Urban rail transit virtual marshalling passenger and freight collaborative operation optimization method and device

Abstract

The invention provides an urban rail transit virtual marshalling passenger and freight collaborative operation optimization method and device, and belongs to the field of rail transit, and the method comprises the steps: determining the total travel cost of passengers and the total operation cost of trains; determining the carbon emission of the train according to the electric power and the operation power of the train operation, and determining the carbon emission cost of the train; based on the passenger flow volume and the cargo flow volume of the target track and the marshalling mode, the passenger carrying capacity, the cargo carrying capacity and the carbon emission of the train, a plurality of constraint conditions of train operation are determined; respectively determining the correlation between the total travel cost, the total operation cost and the carbon emission cost and the train operation parameters; an NSGA-II algorithm is constructed; and carrying out iteration on the NSGA-II algorithm to obtain an optimized train operation scheme. In this way, the cost problem of the train operation scheme is comprehensively considered from multiple angles, multiple optimization targets and constraint conditions are combined, the train operation scheme can be fully optimized, and the ideal optimization effect can be achieved.

Description

Technical Field

[0001] The present invention belongs to the field of rail transit, and in particular relates to a method and device for virtual marshaling passenger-freight coordinated operation of urban rail transit. Background Art

[0002] In recent years, urban congestion and environmental pollution have become hot issues in modern logistics research, and the key to solving this problem is how to alleviate the pressure of cargo transportation on urban traffic. With the advancement of urban informatization, many cities have transformed from material production and cost centers to sales and profit centers, which has greatly accelerated the development of urban logistics. However, urban logistics with intra-city distribution as the core still mainly uses traditional truck transportation. Although it reduces warehousing management costs and ensures the timeliness of cargo transportation, trucks have the characteristics of occupying a large proportion of urban road space resources, heavy loads, long transportation distances, many turnovers, and high frequency of use, which makes the proportion of in-transit vehicles high and more likely to cause traffic congestion and pollute the environment. Therefore, enhancing the capacity of cargo transportation and improving its operating efficiency are particularly important for alleviating the pressure of urban traffic.

[0003] Among the existing technologies, M-ULS (metro freight system) technology is still in its infancy, and there is little research on the organization and optimization of rail transit passenger and freight coordinated transportation, which limits the optimization effect of the solution; urban rail transit transportation technology has not been optimized from the perspective of emerging technologies, such as marshalling technology, and the role of technological development in the organizational model of urban rail transit transportation has been ignored; there are also related studies that take into account the operating costs of enterprises in passenger and freight transport technology, and improve the train transportation capacity at a certain enterprise cost, but ignore the train energy consumption and its related impacts, resulting in high energy consumption. At present, the ideal optimization effect of low-carbon operation of rail transit has not been achieved. Summary of the invention

[0004] In order to solve the above-mentioned optimization problem of low-carbon operation of rail transit, the present invention provides a method and device for optimizing the coordinated operation of passengers and freight in virtual marshaling of urban rail transit.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] First, a method for optimizing the coordinated operation of passengers and freight in virtual marshaling of urban rail transit is provided, the method comprising:

[0007] According to the starting point of the passenger's trip and the distance of the train's operation in the target track where the train travels back and forth, based on different travel plans, the total travel cost of the passenger and the total operating cost of the train are determined respectively;

[0008] Determine the train's carbon emissions based on the train's electrical power and operating power, and determine the train's carbon emission costs based on the current carbon emission quota and cost within the quota;

[0009] Determine multiple constraints on train operation based on the passenger and freight flows of the target track, as well as the train marshaling mode, passenger capacity, freight capacity, and preset carbon emissions;

[0010] Determining the correlation between the total travel cost, the total operating cost and the carbon emission cost and the train operating parameters respectively; the train operating parameters include the train formation type, the departure interval and the number of carriages allocated;

[0011] Constructing the NSGA-II algorithm, including: building the corresponding chromosome code according to the train formation type, departure interval and number of carriages allocated, and generating the initial population according to multiple constraints;

[0012] The NSGA-II algorithm is iterated to minimize the total travel cost, the total operation cost and the carbon emission cost, and the iterative results are screened through the correlation to obtain an optimized train operation plan.

[0013] Optionally, the determining of the total travel cost of the passenger and the total running cost of the train based on different travel plans according to the starting point of the passenger's travel and the distance of the train's travel in the target track includes:

[0014] Determine the first waiting time for passengers who do not transfer and the second waiting time for passengers who transfer respectively according to the starting point;

[0015] The first waiting time and the second waiting time are determined based on the probability of not transferring, and a transfer penalty coefficient is introduced. The total waiting time of the passenger is determined according to the first waiting time, the second waiting time and the transfer penalty coefficient; the transfer penalty coefficient represents the proportion of the transfer time in the second waiting time; the probability of not transferring is the probability of the passenger not transferring obtained based on data statistics;

[0016] Determine the total travel cost of the passenger based on the total waiting time and the per capita time value of the unit waiting time;

[0017] The total operating cost of the train is determined based on the unit operating cost per kilometer and the unit purchase cost per hour of the train.

[0018] Optionally, the first waiting time and the second waiting time are determined based on the probability of not transferring, a transfer penalty coefficient is introduced, and the total waiting time of the passenger is determined according to the first waiting time, the second waiting time and the transfer penalty coefficient, including:

[0019] For passengers who do not transfer, the sum of the products of the number of waiting passengers and the length of the discrete time segment in all discrete time intervals is determined as the first waiting time; the discrete time segment is the interval time between trains of the same route in adjacent sequences passing the starting point;

[0020] For transfer passengers, the sum of the waiting time of multiple transfer sections is determined as the second waiting time, and the transfer time is determined according to the transfer penalty coefficient and the second waiting time, and the sum of the second waiting time and the transfer time is determined as the waiting time of the transfer passenger;

[0021] Determine a total waiting time which is the sum of the first waiting time and the waiting time of the transfer passenger.

[0022] Optionally, determine the train electric power P de The formula is:

[0023]

[0024] in, are decision variables, namely the total number of passenger cars and freight cars assigned to the train, ρ air is the number of lamps in a single carriage, S ve is the surface area of ​​the carriage, Te is the temperature difference between the inside and outside of the carriage, and K is the train index;

[0025] Determine the train running power P F The formula is:

[0026]

[0027] Among them, P μ is the vehicle adhesion mass, F μ is the train traction resistance, g is the acceleration of gravity, m h is the number of vehicles in the train, ω0 is the basic resistance of the train, L h is the distance travelled by the train;

[0028] The formula for determining the carbon emissions of a transportation solution based on the train's electrical power and operating power is:

[0029]

[0030]

[0031] Ce=PC sub ·α·E p +PC sub β E q ;

[0032] Among them, E p is the carbon emission factor of rail transit passenger transport, E q is the carbon emission factor of rail transit freight.

[0033] Optionally, the multiple constraints for train operation are determined based on the passenger flow and freight flow of the target track and the marshaling mode, passenger capacity, freight capacity and carbon emissions of the running train, respectively, including:

[0034] Determining the first constraint condition based on the grouping mode includes:

[0035]

[0036] m1+m2≤m max ;

[0037] N1≤N;

[0038]

[0039] Based on the passenger flow of the target track and the passenger capacity of the train, the second constraint is determined to include:

[0040]

[0041]

[0042] Based on the cargo flow of the target track and the cargo capacity of the train, the third constraint is determined to include:

[0043] Determining the fourth constraint condition based on the preset carbon emissions includes:

[0044]

[0045] Among them, h max ,h min is the maximum and minimum safe departure interval of the train; f h is the decision variable, the departure frequency of trains on large and small routes; is the train arrival time, which means the time when train k arrives at station i; is the departure time of train at station i, indicating the time when train k leaves station i; is the stop time of train k at station i; is the running time of train k between stations i-1 and i; m h is the number of vehicles in the train set of large and small routes; h = 1 for large routes, h = 2 for small routes; m max The maximum number of cars in a train.

[0046] Optionally, the NSGA-II algorithm is iterated to minimize the total travel cost, the total operation cost and the carbon emission cost, and the iterative results are screened by the correlation to obtain an optimized train operation plan including:

[0047] The population is sorted according to non-dominance, and for the last selected frontier, the crowding comparison operator is used to sort it in descending order, and the individuals in front are selected by the binary tournament selection operator until the population is filled, so as to obtain a first-generation population; the first-generation population is used as the parent generation, and the second-generation population is generated through selection, crossover and mutation, and it is iterated until the number of iterations reaches a preset number, the iteration is terminated, and the finally obtained population is determined as the optimization result;

[0048] An optimized train operation plan is determined according to the optimization result.

[0049] Secondly, a device for optimizing the coordinated operation of passengers and freight in virtual marshaling of urban rail transit is provided, the device comprising:

[0050] The cost determination module is used to determine the total travel cost of passengers and the total operating cost of the train based on different travel plans according to the starting point of the passenger trip and the distance of the train in the target track where the train travels back and forth; determine the carbon emissions of the train according to the electric power and operating power of the train, and determine the carbon emission cost of the train based on the current carbon emission quota and the cost within the quota;

[0051] A condition determination module is used to determine multiple constraints for train operation based on the passenger flow and freight flow of the target track and the train's marshaling mode, passenger capacity, freight capacity and carbon emissions; respectively determine the correlation between the total travel cost, total operating cost and carbon emission cost and train operation parameters; the train operation parameters include the train's marshaling type, departure interval and number of carriages allocated;

[0052] The construction module is used to construct the NSGA-II algorithm, including: building the corresponding chromosome code according to the train formation type, departure interval and number of carriage allocation, and generating the initial population according to multiple constraints;

[0053] The optimization module is used to iterate the NSGA-II algorithm to minimize the total travel cost, the total operation cost and the carbon emission cost, and to screen the iteration results through the correlation to obtain the optimized train operation plan.

[0054] In addition, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the above-mentioned method for optimizing the coordinated operation of passengers and freight in virtual marshaling of urban rail transit is implemented.

[0055] Finally, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned method for optimizing the coordinated operation of passengers and freight in virtual marshaling of urban rail transit is implemented.

[0056] The present invention provides a method for optimizing the coordinated operation of passengers and freight in virtual marshaling of urban rail transit, which has the following features:

[0057] Beneficial effects:

[0058] First, the passengers' travel costs, train operating costs and carbon emission costs are determined, and the cost issues of the train operation plan are comprehensively considered from multiple angles, which is conducive to solving the problem of effective cost control in subsequent optimization. Secondly, according to the train operation mode, such as the marshaling mode, the train's passenger capacity, cargo capacity and carbon emissions, and combined with the actual passenger and cargo flows, the train operation constraints are comprehensively considered from multiple aspects to avoid problems that do not conform to the actual situation due to over-optimization. Finally, the multi-objective optimization algorithm NSGA-II is used for iteration. In this way, multiple optimization objectives and constraints are taken into account, and the train operation plan can be fully optimized, which is conducive to achieving ideal optimization results. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiment of the present invention and its design scheme, the following briefly introduces the drawings required for this embodiment. The drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 The present invention is a flowchart of a method for low-carbon optimization of passenger and freight collaboration in virtual marshaling of urban rail transit according to an exemplary embodiment of the present invention.

[0061] Figure 2 A schematic diagram of classification of uplink passenger flow and / or cargo flow provided by the present invention according to an exemplary embodiment;

[0062] Figure 3 A schematic diagram of classification of downlink passenger flow and / or cargo flow provided by the present invention according to an exemplary embodiment;

[0063] Figure 4 A schematic diagram of chromosome coding design provided by the present invention according to an exemplary embodiment;

[0064] Figure 5 A schematic diagram of screening in the initial population construction process provided by the present invention according to an exemplary embodiment;

[0065] Figure 6 A schematic diagram of an algorithm flow chart provided by the present invention according to an exemplary embodiment;

[0066] Figure 7 A block diagram of a low-carbon optimization device for passenger and freight collaboration in virtual marshaling of urban rail transit provided by the present invention according to an exemplary embodiment. DETAILED DESCRIPTION

[0067] In order to enable those skilled in the art to better understand the technical solution of the present invention and implement it, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the scope of protection of the present invention.

[0068] The technical solutions provided by various embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0069] First, the present invention provides a method for optimizing the coordinated operation of passengers and freight in virtual marshaling of urban rail transit, specifically: Figure 1 As shown, the following steps are included:

[0070] S101. According to the starting point of the passenger's trip and the distance of the train's operation in the target track for the train's round trip, based on different travel plans, the total travel cost of the passenger and the total operating cost of the train are determined respectively.

[0071] In this step, the target track is a fixed track for round trip travel of trains, wherein the round trip trains include large-route trains that travel back and forth on the entire section and small-route trains that travel back and forth on the middle section.

[0072] Specifically, the first waiting time for passengers who do not transfer and the second waiting time for passengers who transfer are determined respectively according to the starting point; the first waiting time and the second waiting time are determined respectively based on the probability of not transferring, a transfer penalty coefficient is introduced, and the total waiting time for passengers is determined according to the first waiting time, the second waiting time and the transfer penalty coefficient; the transfer penalty coefficient represents the proportion of transfer time in the second waiting time; the probability of not transferring is the probability of passengers not transferring obtained based on data statistics; the total travel cost of passengers is determined according to the per capita time value of the total waiting time and the unit waiting time; the total operating cost of the train is determined according to the unit operating cost per kilometer and the unit purchase cost per hour of the train.

[0073] The present invention takes the composite urban rail transit line as the research object, and denotes the urban rail transit line L = {S, E}, where S and E d represent the station set and the section set of the line respectively, and S = {1, 2, ..., S a ,...,i,...,j,...,S b ,...,S (n)}, the total number of lines is S (n) The study period is divided into several operating periods of equal length, and T is defined as {1,2,...,t,...,T (n)} is the set of operating periods, T (n) is the total number of time periods, σ is the length of the unit operating period, min.

[0074] The symbols and parameters involved in the present invention and their related definitions are shown in Table 1. The decision variables for constructing the model of the present invention include:

[0075] (1) Train operation-related decision variables and intermediate variables:

[0076] f h ——Decision variables, the departure frequency of trains on major and minor routes;

[0077] λ——decision variable, the ratio of trains running on large and small routes;

[0078] χ i ——Decision variable, indicator variable of the small-route turnaround station, which is 1 if yes, and 0 otherwise;

[0079] ——Train arrival time, which means the time when train k arrives at station i;

[0080] ——Train departure time, which means the time when train k leaves station i;

[0081] ——Train departure indicator variable, indicating whether train k has left station i at time t, if yes, it is 1, otherwise it is 0.

[0082] (2) Train formation-related decision variables and intermediate variables:

[0083] m h ——Number of vehicles in the trains of both large and small routes, vehicles;

[0084] ——Decision variables, the total number of passenger cars and freight cars allocated to route h trains, vehicles.

[0085] (3) Decision variables and intermediate variables related to passenger and cargo flow loading:

[0086] ——passenger and cargo capacity of train k when it leaves station i, persons / boxes;

[0087] ——passenger and cargo loading of train k at station i, persons / boxes;

[0088] ——The amount of cargo loaded by train k when it leaves station i with station j as its destination, in people / boxes;

[0089] ——The number of passengers and cargoes unloaded when train k arrives at station i, people / boxes;

[0090] ——The number of passengers waiting for station j at station i at time t;

[0091] ——The number of goods waiting for station j at station i at time t.

[0092] Table 1 Symbols and parameters representation and definition

[0093]

[0094] (4) Decision variables and intermediate variables related to train carbon emissions:

[0095] ξ i ——0-1 decision variable, indicating the selection of carbon emission policy, i=1,2,3, representing mandatory carbon emission policy, carbon tax policy, and carbon trading policy respectively;

[0096] PC sub ——Train power consumption during urban rail operation, including train traction power consumption and ventilation, air conditioning, lighting, and signaling system power consumption, kw / h;

[0097] P F ——Train traction power, kw; P de ——Electric power of train air conditioning, lighting and signaling systems, kw;

[0098] P μ ——Vehicle adhesion mass, tons; g——Gravity acceleration, the present invention takes 9.81m / s 2 ;

[0099] ρ1, ρ2, ρ3——Empirical coefficients of basic resistance, the present invention takes 2.4, 0.014, 0.001293;

[0100] v——train speed, km / h; ρ air ——Number of lamps in a single carriage, groups;

[0101] s ve ——Inner surface area of ​​the carriage, m 2 ; Te——Temperature difference between inside and outside of the car, ℃.

[0102] Based on the above research object, the present invention proposes a subway passenger and freight coordinated transportation process design method that meets the network operation conditions. When the virtual marshaling train runs on large and small routes, the virtual double-unit train carriages are independent of each other. The departure frequencies in large routes and small routes are different. Therefore, it is necessary to classify passenger and freight flows according to the passenger travel paths. Figure 2 and Figure 3 It is divided into 12 categories: categories 1-6 are passenger (cargo) flows in the upward direction, and categories 7-12 are passenger (cargo) flows in the downward direction.

[0103] Taking the uplink direction as an example, the starting points of passenger flows of categories 1, 2, 3 and 6 are located in the large route section, and they can only take large route trains; the starting and ending points of passenger flows of category 4 are both located in the small route section, and they can take both large and small route trains; the starting point of passenger flows of category 5 is located in the small route section and the ending point is located in the large route section. Passengers can choose to take the direct plan of large route trains, or take the small route train to S b The transfer plan for transferring to a large-route train after the station. The passenger (freight) flow in the downward direction also has the same characteristics.

[0104] Based on the above, the total travel cost of passengers and the total operating cost of trains can be constructed as follows:

[0105] The passenger travel cost is composed of the passenger time cost (waiting time, riding time, transfer time) and the ticket purchase cost. The urban rail determines the ticket price based on the passenger's starting and ending points. Therefore, the ticket price is relatively fixed when the OD (starting point and ending point) remains unchanged; when the train running speed is constant, the riding time of a single route and a large and small route under the same OD will not change, and the transfer time is expressed by the transfer penalty coefficient multiplied by the waiting time. Therefore, the present invention mainly considers the waiting time of passengers.

[0106] Specifically, for passengers who do not transfer, the first waiting time is determined as the sum of the product of the number of waiting passengers in all discrete time intervals and the length of the discrete time segment; the discrete time segment is the interval time between adjacent sequences of trains of the same route passing through the starting point; for passengers who transfer, the second waiting time is determined as the sum of the waiting times of multiple transfer sections, and the transfer time is determined based on the transfer penalty coefficient and the second waiting time, and the sum of the second waiting time and the transfer time is determined as the waiting time for transfer passengers; the total waiting time is determined as the sum of the first waiting time and the waiting time of transfer passengers.

[0107] Taking the uplink direction as an example, passengers of categories 1, 2, 3, 4, and 6 can take any arriving train to reach the destination station without transferring. Therefore, the waiting time of passengers can be expressed as the sum of the product of the number of waiting passengers in all discrete time intervals and the length of the discrete time segment, that is,

[0108]

[0109] The starting point of the 5th type of passenger flow is located in the small route section and the end point is located in the large route section. The 5(a) type of passenger flow chooses to take the large route train to reach S directly, and the 5(b) type of passenger flow chooses to take the small route train to reach S b The transfer plan for transferring to a long-distance train after the station. The passenger choice preference coefficient μ (0≤μ≤1) is introduced to represent the probability that the passenger chooses the direct plan. The waiting time is:

[0110]

[0111] For the transfer plan, passengers depart from the first arriving train k' and arrive at S b The passengers get off at the S station and wait for the long-distance train to transfer. At this time, the starting and ending points of this part of the passenger flow are located in the short-distance section, which belongs to the 5th (a) type of passenger flow; then the passengers go to S b The starting and ending points of this part of passenger flow are both located in the large route section, belonging to the sixth type of passenger flow, which can be converted into the newly added Arrival time at station S b Passenger flow demand in:

[0112]

[0113] Therefore, the total waiting time for passenger flow type 5(b) is:

[0114]

[0115] In daily life, transfers usually bring additional travel costs to passengers, so the transfer waiting time penalty coefficient ν is introduced to represent the inconvenience caused by transfers to passengers. The total waiting time for the fifth type of passenger flow is

[0116]

[0117] The passenger flow in the downlink direction is the same as above. In order to convert the waiting time into travel cost, the present invention uses the value of non-working time to measure the value of passenger time. The travel cost F1 of passengers on large and small routes is the product of the total waiting time and the time value of the per capita non-working time in the city, that is:

[0118] F1=c1×T;

[0119]

[0120] In the formula, c1 is the time value of non-working time per capita in the city, yuan / min; GDP is the gross national product of a city in the current year, ten thousand yuan; R is the number of employees in a city in the current year, ten thousand people; T represents the average working hours of the employed population in the city, h.

[0121] The total operating cost of a train mainly includes the train operating cost related to the train mileage and the train purchase cost amortized to each hour, namely:

[0122] F2=F 21 +F 22 ;

[0123]

[0124] Among them, F 21is the train purchase cost, calculated using the number of vehicles such as N1. The train operating cost is the product of the train operating cost per kilometer and the train running kilometers, that is, F 22 .

[0125] S102. Determine the carbon emissions of the train according to the electric power and operating power of the train, and determine the carbon emission cost of the train based on the current carbon emission quota and the cost within the quota.

[0126] The carbon emissions that can be optimized during subway operations are mainly concentrated on the impacts of subway train traction, ventilation and air-conditioning, lighting, and signal systems. Under the conditions of coordinated passenger and freight transportation, freight can reduce the carbon emissions generated by freight cars by reducing the number of train lighting equipment used, etc., without affecting its transportation and counting operations, which can effectively promote the construction of "green" transportation.

[0127] Specifically, the carbon emission Ce can be expressed as:

[0128] Ce=PC sub ·α·E p +PC sub β E q ;

[0129] in,

[0130] Among them, α and β are the proportion of passenger and freight in the transportation plan.

[0131] The running power of the train is P F , the electric power of the train air conditioning, lighting and signaling system is P de :

[0132]

[0133] ω0=ρ1+ρ2·v+ρ3·v 2 ;

[0134]

[0135] Under different carbon emission policies, the carbon emission costs are different. Under the mandatory carbon emission policy, the government requires enterprises to have carbon emissions within a certain limit. t Within the quota, the carbon emission cost is a fixed value Ce1; under the carbon tax policy, the government collects a fixed carbon tax φ on the unit carbon emissions generated by the enterprise; under the carbon trading policy, if the carbon emissions exceed the limit, the emission quota must be purchased separately, and the enterprise can sell the saved carbon emission quota to the outside. The present invention introduces a 0-1 decision variable to ensure a single carbon emission policy background, so the carbon emission cost can be expressed as:

[0136]

[0137] Among them, ξ i is a 0-1 decision variable, indicating the selection of carbon emission policy, i=1,2,3, representing mandatory carbon emission policy, carbon tax policy, and carbon trading policy, respectively. is the purchase price per unit of carbon emissions, is the excess indicator variable under the carbon trading policy, and φ is the carbon tax collected by the government on the unit carbon emissions generated by the enterprise.

[0138] S103, determining multiple constraints on train operation based on the passenger flow and freight flow of the target track, as well as the train's marshaling mode, passenger capacity, freight capacity, and preset carbon emissions.

[0139] Specifically, determining the first constraint condition based on the grouping mode includes:

[0140] This formula is the time constraint for the departure interval of large and small trains.

[0141] and It is the association constraint between the train arrival and departure time, the stop time, and the interval running time.

[0142] h=1,2; This formula constructs the upper limit constraint for the allocation of freight cars on large and small route trains.

[0143] This formula constructs the association constraint on the number of passenger and freight cars on the train.

[0144] m1+m2≤m max ; This formula ensures the feasibility of the maximum number of trains in virtual marshaling conditions taking into account the platform length;

[0145] N1≤N; this formula constructs the constraint on the number of available trains.

[0146] M h The constraints on the number of available vehicles and the relationship between the operating vehicles and the running frequency and marshaling type are constructed.

[0147] This formula ensures the feasibility of small-route reversal.

[0148] Among them, χ i is the decision variable, the indicator variable of the small route turnaround station, which is 1 if yes, and 0 otherwise.

[0149] Based on the passenger flow of the target track and the passenger capacity of the train, the second constraint is determined to include:

[0150] This formula constructs the constraints related to the remaining passenger capacity of the train.

[0151] and The loading constraints for passenger boarding and alighting are constructed.

[0152] This formula constructs the association constraint on the number of passengers waiting at the station.

[0153] This formula constructs the train departure indicator variable.

[0154] Based on the cargo flow of the target track and the cargo capacity of the train, the third constraint is determined to include:

[0155] This method ensures that all cargo transportation needs are met.

[0156]

[0157] and The constraint relationship between the quantity of cargo loading and unloading is established.

[0158] This formula constructs the constraints for cargo transportation on vehicles.

[0159] This formula constructs the association constraint between train cargo capacity and marshaling type.

[0160] Determining the fourth constraint condition based on the preset carbon emissions includes:

[0161] This formula ensures that the model can only select one carbon emission policy.

[0162] This formula is the carbon emission limit constraint under the mandatory carbon emission policy, and M is a very large constant.

[0163] This formula is a set of mutually exclusive constraints, which constructs the sales quota constraints and purchase quota constraints under the carbon trading policy.

[0164] This formula constructs an indicator variable for whether an enterprise exceeds the quota under the carbon trading policy.

[0165] S104, constructing an NSGA-II algorithm, including: building a corresponding chromosome code according to the train formation type, departure interval and the number of carriage allocations, and generating an initial population according to multiple constraints.

[0166] In this step, the correlation between the total travel cost, the total operating cost and the carbon emission cost and the train operating parameters can be determined respectively, wherein the train operating parameters include the train formation type, the departure interval and the number of carriages allocated.

[0167] The model constructed by the present invention is a multi-objective linear integer programming model. If the waiting time of passengers is reduced, the train departure frequency and the number of passenger cars allocated will inevitably increase, causing an increase in carbon emissions; if the carbon emissions are reduced, the train departure interval will increase and the number of freight cars will increase, which in turn increases the waiting time of passengers. That is, an increase in the number of passenger cars will reduce the operating costs of the enterprise, but will increase the carbon emission costs and passenger travel costs. On the contrary, when the number of passenger cars is at a low level, the carbon emission costs and passenger travel costs are low, and at the same time, the operating costs of the enterprise will increase accordingly. A higher level of train departure frequency in the operation plan will reduce passenger travel costs, but will increase the operating costs and carbon emission costs of the enterprise. There are contradictions between the objectives, and it is impossible to obtain the optimal solution at the same time when solving.

[0168] The NSGA-II algorithm is a non-dominated sorting genetic algorithm with a business strategy, which has good convergence performance and high running speed. Therefore, the present invention uses the improved fast non-dominated sorting genetic algorithm NSGA-II to solve the model.

[0169] Since the model constructed by the present invention is a mixed integer programming model, the traditional NSGA-II algorithm cannot adapt to this model well. Therefore, the present invention makes some improvements to the algorithm, mainly including the following two aspects:

[0170] (1) Chromosome coding design. The chromosome length designed by the present invention is 3*S (n) ,like Figure 4 As shown, each 3-bit code represents the relevant parameter design of a train, among which the first bit is a binary code, indicating the train formation type, 0 is a small formation type, and 1 is a large formation type; the second bit is a decimal code, indicating the departure interval between this train and the previous train; the third bit is a decimal code, indicating the number of freight cars allocated to the train.

[0171] (2) Generate an initial population. The present invention incorporates some model constraints into the process when generating the initial population to improve the efficiency of the algorithm. First, an initial population of size npop is randomly generated, and then the first round of screening is performed on it. The screening criteria are that the number of freight cars allocated to each train is not greater than the total number of cars of the train formation type, and the train departure interval meets the safety driving requirements. Individuals in the population that do not meet the constraints are screened out of the population. Assume that the number of individuals screened out in the mth time is nc m (nc m ≤npop) After the screening is completed, nc is randomly generated again mThe above steps are repeated until the screening is completed. After the population size is npop, the initial population generation is completed. Figure 5 .

[0172] S105, with the goal of minimizing the total travel cost, the total operating cost and the carbon emission cost, iterate the NSGA-II algorithm to obtain an optimized train operation plan.

[0173] In this step, the NSGA-II algorithm can be iterated to minimize the total travel cost, total operating cost and carbon emission cost. The iterative results are screened through the correlation to obtain the optimized train operation plan.

[0174] Specifically, the sum of the total travel cost, the total operating cost and the carbon emission cost may be used as the objective function, and the minimum value of the objective function is sought in each iteration until a preset number of iterations is reached.

[0175] For example, Figure 6 As shown, the population can be sorted according to non-dominance, and the last selected frontier is sorted in descending order using a crowding comparison operator, and the front individuals are selected through a binary tournament selection operator until the population is filled, thereby obtaining a first-generation population; the first-generation population is used as the parent generation, and a second-generation population is generated through selection, crossover, and mutation, and the iteration is performed until the number of iterations reaches a preset number, at which time the iteration is terminated and the final population is determined to be the optimization result; the optimization result is combined with the correlation to obtain an optimized train operation plan.

[0176] It should be noted that the NSGA-II algorithm is iterated, and the population is sorted according to non-dominance. The solution belonging to the best non-dominated set is the best solution in the combined population and must be closer to the target than any other solution in the combined population. In order to accurately select population members, the present invention uses a crowding comparison operator to sort the solutions of the last frontier in descending order and selects the best solution required to fill all populations. The new size population is now used for selection, crossover and mutation to create a new size population. It is worth noting that the present invention uses a binary tournament selection operator, but the selection criteria are now based on a crowding comparison operator. Since the operator requires the rank and crowding distance of each solution in the population, the present invention calculates the objective function values ​​of the adjacent individuals before and after sorting while forming the population. The present invention takes the non-dominated sorted population after 300 population iterations as the optimization result, and obtains the optimized train operation plan in combination with the correlation.

[0177] Using the above method, we first determine the passenger travel cost, train operation cost and carbon emission cost, and comprehensively consider the cost of the train operation plan from multiple angles, which is conducive to solving the problem of effective cost control in subsequent optimization. Secondly, according to the train operation mode, such as the marshaling mode, the train's passenger capacity, cargo capacity and carbon emissions, and combined with the actual passenger and cargo flows, the train operation constraints are comprehensively considered from multiple aspects to avoid problems that do not conform to the actual situation due to over-optimization. Finally, the multi-objective optimization algorithm NSGA-II is used for iteration. In this way, multiple optimization objectives and constraints are considered, and the train operation plan can be fully optimized, which is conducive to achieving ideal optimization results.

[0178] Based on the above method steps, the present invention further provides a specific embodiment in combination with actual conditions:

[0179] The present invention conducts case analysis based on a specific subway line to verify the effectiveness of the model. Consider that 8 stations, 1, 5, 8, 11, 14, 19, 20, and 21, are urban rail transit freight business handling stations. 7 stations, 1, 5, 9, 12, 15, 20, and 21, have turnaround conditions. The research period is selected as 11:00-14:00 on a certain weekday, which are all non-peak passenger flow periods, and the time segment granularity is set to 1min. The passenger flow data is historical data derived from the automatic ticket vending and checking system (AFC). The freight flow data cannot be obtained in actual operation. The data is a numerical value given based on experience in combination with actual laws. The vehicle model is B1, the basic length of the vehicle is 19 meters, the basic width is 2.8 meters, and the height is 3.8 meters. The present invention assumes that the train stop time is 1 minute, and the interval running time set is {2, 2, 3, 2, 2, 3, 2, 2, 2, 2, 2, 2, 2, 3, 2, 3, 2, 2}. The model parameters are shown in Table 2.

[0180] Table 2 Parameter values

[0181] symbol Value and single symbol Value and single symbol Value and unit symbol Value and unit φ USD 35 / ton <![CDATA[E p ]]> <![CDATA[0027

[35] ]]> <![CDATA[h min ]]> 3min <![CDATA[F (p) ]]> 230 people <![CDATA[D1]]> 28.493km <![CDATA[E q ]]> <![CDATA[0.008

[35] ]]> <![CDATA[h max ]]> 10min <![CDATA[F (q) ]]> 40 boxes θ 0.5 <![CDATA[ST i ]]> 1min <![CDATA[s ve ]]> <![CDATA[27208m 2 ]]> v 75km / h σ 1min Te 15℃ <![CDATA[ρ air ]]> 10 μ 0.7

[0182] Based on the model and algorithm of the present invention, under the carbon tax policy, the present invention uses Matlaba2021 to solve the model of the present invention, with a population size of 200 and 200 iterations, and obtains the optimal operation plan as shown in Table 3.

[0183] Table 3 Optimal operation plan

[0184]

[0185] As shown in Table 3, compared with the single passenger flow transport mode, the average full load rate of passenger-freight coordinated transport under the conditions of fixed marshaling operation scheme, fixed marshaling large and small route operation scheme and virtual marshaling train operation scheme was increased by 6.5%, 5.0% and 6.8% respectively, which verified the effectiveness of passenger-freight coordinated transport on the airport line of composite urban rail transit. Compared with the single route mode, the passenger travel cost under the passenger-freight coordinated transport scheme of virtual marshaling train large and small route operation mode increased by about 2.5%, the average waiting time per person increased by about 0.46min, and the carbon emission cost and enterprise operating cost decreased by about 16.1% and 37.6% respectively; compared with the passenger-freight coordinated transport scheme of fixed marshaling large and small route, the passenger travel cost increased by about 1.5%, the average waiting time per person increased by about 0.28min, and the carbon emission cost and enterprise operating cost decreased by about 24.1% and 56.8% respectively, which verified the superiority of the virtual marshaling train large and small route transport scheme. The present invention compares the section load rate and the average waiting time of passengers at stations between the current transportation scheme and the optimization scheme of the present invention. In different sections, the optimization scheme can increase the section load rate by about 8%-50%, which is particularly obvious in small route sections. The average waiting time of passengers at different stations increases by about 0%-1%, which is more obvious in passenger and freight collaborative operation stations in small route sections.

[0186] The opening frequency and other parameters of the large and small route schemes (optimal scheme) mentioned above are fixed, and different stations with turnaround capability are considered to construct small routes to explore the influence of the turnaround station location on the operation scheme. The calculation results are shown in Table 3.

[0187] As shown in Table 3, Scheme 2 and Scheme 3 select the sections where the passenger and freight demand accounts for more than 60% to open small routes. Due to the high proportion of passenger and freight flow in the small routes, the waiting time of a large number of passengers in the small routes is reduced, resulting in a lower passenger travel cost than the optimal scheme, indicating that when selecting the location of the turnaround station, the proportion of passenger flow in the small routes should be increased; and in Scheme 2, the small routes do not include the No. 14 and No. 15 stations with the highest passenger flow, and the average cross-section full load rate increases by 2.6%, while the passenger travel cost increases by 18.7%, indicating that when selecting the turnaround station location, the small routes should cover the sections with high cross-section passenger flow in the up and down directions. Combining the six schemes, it can be seen that as the length of the small routes increases, the passenger travel cost decreases accordingly. Factors that affect carbon emissions and corporate costs include frequency of operation, route distance, and operating time. Scheme 6 selected the smallest route distance, but compared with the optimal solution, passenger travel costs increased by 6.7%, and carbon emission costs and corporate costs were only saved by 5.2% and 5.5%, respectively. This shows that when choosing routes, operating companies cannot blindly pursue their own interests. At the same time, running large and small routes also requires a certain passenger flow base. The route distance gap between Scheme 4 and Scheme 3 is significant, but carbon emissions and corporate operating costs do not decrease linearly with the increase in routes, indicating that the length of small routes is not as long as possible.

[0188] Based on this, when designing train route plans, enterprises cannot use a single indicator as the determining factor for making route plans. They should consider multiple aspects and reasonably balance the weight of each goal. For example, the influence of the location of the turnaround station should also be considered, as shown in Table 4. At the same time, small routes during peak hours should cover sections with high passenger and freight flows in the up and down sections, thereby alleviating the passenger and freight flow pressure in these sections.

[0189] Table 4 Effect of the turnaround station location on the objective function

[0190]

[0191] Based on the optimal large and small route operation plan in the above content, other parameters are fixed, and only the frequency ratio of large and small route train operation is changed to observe the impact of the operation ratio on the optimal operation plan.

[0192] When the marshaling scheme and the route scheme are fixed, as the frequency of the small route trains continues to increase, the passenger travel cost gradually decreases, but at the same time, the increase in the frequency of train operation leads to an increase in carbon emissions and corporate operating costs, and the average full load rate gradually decreases. Since the time value of non-working time per capita in the case cities selected by the present invention is relatively large, the curvature of passenger travel costs changes the most with the increase in the frequency of small route trains, indicating that for cities with a higher time value of non-working time per capita, the change in the ratio of the frequency of large and small routes has a greater impact on the passenger travel cost. Decision makers can choose the appropriate frequency ratio according to actual needs to adjust the three factors of passenger interests, energy conservation and emission reduction, and corporate costs.

[0193] Secondly, the present invention also provides a rail transit optimization device, such as Figure 7 As shown, including:

[0194] The cost determination module 701 is used to determine the total travel cost of passengers and the total operating cost of the train based on different travel plans according to the starting point of the passenger trip and the distance of the train in the target track for the train's round-trip travel; determine the carbon emissions of the train based on the electric power and operating power of the train, and determine the carbon emission cost of the train based on the current carbon emission quota and the cost within the quota.

[0195] The condition determination module 702 is used to determine multiple constraints on train operation based on the passenger flow and freight flow of the target track and the train's marshaling mode, passenger capacity, freight capacity and carbon emissions; determine the correlation between the total travel cost, total operating cost and carbon emission cost and the train operation parameters; the train operation parameters include the train's marshaling type, departure interval and number of carriage allocations.

[0196] The construction module 703 is used to construct the NSGA-II algorithm, including: building corresponding chromosome codes according to the train formation type, departure interval and number of carriage allocations, and generating an initial population according to multiple constraints.

[0197] The optimization module 704 is used to iterate the NSGA-II algorithm to minimize the total travel cost, the total operation cost and the carbon emission cost, and to screen the iteration results through the correlation to obtain an optimized train operation plan.

[0198] The present invention also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figure 1 The steps of a low-carbon optimization method for passenger and freight collaboration in virtual marshaling of urban rail transit are provided.

[0199] The present invention also provides a computer device. At the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory. Of course, it may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 1 The steps of a low-carbon optimization method for passenger and freight collaboration in virtual marshaling of urban rail transit are provided.

[0200] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0201] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as a combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0202] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0203] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0204] It should be noted that the above specific implementation method can enable those skilled in the art to understand the invention more comprehensively, but does not limit the invention in any way. Therefore, although the present invention has been described in detail in this specification, those skilled in the art should understand that the invention can still be modified or replaced by equivalents; and all technical solutions and improvements that do not deviate from the spirit and scope of the invention are included in the protection scope of the patent for the invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

Claims

1. A method for optimizing the coordinated operation of passenger and freight in virtual marshaling of urban rail transit, characterized in that: The method comprises: According to the starting point of the passenger's trip and the distance of the train's operation in the target track where the train travels back and forth, based on different travel plans, the total travel cost of the passenger and the total operating cost of the train are determined respectively; Determine the train's carbon emissions based on the train's electrical power and operating power, and determine the train's carbon emission costs based on the current carbon emission quota and cost within the quota; Determine multiple constraints on train operation based on the passenger and freight flows of the target track, as well as the train marshaling mode, passenger capacity, freight capacity, and preset carbon emissions; Determining the correlation between the total travel cost, the total operating cost and the carbon emission cost and the train operating parameters respectively; the train operating parameters include the train formation type, the departure interval and the number of carriages allocated; Constructing the NSGA-II algorithm, including: building the corresponding chromosome code according to the train formation type, departure interval and number of carriages allocated, and generating the initial population according to multiple constraints; The NSGA-II algorithm is iterated to minimize the total travel cost, the total operation cost and the carbon emission cost, and the iterative results are screened through the correlation to obtain an optimized train operation plan.

2. The method for optimizing the coordinated operation of passenger and freight in virtual marshaling of urban rail transit according to claim 1, characterized in that: The method of determining the total travel cost of passengers and the total running cost of trains based on different travel plans according to the starting point of passenger travel and the running distance of trains in the target track includes: Determine the first waiting time for passengers who do not transfer and the second waiting time for passengers who transfer respectively according to the starting point; The first waiting time and the second waiting time are determined based on the probability of not transferring, and a transfer penalty coefficient is introduced. The total waiting time of the passenger is determined according to the first waiting time, the second waiting time and the transfer penalty coefficient; the transfer penalty coefficient represents the proportion of the transfer time in the second waiting time; the probability of not transferring is the probability of the passenger not transferring obtained based on data statistics; Determine the total travel cost of the passenger based on the total waiting time and the per capita time value of the unit waiting time; The total operating cost of the train is determined based on the unit operating cost per kilometer and the unit purchase cost per hour of the train.

3. The method for optimizing the coordinated operation of passenger and freight in virtual marshaling of urban rail transit according to claim 2 is characterized in that: The first waiting time and the second waiting time are determined based on the probability of not transferring, and a transfer penalty coefficient is introduced. The total waiting time of the passenger is determined according to the first waiting time, the second waiting time and the transfer penalty coefficient, and includes: For passengers who do not transfer, the sum of the products of the number of waiting passengers and the length of the discrete time segment in all discrete time intervals is determined as the first waiting time; the discrete time segment is the interval time between trains of the same route in adjacent sequences passing the starting point; For transfer passengers, the sum of the waiting time of multiple transfer sections is determined as the second waiting time, and the transfer time is determined according to the transfer penalty coefficient and the second waiting time, and the sum of the second waiting time and the transfer time is determined as the waiting time of the transfer passenger; Determine a total waiting time which is the sum of the first waiting time and the waiting time of the transfer passenger.

4. The method for optimizing the coordinated operation of passenger and freight in virtual marshaling of urban rail transit according to claim 1, characterized in that: Determine the train electric power P de The formula is: in, are decision variables, namely the total number of passenger cars and freight cars assigned to the train, ρ air is the number of lamps in a single carriage, S ve is the surface area of ​​the carriage, Te is the temperature difference between the inside and outside of the carriage, and K is the train index; Determine the train running power P F The formula is: Among them, P μ is the vehicle adhesion mass, F μ is the train traction resistance, g is the acceleration of gravity, m h is the number of vehicles in the train, ω0 is the basic resistance of the train, L h is the distance travelled by the train; The formula for determining the carbon emissions of a transportation solution based on the train's electrical power and operating power is: Ce=PC sub ·a·E p +PC sub ·b·E q ; Among them, PC sub is the power consumption of train operation, α and β are the proportion of passenger and freight in the transportation plan, E p is the carbon emission factor of rail transit passenger transport, E q is the carbon emission factor of rail transit freight.

5. The method for optimizing the coordinated operation of passenger and freight in virtual marshaling of urban rail transit according to claim 1, characterized in that: The multiple constraints for train operation are determined based on the passenger flow and freight flow of the target track and the marshaling mode, passenger capacity, freight capacity and carbon emissions of the running train, respectively, including: Determining the first constraint condition based on the grouping mode includes: m1+m2≤m max ; N1≤N; Based on the passenger flow of the target track and the passenger capacity of the train, the second constraint is determined to include: Based on the cargo flow of the target track and the cargo capacity of the train, the third constraint is determined to include: Determining the fourth constraint condition based on the preset carbon emissions includes: Among them, h max ,h min is the maximum and minimum safe departure interval of the train; f h , λ, χ i and i are all decision variables; is the train arrival time, which means the time when train k arrives at station i; is the departure time of train at station i, indicating the time when train k leaves station i; is the stop time of train k at station i; is the running time of train k between stations i-1 and i; m h is the number of vehicles in the train set of large and small routes; h = 1 for large routes, h = 2 for small routes; m max is the maximum number of trains, Q s For the carbon emission quota purchased, are the passenger boarding and cargo loading of train k at station i, are the passenger and cargo capacity of train k when it leaves station i, is the cargo volume of train k when it leaves station i with station j as its destination, are the passenger disembarkation and cargo unloading quantities of train k when it arrives at station i, is the number of passengers waiting for station j at time t at station i, is the number of goods waiting for station j at station i at time t, is an indicator variable indicating whether the train has left the station or not.

6. The method for optimizing the coordinated operation of passenger and freight in virtual marshaling of urban rail transit according to claim 1, characterized in that: The NSGA-II algorithm is iterated to minimize the total travel cost, the total operation cost and the carbon emission cost. The iterative results are screened by the correlation relationship to obtain the optimized train operation plan including: The population is sorted according to non-dominance, and for the last selected frontier, the crowding comparison operator is used to sort it in descending order, and the individuals in front are selected by the binary tournament selection operator until the population is filled, so as to obtain a first-generation population; the first-generation population is used as the parent generation, and the second-generation population is generated through selection, crossover and mutation, and it is iterated until the number of iterations reaches a preset number, the iteration is terminated, and the finally obtained population is determined as the optimization result; An optimized train operation plan is determined according to the optimization result.

7. A rail transit optimization device, characterized in that: The device comprises: The cost determination module is used to determine the total travel cost of passengers and the total operating cost of the train based on different travel plans according to the starting point of the passenger trip and the distance of the train in the target track where the train travels back and forth; determine the carbon emissions of the train according to the electric power and operating power of the train, and determine the carbon emission cost of the train based on the current carbon emission quota and the cost within the quota; A condition determination module is used to determine multiple constraints for train operation based on the passenger flow and freight flow of the target track and the train's marshaling mode, passenger capacity, freight capacity and carbon emissions; respectively determine the correlation between the total travel cost, total operating cost and carbon emission cost and train operation parameters; the train operation parameters include the train's marshaling type, departure interval and number of carriages allocated; The construction module is used to construct the NSGA-II algorithm, including: building the corresponding chromosome code according to the train formation type, departure interval and number of carriage allocation, and generating the initial population according to multiple constraints; The optimization module is used to iterate the NSGA-II algorithm to minimize the total travel cost, the total operation cost and the carbon emission cost, and to screen the iteration results through the correlation to obtain the optimized train operation plan.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

Images