total number of rail transit lines that need to be adjusted, T is the number of time periods in a day, and R ikt is the revenue of the kth route at time t, and is obtained based on the preliminary predicted passenger flow, O k is the operating cost of the kth line, including personnel costs, etc., f ikt is the departure frequency of the kth line at time t, that is, the number of departures per unit time.

[0089] R ikt The passenger flow based on the preliminary forecast can be obtained by referring to the prior art. As an option, the historical average ticket price is multiplied by the preliminary forecast passenger flow.

[0090] Constraints include: passenger flow carrying capacity constraints, passenger flow and demand matching constraints, and operating time constraints

[0091] The passenger flow carrying capacity constraint is:

[0092]

[0093] in, is the predicted passenger flow from station i to station j at time t in the nth candidate rail transit strategy, C k is the vehicle capacity of the kth route; is the departure frequency of the kth line in the nth candidate rail transit strategy at time t;

[0094] The constraints for matching passenger flow with demand are:

[0095]

[0096] Among them, α ij is the adjustment factor, The passenger flow initially predicted from station i to station j at time t is obtained based on the passenger flow initially predicted by the first passenger flow prediction model;

[0097] The operating time constraints are:

[0098]

[0099] Among them, t start and t end are the operation start time and operation end time of line k respectively.

[0100] Step S23: Solve the objective function and output the optimal solution for generating a rail transit adjustment strategy.

[0101] The optimal solution is to find the optimal rail transit adjustment strategy, which includes the total number of trips, departure frequency and total operating time of each rail transit line.

[0102] The method of solving the optimal solution of the objective function can refer to existing technologies, such as particle swarm optimization.

[0103] Step S3: Based on the preliminary predicted passenger flow and historical data of roads around the activity area, predict the congestion level of self-driving routes around the activity area during the activity, and generate and publish detour recommended routes based on the predicted congestion level.

[0104] Historical data of roads around the activity area, including but not limited to road speeds, accident reports, construction information, etc.

[0105] Step S3 specifically includes the following sub-steps:

[0106] Step S31: predicting the congestion level of the road corresponding to the rail transit platform i to the platform j at time t by the following formula;

[0107] C ijt =λ(1-V t V max )F ijt +μA t +νS t

[0108] Among them, Cijt is the congestion index of the road from rail transit platform i to platform j at time t, λ is the congestion sensitivity coefficient, V t is the average speed of the self-driving vehicle at time t, V max is the maximum average speed of vehicles in the historical data without congestion, μ is the accident impact coefficient, A t is the number of accident reports at time t in the historical data, ν is the construction impact coefficient, S t is the number of construction information during the activity period, F ijt It is the self-driving traffic flow from rail transit platform i to platform j on the corresponding road at time t, and is obtained based on the initial predicted passenger flow;

[0109] Step S32: Based on the congestion level of the self-driving route, minimizing the self-driving time is used as the objective function, and the Dijkstra algorithm is used to solve the objective function and output the optimal detour route;

[0110] T ijt =D ij V t +ΔT(C ijt )

[0111] Among them, T ijt is the driving time from rail transit station i to station j at time t, D ij is the distance from the rail transit platform i to the corresponding road of platform j, ΔT(C ijt ) is based on C ijt The extended self-driving time can be a nonlinear function, reflecting the impact of the congestion index on the self-driving time.

[0112] The specific process of solving the optimal detour path is as follows:

[0113] If G = (V, E) represents a road network graph, where V is a set of nodes and E is a set of edges, then the path with the minimum travel time can be obtained by the following optimization problem:

[0114]

[0115] Among them, P ij are all possible paths from point i to point j, w e is the travel time weight on edge e, which can be based on T ijt calculate.

[0116] Step S4: Based on the actual activity data during the activity, the passenger flow is accurately predicted by the constructed second passenger flow prediction model, and the initial rail transit strategy is adjusted and announced based on the accurately predicted passenger flow.

[0117] In step S4, the mathematical expression of the constructed second passenger flow prediction model is:

[0118] N(t)=αf a (t)+βf w (t)+γf h (t)+∈

[0119] f a (t) = W A A(t)+W S S(t)+W T T(t)+W D D(t)+W L L(t)

[0120] f w (t) = W t t(t)+W P P(t)

[0121] f h (t)=K+W1H(t)+W2Th(t)+W3W(t)+W4SW(t)+W5Dh(t)

[0122] Among them, N(t) is the accurate predicted passenger flow at time t, f a (t) is the activity characteristic function, f w (t) is the weather condition function, f h (t) is the holiday effect function, ∈ is the random error term, α, β and γ are f a (t), f w (t) and f h (t) corresponding weight coefficient; A(t) is the actual activity attention at time t, S(t) is the actual activity scale at time t, T(t) is a categorical variable representing the activity type, D(t) is the activity duration at time t, L(t) is the location accessibility, W A , W S , W T , W D and W L are the weights corresponding to A(t), S(t), T(t), D(t) and L(t), t(t) is the temperature at time t, W t and W Pare the weights corresponding to f(t) and P(t), P(t) is the precipitation at time t, K is the average passenger flow under daily conditions, H(t) is a binary variable indicating whether it is a holiday, Th(t) is a categorical variable about the type of holiday, W(t) is a binary variable indicating whether it is a weekday, SW(t) is a categorical variable indicating whether it is a special weekday (for example, the last day of a long holiday), Dh(t) is the duration of the holiday, W1, W2, W3, W4 and W5 are the weights corresponding to H(t), Th(t), W(t), SW(t) and Dh(t), respectively.

[0123] In step S4, the initial rail transit strategy is adjusted and announced based on the accurately predicted passenger flow, including the adjustment of the total number of rail transit trains and the adjustment of the rail transit operation hours.

[0124] The adjustments to the total frequency of rail transit are as follows:

[0125] B(t)=B0+ΔB=B0+δN(t)

[0126] Wherein, B(t) is the total frequency of rail transit after adjustment, B0 is the original total frequency of rail transit in the initial rail transit strategy, δ is the frequency adjustment coefficient, and N(t) is the passenger flow accurately predicted based on the second passenger flow prediction model;

[0127] The adjustment of rail transit operation hours is as follows:

[0128]

[0129] Among them, ΔT is the extension value of rail transit operation time, N threshold In order to accurately predict the threshold of passenger flow, τ is the operating time adjustment coefficient.

[0130] Of course, in this embodiment, the rail transit strategy can be released through multiple channels after adjustment, such as mobile applications, social media, official websites, station display screens, broadcasts and SMS notifications, etc.

[0131] The adjusted rail transit strategy can be converted into icons, timetables or map indicators; it can also be used to subsequently improve the information release strategy based on passenger feedback.

[0132] Example 2

[0133] like Figure 2 As shown, a rail transit management system is provided, comprising:

[0134] A preliminary prediction module, used to obtain current activity information and use a first passenger flow prediction model established based on historical activity information to preliminarily predict passenger flow during the activity;

[0135] The rail transit strategy generation module is used to design the objective function based on the preliminary predicted passenger flow to maximize the rail transit revenue, generate the initial rail transit strategy during the activity period, and publish it;

[0136] A detour route generation module is used to predict the congestion level of self-driving routes around the activity area during the activity period based on the preliminary predicted passenger flow and historical data of roads around the activity area, and to generate and publish detour recommended routes based on the predicted congestion level;

[0137] The rail transit strategy optimization module is used to accurately predict the passenger flow based on the actual activity data during the activity by means of the constructed second passenger flow prediction model, and to adjust and announce the initial rail transit strategy based on the accurately predicted passenger flow.

[0138] For more specific processes of the above method, please refer to the corresponding contents disclosed in the aforementioned embodiments, which will not be repeated here.

[0139] Example 3

[0140] The present invention provides a computer device, comprising a processor and a memory; wherein the processor implements the steps of the above-mentioned rail transit management method when executing a computer program stored in the memory.

[0141] For more specific processes of the above method, please refer to the corresponding contents disclosed in the aforementioned embodiments, which will not be repeated here.

[0142] Example 4

[0143] The present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, the steps of the above-mentioned rail transit management method are implemented.

[0144] For more specific processes of the above method, please refer to the corresponding contents disclosed in the aforementioned embodiments, which will not be repeated here.

[0145] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system, device, and storage medium disclosed in the embodiment, since they correspond to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0146] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention or some parts of the embodiments.

[0147] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A rail transit management method, characterized in that: include: Step S1: obtaining current activity information, and using a first passenger flow prediction model established based on historical activity information to preliminarily predict passenger flow during the activity; Step S2: Based on the preliminary predicted passenger flow, the objective function is designed to maximize the rail transit revenue, and the initial rail transit strategy during the activity period is generated and announced; Step S3: Based on the initially predicted passenger flow and historical data of roads around the activity area, predict the congestion level of self-driving routes around the activity area during the activity period, and generate and publish detour recommended routes based on the predicted congestion level; Step S4: Based on the actual activity data during the activity, the passenger flow is accurately predicted by the constructed second passenger flow prediction model, and the initial rail transit strategy is adjusted and announced based on the accurately predicted passenger flow.

2. The rail transit management method according to claim 1, characterized in that: In step S1, the mathematical expression of the first passenger flow prediction model is: P(t)=F S (S)·F T (t,t0,T day )·F D (t,t1,D)·F C (C) F S (S)=k S ·S+P0 F D (t,t1,D)=exp(-k D (t-t1)) F C (C)=1-k C ·C Among them, P(t) is the passenger flow initially predicted at time t, F S (S) is the activity scale impact function, F T (t,t0,T day ) is the activity time influence function, F D (t, t1, D) is the activity cycle influence function, F C (C) is the construction condition influence function, k S is the sensitivity coefficient of the activity scale, S is the activity scale, P0 is the basic passenger flow when there is no activity, A is the maximum amplitude of passenger flow fluctuation, t0 is the start time of the activity, T day is the length of a day, P avg is the average passenger flow, k D is the decay rate, t1 is the start time of the activity, D is the duration of the activity, k C is the construction impact coefficient, C is a binary variable of the construction situation. When there is no construction, C takes the value of 0, and when there is construction, C takes the value of 1.

3. The rail transit management method according to claim 1, characterized in that: The step S2 specifically includes: Based on the preliminary predicted passenger flow and historical data, obtain the rail transit lines that need to be adjusted; Set the objective function to maximize the total revenue of rail transit operation and establish constraints; Solve the objective function and output the optimal solution to realize the generation of rail transit adjustment strategy.

4. The rail transit management method according to claim 3, characterized in that: The objective function is: Among them, J is the total operating revenue of rail transit, N is the total number of rail transit lines that need to be adjusted, T is the number of time periods in a day, and R ikt is the revenue of the kth route at time t, and is obtained based on the preliminary predicted passenger flow, O k is the operating cost of the kth line, f ikt is the departure frequency of the kth line at time t; Constraints include: passenger flow carrying capacity constraints, passenger flow and demand matching constraints, and operating time constraints; The passenger flow carrying capacity constraint is: in, is the predicted passenger flow from station i to station j at time t in the nth candidate rail transit strategy, C k is the vehicle capacity of the kth route; is the departure frequency of the kth line in the nth candidate rail transit strategy at time t; The constraints for matching passenger flow with demand are: Among them, α ij is the adjustment factor, The passenger flow initially predicted from station i to station j at time t is obtained based on the passenger flow initially predicted by the first passenger flow prediction model; The operating time constraints are: Among them, t start and t end are the operation start time and operation end time of line k respectively.

5. The rail transit management method according to claim 1, characterized in that: The specific process of step S3 is as follows: The following formula is used to predict the congestion level of the road corresponding to the rail transit platform i to platform j at time t; C ijt =λ(1-V t V max )F ijt +μA t +νS t Among them, C ijt is the congestion index of the road from rail transit platform i to platform j at time t, λ is the congestion sensitivity coefficient, V t is the average speed of the self-driving vehicle at time t, V max is the maximum average speed of vehicles in the historical data without congestion, μ is the accident impact coefficient, A t is the number of accident reports at time t in the historical data, v is the construction impact coefficient, S t is the number of construction information during the activity period, F ijt It is the self-driving traffic flow from rail transit platform i to platform j on the corresponding road at time t, and is obtained based on the initial predicted passenger flow; Based on the congestion level of the self-driving route, the objective function is to minimize the self-driving time. The Dijkstra algorithm is used to solve the objective function and output the optimal detour route. T ijt =D ij V t +ΔT(C ijt ) Among them, T ijt is the driving time from rail transit station i to station j at time t, D ij is the distance from the rail transit platform i to the corresponding road of platform j, ΔT(C ijt ) is based on C ijt Extended self-driving time.

6. The rail transit management method according to claim 1, characterized in that: In step S4, the mathematical expression of the constructed second passenger flow prediction model is: N(t)=αf a (t)+βf w (t)+γf h (t)+∈ f a (t)=W A A(t)+W S S(t)+W T T(t)+W D D(t)+W L L(t) f w (t)=W t t(t)+W P P(t) f h (t)=K+W1H(t)+W2Th(t)+W3W(t)+W4SW(t)+W5Dh(t) Among them, N(t) is the accurate predicted passenger flow at time t, f a (t) is the activity characteristic function, f w (t) is the weather condition function, f h (t) is the holiday effect function, ∈ is the random error term, α, β and γ are f a (t), f w (t) and f h (t) corresponding weight coefficient; A(t) is the actual activity attention at time t, S(t) is the actual activity scale at time t, T(t) is a categorical variable representing the activity type, D(t) is the activity duration at time t, L(t) is the location accessibility, W A , W S , W T , W D and W L are the weights corresponding to A(t), S(t), T(t), D(t) and L(t), t(t) is the temperature at time t, W t and W P are the weights corresponding to t(t) and P(t), P(t) is the precipitation at time t, K is the average passenger flow under daily conditions, H(t) is a binary variable indicating whether it is a holiday, Th(t) is a categorical variable about the type of holiday, W(t) is a binary variable indicating whether it is a weekday, SW(t) is a categorical variable indicating whether it is a special weekday, Dh(t) is the duration of the holiday, W1, W2, W3, W4 and W5 are the weights corresponding to H(t), Th(t), W(t), SW(t) and Dh(t), respectively.

7. The rail transit management method according to claim 1, characterized in that: The adjustment and publication of the initial rail transit strategy based on the accurate predicted passenger flow, including the adjustment of the total frequency of rail transit and the adjustment of rail transit operation hours; The adjustment to the total frequency of rail transit is as follows: B(t)=B0+ΔB=B0+δN(t) Wherein, B(t) is the total frequency of rail transit after adjustment, B0 is the original total frequency of rail transit in the initial rail transit strategy, δ is the frequency adjustment coefficient, and N(t) is the passenger flow accurately predicted based on the second passenger flow prediction model; The adjustments to rail transit operating hours are as follows: Among them, ΔT is the extension value of rail transit operation time, N threshold In order to accurately predict the threshold of passenger flow, τ is the operating time adjustment coefficient.

8. A rail transit management system, characterized in that: include: A preliminary prediction module, used to obtain current activity information and use a first passenger flow prediction model established based on historical activity information to preliminarily predict passenger flow during the activity; The rail transit strategy generation module is used to design the objective function based on the preliminary predicted passenger flow to maximize the rail transit revenue, generate the initial rail transit strategy during the activity period, and publish it; A detour route generation module is used to predict the congestion level of self-driving routes around the activity area during the activity period based on the preliminary predicted passenger flow and historical data of roads around the activity area, and to generate and publish detour recommended routes based on the predicted congestion level; The rail transit strategy optimization module is used to accurately predict the passenger flow based on the actual activity data during the activity by means of the constructed second passenger flow prediction model, and to adjust and announce the initial rail transit strategy based on the accurately predicted passenger flow.

9. A computer device, characterized in that: It comprises a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the steps of the rail transit management method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: Used to store computer programs; when the computer programs are executed by the processor, the steps of the rail transit management method described in any one of claims 1-7 are implemented.

Images