Railway freight train operation scheme optimization method and device under unbalanced traffic capacity condition

Abstract

The embodiment of the application provides a railway freight train operation scheme optimization method and device under unbalanced transport capacity conditions, abstracts a station as a point, abstracts trunk line transportation, transfer loading and unloading, car bottom turnover and transportation demand as arcs, constructs a transportation service network according to the point and the arc, and is used for describing the transportation turnover process of a train; and based on the transportation service network, under a preset constraint condition, takes the maximum transportation net profit as a target function, adopts an adaptive disturbance mechanism to determine a train operation scheme, realizes optimization of the railway freight train operation scheme under the unbalanced transport capacity conditions, and improves railway transportation resource utilization.

Description

Technical Field

[0001] This invention relates to the field of railway transportation technology, specifically to a method and apparatus for optimizing railway freight train operation schemes under conditions of unbalanced transport capacity. Background Technology

[0002] Train operation plans are the core of railway transport organization and the foundation for freight train timetable compilation. They are based on freight demand in all directions along the railway network, determining information such as the operating sections, rolling stock formations, and number of trains for each freight train. Given the limited capacity of railway transport, the rational design of train operation plans is crucial for improving railway transport revenue and enhancing the railway's competitiveness in the freight market.

[0003] Existing train operation plans can be broadly categorized into four types: First, manual decision-making, primarily based on qualitative analysis and experience-driven judgment, with plans manually formulated based on the actual experience of the staff; second, metaheuristic algorithms, such as the genetic algorithm with biased information designed by Zhang Yuzhao et al. based on the characteristics of the model to solve the optimization model of the operation plan; third, mathematical programming algorithms, such as the solution method based on Lagrange relaxation designed by Li Xinyi to integrate the optimization of the operation plan and the turnover of train carriages; and fourth, using commercial solvers, such as the multi-objective optimization model constructed by Wu Kaixin et al. to coordinate the optimization of the train operation plan and passenger flow control, and directly solving the problem by calling the GAMS solver. Among these methods, manual decision-making is not only inefficient but also has very limited consideration of factors, making it difficult to guarantee the quality of the proposed route optimization schemes. Metaheuristic algorithms (such as genetic algorithms and ant colony algorithms) are mostly inspired by natural phenomena, imitating physical phenomena in nature and the group survival behavior of organisms or animals. They are generally versatile, but they are not well applied to some route optimization problems. Mathematical programming algorithms have high requirements for solution resources and consume a lot of computer memory. Commercial solvers are generally expensive, and the time cost of solving large-scale problems is generally unacceptable.

[0004] Competition in the freight market is becoming increasingly fierce, yet railway freight turnover accounts for only 15.1% of the total national freight turnover, with bulk cargo transportation accounting for over 90% of the total railway freight volume. Due to my country's vast territory and the uneven distribution of resources and industries, the transportation of bulk commodities such as coal exhibits a "one-way" characteristic, resulting in many lines having trains heavily loaded in one direction and lightly loaded in the other, leading to a significant waste of railway transportation resources. Efficiently and rationally utilizing the surplus capacity outside of bulk cargo transportation is crucial for the rational allocation of railway transport capacity resources and improving the competitiveness of the railway market. However, this "one-way" characteristic means that in many bulk cargo railway transport sections, a large portion of the capacity in one direction is occupied by bulk cargo transportation, resulting in an "imbalance between forward and reverse capacity" for the available surplus capacity. The uneven transport capacity in both directions within a single railway freight transport section is very common. However, existing methods do not address this imbalance in train operation schemes, resulting in low utilization of railway transport resources and consequently low profits for railway companies. Consequently, railways are less competitive in the freight market, making it imperative to optimize railway freight train operation schemes. Summary of the Invention

[0005] This invention provides a method and apparatus for optimizing railway freight train operation schemes under unbalanced transport capacity conditions, in order to solve the problem of low utilization rate of railway transport resources under existing railway freight train operation schemes.

[0006] In a first aspect, embodiments of the present invention provide a method for optimizing railway freight train operation schemes under conditions of unbalanced transport capacity, including:

[0007] The stations are abstracted as points, and the trunk transportation, transshipment loading and unloading, wagon turnover and transportation demand are abstracted as arcs. A transportation service network is constructed based on the points and arcs to describe the transportation turnover process of the train.

[0008] Based on the transportation service network, and under preset constraints, the train operation plan is determined by using the maximization of net transportation profit as the objective function and an adaptive disturbance mechanism.

[0009] In one embodiment, the transportation service network includes: a two-way bidirectional forward transportation service network, a two-way bidirectional reverse transportation service network, an empty return train transportation service network, and a freight transportation service network.

[0010] In one embodiment, the preset constraints include:

[0011] The inflow and outflow of goods at each node must satisfy the flow conservation relationship;

[0012] The inflow and outflow of vehicles at each node must satisfy the flow conservation relationship;

[0013] Two-way freight trains can run in pairs between selectable origin and destination stations;

[0014] The sum of the cargo flow on the arc segment must be less than the maximum flow allowed under the vehicle on the arc segment;

[0015] The total number of vehicles used in a train service must be less than or equal to the total number of available vehicles.

[0016] Each shipment can only be carried by one train and no transshipment is allowed during transport;

[0017] Two-way freight trains can only choose one transport arc for both the forward and reverse directions and do not stop at intermediate stations.

[0018] In one embodiment, the objective function is determined according to the following expression:

[0019]

[0020] Where Max represents maximization, Z represents net transportation profit, d represents transportation demand, D represents the set of transportation demands, a represents an arc, and A represents the maximum value of the arc. xstart Let A represent the set of virtual initial arcs. d Represents the set of demand arcs. This indicates that the transportation demand d is allocated to arc a, r d A represents the freight revenue representing the transportation demand d. y Let c represent the set of running arcs. a v w represents the unit cost of vehicle use. d The weight represents the transport demand d, M represents the number of train formations, and c represents the weight of the transport demand d. a g Indicates unit transportation cost, l a c represents the length of the arc segment a. x N represents the idle cost per unit of vehicle. ping A represents the total number of usable vehicles that make up a two-way freight train service. pz V represents the set of forward arcs of bidirectional train services. p Let y represent the set of two-way, bidirectional freight trains. v a This indicates that train v selected arc segment a as its running path.

[0021] In one embodiment, determining the train operation plan using an adaptive disturbance mechanism includes:

[0022] Initialize the algorithm parameters, including the initial feasible solution set size, the number of iterations, and the probability of random perturbation;

[0023] Transportation routes with transportation revenue less than or equal to transportation cost are removed from the basic data to obtain preprocessed data;

[0024] Determine the encoding method and generate an initial feasible solution set according to the size of the initial feasible solution set based on the preprocessed data;

[0025] The initial feasible solution set is updated iteratively based on the random perturbation probability to obtain the updated feasible solution set.

[0026] When the number of iterations is reached, the train operation plan is determined based on the updated feasible solution set.

[0027] In one embodiment, the differential iterative update of the initial feasible solution set based on the random perturbation probability includes:

[0028] Determine the evaluation function value for each solution within the feasible solution set;

[0029] The feasible solution set is divided into a first solution set and a second solution set based on the magnitude of the evaluation function value. The evaluation function value of the solution in the first solution set is greater than that of the solution in the second solution set.

[0030] An adaptive perturbation mechanism is used to update the first solution set, resulting in the updated first solution set.

[0031] A solution is randomly selected from the first solution set to replace the corresponding solution in the second solution set, and the second solution set is updated based on the evaluation function value of the replaced solution to obtain the updated second solution set.

[0032] Mix the updated first solution set and the updated second solution set.

[0033] In one embodiment, updating the first solution set using an adaptive perturbation mechanism includes:

[0034] When the random disturbance probability is met, the initial priority of each running path is calculated, and the initial priority is determined based on the total freight cost of the corresponding running path.

[0035] The initial priority of each selected transportation route is adjusted to obtain the adjusted priority;

[0036] The element value at the perturbation position after the perturbation is determined using the roulette wheel algorithm based on the adjusted priority.

[0037] Secondly, embodiments of the present invention provide a device for optimizing railway freight train operation schemes under conditions of unbalanced transport capacity, comprising:

[0038] The module is used to abstract stations as points, trunk transportation, transshipment loading and unloading, wagon turnover and transportation demand as arcs, and to build a transportation service network based on points and arcs to characterize the transportation turnover process of the train.

[0039] The optimization module is used to determine train operation schemes based on the transportation service network, under preset constraints, with the objective function of maximizing transportation net profit, and employing an adaptive disturbance mechanism.

[0040] Thirdly, embodiments of the present invention provide an electronic device, comprising:

[0041] At least one processor and memory;

[0042] The memory stores the instructions that the computer executes;

[0043] At least one processor executes computer execution instructions stored in memory, causing the at least one processor to execute the railway freight train operation optimization method under unbalanced transport capacity conditions as described in any of the first aspects.

[0044] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the railway freight train operation optimization method under unbalanced transport capacity conditions as described in any of the first aspects.

[0045] The method and apparatus for optimizing railway freight train operation schemes under unbalanced transport capacity conditions provided in this invention abstracts stations as points and trunk line transportation, transshipment loading and unloading, wagon turnover, and transport demand as arcs. A transport service network is constructed based on the points and arcs to characterize the transport turnover process of the trains. Based on the transport service network, and under preset constraints, the method uses the maximization of net transport profit as the objective function and employs an adaptive disturbance mechanism to determine the train operation scheme. This achieves optimization of railway freight train operation schemes under unbalanced transport capacity conditions and improves the utilization rate of railway transport resources. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0047] Figure 1 This is a flowchart of a method for optimizing railway freight train operation under unbalanced transport capacity conditions, provided by an embodiment of the present invention.

[0048] Figure 2 This is a schematic diagram of a bidirectional, symmetrical, forward transport service network provided in an embodiment of the present invention;

[0049] Figure 3 This is a schematic diagram of a bidirectional, symmetrical, reverse transport service network provided in an embodiment of the present invention;

[0050] Figure 4 This is a schematic diagram of an empty train transportation service network provided in an embodiment of the present invention;

[0051] Figure 5 This is a schematic diagram of a cargo transportation service network provided in an embodiment of the present invention;

[0052] Figure 6 A flowchart of a method for optimizing railway freight train operation under unbalanced transport capacity conditions, as provided in another embodiment of the present invention;

[0053] Figure 7 This is a schematic diagram of the structure of a railway freight train operation optimization device under unbalanced transport capacity conditions according to an embodiment of the present invention;

[0054] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0055] The accompanying drawings have illustrated specific embodiments of the invention, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the invention in any way, but rather to illustrate the concept of the invention to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0056] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0057] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0058] The serial numbers assigned to components in this document, such as "first" and "second," are used only to distinguish the described objects and have no sequential or technical meaning. The terms "connection" and "linkage" used in this application, unless otherwise specified, include both direct and indirect connections (linkages).

[0059] The railway freight system is a comprehensive, complex, and dynamic system. Optimizing its train operation schemes is a typical NP-class problem with extremely high complexity. However, in actual transportation operations, the acceptable time for making these decisions is relatively short. Therefore, there is an urgent need for a method that closely matches practical application scenarios, is highly adaptable to real-world problems, has low computational cost requirements, and a short solution time to optimize train operation schemes.

[0060] This application addresses a scenario of uneven forward and reverse railway transport capacity, assuming a tight forward transport capacity under all conditions. Bulk cargo transport occupies the majority of the forward transport capacity, so the available capacity mainly comes from two sources: first, utilizing the remaining forward capacity excluding bulk cargo transport; and second, utilizing the substantial empty return capacity from reverse bulk cargo transport. Due to the severe tightness of forward capacity, to accommodate transport demands other than bulk cargo transport while minimizing the impact of station stops on line capacity, point-to-point bidirectional freight trains (hereinafter referred to as bidirectional trains) are operated using the remaining forward capacity. The forward transport demand is met by the forward capacity of the bidirectional trains, while the reverse transport demand is met by the reverse capacity of the bidirectional trains and the substantial empty return capacity of bulk cargo transport trains (hereinafter referred to as empty return trains). This application closely follows practical application scenarios in the field of railway transportation, establishing an optimization model for train operation schemes under conditions of unbalanced transport capacity. It employs an adaptive perturbation search algorithm and a differentiated iterative update mechanism to output the optimal train operation scheme, applicable to the optimization problem of railway freight train operation schemes under scenarios of unbalanced forward and reverse transport capacity. Specific embodiments will be used for detailed explanation below.

[0061] For ease of description, before the specific embodiments, the symbols involved in this application are first explained: Table 1 is an explanation of the sets and elements involved in this application, Table 2 is an explanation of the parameters involved in this application, and Table 3 is an explanation of the decision variables involved in this application.

[0062] Table 1

[0063]

[0064] Table 2

[0065]

[0066] Table 3

[0067]

[0068] Figure 1 This is a flowchart illustrating a method for optimizing railway freight train operation under unbalanced transport capacity conditions, as provided in an embodiment of the present invention. Figure 1 As shown in this embodiment, the method for optimizing railway freight train operation under unbalanced transport capacity conditions may include:

[0069] S101. Abstract the stations as points, and the trunk line transportation, transshipment loading and unloading, wagon turnover and transportation demand as arcs. Construct a transportation service network based on the points and arcs to characterize the transportation turnover process of the train.

[0070] To facilitate the establishment of the mathematical model, the data is first abstracted. In this embodiment, stations are abstracted as points, and trunk line transportation and transshipment / loading / unloading operations, wagon turnover, and the start and end points of transportation demand are abstracted as arc segments. A transportation service network is formed by points and arcs to characterize the transportation turnover process of the train service. The transportation service network in this embodiment may include: a two-way bidirectional train forward transportation service network, a two-way bidirectional train reverse transportation service network, an empty return train transportation service network, and a freight transportation service network. The following uses four stations (Station A, Station B, Station C, and Station D) as examples to illustrate the above four types of transportation service networks. Taking the route from Station A to Station D as the forward direction, Stations A, B, and D are assumed to be the starting and ending stations for point-to-point freight trains.

[0071] First, the turnaround process of the two-way freight trains is characterized. The two-way freight trains are mainly limited by the transport capacity of the forward line. Therefore, based on the constraints of the line's transport capacity, the maximum possible number of two-way trains can be determined, generating a candidate set of trains. All possible routes for each train in the candidate set are listed. Point-to-point two-way freight trains do not consider stops along the way. The constructed two-way freight train forward transport service network and reverse transport service network are as follows: Figure 2 and Figure 3 As shown, this network has four types of nodes: virtual start and virtual end points, representing the origin and destination of wagon turnover. Each station has two types of nodes, IN and OUT, to represent different states before and after the operation is completed. OUT indicates that the station operation, such as loading and unloading, is not completed, while IN indicates that the operation is completed. The IN and OUT nodes connecting the same station form a transfer loading and unloading arc, representing the transfer loading and unloading operation of goods when the train is at a certain station, and also representing the turnaround operation of the wagon at the destination station. A virtual start arc is added to connect the virtual start point and the OUT point of the wagon's originating station, representing the generation of the wagon in the network. A virtual end arc connects the virtual end point and the OUT point of the wagon's destination station, representing the disappearance of the wagon. The virtual end point and virtual start point are connected to form a train super arc, indicating that the alternative train has not been operated. There is a running arc connecting every two stations to represent the trunk transportation process of goods. Constraints are added to restrict the origin and destination stations of forward and reverse trains from being the same.

[0072] The capacity of empty return trains for bulk cargo transportation is ample, and the reverse transportation demand can be fully met. To save computation time, the number of empty return trains is assumed to be one, and their capacity is set to infinite. Similar to the bidirectional freight train service network, the empty return train transportation service network also has four types of arcs: virtual arcs, transfer / loading / unloading arcs, running arcs, and super arcs; and four types of nodes: virtual origin, virtual destination, IN point, and OUT point. Their meanings are exactly the same as those of nodes and arcs in the bidirectional freight train service network. The special feature is that, because the reverse transportation capacity is ample and the empty return trains need to return empty to the bulk cargo transportation origin station, the origin and destination stations are already determined, and stops can be made en route. A schematic diagram of the empty return train transportation service network can be found here. Figure 4 .

[0073] For transportation demand, virtual origin and virtual destination points represent the start and end points of the demand. Virtual nodes are connected to station nodes via virtual arcs, representing the generation and disappearance of transportation demand (freight flow) in the network. Furthermore, to ensure a balance in freight flow within the network, a super arc is constructed from the virtual origin to the destination. When freight flow is not transported (transport demand is not met), it is transported to the destination via the super arc; in this case, the freight flow's transportation revenue is 0. A schematic diagram of the freight transportation service network for an AC (Accounting Center) can be found here. Figure 5 .

[0074] S102. Based on the transportation service network, under preset constraints, the train operation plan is determined by using the maximum net transportation profit as the objective function and an adaptive disturbance mechanism.

[0075] On the constructed transportation service network, solve for the train operation scheme that maximizes the net profit of transportation while satisfying the preset constraints.

[0076] The preset constraints in this embodiment may include:

[0077] The inflow and outflow of goods at each node must satisfy the flow conservation relationship;

[0078] The inflow and outflow of vehicles at each node must satisfy the flow conservation relationship;

[0079] Two-way freight trains can run in pairs between selectable origin and destination stations;

[0080] The sum of the cargo flow on the arc segment must be less than the maximum flow allowed under the vehicle on the arc segment;

[0081] The total number of vehicles used in a train service must be less than or equal to the total number of available vehicles.

[0082] Each shipment can only be carried by one train and no transshipment is allowed during transport;

[0083] Two-way freight trains can only choose one transport arc for both the forward and reverse directions and do not stop at intermediate stations.

[0084] The constraints for balancing cargo flow are as follows:

[0085]

[0086] This indicates that the inflow and outflow of goods at each node must satisfy the flow conservation relationship, where... Using 0-1 variables ensures the indivisibility of the same shipment flow on a unique spatial path.

[0087] The constraints for traffic flow balance include constraints on empty return train traffic flow, constraints on forward traffic flow of bidirectional trains, and constraints on reverse traffic flow of bidirectional trains.

[0088] The constraints for the empty train flow are as follows:

[0089]

[0090] The constraints for the forward flow of bidirectional freight trains are:

[0091]

[0092] The constraints for the reverse flow of trains operating in both directions are:

[0093]

[0094] This indicates that the inflow and outflow of vehicles at each node must satisfy the flow conservation relationship, where... Using 0-1 variables ensures that the vehicle flow remains continuous on a unique spatial path, i.e., the vehicle body is fixed in formation.

[0095] The constraint condition for bidirectional trains to operate in pairs is expressed as follows:

[0096]

[0097] This indicates that the trains run in pairs between selectable origin and destination stations. In this case, the starting station of the train in the forward direction must be the destination station of the train in the reverse direction, and the destination station of the train in the forward direction must be the starting station of the train in the reverse direction.

[0098] The constraints on vehicle capabilities are:

[0099]

[0100] This means that the sum of the cargo flow on the arc segment must be less than the maximum flow allowed under the car body on the arc segment (determined by the number of cars in the undercarriage passing through the arc segment and the load capacity of each car).

[0101] The constraint on the number of vehicle bodies is:

[0102]

[0103] This constraint limits the total number of vehicles required to form a two-way train service; the sum of the vehicles used in each train service must be less than or equal to the total number of available vehicles.

[0104] The constraint that each cargo can only be served by a maximum of one train is:

[0105]

[0106]

[0107] This constraint means that each shipment can only be transported by one train, and no transshipment is allowed during transport. The preceding expression represents... and The correspondence is such that if a cargo is assigned to an arc of a certain train, it means that the cargo is served by that train; the latter expression means that each cargo can be served by at most one train.

[0108] The constraints for point-to-point bidirectional freight train services are:

[0109]

[0110] This constraint means that point-to-point bidirectional freight trains can only choose one transport arc for both the forward and reverse directions, and can operate without stopping at intermediate stations.

[0111] The restrictions on the cargo attributes that a vehicle can transport are as follows:

[0112]

[0113] Some goods can only be transported by either a two-way train or an empty return train. Let m v For vehicle attributes, m d The cargo attributes are as follows: cargo with attribute 1 can only be transported by round-trip trains, cargo with attribute 2 can only be transported by empty return trains, and cargo with attribute 0 can be transported by either type; the attribute for round-trip trains is 1, and the attribute for empty return trains is 2.

[0114] In this embodiment, maximizing the net profit of transportation is the optimization objective, and the net profit is determined based on revenue and cost. The freight revenue for each transportation demand is appended to the virtual starting arc of the freight demand. The total revenue can be expressed as... The cost consists of three parts: transportation cost, vehicle usage cost, and vehicle idle cost. Adding the transportation cost and vehicle usage cost to the operating arc, the transportation cost can be expressed as: Vehicle usage costs can be expressed as Vehicles being idle will incur certain idle costs, which can be expressed as...

[0115] In one alternative implementation, the objective function is determined according to the following expression:

[0116]

[0117] Where Max represents maximization, Z represents net transportation profit, d represents transportation demand, D represents the set of transportation demands, a represents an arc, and A represents the maximum value of the arc. xstart Let A represent the set of virtual initial arcs. d Represents the set of demand arcs. This indicates that the transportation demand d is allocated to arc a, r d A represents the freight revenue representing the transportation demand d. y Let c represent the set of running arcs. a v w represents the unit cost of vehicle use. d The weight represents the transport demand d, M represents the number of train formations, and c represents the weight of the transport demand d. a g Indicates unit transportation cost, l a c represents the length of the arc segment a. x N represents the idle cost per unit of vehicle. ping A represents the total number of usable vehicles that make up a two-way freight train service. pz V represents the set of forward arcs of bidirectional train services. p Let y represent the set of two-way, bidirectional freight trains. v a This indicates that train v selected arc segment a as its running path.

[0118] The model constructed in this application is a 0-1 integer programming model. As the number of transportation demands and stations increases, the massive computational complexity of combining network arcs, transportation demands, and different trains can significantly increase the complexity of the solution. Both precise solution algorithms and commercial solvers struggle to provide high-quality solutions in a short time. Therefore, this section designs a heuristic algorithm to solve the train operation optimization model under unbalanced capacity conditions, considering the characteristics of the problem and the model. This problem involves optimizing train operation schemes under unbalanced capacity conditions. It assumes that forward capacity is relatively limited, and its forward freight demand can only be met by bidirectional trains, while reverse empty trains have sufficient capacity to meet all reverse transportation demands. It can be seen that the core of solving this problem lies in how to select the origin and destination stations of bidirectional trains based on transportation demand. Since the origin and destination points of the demand are already determined, once the origin and destination points of a train are determined, the transportation demand allocated to that train will also be determined. In this embodiment, based on the constructed transportation service network, under preset constraints, the objective function is to maximize net transportation profit, and an adaptive perturbation mechanism is used to determine the train operation scheme.

[0119] The method for optimizing railway freight train operation under unbalanced transport capacity conditions provided in this embodiment abstracts stations as points and trunk line transport, transshipment loading and unloading, wagon turnover, and transport demand as arcs. A transport service network is constructed based on the points and arcs to characterize the transport turnover process of the trains. Based on the transport service network, and under preset constraints, the method uses the maximization of net transport profit as the objective function and employs an adaptive disturbance mechanism to determine the train operation plan. This method optimizes railway freight train operation under unbalanced transport capacity conditions and improves the utilization rate of railway transport resources.

[0120] Based on the above embodiments, the following provides a detailed explanation of how to use an adaptive disturbance mechanism to determine the train operation plan.

[0121] In this application, the optimization problem of railway freight train operation schemes under uneven transport capacity conditions can be decomposed into two sub-problems: first, which origin-destination trains should be operated; and second, what transport demands each train should meet. For sub-problem one, since the maximum number of trains that can be operated can be determined according to the line capacity, and the operating routes that each train can choose are also limited, this invention designs an adaptive perturbation search algorithm framework to solve the problem based on the characteristics of the model. For sub-problem two, once the operating routes of the trains are determined, the transport demands that they can serve are also determined. At this point, it is only necessary to select appropriate transport demands from the limited load capacity of each train to maximize its benefits.

[0122] Based on the above embodiments, the method for optimizing railway freight train operation schemes under unbalanced transport capacity conditions provided in this embodiment, specifically including the use of an adaptive disturbance mechanism to determine the train operation scheme, may include:

[0123] Step 1: Initialize the algorithm parameters, including the initial feasible solution set size, the number of iterations, and the probability of random perturbation.

[0124] Initialize the algorithm parameters. For example, you can set the initial feasible solution set size to 50, the number of iterations to 30, and the random perturbation probability to 0.3.

[0125] Step 2: Remove transportation routes whose transportation revenue is less than or equal to their transportation cost from the basic data to obtain preprocessed data.

[0126] To reduce the computational scale and improve data processing efficiency, the basic data can be preprocessed. All transportation demands and feasible transportation routes are known conditions. When the transportation demand on some routes is low or nonexistent, and their transportation revenue is less than or equal to their transportation cost, these routes cannot be selected. Eliminating these routes first can effectively reduce the computational scale. Let d∈{0,1} denote whether each route is eliminated, where d=0 indicates the route is eliminated, and d=1 indicates the route is not eliminated. Let a transportation route be numbered j, w... i For the weight of the cargo required for transport along the route (including both forward and reverse directions) numbered i, l i c is the path length. g For transportation costs, c v c represents the unit cost of vehicle use. x This represents the vehicle idle cost, M is the number of trains, α1 represents base price one, α2 represents base price two, and whether a transportation route is excluded can be represented as:

[0127]

[0128] Step 3: Determine the encoding method and generate an initial feasible solution set according to the size of the initial feasible solution set based on the preprocessed data.

[0129] First, determine the encoding method: As analyzed above, the solution to Problem 1 is to select which route each train should take. Since bidirectional trains operate point-to-point, once a forward route is selected, its reverse route is also determined. Therefore, an integer encoding method is used, starting from 1 to number each possible forward route scheme, with 0 added to indicate that the bidirectional train is not operating. The dimension of the solution space is the maximum number of trains that can be operated due to the forward throughput capacity limitation. Let the maximum forward throughput capacity be n (a maximum of n trains can be operated), and the total number of selectable train routes be m, x i This indicates that the i-th train selected the train numbered x. i If the running path is given, then the solution X can be represented as X = {x1, x2, ..., x...} n}, x i∈(0,1,…,m). Based on this encoding method, a certain number of initial feasible solutions are randomly generated in the feasible solution space as the basis for iterative optimization. The number of solutions included in the initial feasible solution set is equal to the size of the initial feasible solution set.

[0130] Step 4: Perform differentiated iterative updates on the initial feasible solution set based on the random perturbation probability to obtain the updated feasible solution set.

[0131] To preserve the characteristics of the better solutions in each generation while avoiding the algorithm getting trapped in local optima, two differentiated iterative strategies are designed based on the evaluation function values ​​of different solutions. The specific steps are as follows:

[0132] Calculate the evaluation function value for each solution in the feasible solution set: The objective of this model is to maximize transportation revenue. The evaluation function (which is the objective function of the model mentioned above) is designed accordingly, including freight revenue, freight transportation costs, vehicle usage costs, and vehicle idle costs. A larger evaluation function value indicates greater transportation revenue. For each feasible solution X = {x1, x2, ..., x...} n First, based on the corresponding transportation demand, a dynamic programming algorithm must be used to determine the service arc (running path) x of the train. i The principle is to maximize the transportation revenue of each train without exceeding its total load capacity, and then calculate the evaluation function for each solution.

[0133] The specific steps for calculating individual fitness are as follows:

[0134] First, calculate the optimal profit for each train corresponding to each element in the solution (X = {x1, x2, ..., x...). n Each running path x in} i The corresponding optimal benefit), the subproblem is modeled as follows: Suppose that there are k transportation demands that can be satisfied in the selected route of a certain train, and each transportation demand has only two states: satisfied or not satisfied. Let the decision variable be c. i = {0, 1}, indicating whether the i-th transportation demand is satisfied. The freight revenue from transportation demand is calculated from the freight rates published by the railway. Let p be the freight revenue from the i-th transportation demand. i The total weight w of the i-th transportation demand i The total load capacity of the train is W, and the model is established as follows:

[0135]

[0136] We use dynamic programming to solve the subproblems and calculate the optimal revenue for each train. We define V(i,r) as the maximum train revenue corresponding to the optimal combination of the first i transport demands, given a current knapsack capacity of r. The transport revenue for the i-th transport demand is p. i The total weight w of the i-th transportation demandi The recurrence relation is:

[0137]

[0138] After traversing all available transportation demands, the optimal solutions to the subproblems are obtained by backtracking from the optimal solution. Since each transportation demand can only be served once, the transportation demand corresponding to the optimal solution is marked as served.

[0139] Calculate the fitness function of all frog individuals in the population based on the solution of the above formula. When the solution X = {x1, x2, ..., x...} n Each x in} i When the optimal revenue for each train (i = 1, 2, ..., n) is calculated, the transportation demand that can be met by all trains running in opposite directions is determined. All remaining reverse transportation demand is met by the capacity of the empty trains running in opposite directions. The final total revenue, i.e., the fitness of an individual, is then calculated.

[0140] Group the solutions in the solution set. Sort all solutions in the solution set from highest to lowest according to their evaluation function values, and record the optimal solution X for this iteration. best The sorted results are divided into two subsets at the 50% mark: a set of better solutions and a set of worse solutions. For the top 50% of the better solutions, an adaptive perturbation mechanism is designed to add random perturbations with a certain probability to prevent them from getting trapped in local optima during iteration. For the bottom 50% of the worse solutions, to absorb the characteristics of the better solutions, elements are swapped with solutions from the better solution set, and the update method is determined by the evaluation function value of the swapped solution. The following sections explain how to update the better and worse solution sets respectively.

[0141] For updating the optimal solution set: To avoid the results getting trapped in local optima, this embodiment designs an adaptive perturbation mechanism, adding random perturbations to the feasible solutions whose evaluation function values ​​are in the top 50% of the solution set: In the solution X = {x1, x2, ..., x...} n Randomly select a perturbation position s∈[1,n] in the array, and for the element x at that position... s Changes are made. In fact, once a feasible solution X of the model is determined, it represents the selected transportation path {x1, x2, ..., x...}. n The transportation routes have been determined, and correspondingly, the transportation demands for these routes have been met. Similarly, in X, except for the perturbation bit x... s When other elements besides the given information are present, it indicates that some transportation demands have already been met. The number of transportation demands varies on different routes, and the priority of different routes in selection also differs. During random perturbations, some route options become unnecessary to select. To improve algorithm efficiency, this algorithm designs an adaptive perturbation mechanism, selecting a more suitable perturbation direction based on individual characteristics.

[0142] The specific steps of the adaptive perturbation mechanism are as follows:

[0143] The probability determines whether a random perturbation needs to be added to the individual. If the random perturbation probability is met, the process proceeds to the next step; otherwise, it determines whether the next individual needs to be added to the random perturbation, continuing until all solutions in the better solution set have been traversed.

[0144] Calculate the initial priority of each selected route. The initial priority is determined by the total freight cost of that route. Let the total number of selectable train routes be m, and let the initial priority of the j-th route be U. j Suppose that there are k transport demands that can be satisfied on the j-th transport route, w i , l i The meanings of α1 and α2 are the same as those of S2. Definition:

[0145]

[0146] The initial priority of each selected transportation route is adjusted based on the characteristics of the solution. Let the disturbance bit be s (s∈[1,n]), since only the individual X={x1,x2,…,x n An element x in} s The elements in X are the path numbers that can be selected (x i =0,1,2…m), the remaining genes can be understood as the arc segments with these numbers already selected, then the corresponding U j Adjustments are needed; perturb the position elements and x. i When the value is 0 (no bidirectional train with serial number i is running), no adjustment is needed. Let the total load of a train be W, and the adjustment method is as follows:

[0147] U j = j -,j∈X,j≠x b , j≠0.

[0148] The roulette wheel algorithm is used to determine the perturbed x based on the current priority of each transportation route. s The roulette wheel selection algorithm's steps are: the probability that the j-th running path is selected. Then calculate the cumulative probability that the j-th running path is selected. Generate a random number r between 0 and 1. Determine which interval of the cumulative probability r falls into. If it falls into that interval, the corresponding transportation route is selected. Replace the element x at the disturbance position with the selected route number. s .

[0149] For updating the poor solution set: Let the i-th feasible solution in the poor solution set be Xiworse, and the i-th feasible solution in the better solution set be Xibetter. According to the method for calculating the transportation revenue of each path in S4.1.1, find the running path xiworse with the minimum revenue within Xiworse. Randomly select one of the transportation paths xibetter in the better solution set Xibetter and replace xiworse with xibetter to generate a new feasible solution Xnew1. Calculate the evaluation function value of Xnew1. If the evaluation function value of the new feasible solution Xnew1 generated after replacement is worse than the original individual Xiworse, then replace the original feasible solution Xiworse in the poor solution set with Xnew1; if the new feasible solution after replacement is worse than the original feasible solution, then randomly select the transportation path xbest in the globally optimal feasible solution Xbest and replace xibetter with xbest to generate a new feasible solution Xnew2. Calculate the evaluation function value of Xnew2. If the evaluation function value of the new feasible solution Xnew2 generated after the replacement is better than that of the original individual xiworse, then replace the original feasible solution Xiworse in the poor solution set with Xnew2; otherwise, randomly generate a feasible solution Xnew3 in the feasible solution space to replace the original feasible solution Xiworse in the poor solution set.

[0150] One optional implementation example, which involves iteratively updating the initial feasible solution set based on random perturbation probabilities, may specifically include:

[0151] Determine the evaluation function value for each solution within the feasible solution set;

[0152] Based on the magnitude of the evaluation function value, the feasible solution set is divided into a first solution set (better solution set) and a second solution set (poorer solution set). The evaluation function value of the solution in the first solution set is greater than the evaluation function value of the solution in the second solution set.

[0153] An adaptive perturbation mechanism is used to update the first solution set, resulting in the updated first solution set.

[0154] A solution is randomly selected from the first solution set to replace the corresponding solution in the second solution set, and the second solution set is updated based on the evaluation function value of the replaced solution to obtain the updated second solution set.

[0155] Mix the updated first solution set and the updated second solution set.

[0156] Specifically, updating the first solution set using an adaptive perturbation mechanism may include:

[0157] When the random disturbance probability is met, the initial priority of each running path is calculated, and the initial priority is determined based on the total freight cost of the corresponding running path.

[0158] The initial priority of each selected transportation route is adjusted to obtain the adjusted priority;

[0159] The element value at the perturbation position after the perturbation is determined using the roulette wheel algorithm based on the adjusted priority.

[0160] Step 5: When the number of iterations is reached, determine the train operation plan based on the updated feasible solution set.

[0161] When the set number of iterations is reached, the optimal solution X is output. best This is the optimal train operation plan.

[0162] For a detailed flowchart of the above algorithm, please refer to [link / reference]. Figure 6 .

[0163] This embodiment provides a method for optimizing railway freight train operation schemes under uneven transport capacity conditions. Building upon the previous embodiment, it further designs an adaptive perturbation search algorithm based on the characteristics of the problem and model. The differentiated iterative update mechanism preserves the characteristics of optimal solutions while preventing the algorithm from getting trapped in local optima. The adaptive perturbation mechanism reduces dependence on initial feasible solutions while conducting guided local optimization, accelerating the algorithm's convergence speed. Closely aligned with practical applications in the railway transportation field, it establishes an optimization model for operation schemes under uneven transport capacity conditions. This model is applicable to optimizing railway freight train operation schemes under scenarios of uneven forward and reverse transport capacity, improving the utilization rate of railway transportation resources, increasing railway enterprise profits, and enhancing the railway's competitiveness in the freight market. This is particularly significant in the context of my country's economic restructuring and industrial transformation and upgrading. The adaptive perturbation search algorithm provided in this embodiment outperforms traditional metaheuristic algorithms (genetic algorithms) and significantly reduces solution time compared to commercial solvers.

[0164] Figure 7 This is a schematic diagram of a railway freight train operation optimization device under unbalanced transport capacity conditions, provided in an embodiment of the present invention. Figure 7 As shown, the railway freight train operation optimization device 70 under unbalanced transport capacity conditions provided in this embodiment may include: a construction module 701 and an optimization module 702.

[0165] Module 701 is used to abstract stations as points, and trunk transportation, transshipment loading and unloading, wagon turnover and transportation demand as arcs. Based on the points and arcs, a transportation service network is constructed to characterize the transportation turnover process of the train.

[0166] The optimization module 702 is used to determine the train operation plan based on the transportation service network, under preset constraints, with the objective function of maximizing the net profit of transportation, and using an adaptive disturbance mechanism.

[0167] The apparatus of this embodiment can be used to perform Figure 1 The technical solutions of the method embodiments shown are similar in principle and in effect, and will not be described again here.

[0168] This invention also provides an electronic device, please refer to [link to relevant documentation]. Figure 8 As shown, the embodiments of the present invention are only used as examples. Figure 8 The examples are provided for illustration only and do not imply that the invention is limited to these examples. Figure 8 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present invention. Figure 8 As shown, the electronic device 80 provided in this embodiment may include: a memory 801, a processor 802, and a bus 803. The bus 803 is used to connect the various components.

[0169] The memory 801 stores a computer program, which, when executed by the processor 802, can implement the technical solutions of any of the above method embodiments.

[0170] The memory 801 and processor 802 are electrically connected directly or indirectly to enable data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses or signal lines, such as bus 803. The memory 801 stores a computer program for optimizing railway freight train operation schemes under unbalanced transport capacity conditions, including at least one software functional module that can be stored in the memory 801 in the form of software or firmware. The processor 802 executes various functional applications and data processing by running the software program and modules stored in the memory 801.

[0171] The memory 801 may be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory 801 stores programs, which are then executed by the processor 802 upon receiving execution instructions. Furthermore, the software programs and modules within the memory 801 may include an operating system, which may include various software components and / or drivers for managing system tasks (e.g., memory management, storage device control, power management, etc.) and can communicate with various hardware or software components to provide an operating environment for other software components.

[0172] Processor 802 can be an integrated circuit chip with signal processing capabilities. The aforementioned processor 802 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. It is understood that... Figure 8 The structure shown is for illustrative purposes only and may include more... Figure 8 The more or fewer components shown, or having the same Figure 8 The different configurations shown. Figure 8 The components shown can be implemented in hardware and / or software.

[0173] This invention also provides a computer-readable storage medium storing a computer program thereon, which is executed by a processor to implement the technical solutions of any of the above method embodiments.

[0174] The various embodiments in this disclosure are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0175] The scope of protection of this disclosure is not limited to the embodiments described above. Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from its scope and spirit. If such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, then the intent of this disclosure also includes such modifications and variations.

Claims

1. A method for optimizing railway freight train operation under conditions of unbalanced transport capacity, characterized in that, include: The stations are abstracted as points, and the trunk transportation, transshipment loading and unloading, wagon turnover and transportation demand are abstracted as arcs. A transportation service network is constructed based on the points and the arcs to characterize the transportation turnover process of the train. Based on the aforementioned transportation service network, and under preset constraints, with the objective function being the maximization of net transportation profit, an adaptive disturbance mechanism is employed to determine the train operation plan. The transportation service network includes: a two-way bidirectional forward transport service network, a two-way bidirectional reverse transport service network, an empty return train transport service network, and a freight transport service network. The preset constraints include: The inflow and outflow of goods at each node must satisfy the flow conservation relationship; The inflow and outflow of vehicles at each node must satisfy the flow conservation relationship; Two-way freight trains can run in pairs between selectable origin and destination stations; The sum of the cargo flow on the arc segment must be less than the maximum flow allowed under the vehicle on the arc segment; The total number of vehicles used in a train service must be less than or equal to the total number of available vehicles. Each shipment can only be carried by one train and no transshipment is allowed during transport; Two-way freight trains can only choose one transport arc for both the forward and reverse directions and do not stop at intermediate stations; The objective function is determined according to the following expression: Where Max represents maximizing, Z represents net transportation profit, d represents transportation demand, D represents the set of transportation demands, and a represents an arc. Represents the set of virtual initial arcs. Represents the set of demand arcs. This indicates that the transportation demand d is allocated to arc a. The freight revenue represents the transportation demand d. Represents the set of running arcs. This indicates the unit cost of vehicle use. The weight represented by transport demand d, and M represent the number of train formations. Indicates unit transportation cost, This represents the length of the arc segment 'a'. This represents the idle cost per unit of vehicle. This indicates the total number of usable vehicles that make up a two-way, reciprocating train service. This represents the set of forward arcs for bidirectional, two-way train services. This represents a set of trains that operate in both directions. This indicates that train v selected arc segment a as its running path; The method of using an adaptive disturbance mechanism to determine the train operation plan includes: Initialize the algorithm parameters, including the initial feasible solution set size, the number of iterations, and the probability of random perturbation; Transportation routes with transportation revenue less than or equal to transportation cost are removed from the basic data to obtain preprocessed data; Determine the encoding method and generate an initial feasible solution set according to the size of the initial feasible solution set based on the preprocessed data; The initial feasible solution set is iteratively updated based on the random perturbation probability to obtain the updated feasible solution set; When the number of iterations is reached, a train operation plan is determined based on the updated feasible solution set.

2. The method according to claim 1, characterized in that, The step of performing differentiated iterative updates on the initial feasible solution set based on the random perturbation probability includes: Determine the evaluation function value for each solution within the feasible solution set; The feasible solution set is divided into a first solution set and a second solution set according to the magnitude of the evaluation function value, wherein the evaluation function value of the solution in the first solution set is greater than the evaluation function value of the solution in the second solution set; An adaptive perturbation mechanism is used to update the first solution set to obtain the updated first solution set; A solution is randomly selected from the first solution set to replace the corresponding solution in the second solution set, and the second solution set is updated according to the evaluation function value of the replaced solution to obtain the updated second solution set. The updated first solution set and the updated second solution set are then mixed.

3. The method according to claim 2, characterized in that, The step of updating the first solution set using an adaptive perturbation mechanism includes: When the random disturbance probability is met, the initial priority of each running path is calculated, and the initial priority is determined based on the total freight cost of the corresponding running path. The initial priority of each selected transportation route is adjusted to obtain the adjusted priority; The element value at the perturbation position after the perturbation is determined using the roulette wheel algorithm based on the adjusted priority.

4. A device for optimizing railway freight train operation schemes under unbalanced transport capacity conditions, used to implement the optimization method for railway freight train operation schemes under unbalanced transport capacity conditions as described in any one of claims 1-3, characterized in that, include: The module is used to abstract stations as points, and trunk transportation, transshipment loading and unloading, wagon turnover and transportation demand as arcs. Based on the points and arcs, a transportation service network is constructed to characterize the transportation turnover process of the train. The optimization module is used to determine the train operation plan based on the transportation service network, under preset constraints, with the objective function of maximizing the net profit of transportation, and employing an adaptive disturbance mechanism.

5. An electronic device, characterized in that, include: At least one processor and memory; The memory stores computer-executed instructions; The at least one processor executes the computer execution instructions stored in the memory, causing the at least one processor to execute the railway freight train operation optimization method under unbalanced transport capacity conditions as described in any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the railway freight train operation optimization method under unbalanced transport capacity conditions as described in any one of claims 1-3.

Images