A method for selecting a strategy of collaborative operation organization of a city rail train and a city rail train

Abstract

This invention relates to a method for selecting a coordinated operation strategy for urban rail transit and urban rail transit trains, comprising: constructing a set of alternative operation strategies for urban rail transit and urban rail transit trains from the perspectives of routes, stops, and train formations; determining feasible strategies suitable for urban rail transit and urban rail transit trains based on the set of alternative strategies; and establishing a coordinated optimization model for the operation schemes of urban rail transit and urban rail transit trains based on the feasible strategies, and solving for the route departure frequency, stop scheme, and train formation number under each feasible strategy. This invention provides a clear answer to "which operation organization level strategy to adopt" in specific situations of multi-mode rail transit, thereby making up for the lack of a set of operation organization level strategy selection methods for urban rail transit and urban rail transit to cope with variable passenger flows.

Description

Technical Field

[0001] This invention relates to the field of urban rail transit technology, and in particular to a method for selecting a coordinated operation strategy for urban rail and suburban rail trains. Background Technology

[0002] With the acceleration of urbanization in my country, regional coordinated development has become a new trend. Within urban agglomerations and metropolitan areas, regional rail transit composite network systems composed of multiple rail transit systems have become the main force driving the development of urban clusters within the region, especially urban rail transit and suburban rail transit as typical examples.

[0003] The operation organization of urban rail transit and suburban rail transit exhibits diverse characteristics, with varying needs for operation organization in different scenarios. While existing research on rail transit operation organization strategies is relatively in-depth, providing advantages, disadvantages, and clear applicable scopes for each strategy, current research on urban rail transit is limited. Few studies delve into the selection of multi-mode rail transit operation organization strategies, lacking a clear method for selecting operation organization strategies to cope with the variable passenger flow of urban rail transit and suburban rail transit. It is difficult to provide a clear answer to "which specific collaborative operation organization strategy should be adopted in specific situations." Summary of the Invention

[0004] The purpose of this invention is to provide a method for selecting a coordinated operation strategy for urban rail transit and suburban rail trains, which optimizes the rail transit network from the perspective of operation organization, realizes the deep integration of urban rail transit and suburban rail transit, and thus improves the quality of passenger travel.

[0005] The technical solution to achieve the purpose of this invention is as follows:

[0006] A method for selecting a coordinated operation strategy for urban rail transit and suburban rail trains, the method comprising:

[0007] S1 constructs a set of alternative strategies for the operation organization of urban rail and suburban rail trains from the perspectives of routes, stops, and train formation;

[0008] S2, based on the set of alternative strategies in S1, determines feasible strategies suitable for urban rail and suburban rail trains.

[0009] S3, based on feasible strategies of S2, aims to minimize passenger travel time costs and enterprise total operating costs. It establishes a collaborative optimization model for the operation schemes of urban rail and suburban rail trains, and solves for the route departure frequency, station stopping scheme and train formation number under each feasible strategy.

[0010] Furthermore: the S1 includes three strategies for route planning: cross-line route planning, long-distance route planning, and single route planning; two strategies for stopping at each station and non-stopping at each station; and two strategies for train formation: fixed train formation and mixed multi-train formation.

[0011] Furthermore, the steps for determining feasible strategies for the coordinated operation of urban rail and suburban rail in S2 include:

[0012] Step 1: Determine if the cross-line condition is met. If it is, proceed to Step 2; otherwise, proceed to Step 4. The cross-line condition is the cross-line passenger flow Q. 跨 Both the cross-line passenger flow ratio λ and the cross-line passenger flow ratio λ are greater than the corresponding set cross-line passenger flow threshold and cross-line passenger flow ratio threshold.

[0013] Step 2: If the imbalance coefficient β of the line section is greater than the set imbalance threshold, then it is determined that the passenger flow distribution is uneven and proceeds to Step 3; otherwise, proceeds to Step 6.

[0014] Step 3: If the number of stations with a corresponding passenger flow coefficient γ greater than or equal to the set passenger flow peak value or less than or equal to the set passenger flow trough value exceeds the set first proportion, it is recommended to operate a non-stop-at-station strategy; if the number of stations with a corresponding cross-section load factor η greater than or equal to the set cross-section load factor peak value or less than or equal to the set cross-section load factor trough value exceeds the set second proportion, it is recommended to operate a multi-train mixed-run strategy; otherwise, proceed to Step 6.

[0015] Step 4: If the imbalance coefficient β of the line section is greater than the set imbalance threshold, then it is determined that the passenger flow distribution is uneven and proceeds to Step 5; otherwise, proceeds to Step 6.

[0016] Step 5: If the number of stations with a corresponding passenger flow coefficient γ greater than or equal to the set passenger flow peak value or less than or equal to the set passenger flow trough value exceeds the set first proportion, and further judgment is made: if such stations are concentrated in the same section formed by two turnaround stations, it is recommended to operate a long-distance and short-distance route strategy; if such stations are scattered in different sections formed by two turnaround stations, it is recommended to operate a non-stop-at-station strategy; except for the above two cases, it is recommended to operate a long-distance and short-distance route strategy and a non-stop-at-station strategy.

[0017] If the number of stations with a corresponding cross-section load factor η greater than or equal to the set cross-section load factor peak value or less than or equal to the set cross-section load factor valley value exceeds the set second proportion, then it is recommended to implement a multi-train mixed-run strategy.

[0018] Otherwise, proceed to Step 6;

[0019] Step 6: Maintain the existing operational organizational strategy.

[0020] Furthermore: The specific calculation formula for the cross-line passenger flow ratio λ in Step 1 is as follows:

[0021]

[0022] Among them, Q 跨 To calculate cross-line passenger flow within a given time period, Q 本 This refers to the passenger flow on this line within the statistical period.

[0023] The specific formula for calculating the cross-sectional imbalance coefficient β is as follows:

[0024]

[0025] Among them, Q max M represents the maximum cross-sectional passenger flow, and Q represents the number of one-way line sections. m Let m be the passenger flow at the m-th cross-section;

[0026] Station i Passenger flow coefficient γ i The specific calculation formula is as follows:

[0027]

[0028] in, Q represents the maximum hourly passenger volume of station i. u Let k be the hourly passenger / passage distance for station u, and k be the number of stations.

[0029] The specific formula for calculating the cross-sectional load factor η is as follows:

[0030]

[0031] in, N represents the passenger flow at section m within the statistical time period; N represents the number of trains passing through section m in a certain direction within the statistical time period; and C represents the train's passenger capacity.

[0032] Furthermore: the collaborative optimization model for the operation schemes of urban rail and suburban rail trains in S3 includes:

[0033] The objective function is:

[0034] min Z1=t w ·a1·p w

[0035] min Z2+Z3

[0036]

[0037]

[0038] The constraints are:

[0039]

[0040] In the formula: Z1 represents the passenger travel cost; t w Let OD be the travel time for passenger group w; a1 be the passenger time cost coefficient; p w Z1 represents the number of passengers in passenger group w at OD; Z2 represents the cost per kilometer of train travel. For overpass and short-pass routes; This is the length of the intersection of this line; This refers to the number of train formations on this line. a1 represents the number of train formations on cross-line and short-line routes; a2 represents the operating cost coefficient per unit length. This refers to the frequency of train services on this line. For train frequency on cross-line and short-line routes; The minimum and maximum permitted train departure frequencies for cross-line and short-line routes; These are the minimum and maximum train departure frequencies allowed for this line. The maximum cross-sectional passenger flow for overpass routes and local routes; P represents the maximum cross-sectional passenger flow on this line; η represents the carriage capacity; SE represents the train occupancy rate; and SE represents the interval set. The value is 1 if the line passes through interval se, and 0 otherwise. The value t indicates that the cross-line intersection and short-line intersection pass through the interval se, and is 1 if they are both 1 or 0 otherwise; min To track train intervals; The variable is a 0-1 variable indicating whether the two turnaround stations of the cross-line route stop; C is the set of routes other than the main route; Z3 is the train stopping cost; a3 is the single stopping cost coefficient.

[0041] Furthermore, the NSGA-II algorithm is used to solve the collaborative optimization model of the operation schemes of urban rail and suburban rail trains, and the departure frequency, stopping scheme and train formation number under each feasible strategy are obtained.

[0042] Furthermore, in the NSGA-II algorithm, decision variables include route departure frequency, station stopping scheme, and number of train groups. Each chromosome represents a specific operation scheme. The route departure frequency and number of train groups are encoded using real numbers, while the station stopping scheme is encoded using 0-1.

[0043] Furthermore, the crossover and mutation probabilities in the NSGA-II algorithm are adaptively adjusted using the Q-Learning algorithm.

[0044] Where the reward function r c Adjusting the crossover probability P c Through the reward function r m Adjusting the mutation probability P m :

[0045]

[0046] In the formula, For the i-th chromosome of the t-th generation population fitness value; For the i-th chromosome of the (t+1)-th generation population The fitness value; Popsize is the population size; The optimal fitness of the t generation population;

[0047] The optimal fitness of the population in generation t+1 is given.

[0048] Furthermore: the number of rows in the Q-value table corresponds to the number of environment states in NSGA-II, and the number of columns corresponds to the number of actions in action set A; initially, the Q-value table is set as an all-zero matrix.

[0049] Furthermore: the t-th generation environmental state s of NSGA-II t for;

[0050] s t =w1·fit * +w2·div * +w3·Best * (w1+w2+w3=1)

[0051]

[0052]

[0053] Among them, fit * The average fitness of the population; div * For population diversity; Best * The optimal fitness of the population; w1, w2, and w3 are respectively the optimal fitness of the population. * div * Best * The weighting coefficients.

[0054] In addition, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for selecting the coordinated operation organization strategy of urban rail and suburban rail trains.

[0055] The present invention also provides an electronic device, comprising:

[0056] Memory, used to store computer programs;

[0057] A processor is used to execute the computer program to implement the steps of the above-mentioned method for selecting the coordinated operation organization strategy of urban rail and suburban rail trains.

[0058] The significant advantages of this invention compared to existing technologies are:

[0059] (1) This invention proposes a selection method for the coordinated operation organization strategy of urban rail transit and urban rail transit from three aspects: selection ideas, basis indicators, and specific processes. It selects feasible strategies that conform to the actual situation of the line and provides a clear answer to "what kind of operation organization strategy to adopt" in the specific situation of multi-mode rail transit. This makes up for the lack of a set of operation organization strategy selection methods for urban rail transit and urban rail transit to cope with the variable passenger flow.

[0060] (2) This invention is applicable to train operation organization under complex operation modes. In the process of coordinating the optimization of the operation scheme of urban rail and urban rail trains, the frequency of intersection, station stop scheme and number of trains involved in urban rail and urban rail are all taken into account. At the same time, the algorithm proposed in this invention has a faster solution efficiency. Based on the NSGA-II algorithm framework, it combines reinforcement learning to adaptively adjust the crossover mutation probability of the NSGA-II algorithm. Attached Figure Description

[0061] Figure 1 This is a schematic diagram illustrating the steps of a method for selecting a coordinated operation organization strategy for urban rail and suburban rail trains as described in this invention.

[0062] Figure 2 This is a schematic diagram illustrating the thought process behind selecting feasible strategies for collaborative operation organizations in this invention.

[0063] Figure 3 This is a schematic diagram illustrating the specific process of selecting feasible strategies for collaborative operation organizations in this invention.

[0064] Figure 4 The flowchart shows the algorithm's solution process.

[0065] Figure 5 This is a schematic diagram illustrating the fusion of genetic algorithms and reinforcement learning.

[0066] Figure 6 This is a chart showing the passenger boarding and alighting coefficients for each station on Lines S7, S1, and S3 from 7:00 AM to 8:00 AM.

[0067] Figure 7 The full load rate diagram for each section of Lines S7-S1-3 from 7:00 to 8:00 AM. Detailed Implementation

[0068] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0069] This invention provides a method for selecting a coordinated operation strategy for urban rail transit and urban rail trains. Based on the analysis of common operation strategies, a candidate set of coordinated operation strategies for urban rail transit and urban rail transit is constructed. Furthermore, from three aspects—selection approach, basis indicators, and specific procedures—a method for selecting feasible coordinated operation strategies for urban rail transit and urban rail transit is proposed, screening out feasible strategies that conform to the actual conditions of the lines. Based on the obtained feasible strategies, a coordinated optimization model for the operation schemes of urban rail transit and urban rail trains is established to determine the specific route operation frequency, stops, and train formation number under each feasible strategy. This invention aims to solve the problem of selecting a coordinated operation strategy for urban rail transit and urban rail transit, and has broad application prospects in regional rail transit intermodal transport models.

[0070] To achieve the above objectives, such as Figure 1 As shown, the present invention adopts the following technical solution:

[0071] S1, based on existing common operation organization models, constructs a set of alternative strategies for the operation organization of urban rail and suburban rail trains from the perspectives of routes, stops, and train formation.

[0072] S2, based on the alternative strategy set of S1, proposes a selection method for feasible strategies of collaborative operation organization from three aspects: selection ideas, basis indicators, and specific processes, and determines feasible strategies suitable for urban rail transit and suburban rail transit.

[0073] Based on the feasible strategies determined in S2, S3 establishes a collaborative optimization model for the operation of urban rail and suburban rail trains, detailing the specific departure frequency, stops, and train formation numbers for each feasible strategy. With the goal of minimizing passenger travel time costs and enterprise operating costs, and considering constraints such as train occupancy rates and line capacity, a collaborative optimization model for the operation of urban rail and suburban rail trains is established.

[0074] S4. Design a solution algorithm. Based on the NSGA-II algorithm framework, combine reinforcement learning to adaptively adjust the crossover and mutation probability of the NSGA-II algorithm, solve the model established in S3, and decide the specific route departure frequency, stop scheme, and number of trains under each feasible strategy.

[0075] Furthermore, S1 specifically includes:

[0076] S101 collects existing common operation organization strategies for urban rail transit and suburban rail transit. For routes, it includes three strategies: cross-line routes, short-distance routes, and single-line routes. For station stopping schemes, it includes two strategies: all-stop and non-all-stop. For train formation modes, it includes two strategies: fixed formation and mixed multi-formation operation. Note that cross-line routes and short-distance routes not being considered on the same line can form 3*2*2=12 combinations. A candidate set of operation organization strategies is constructed, including strategies 1-12, as shown in Table 1.

[0077] Table 1. Alternative Set of Operational Organizational Strategies

[0078]

[0079] Furthermore, S2 specifically includes:

[0080] S201 proposes a selection strategy for feasible organizational tactics of coordinated operation of urban rail and suburban rail, such as... Figure 2 As shown.

[0081] A hierarchical analysis method from large to small is adopted. First, the connection between different types of tracks is considered at the network level. Then, the cross-line passenger flow and proportion are used to determine whether cross-line operation is necessary.

[0082] When a line operates across multiple lines, an imbalance coefficient is further calculated to determine the balance of passenger flow distribution. When the passenger flow distribution is relatively balanced, the existing operation organization strategy can be maintained, and optimization can be achieved by changing the train departure frequency. When the passenger flow distribution is uneven, a stopping strategy is formulated based on the passenger flow coefficient of each station. In addition, the train formation strategy is determined based on the load factor of different sections.

[0083] When a line operates on a single track, an imbalance coefficient is calculated to assess the balance of passenger flow distribution. If passenger flow is relatively balanced, the existing operational strategy can be maintained, with optimization achieved by adjusting train departure frequencies. If passenger flow is uneven, the distribution of passenger volume at each station is calculated. If stations are concentrated, the need for additional long and short routes on the single-track line is considered. If stations are dispersed, a stopping strategy is developed based on the passenger flow coefficients of each station. If there is no clear pattern in station distribution, both factors can be combined. Furthermore, when passenger flow is uneven, the train formation strategy must be determined based on the load factor at different cross-sections.

[0084] S202 proposes indicators as the basis for selecting feasible strategies for the coordinated operation of urban rail and suburban rail.

[0085] 1) Calculate cross-line passenger flow and its proportion. Cross-line passenger flow Q 跨 This represents passenger flow where the origin and destination are located on two different routes, and can be obtained from the passenger flow OD table. The cross-line passenger flow ratio λ is the ratio of cross-line passenger flow to the total passenger flow on a given route. The total passenger flow consists of cross-line passenger flow and local line passenger flow. The specific calculation formula is as follows:

[0086]

[0087] Among them, Q 跨 To calculate cross-line passenger flow within a given time period, Q 本 This refers to the passenger flow on this line within the statistical period.

[0088] In this invention, the cross-line passenger flow Q required for the operation of cross-line routes is... 跨 The passenger flow must be ≥3240 people / hour and the cross-line passenger flow must account for more than 50%. If either of these two conditions is not met, cross-line operation is not allowed.

[0089] 2) Calculate the line cross-sectional imbalance coefficient β. The line cross-sectional imbalance coefficient β is used to determine whether the passenger flow distribution is balanced. The specific calculation formula is as follows:

[0090]

[0091] Among them, Q max M represents the maximum cross-sectional passenger flow, and Q represents the number of one-way line sections. m Let m be the passenger flow at the m-th cross-section.

[0092] In this invention, when the coefficient β ≤ 1.5, it indicates that there is no imbalance in passenger flow distribution, and the existing strategy can be maintained. Otherwise, adjustments need to be made in terms of routes, stops, and train formations.

[0093] 3) Calculate the passenger flow coefficient γ for each station. The station passenger flow coefficient is used to describe the passenger flow level at different stations. The specific calculation formula is as follows:

[0094]

[0095] in, Q represents the maximum hourly passenger volume of station i. u Let u be the maximum hourly passenger volume at station u, and k be the number of stations.

[0096] In this invention, when the passenger drop-off coefficient γ≤0.5, it indicates that the number of passengers getting on and off at the station is too small; when the passenger drop-off coefficient γ≥1.5, it indicates that the number of passengers getting on and off at the station is too large. By observing the distribution of stations with passenger drop-off coefficients γ≥1.5 or γ≤0.5, it is determined whether to adopt a long-distance / short-distance route and non-stop-at-station strategy.

[0097] 4) Calculate the load factor η for each section. The load factor represents the ratio of the actual passenger volume carried by the section to the actual operating capacity of the section, reflecting the utilization efficiency of the transport capacity within a section.

[0098]

[0099] in, N represents the passenger flow at section i within the statistical time period; N represents the number of trains passing through this section in a certain direction within the statistical time period; and C represents the train's passenger capacity.

[0100] In this invention, when the average full load rate η of multiple sections of the line is ≥120% or η≤20%, it indicates that the full load rate of each section of the line is different. In this case, a multi-grouping strategy can be selected to achieve a faster relief effect.

[0101] S203 proposes specific steps for a feasible strategy to organize the coordinated operation of urban rail transit and suburban rail transit, such as... Figure 3 As shown.

[0102] Step 1: Determine if the cross-line conditions are met, and calculate the cross-line passenger flow Q. 跨 And the cross-line passenger flow ratio λ. If all conditions are met, proceed to Step 2; if any condition is not met, proceed to Step 4.

[0103] Step 2: Under cross-line operation conditions, calculate the line cross-section imbalance coefficient β, determine whether the passenger flow distribution is balanced, and determine whether the imbalance coefficient satisfies β>1.5. If it satisfies, proceed to Step 3; if it does not, proceed to Step 6.

[0104] Step 3: Calculate the passenger flow coefficient γ and cross-sectional load factor η for each station. If the number of stations with γ≥1.5 or γ≤0.5 exceeds the set threshold, it is recommended to operate non-stop trains. If the number of stations with η≥120% or η≤20% exceeds the set threshold, it is recommended to operate multi-train mixed trains. Otherwise, proceed to Step 6.

[0105] Step 4: Under single-line operation conditions, determine whether the passenger flow distribution is balanced, calculate the line cross-section imbalance coefficient β, and determine whether the imbalance coefficient satisfies β>1.5. If it satisfies, proceed to Step 5; otherwise, proceed to Step 6.

[0106] Step 5: Calculate the passenger flow coefficient γ and the cross-sectional load factor η for each station;

[0107] If the number of stations satisfying γ≥1.5 or γ≤0.5 exceeds the set threshold, and: if these stations are concentrated in the same section formed by two turnaround stations, then a long-distance / short-distance route strategy is recommended; if these stations are scattered in different sections formed by two turnaround stations, then a non-stop-at-station strategy is recommended; apart from the above two cases, if the station distribution is disordered and has no obvious regularity, then a long-distance / short-distance route strategy and a non-stop-at-station strategy are recommended.

[0108] If η≥120% or η≤20% of stations exceed the set threshold, a multi-group mixed-run strategy is recommended.

[0109] Otherwise, proceed to Step 6.

[0110] Step 6: Maintain the existing operational organization strategy and design train departure frequencies.

[0111] Furthermore, S3 specifically includes:

[0112] S301, calculate the travel time cost for passengers.

[0113] The passenger's travel time cost equals the product of the passenger's travel time and the time cost coefficient. The passenger's travel time includes the waiting time after entering the station, the time on the train, and the time required for transfers.

[0114]

[0115] Where, p w Let OD be the number of passengers in passenger group w; Z1 be the passenger travel cost; a1 be the passenger time cost coefficient; t w For OD to the travel time of the w passenger group; The average time spent on the vehicle per passenger in the OD group; The average transfer time per passenger for the OD group; The average waiting time per passenger for the OD group.

[0116] On-train time refers to the total travel time required for a specific origin-destination (OD) passenger group across all sections of its travel route, plus the total dwell time at all intermediate stops excluding the origin and destination. The travel time within each section and the dwell time at each intermediate station can be obtained from operational data.

[0117]

[0118] Among them, t se γ represents the travel time within the section; γ represents the train's station dwell time. For variables of 0-1, OD represents the number of passengers (w) traveling on route c. n l m The value of a train traveling within the se section is 1, and the value of a train traveling outside the se section is 0.

[0119] Waiting time refers to the total waiting time for passengers on different routes throughout their entire journey.

[0120]

[0121] in, This refers to the frequency of train services on this line. The frequency of train services on cross-line and short-distance routes.

[0122] Transfer time refers to the total time required for a passenger to transfer from one line to another during their journey. Transfer time is equal to the product of the number of transfers and the walking time for each transfer, where the number of transfers is equal to the number of routes traversed by the passenger during their trip minus 1.

[0123]

[0124] Among them, t walk a1 represents the travel time for transfers. a1 is the passenger time cost coefficient.

[0125] S301, calculate the total operating cost of the enterprise.

[0126] The total operating cost of an enterprise includes the cost per kilometer traveled by trains and the cost per kilometer spent at stations. The cost per kilometer traveled by trains is expressed as the product of the total kilometers traveled by all trains in operation and the cost per kilometer traveled by the trains.

[0127]

[0128] Where Z2 is the cost per kilometer traveled by the train; For overpass lines and short-distance routes; This is the length of the intersection of this line; This refers to the number of train formations on this line. a1 represents the number of train formations on cross-line and short-line routes; a2 represents the operating cost coefficient per unit length.

[0129] Train stop costs refer to the expenses incurred by a train during its operation due to stops at stations, and are closely related to the total number of stops. Train stop costs equal the number of stops multiplied by the cost coefficient per stop.

[0130]

[0131] Where Z3 is the train stopping cost; a3 is the single-trip stopping cost coefficient.

[0132] S303, Set model constraints.

[0133] 1) Upper and lower limits of train departure frequency for each route

[0134] The formulation of train operation plans needs to take into account the throughput capacity of the line and the turnaround capacity of the turnaround station, and upper and lower limits should be set for the departure frequency of each route.

[0135]

[0136] in, Minimum and maximum permitted train departure frequencies for cross-line and short-line routes; Minimum and maximum train departure frequencies permitted on this line;

[0137] 2) Train occupancy rate constraints

[0138] When formulating train operation plans, ensure that sufficient operating capacity is provided for passengers to avoid passenger congestion, and the maximum cross-sectional passenger flow should not exceed the capacity provided by the train.

[0139]

[0140] in, Maximum cross-sectional passenger flow for cross-line and local routes; Maximum cross-sectional passenger flow on this line; P carriage capacity; η train load factor.

[0141] 3) Line capacity constraints

[0142] The network operates cross-line routes, short routes, and long routes on the same line in parallel. In sections where multiple routes operate together, it is necessary to ensure that the sum of the departure frequencies of all routes passing through that section does not exceed the line's capacity.

[0143]

[0144] in, The value is 1 if the line passes through interval se, and 0 otherwise. The value t indicates that the cross-line intersection and short-line intersection pass through se, and is 1 if they do not; otherwise, it is 0. min Track train intervals.

[0145] 4) Stop constraints

[0146] Based on operational needs, trains must stop at turnaround points on the route. For cross-line lines, trains may choose to stop at intermediate stations, in addition to the mandatory stops at the first and last stations.

[0147]

[0148] in, A 0-1 variable indicating whether the two turnaround stations at both ends of an overpass route stop.

[0149] 5) Grouping Constraints

[0150]

[0151] That is, the number of sections can only be four, six, or eight.

[0152] Furthermore, S4 specifically includes:

[0153] S401, set up the NSGA-Ⅱ algorithm framework.

[0154] The NSGA-II algorithm begins by initializing the population and calculating the objective function value. It then performs fast non-dominated sorting and crowding calculation to select parent populations. During iterations, reinforcement learning is used to dynamically adjust crossover and mutation probabilities. Offspring are generated through selection, crossover, and mutation operations, and then merged with the parent population. Non-dominated sorting and crowding calculation are performed again to form new parent populations until a preset number of iterations is reached. Finally, a set of Pareto optimal solutions is output. The algorithm framework is as follows: Figure 4 As shown.

[0155] 1) Chromosome Encoding. Decision variables include route departure frequency, stop schedule, and number of train formations. Each chromosome represents a specific operation plan, using a combination of real number encoding and 0-1 encoding. Route departure frequency and number of train formations are encoded using real numbers, while stop schedules are encoded using 0-1 codes.

[0156] 2) Fitness function selection. Using fast non-dominated sorting, all individuals in the population are stratified according to their quality, with the individuals in the first stratum being the best, and so on; the crowding operator is used to compare the quality of individuals in the same stratum, with the larger the crowding distance, the better the individual.

[0157] 3) Genetic operations. Tournament selection is a commonly used selection strategy, which prioritizes individuals with smaller non-dominated levels. When two individuals have the same non-dominated level, the individual with a larger crowding distance is selected. The probabilities of crossover and mutation are obtained by training and optimizing a reinforcement learning model, and the probabilities of crossover and mutation are dynamically adjusted.

[0158] 4) Algorithm termination condition. The algorithm stops when the preset maximum number of iterations is reached.

[0159] S402, adaptively adjusts the crossover mutation probability.

[0160] By combining Q-learning and the NSGA-II algorithm, a method for adaptively adjusting crossover and mutation probabilities is provided, such as... Figure 5 As shown. The obtained current state s t The input is fed into the Q-learning algorithm, which uses its internal Q-value table to estimate the value of taking different actions (adjusting crossover and mutation probabilities) under different states; the Q-value table is updated based on the output of the Q-learning algorithm; and the optimal action 'a' is selected based on the updated Q-value table. t That is, the optimal scheme for adjusting the crossover and mutation probabilities, which leads to a change in the environmental state and the generation of a new state s. t+1 This process is repeated.

[0161] 1) Parameter initialization

[0162] The parameters include: NSGA-Ⅱ environment state set S, action set A, and Q value table;

[0163] Initially, the Q-value table is set as an all-zero matrix, with the number of rows corresponding to the number of states in NSGA-II and the number of columns corresponding to the number of actions in action set A.

[0164] 2) Calculate the environment state s of the t-th generation of the NSGA-II algorithm. t ;

[0165] s t =w1·fit* +w2·div * +w3·Best * (w1+w2+w3=1)

[0166]

[0167] in, For the i-th chromosome in the t-th generation population; fit * The average fitness of the population; div * For population diversity; Best * For optimal fitness of the population; Chromosomes fitness value; The optimal fitness of the t-th generation population; w1, w2, and w3 are respectively the optimal fitness of the t-th generation population. * div * Best * The weighting coefficients; Popsize is the population size.

[0168] 3) Calculate the reward function r t+1 Adjusting the reward function value r corresponding to the crossover probability and the mutation probability. t+1 The difference lies in the reward function r when the action is to adjust the mutation probability. t+1 That is When the action performed is to adjust the crossover probability, the corresponding reward function r t+1 That is

[0169]

[0170] in, This refers to the i-th chromosome in the (t+1)-th generation population. The optimal fitness of the population in generation t+1 is given.

[0171] 4) Select action a based on the ε-greed strategy. t+1 ;

[0172]

[0173] Where a is the action variable, a∈A; a random This means randomly selecting an action from A; r 0-1 This indicates that a random number in the range [0,1] is generated; ε is the greedy rate of the strategy.

[0174] 5) Perform action a t+1 Observe the reward function value r t+1 and new state s t+1 ;

[0175] 6) Calculate and update Q(s) t ,a t )value;

[0176] Q(s t ,a t )=(1-α)Q(s t ,a t )+α(r t+1 +γmaxQ(s t+1 ,a t+1 ))

[0177] Where α is the learning rate; γ is the learning discount rate; r t+1 The reward function value varies depending on the action performed.

[0178] 7) Based on current action a t+1 Change the crossover and mutation probabilities P c P m .

[0179] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. To more intuitively illustrate the method for selecting the coordinated operation organization strategy of urban rail and suburban rail trains, the present invention is demonstrated through an application example.

[0180] Taking the Nanjing Metro Lines S7-S1-3 regional rail transit network as an example for verification, focusing on the inbound direction, the 7:00-8:00 AM time slot was selected to obtain the specific operational organization strategy for the lines during this time slot. The passenger boarding factor and cross-sectional load factor for each station during the 7:00-8:00 AM time slot are shown below. Figure 6 , Figure 7 As shown.

[0181] The passenger flow characteristic coefficients of lines S1 and S7 from 7:00 to 8:00 am were calculated and used as the basis for selecting feasible operation organization strategies. The results are shown in Table 2.

[0182] Table 2 shows the calculation of the S1 and S7 lines from 7:00 AM to 8:00 AM based on indicators.

[0183]

[0184]

[0185] First, determine if the cross-line conditions are met. The cross-line passenger flow is 3423 people / hour, and the cross-line passenger flow ratio is 82.2%, which meets the conditions, so it is recommended to operate cross-line routes. Next, determine the cross-section imbalance coefficient, which is 2.24, greater than 1.5. Then, determine the station boarding and alighting coefficients. For stations with excessively large or small boarding and alighting coefficients, it is recommended to stop at every station. Finally, determine the section cross-section load factor. If the section load factor varies too much, a multi-train mixed operation strategy is recommended. Based on the strategy selection method, three feasible strategies are generated for lines S1 and S7:

[0186] ① Cross-line + non-stop; ② Cross-line + multi-train mixed operation; ③ Cross-line + non-stop + multi-train mixed operation.

[0187] The passenger flow characteristic coefficient of Line 3 from 7:00 to 8:00 am was calculated and used as the basis for selecting the operation organization strategy. The results are shown in Table 3.

[0188] Table 3. Line 3 Calculations based on indicators from 7:00 AM to 8:00 AM

[0189]

[0190] First, determine if the crossing condition is met; it is not. Next, assess the cross-sectional imbalance coefficient; it is 1.68, greater than 1.5. Then, determine the station passenger / drop coefficient; stations with excessively high coefficients are clustered within the Linchang-Shengtai West Road section, recommending the operation of both long and short routes. Finally, assess the section load factor; the differences in load factors are too large, with sections having a load factor greater than 1.2 being too concentrated, recommending a multi-train mixed-operation strategy. Based on the strategy selection method, three feasible strategies are generated for Line 3:

[0191] ①Large and small routes; ②Multiple train sets running together; ③Large and small routes plus multiple train sets running together.

[0192] Considering that the difference in load factor within each area of ​​Line 3 is small, but the difference between areas is large, the multi-train mixed operation mode is not suitable for Line 3 when only one long-distance route is opened; if S1-S7 only adopt cross-line + multi-train mixed operation and stop at each station, there will still be a waste of transport capacity due to the small number of passengers entering the station, so cross-line + multi-train mixed operation is not suitable.

[0193] Therefore, the operation organization strategies obtained based on the strategy selection for the urban rail and suburban rail lines S7-S1-3 are shown in Table 4.

[0194] Table 4. Feasible operational organizational strategies obtained from strategy selection.

[0195]

[0196] For the four feasible strategies mentioned above, a collaborative optimization of train operation plans is performed. Using the time and operating costs of the current actual plans as benchmarks, the optimal solution with the lower cost is found. Under each strategy, three Pareto optimal solutions are selected. It should be noted that all selected solutions are considered optimal train operation plans, and the optimal solutions are summarized in Table 5.

[0197] Table 5 Optimization results of four strategies

[0198]

[0199] The optimization effects varied under different feasible strategies. Among them, the third option of Strategy 1 showed the most outstanding performance in terms of operating costs. Specifically, during the morning peak hours of 7:00-8:00, Line 3's short-route operated one 6-car train; Line 3's long-route operated 12 6-car trains; Line S1's main route operated 6 4-car trains; Line S7's main route operated 4 4-car trains; and the S1-S7 cross-line operation operated one 4-car train, stopping at Nanjing South Station, Cuipingshan, Focheng West Road, Zhengfang Middle Road, Xiangyu Road North, Airport Lishui Station, Wolonghu, Zhongshanhu, Xingzhuang, and Wuxiangshan. This specific operation organization strategy reduced operating costs by 10.99%. This significant cost saving is mainly due to the fine-tuning of train formation, stop arrangements, and departure frequency, which made resource utilization more efficient and reduced unnecessary energy consumption and labor costs.

[0200] On the other hand, the sixth option of Strategy Two achieved the greatest reduction in time costs. Specifically, during the morning rush hour (7:00-8:00), Line 3's short-route operates two 6-car trains; Line 3's long-route operates twelve 4-car trains; Line S1's main route operates ten 4-car trains; Line S7's main route operates six 4-car trains; and the S1-S7 cross-line operation operates one 4-car train, stopping at Nanjing South Station, Cuipingshan, Focheng West Road, Xiangyu Road South, Lukou Airport, Qunli, Lishui, Zhongshan Lake, Xingzhuang, and Wuxiangshan stations. This specific operational strategy achieved a time cost reduction of 8.26%.

[0201] Operators can choose the most suitable strategy based on the actual situation to achieve the best balance between time and operating costs, thereby improving overall operational efficiency and service quality.

[0202] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

[0203] This application also provides an electronic device, including: a memory and a processor; the memory stores a computer program, and when the computer program is executed by the processor, it implements the above-described method for selecting a coordinated operation organization strategy for urban rail and suburban rail trains.

[0204] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the steps of the aforementioned method for selecting a coordinated operation organization strategy for urban rail and suburban rail trains. The computer-readable storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0205] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

Claims

1. A method for selecting a coordinated operation strategy for urban rail transit and suburban rail trains, characterized in that: The method includes: S1 constructs a set of alternative strategies for the operation organization of urban rail and suburban rail trains from the perspectives of routes, stops, and train formation; S2, based on the set of alternative strategies in S1, determines feasible strategies suitable for urban rail and suburban rail trains. S3, based on the feasible strategy of S2, aims to minimize the passenger travel time cost and the enterprise's total operating cost. It establishes a collaborative optimization model for the operation scheme of urban rail and suburban rail trains, and solves the route departure frequency, station stopping scheme and train formation number under each feasible strategy. The steps for determining feasible strategies for the coordinated operation of urban rail and suburban rail in S2 include: Step 1: Determine if the cross-line condition is met. If it is, proceed to Step 2; otherwise, proceed to Step 4. The cross-line condition is the cross-line passenger flow. and cross-line passenger flow ratio All of them are greater than the corresponding set thresholds for cross-line passenger flow and cross-line passenger flow ratio. Step 2: If the line cross-section imbalance coefficient If the passenger flow exceeds the set imbalance threshold, it is determined that the passenger flow distribution is uneven and proceeds to Step 3; otherwise, proceeds to Step 6. Step 3: If the corresponding passenger flow boarding and alighting coefficient If the number of stations with a peak passenger flow greater than or equal to the set peak passenger flow or a trough passenger flow less than or equal to the set trough passenger flow exceeds the set first percentage, then a non-stop-at-everything (non-stop) strategy is recommended; if the corresponding section's occupancy rate... If the number of stations with a peak load factor greater than or equal to the set cross-section load factor or a trough load factor less than or equal to the set cross-section load factor exceeds the set second percentage, then a multi-train mixed-run strategy is recommended; otherwise, proceed to Step 6. Step 4: If the line cross-section imbalance coefficient If the passenger flow exceeds the set imbalance threshold, it is determined that the passenger flow distribution is uneven and proceeds to Step 5; otherwise, proceeds to Step 6. Step 5: If the corresponding passenger flow boarding and alighting coefficient If the number of stations with a peak passenger flow greater than or equal to the set peak passenger flow or a valley passenger flow less than or equal to the set valley passenger flow exceeds the set first percentage, and further judgment is made: if such stations are concentrated in the same section formed by two turnaround stations, it is recommended to operate a long-distance / short-distance route strategy; if such stations are scattered in different sections formed by two turnaround stations, it is recommended to operate a non-stop-at-station strategy; except for the above two cases, it is recommended to operate a long-distance / short-distance route strategy and a non-stop-at-station strategy. If the corresponding cross-section is full load rate If the number of stations with a peak load factor greater than or equal to the set cross-section load factor or a valley load factor less than or equal to the set cross-section load factor exceeds the set second percentage, then a multi-train mixed-run strategy is recommended. Otherwise, proceed to Step 6; Step 6: Maintain the existing operational organizational strategy.

2. The method according to claim 1, characterized in that: The S1 route includes three strategies: cross-line route, long-distance route, and single route. The station stopping scheme includes two strategies: stopping at every station and not stopping at every station. The train formation mode includes two strategies: fixed train formation and mixed multi-train formation.

3. The method according to claim 1, characterized in that: The cross-line passenger flow ratio in Step 1 The specific calculation formula is as follows: ， in, To count cross-line passenger flow within a given time period, This refers to the passenger flow on this line within the statistical period. Line cross-sectional imbalance coefficient The specific calculation formula is as follows: ， in, To maximize passenger flow at the cross-section, This refers to the number of cross-sections of a one-way line. Let m be the passenger flow at the m-th cross-section; Site Passenger flow boarding factor The specific calculation formula is as follows: ， in, For the site Maximum passenger capacity per hour For the site hourly passenger volume Number of stations; Section full load rate The specific calculation formula is as follows: ， in, Cross-section within the statistical time period Passenger flow; For the cross-section passing through within the statistical time period The number of trains in a certain direction; The number of passengers allowed on the train.

4. The method according to claim 1, characterized in that: The S3-based collaborative optimization model for the operation of urban rail and suburban rail trains includes: The objective function is: ， ， ， ， The constraints are: ， ， ， ， ， ， In the formula: For passenger travel costs; For OD to the travel time of the w passenger group; This is the passenger time cost coefficient; The number of passengers in the OD to w passenger group; Cost per kilometer traveled by the train; For overpass and short-pass routes; This is the length of the intersection of this line; This refers to the number of train formations on this line. For train formations on cross-line and short-line routes; This is the operating cost coefficient per unit length; This refers to the frequency of train services on this line. For train frequency on cross-line and short-line routes; , The minimum and maximum permitted train departure frequencies for cross-line and short-line routes; , These are the minimum and maximum train departure frequencies allowed for this line. The maximum cross-sectional passenger flow for overpass routes and local routes; P represents the maximum cross-sectional passenger flow on this line; P represents the carriage capacity. This refers to the train's occupancy rate. It is a set of intervals; The value is 1 if the line passes through interval se, and 0 otherwise. The value is 1 if the cross-line intersection and short-line intersection pass through the interval se, and 0 otherwise. To track train intervals; , A 0-1 variable indicating whether the two turnaround stations at both ends of an overpass route stop; This refers to the set of routes other than the main routes on this line. Costs associated with train stops; This represents the cost coefficient for a single stop.

5. The method according to claim 1, characterized in that: The NSGA-II algorithm is used to solve the collaborative optimization model of the operation scheme of urban rail and suburban rail trains, and the departure frequency, stopping scheme and number of trains under each feasible strategy are obtained.

6. The method according to claim 5, characterized in that: In the NSGA-II algorithm, decision variables include route departure frequency, stop plan, and number of trains. Each chromosome represents a train operation plan. The route departure frequency and number of trains are encoded using real numbers, while the stop plan is encoded using 0-1.

7. The method according to claim 5, characterized in that: The Q-Learning algorithm is used to adaptively adjust the crossover and mutation probabilities in the NSGA-II algorithm. Among them, through the reward function Adjusting crossover probability Through the reward function Adjusting mutation probability : ， ， In the formula, For the first The i-th chromosome of the generation population fitness value; For the i-th chromosome of the (t+1)-th generation population fitness value; Population size; for Optimal fitness of the generation population; The optimal fitness of the population in generation t+1 is given.

8. The method according to claim 7, characterized in that: The number of rows in the Q-value table corresponds to the number of environment states in NSGA-II, and the number of columns corresponds to the number of actions in action set A; initially, the Q-value table is set as an all-zero matrix.

9. The method according to claim 7, characterized in that: NSGA-II's environmental status for; ， ， ， ， in, The average fitness of the population; For population diversity; The optimal fitness of the population; , , They are , , The weighting coefficients.

Images