A train diagram optimization method, device and equipment and storage medium

Abstract

The application discloses a train operation diagram optimization method, device and equipment and a storage medium. The method comprises the following steps: for each train, generating an actual disturbance stay scene of the train at each station according to the historical actual stay duration of the train and the planned stay duration in an initial train operation diagram; taking the minimum expected value of the total net traction energy consumed by all power supply partitions under the actual disturbance stay scene as the target, constructing a train operation diagram optimization model, solving the train operation diagram optimization model, and optimizing the initial train operation diagram according to the solving result to obtain an optimized train operation diagram. The optimization effect of the train operation diagram can be improved through the application, and the energy consumption of urban rail transit can be reduced to the maximum extent according to the optimized train operation diagram.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of urban rail transit operation diagram optimization, and in particular to a train operation diagram optimization method, device, equipment and storage medium. BACKGROUND

[0002] With the continuous expansion of urban rail transit network, its energy consumption also increases rapidly, and the train traction energy consumption accounts for more than half of the total energy consumption, so reducing train traction energy consumption is the key to reducing urban rail transit energy consumption. Because the train traction energy consumption mainly depends on train performance, line conditions and train operation behavior, if the train operation is adjusted from the perspective of reasonably allocating the interval running time and adjusting the train arrival and departure time from the operation diagram level, the train traction energy consumption can be reduced and the regenerative energy utilization rate can be improved, so as to achieve the purpose of reducing urban rail transit energy consumption without additional investment.

[0003] However, the existing train operation diagram optimization method only optimizes the allocation of train interval running time from the perspective of reducing train traction energy consumption, ignoring the recycling of regenerative braking energy generated during train braking. In addition, in the actual urban rail transit operation process, factors such as passenger flow congestion will cause train station time delay, so that the actual train station time deviates from the planned station time. However, the existing technology often only targets fixed operation scenarios when optimizing the train operation diagram, ignoring the uncertainty of the actual train station time, which has strong limitations and poor anti-interference ability, resulting in low optimization effect of the train operation diagram. Therefore, there is an urgent need for a train operation diagram optimization method to improve the optimization effect of the train operation diagram and minimize the energy consumption of urban rail transit. SUMMARY

[0004] The present application provides a train operation diagram optimization method, device, equipment and storage medium to solve the technical problem of poor optimization effect of the train operation diagram in the existing train operation diagram optimization method.

[0005] To solve the above technical problems, the present application provides a train operation diagram optimization method, which comprises:

[0006] An initial operation diagram of a train is obtained, wherein the initial operation diagram comprises: a planned stay duration of each train at each station;

[0007] For each train, a historical actual stay duration of the train at each station is obtained, and an actual disturbance stay scenario of the train at each station is generated according to the historical actual stay duration and the planned stay duration;

[0008] An optimization model of a train diagram is constructed with an objective of minimizing an expected value of total net traction energy consumed by all power supply partitions under the actual disturbance stay scenario;

[0009] The optimization model of the train diagram is solved to obtain a departure interval duration of the train and a next adjacent train at a departure station, and a predicted running speed level of the train in each section, and then the initial train diagram is optimized according to the departure interval duration and the predicted running speed level to obtain an optimized train diagram;

[0010] Wherein, for each power supply partition, the net traction energy consumed by the power supply partition under the actual disturbance stay scenario is calculated by the following way:

[0011] The traction energy consumed and the braking energy generated by the train at each time during running are calculated according to the running speed level of the train between adjacent stations;

[0012] The total traction energy consumed and the total braking energy generated by the power supply partition at each time are calculated according to the traction energy consumed and the braking energy generated by the train at each time during running, and then the net traction energy consumed by the power supply partition at each time is calculated according to the total traction energy and the total braking energy; wherein, the power supply partition corresponds to several stations.

[0013] As a preferred solution, the historical actual stay duration of the train at each station is obtained, and the actual disturbance stay scenario of the train at each station is generated according to the historical actual stay duration and the planned stay duration, including:

[0014] A preset stay duration interval is obtained, and a stay duration is randomly extracted from the stay duration interval as a disturbance stay duration;

[0015] The predicted stay duration of the train at each station is obtained according to the disturbance stay duration and the planned stay duration;

[0016] The historical actual stay duration of the train at each station is obtained, and the occurrence probability of the predicted stay duration is calculated according to the historical actual stay duration;

[0017] The actual disturbance stay scenario of the train at each station is generated according to the predicted stay duration of the train at each station and the occurrence probability of the predicted stay duration.

[0018] As a preferred solution, the traction energy consumed and the braking energy generated by the train at each time during running are calculated according to the running speed level of the train between adjacent stations, including:

[0019] The resistance of the train at each time during the running process is calculated according to the running speed level of the train between adjacent stations, wherein the resistance includes a basic resistance of the train, an additional resistance of a slope and an additional resistance of a curve.

[0020] The traction force and the braking force of the train at each time during the running process are calculated according to the running speed level and the resistance.

[0021] The traction energy consumed by the train at each time during the running process is calculated according to the traction force, and the braking energy generated by the train at each time during the running process is calculated according to the braking force.

[0022] As a preferred solution, the train diagram optimization model is constructed with the minimum expected value of the total net traction energy consumed by all power supply partitions under the actual disturbance stay scene as the target, including:

[0023] Decision variables of the train diagram optimization model are obtained, wherein the decision variables include a departure interval duration of the train and a next adjacent train at a starting station, and a predicted running speed level of the train in each section;

[0024] Constraint conditions of the train diagram optimization model are obtained, wherein the constraint conditions include a constraint of the departure interval duration, an integer constraint of the departure interval duration, a constraint of an operation service duration of the train, a constraint of a total running duration of the train from a starting station to a terminal station, a constraint of an interval duration between a departure of the train from a station and an arrival of a next adjacent train at the station, and a selection constraint of the predicted running speed level;

[0025] The train diagram optimization model is constructed with the minimum expected value of the total net traction energy consumed by all power supply partitions under the actual disturbance stay scene as the target according to the decision variables and the constraint conditions.

[0026] As a preferred solution, the train diagram optimization model is solved to obtain the departure interval duration of the train and the next adjacent train at the starting station, and the predicted running speed level of the train in each section, including:

[0027] The train diagram optimization model is solved according to the adaptive large neighborhood search algorithm to generate an initial solution of the departure interval duration of the train and the next adjacent train at the starting station, and the predicted running speed level of the train in each section;

[0028] Initial parameters of the adaptive large neighborhood search algorithm are obtained, wherein the initial parameters include rules of a destroy operator, rules of a repair operator, scores of operators, weights of operators, an initial temperature of simulated annealing, a termination temperature of simulated annealing, a cooling speed of simulated annealing, an iteration number of the adaptive large neighborhood search algorithm and an iteration number of the simulated annealing algorithm;

[0029] assign the initial solution to a current solution and an optimal solution of the train diagram optimization model, then optimize the current solution according to the initial parameters to obtain the optimal solution of the train diagram optimization model, and take the optimal solution as the departure interval time length of the train and the next adjacent train at a departure station and the predicted running speed level of the train at each section.

[0030] As a preferred solution, the optimizing the current solution according to the initial parameters to obtain the optimal solution of the train diagram optimization model comprises:

[0031] The optimization operation is performed on the current solution according to the initial parameters until a preset termination condition is met, and the optimal solution of the train diagram optimization model is obtained.

[0032] The optimization operation comprises:

[0033] According to the weight of the operator, a roulette method is used to select a corresponding operator as a perturbation operator, and then the current solution of the train diagram optimization model is perturbed according to the perturbation operator to obtain a new solution of the train diagram optimization model.

[0034] The new solution is judged in relation to the current solution and the optimal solution, and the current solution, the optimal solution and the operator score are updated according to the relation.

[0035] When it is detected that the updated operator score reaches a preset weight update condition, the operator weight is updated and the operator score is reset.

[0036] The initial temperature and the iteration number of simulated annealing are updated according to the new solution.

[0037] On the basis of the above-mentioned embodiments, another embodiment of the present application provides a train diagram optimization device, characterized in that it comprises an initial train diagram acquisition module, an actual perturbation stay scene generation module, a train diagram optimization model construction module, an initial train diagram optimization module and a net traction energy calculation module.

[0038] The initial train diagram acquisition module is configured to acquire an initial train diagram of a train, wherein the initial train diagram comprises a planned stay time length of each train at each station.

[0039] The actual perturbation stay scene generation module is configured to acquire, for each train, a historical actual stay time length of the train at each station, and generate an actual perturbation stay scene of the train at each station according to the historical actual stay time length and the planned stay time length.

[0040] The train diagram optimization model construction module is configured to construct a train diagram optimization model with the minimum expected value of total net traction energy consumed by all power supply partitions under the actual disturbance stay scenario as the target;

[0041] The initial train diagram optimization module is configured to solve the train diagram optimization model to obtain a departure interval duration of a train and a next adjacent train at a departure station and a predicted running speed level of the train in each section, and then optimize the initial train diagram according to the departure interval duration and the predicted running speed level to obtain an optimized train diagram.

[0042] The net traction energy calculation module is configured to calculate the net traction energy consumed by a power supply partition under the actual disturbance stay scenario for each power supply partition by: calculating the traction energy consumed and the braking energy generated by the train at each moment during running according to the running speed level of the train between adjacent stations; calculating the total traction energy consumed and the total braking energy generated by the power supply partition at each moment according to the traction energy consumed and the braking energy generated by the train at each moment during running, and then calculating the net traction energy consumed by the power supply partition at each moment according to the total traction energy and the total braking energy; wherein the power supply partition corresponds to a plurality of stations.

[0043] On the basis of the above-mentioned embodiments, a further embodiment of the application provides a train diagram optimization device, which comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the train diagram optimization method of the above-mentioned embodiments of the application when executing the computer program.

[0044] On the basis of the above-mentioned embodiments, a further embodiment of the application provides a storage medium, which comprises a stored computer program, wherein the computer program controls a device in which the computer readable storage medium is located to execute the train diagram optimization method of the above-mentioned embodiments of the application when running.

[0045] Compared with the prior art, the embodiments of the application have the following beneficial effects:

[0046] The train diagram optimization method provided by the application firstly constructs an actual disturbance stay scenario of a train at each station according to the historical actual stay duration of the train at each station and the planned stay duration in the initial train diagram for each train service, and then optimizes the train diagram under the actual disturbance stay scenario, so that the finally optimized train diagram is more in line with the actual operation situation and has a better optimization effect.

[0047] In addition, in the calculation of the energy consumption of the train, the traction energy consumed and the braking energy generated by each train in the driving process at each time are comprehensively considered, the total traction energy consumed and the total braking energy generated by the power supply partition at each time are calculated according to the traction energy consumed and the braking energy generated by the train at each time in the driving process, and then the net traction energy consumed by the power supply partition at each time is calculated according to the total traction energy and the total braking energy, and the net traction energy is taken as the actual consumption of the traction of each power supply partition. The expected value of the total net traction energy consumed by all power supply partitions under the actual disturbance stay scene is minimized as the target, a train working diagram optimization model is constructed, the recovery and utilization of the regenerative braking energy generated by the train during braking are also comprehensively considered in the case of fitting the actual operation, and the optimization effect of the train working diagram is effectively improved. According to the optimized train working diagram, the energy consumption of urban rail transit can be reduced to the maximum extent. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 is a flowchart of a train working diagram optimization method provided by an embodiment of the present application;

[0049] Figure 2 is a curve diagram of the driving speed level of the train between adjacent stations;

[0050] Figure 3 is a schematic diagram of energy transmission in the same power supply partition;

[0051] Figure 4 is a flowchart of the solution of the train working diagram optimization model;

[0052] Figure 5 is an optimized train working diagram;

[0053] Figure 6 is a structural schematic diagram of a train working diagram optimization device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0055] Embodiment one

[0056] Please refer to Figure 1 is a flowchart of a train working diagram optimization method provided by an embodiment of the present application, including the following specific steps:

[0057] S1, obtaining an initial train diagram of a train; wherein the initial train diagram comprises: a planned stay duration of each train at each station;

[0058] Firstly, initial train diagram, train parameters and line condition data are obtained; the initial train diagram data comprises: the number of trains, and the planned arrival and departure time of each train at each station, from which the planned stay duration of each train at each station can be known; the train parameter data comprises: the mass of the train, the Davis equation coefficient, the maximum acceleration, the traction characteristic curve and the braking characteristic curve; and the line condition data comprises: station data, slope data, curve data and the corresponding speed limit of the road section.

[0059] S2, for each train, obtaining the historical actual stay duration of the train at each station, and generating an actual disturbance stay scenario of the train at each station according to the historical actual stay duration and the planned stay duration;

[0060] Preferably, the obtaining of the historical actual stay duration of the train at each station and the generation of the actual disturbance stay scenario of the train at each station according to the historical actual stay duration and the planned stay duration comprise: obtaining a preset stay duration interval, and randomly extracting a stay duration from the stay duration interval as a disturbance stay duration; obtaining a predicted stay duration of the train at each station according to the disturbance stay duration and the planned stay duration; obtaining the historical actual stay duration of the train at each station, and calculating the occurrence probability of the predicted stay duration according to the historical actual stay duration; and generating the actual disturbance stay scenario of the train at each station according to the predicted stay duration of the train at each station and the occurrence probability of the predicted stay duration.

[0061] In the actual operation of urban rail transit, train station delays caused by passenger flow congestion and other factors will cause the actual stay duration of the train at the station to deviate from the planned stay duration. The predicted stay duration of the train at each station can be composed of the planned stay duration and the disturbance stay duration, and can be expressed as:

[0062]

[0063] In the above formula, is the predicted stay duration of train i at station m; is the planned stay duration of the train at station m; is the disturbance stay duration of train i at station m, which is randomly selected within a preset stay duration interval .

[0064] All the predicted dwell time lengths of the trains in the whole operation process form an actual disturbance dwell scenario w, w is composed of an I row and 2M column matrix, wherein each element represents the predicted dwell time length of train i at station m under the actual disturbance dwell scenario w, as follows:

[0065]

[0066] The train diagram optimization will generate W actual disturbance dwell scenarios, and the occurrence probability p of each scenario is w It can be obtained according to historical operation data statistics, satisfying That is, the occurrence probability of the predicted dwell time length can be calculated according to the historical actual dwell time length of the train at each station.

[0067] In this step, for each train, first, according to the historical actual dwell time length of the train at each station and the planned dwell time length in the initial train diagram, the actual disturbance dwell scenario of the train at each station is constructed, and then the train diagram is optimized under the actual disturbance dwell scenario, so that the final optimized train diagram is more in line with the actual operation situation, and the optimization effect is better.

[0068] S3, taking the minimum expected value of the total net traction energy consumed by all power supply partitions under the actual disturbance dwell scenario as the target, a train diagram optimization model is constructed;

[0069] Among them, for each power supply partition, the net traction energy consumed by the power supply partition under the actual disturbance dwell scenario is calculated by the following method:

[0070] The driving speed level of the train between adjacent stations is used to calculate the traction energy consumed and the braking energy generated by the train at each time during driving;

[0071] According to the traction energy consumed and the braking energy generated by the train at each time during driving, the total traction energy consumed and the total braking energy generated by the power supply partition at each time are calculated, and then the net traction energy consumed by the power supply partition at each time is calculated according to the total traction energy and the total braking energy; wherein the power supply partition corresponds to several stations.

[0072] Preferably, the calculation of the traction energy and the braking energy consumed by the train at each moment during the running process according to the running speed level of the train between adjacent stations comprises: calculating the resistance of the train at each moment during the running process according to the running speed level of the train between adjacent stations; wherein the resistance comprises: basic resistance of the train, additional resistance of a slope and additional resistance of a curve; calculating the traction force and the braking force of the train at each moment during the running process according to the running speed level and the resistance; calculating the traction energy consumed by the train at each moment during the running process according to the traction force, and calculating the braking energy generated by the train at each moment during the running process according to the braking force.

[0073] Firstly, the related concepts of the traction energy, the braking energy and the net traction energy are described:

[0074] The traction energy of the train on the whole line refers to the sum of the energy required for the traction of all trains on the line within the operation time; the regenerative braking energy of the train on the whole line refers to the sum of the regenerative braking energy generated by the braking of all trains on the line within the operation time; the regenerative energy utilization of the train on the whole line refers to the sum of the regenerative braking energy used for the traction of all trains on the line within the operation time; and the net traction energy of the train on the whole line is equal to the traction energy of the train on the whole line minus the regenerative energy utilization of the train on the whole line.

[0075] The running status of the train on the line is as follows:

[0076] Let i represent the train identification, i = 1, 2, …, I, and I trains, i.e. I train numbers, are running within the operation period; let m represent the station identification, m = 1, 2, …, M, …, 2M, and the single-line two-way subway line with M physical station sites is equivalent to a line with 2M virtual station sites; the adjacent station interval between the m station and the m+1 station is represented by m', m' = 1, 2, …, M, …, 2M-1.

[0077] Please refer to Figure 2 for the curve diagram of the running speed level of the train between adjacent stations, a plurality of curves are usually preset in the ATO driving system, and the running time and the speed level of different station intervals are divided according to the passenger flow in different time periods. In the present application, k represents the identification of the train running speed curve, k = 1, 2, …, K, and K curves constitute the selectable set of the train running speed curve, and each train number can only select one level of speed curve in each station interval. The present application adopts a typical traction-cruise-idling-braking four-stage control strategy to control the train running, i.e. the train is tractioned with the maximum traction force in the starting stage, is converted to the cruise stage when the target speed is reached, i.e. runs at a constant speed to a certain position, is then converted to idling, and finally is braked with the maximum braking force when idling to a certain distance from the station. and respectively are the time of train i leaving and arriving at the adjacent station m+1 from station m; and respectively are the time of train i switching from traction to cruise, from cruise to coasting and from coasting to braking when running on the speed curve of level k in station section m'.

[0078] The train is regarded as a single point, whose movement is in accordance with Newton's law of motion, then the traction force , the braking force and the resistance of train i during running are calculated as follows:

[0079]

[0080]

[0081]

[0082] The resistance of the train during running includes basic running resistance, slope additional resistance and curve additional resistance. The basic resistance is the resistance of the train under any running condition, which is usually calculated by using an empirical formula obtained from a large number of comprehensive experiments; the slope additional resistance is the resistance of the train running on a slope due to the action of the gravity component, which is approximately equal to the slope gradient in thousandths in the present application; the curve additional resistance is the resistance caused by the friction between the wheel and the rail of the train running on a plane curve, which is calculated by using an empirical formula in the present application. The calculation of the basic running resistance, the slope additional resistance and the curve additional resistance is shown in the following formula:

[0083]

[0084]

[0085]

[0086]

[0087] Further, the utilization of regenerative energy of the train on the line is as follows:

[0088] Please refer to Figure 3 for a schematic diagram of energy transmission in the same power supply section, it is assumed that the regenerative braking energy can be transmitted between multiple trains in the same power supply section in both directions, and the trains in different power supply sections cannot transmit regenerative braking energy. α is the efficiency of converting electrical energy into mechanical energy; β is the efficiency of converting mechanical energy into electrical energy; when train i selects a speed running curve of level k in station section m', then is equal to 1, otherwise it is 0; v cis the critical speed of train regenerative braking and air braking, when the train speed is lower than the critical speed, the train uses air braking, and no regenerative braking energy is generated.

[0089] (1) Calculate the traction energy consumed by train i at time t and the regenerative braking energy generated

[0090]

[0091] (2) Calculate the total traction energy consumed by all trains in the power supply section at time t and the total regenerative braking energy generated

[0092]

[0093]

[0094] In the formula, The value of is 0 or 1, when train i is in power supply section q at time t, The value is 1, otherwise 0.

[0095] (3) Calculate the net traction energy E of the train in the study period:

[0096]

[0097] When considering the immediate recycling of regenerative braking energy, the regenerative energy utilization is the smaller value between the energy required for train traction and the regenerative braking energy generated by braking.

[0098] Preferably, the train diagram optimization model is constructed to minimize the expected value of the total net traction energy consumed by all power supply sections under the actual disturbance stay scene, including: obtaining the decision variables of the train diagram optimization model; wherein the decision variables include: the departure interval time of the train and the next adjacent train at the starting station, and the predicted running speed level of the train in each station interval; obtaining the constraint conditions of the train diagram optimization model; wherein the constraint conditions include: the constraint of the departure interval time, the integer constraint of the departure interval time, the constraint of the operation service time of the train, the constraint of the total running time of the train from the starting station to the terminal station, the constraint of the interval time between the train leaving the station and the adjacent next train arriving at the station, and the selection constraint of the predicted running speed level; according to the decision variables and the constraint conditions, the train diagram optimization model is constructed to minimize the expected value of the total net traction energy consumed by all power supply sections under the actual disturbance stay scene.

[0099] Firstly, determine the decision variable of the train diagram optimization model:

[0100] In the embodiment of the present application, the decision variable includes the departure interval time length h of train i and the next adjacent train i+1 at the departure station i , and the 0-1 variable of train i selecting the speed operation curve with the grade k in the station section m' The selection takes the value of 1, otherwise 0. The interval operation time length corresponding to the train operation speed curve is known, and together with the departure interval time length, determines the arrival and departure time of each train at each station.

[0101] Further, taking the minimum expected value of the total net traction energy consumed by all power supply sub-zones under the actual disturbance stay scene as the target, a target function of the train diagram optimization model is constructed:

[0102]

[0103] In the formula, E p is the expected value of the total net traction energy consumed by all power supply sub-zones, i.e. the total net traction energy of the train on the whole line under the actual disturbance stay scene; E w is the total net traction energy of the train on the whole line under the actual disturbance stay scene w.

[0104] Further, the constraint condition of the train diagram optimization model is determined:

[0105] (1) The constraint of the departure interval time length. If the train departure interval time length is too large, it will cause the passenger waiting time to be too long, affecting the passenger satisfaction; if the train departure interval time length is too small, it does not meet the safety requirement, therefore, the constraint of the departure interval time length is:

[0106] h min ≤h i ≤h max , i≠I

[0107] (2) The constraint of the train operation service time length. The whole line operation service time is defined as the time when train 1 departs from station 1 to the time when train I arrives at station 2M, and the time length thereof needs to be within a certain range to ensure the service quality of the subway operation, therefore, the constraint of the train operation service time length is:

[0108]

[0109] T lmin ≤T l ≤T lmax

[0110] (3) The constraint of the total travel time of the train from the starting station to the terminal station. The total travel time of the train from the starting station to the terminal station needs to meet a certain range constraint to ensure that the car bottom turnover and connection relationship are not affected. The total travel time of the train is the time from station 1 to station 2M, and the turnaround time t from M station to M+1 station tb Fixed, therefore, the constraint of the total travel time of the train from the starting station to the terminal station is:

[0111]

[0112] 0≤T i ≤T imax

[0113] (4) The constraint of the interval time between the train leaving the station and the adjacent next train arriving at the station. In the process of train operation, after the current train leaves the station, the next train can enter the station. Therefore, when the stop time is disturbed to cause delay, in order to ensure the safety tracking interval between the front and rear trains, the time when the front train leaves the station and the time when the rear train arrives at the station must meet a certain arrival and departure interval constraint, therefore, the constraint of the interval time between the train leaving the station and the adjacent next train arriving at the station is:

[0114]

[0115] In addition, the train arrival time, departure time and stop time meet the following equation:

[0116]

[0117] The interval running time of train i running in the interval m' according to the kth level speed curve can be expressed as:

[0118]

[0119] (5) The selection constraint of the predicted travel speed level. Assuming that there are K different levels of speed curves available for selection between each station m', each train can only select one level of speed running curve in each station interval, therefore, the selection constraint of the predicted travel speed level is:

[0120]

[0121] (6) The integer constraint of the departure interval time. The research period is divided into several segments with 1s as the step, and the train departure interval time needs to meet the integer constraint, therefore, the integer constraint of the departure interval time is:

[0122] h i ∈N +

[0123] According to the objective function and the constraints, a train diagram optimization model is constructed, and finally a train diagram optimization model for system energy saving is constructed as follows:

[0124]

[0125] S4, solving the train diagram optimization model to obtain a departure interval duration of a train and a next adjacent train at a departure station and a predicted running speed level of the train in each section, and then optimizing the initial train diagram according to the departure interval duration and the predicted running speed level to obtain an optimized train diagram;

[0126] Preferably, the solving of the train diagram optimization model to obtain the departure interval duration of the train and the next adjacent train at the departure station and the predicted running speed level of the train in each section comprises: solving the train diagram optimization model according to an adaptive large neighborhood search algorithm to generate an initial solution of the departure interval duration of the train and the next adjacent train at the departure station and the predicted running speed level of the train in each section; obtaining initial parameters of the adaptive large neighborhood search algorithm, wherein the initial parameters comprise: rules of a destruction operator, rules of a repair operator, scores of operators, weights of operators, an initial temperature of simulated annealing, a termination temperature of simulated annealing, a cooling speed of simulated annealing, an iteration number of the adaptive large neighborhood search algorithm, and an iteration number of the simulated annealing algorithm; assigning the initial solution to a current solution and an optimal solution of the train diagram optimization model, and then optimizing the current solution according to the initial parameters to obtain the optimal solution of the train diagram optimization model, and taking the optimal solution as the departure interval duration of the train and the next adjacent train at the departure station and the predicted running speed level of the train in each section.

[0127] Preferably, the optimization of the current solution according to the initial parameters to obtain the optimal solution of the train diagram optimization model comprises: performing an optimization operation on the current solution according to the initial parameters until a preset termination condition is met to obtain the optimal solution of the train diagram optimization model; wherein the optimization operation comprises: selecting a corresponding operator as a disturbance operator using a roulette method according to the weights of the operators, and then disturbing the current solution of the train diagram optimization model according to the disturbance operator to obtain a new solution of the train diagram optimization model; judging the superiority-inferiority relationship of the new solution and the current solution and the optimal solution, and updating the current solution, the optimal solution, and the operator scores according to the superiority-inferiority relationship; when it is detected that the updated operator scores reach a preset weight update condition, updating the operator weights and resetting the operator scores; updating the initial temperature and the iteration number of simulated annealing according to the new solution.

[0128] Please refer toFigure 4 The flowchart for solving the train diagram optimization model is shown in Figure 1. The specific steps for solving the train diagram optimization model according to the adaptive large neighborhood search algorithm are as follows:

[0129] S401, generate an initial feasible solution of the train and the next adjacent train departure interval at the starting station, and the predicted running speed level of the train in each station interval, set the initial parameters, and assign the initial solution to the current solution and the optimal solution.

[0130] The feasible solution of the train diagram optimization model is composed of the train and the next adjacent train departure interval at the starting station, and the predicted running speed level of the train in each station interval. Wherein, the departure interval time set is represented as H, and the predicted running speed level set is represented as X, then any feasible solution can be represented as φ = {H, X}, and an initial feasible solution φ' satisfying the constraint condition is randomly generated.

[0131] The initial parameters of the adaptive large neighborhood search algorithm include: the rules of the destruction operator, the rules of the repair operator, the score of the operator, the weight of the operator, the initial temperature of simulated annealing, the termination temperature of simulated annealing, the cooling speed of simulated annealing, the iteration number of adaptive large neighborhood search algorithm and the iteration number of simulated annealing algorithm.

[0132] Calculate the objective function f(φ') of the initial solution, and assign the initial solution to the current solution and the optimal solution, i.e. φ best ←φ←φ',f(φ) best ←f(φ)←f(φ')。

[0133] S402, according to the initial parameters, execute optimization operation on the current solution until the preset termination condition is met, wherein the optimization operation includes:

[0134] (1) according to the weight of the operator, using roulette method to select the corresponding operator as the disturbance operator, and then according to the disturbance operator to disturb the current solution of the train diagram optimization model, to get the new solution of the train diagram optimization model;

[0135] (2) judge the pros and cons of φ' and the current solution φ, the optimal solution φ best , update the current solution, the optimal solution and the operator score;

[0136] (3) when the updated operator score reaches the preset weight update condition, update the operator weight and reset the operator score;

[0137] (4) according to the new solution, update the initial temperature and iteration number of simulated annealing;

[0138] S403, obtaining the optimal solution of the train diagram optimization model: the algorithm terminates, and the optimal solution found is returned.

[0139] Further, three types of customized damage repair operator combinations are proposed in the embodiments of the present application, including damage operators, repair operators and adjustment operators, and the rules are as follows:

[0140] The damage operator removes a train from the train diagram, and two damage rules are set:

[0141] 1) Randomly remove the operation of a train in the train diagram;

[0142] 2) Remove the operation of the train before the minimum headway position;

[0143] The repair operator adds a train in the damaged train diagram, and two repair rules are set:

[0144] 1) Add the operation of a train at a random time point;

[0145] 2) Add the operation of a train in the maximum headway;

[0146] The adjustment operator adjusts the train curve selection, and two adjustment rules are set:

[0147] 1) Randomly select l elements in the feasible solution, and respectively replace the current speed curve of the element with one of the remaining (K-1) running speed curves;

[0148] 2) Select a train that departs later than t o , and adjust the running speed curve level of the train to the curve with speed level K.

[0149] The six operations of the above three types of operators must satisfy the constraint conditions. One is selected from each type of operator, and there are 8 operator combinations. The combination needs to perform damage, repair and adjustment operations.

[0150] Further, the operator score and weight updating method in the embodiments of the present application is as follows:

[0151] Each operator combination has a corresponding score π i and weight w i . The initial score of all operators is 0, and the weight is 1. If the feasible solution is better than the optimal solution, the score χ1 of the operator combination is increased; if the feasible solution is better than the current solution, the score χ2 of the operator combination is increased; if the feasible solution is worse than the current solution, but the new feasible solution is accepted with a certain probability, the score χ3 of the operator combination is increased.

[0152] Each time the algorithm iterates Secondly, the weight of the operator combination is updated, and the score of the operator combination is cleared after the weight is updated, and the weight is updated as shown in the following formula:

[0153]

[0154] In the formula, r is a weight influence factor, is the number of times the operator combination is selected in the first iteration.

[0155] Further, in the embodiment, the new solution is accepted based on the simulated annealing idea to avoid falling into local optimum, and the acceptance criterion is:

[0156] 1) If the new solution φ' after the destruction and repair is better than the current solution, the solution is accepted and the current solution is given, φ best ←φ;

[0157] 2) If the new solution φ' after the destruction and repair is worse than the current solution, it has a certain probability of being accepted, and the probability function is as follows: the selection probability becomes smaller as the simulated annealing temperature T decreases.

[0158]

[0159] 3) If the updated current solution is better than the optimal solution, the current solution is given to the optimal solution, φ ← φ'.

[0160] The train diagram optimization method of the present application is further described below with a specific embodiment:

[0161] Please refer to Figure 5 , the train diagram optimization method is simulated and analyzed based on the data of Guangzhou Metro Line 2 (Guangzhou South Station to Changgang Station), and the optimized train diagram scheme is obtained. The simulation period is set to 7:30 to 9:00, and the number of trains in the uplink and downlink directions is 16 in this period.

[0162] Guangzhou Metro Line 2 starts from Guangzhou South Station and ends at Jiahe Wanguang, with a total length of about 31.8 km, and has 24 stations, all of which are underground stations. Guangzhou South Station to Changgang Station is selected for optimization research. The research section is 13.2 km long, of which Guangzhou South Station to Huijiang Station belongs to the same power supply partition, Huijiang Station to Luoxi Station belongs to the same power supply partition, Luoxi Station to Dongxiaonanlu Station belongs to the same power supply partition, and Dongxiaonanlu Station to Changgang Station belongs to the same power supply partition. Guangzhou Metro Line 2 trains are all 6-car A-type trains with a length of about 140 m and a train mass of 249 t. According to the historical delay data of Line 2, the actual disturbance stay scene is generated, and the initial timetable is optimized. The related parameters of the adaptive large neighborhood search algorithm are set as follows: the initial temperature is set to 100000, the termination temperature is set to 0.01, and the cooling speed is set to .​

[0163] According to the method of S4, the optimized train operation diagram is obtained, the total running time of the optimized train is not changed, and the stop time is in a reasonable range of [30s, 55s]. Compared with the initial operation diagram, the net traction energy consumption of the optimized operation diagram is reduced by about 4% on the whole line.

[0164] As can be seen, the present application provides a train operation diagram optimization method, the actual disturbance stay scene of the train at each station is constructed by the present application, the optimization of the train operation diagram is carried out under the actual disturbance stay scene, so that the train operation diagram obtained after the final optimization is more in line with the actual operation situation, and the optimization effect is better. In addition, when calculating the energy consumption of the train, the traction energy consumed and the braking energy generated by each train at each time during the running process are comprehensively considered, the minimum expected value of the total net traction energy consumed by all power supply partitions under the actual disturbance stay scene is taken as the target, the train operation diagram optimization model is constructed, the recycling of the renewable braking energy generated by the train during braking is considered under the condition of fitting the actual operation, the optimization effect of the train operation diagram is effectively improved, and the energy consumption of urban rail transit can be reduced to the maximum extent according to the optimized train operation diagram.

[0165] Embodiment two

[0166] Please refer to Figure 6 A structure schematic diagram of a train operation diagram optimization device provided by an embodiment of the present application, the device comprises: an initial operation diagram acquisition module, an actual disturbance stay scene generation module, a train operation diagram optimization model construction module, an initial operation diagram optimization module and a net traction energy calculation module.

[0167] The initial operation diagram acquisition module is used for acquiring the initial operation diagram of the train, wherein the initial operation diagram comprises: the planned stay duration of each train at each station.

[0168] The actual disturbance stay scene generation module is used for acquiring the historical actual stay duration of the train at each station for each train, and generating the actual disturbance stay scene of the train at each station according to the historical actual stay duration and the planned stay duration.

[0169] The train operation diagram optimization model construction module is used for constructing the train operation diagram optimization model with the minimum expected value of the total net traction energy consumed by all power supply partitions under the actual disturbance stay scene as the target.

[0170] The initial operation diagram optimization module is configured to solve the train operation diagram optimization model to obtain a departure interval duration of a train and a next adjacent train at a departure station and a predicted running speed level of the train in each section, and then optimize the initial operation diagram according to the departure interval duration and the predicted running speed level to obtain an optimized train operation diagram.

[0171] The net traction energy calculation module is configured to calculate, for each power supply section, the net traction energy consumed by the power supply section in the actual disturbance stay scenario in the following manner: calculate the traction energy consumed and the braking energy generated by the train at each moment in the running process according to the running speed level of the train between adjacent stations; calculate the total traction energy consumed and the total braking energy generated by the power supply section at each moment according to the traction energy consumed and the braking energy generated by the train at each moment in the running process, and then calculate the net traction energy consumed by the power supply section at each moment according to the total traction energy and the total braking energy; wherein the power supply section corresponds to a plurality of stations.

[0172] Embodiment three

[0173] Correspondingly, the embodiment of the present application provides a train operation diagram optimization device, which comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the train operation diagram optimization method of the above-mentioned embodiment of the present application when executing the computer program.

[0174] Embodiment four

[0175] Correspondingly, the embodiment of the present application provides a storage medium, which comprises a stored computer program, wherein the computer program controls the device where the computer readable storage medium is located to execute the train operation diagram optimization method of the above-mentioned embodiment of the present application when running.

[0176] In summary, the present application provides a train operation diagram optimization device, equipment and storage medium, constructs an actual disturbance stay scenario of the train at each station, optimizes the train operation diagram in the actual disturbance stay scenario, so that the finally optimized train operation diagram is more in line with the actual operation situation and has better optimization effect. In addition, when calculating the energy consumption of the train, the traction energy consumed and the braking energy generated by each train in the running process at each moment are comprehensively considered, the minimum expected value of the total net traction energy consumed by all power supply sections in the actual disturbance stay scenario is taken as the target, the train operation diagram optimization model is constructed, the recycling of the renewable braking energy generated by the train during braking is considered in the actual operation situation, the optimization effect of the train operation diagram is effectively improved, and the energy consumption of urban rail transit can be maximally reduced according to the optimized train operation diagram.

[0177] It should be noted that the apparatus embodiments described above are merely illustrative, and the units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs. In addition, the connection relationship between the modules in the apparatus embodiment provided by the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0178] Those skilled in the art can clearly understand that, for the convenience and brevity, the specific working process of the apparatus described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0179] The device can be a desktop computer, a notebook, a palm computer, and a cloud server, etc. The device can include, but is not limited to, a processor, a memory.

[0180] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The processor is the control center of the device, and connects various parts of the device through various interfaces and lines.

[0181] The memory can be used to store the computer program, and the processor realizes various functions of the device by running or executing the computer program stored in the memory and calling data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function, and the like; and the data storage area can store data created according to the use of the mobile phone and the like. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0182] The storage medium is a computer readable storage medium, and the computer program is stored in the computer readable storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be realized. The computer program includes computer program code, which can be in the form of source code, object code, an executable file, or some intermediate form, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, a recording medium, a U disk, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0183] The above is the preferred embodiment of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which are also considered within the scope of protection of the present application.

Claims

1. A method for optimizing a train diagram, characterized in that, The method comprises the following steps: obtaining an initial train diagram, wherein the initial train diagram comprises a planned stay duration of each train at each station; for each train, obtaining a preset stay duration interval, and randomly extracting a stay duration from the stay duration interval as a perturbation stay duration; obtaining a predicted stay duration of the train at each station according to the perturbation stay duration and the planned stay duration; obtaining a historical actual stay duration of the train at each station, and calculating an occurrence probability of the predicted stay duration according to the historical actual stay duration; generating an actual perturbation stay scenario of the train at each station according to the predicted stay duration of the train at each station and the occurrence probability of the predicted stay duration; constructing a train diagram optimization model with the objective of minimizing an expected value of total net traction energy consumed by all power supply partitions under the actual perturbation stay scenario; solving the train diagram optimization model to obtain a departure interval duration of the train and a next adjacent train at a departure station, and a predicted running speed level of the train in each interval, and then optimizing the initial train diagram according to the departure interval duration and the predicted running speed level to obtain an optimized train diagram; wherein, for each power supply partition, the net traction energy consumed by the power supply partition under the actual perturbation stay scenario is calculated in the following manner: calculating the traction energy consumed and the braking energy generated by the train at each time during running according to the running speed level of the train between adjacent stations; calculating the total traction energy consumed and the total braking energy generated by the power supply partition at each time according to the traction energy consumed and the braking energy generated by the train at each time during running, and then calculating the net traction energy consumed by the power supply partition at each time according to the total traction energy and the total braking energy; wherein the power supply partition corresponds to a plurality of stations.

2. The train diagram optimization method of claim 1, wherein, The calculation of the traction energy consumed and the braking energy generated by the train at each time during running according to the running speed level of the train between adjacent stations comprises: calculating the resistance of the train at each time during running according to the running speed level of the train between adjacent stations; wherein the resistance comprises a basic resistance of the train, an additional resistance of a slope and an additional resistance of a curve; calculating the traction force and the braking force of the train at each time during running according to the running speed level and the resistance; calculating the traction energy consumed by the train at each time during running according to the traction force, and calculating the braking energy generated by the train at each time during running according to the braking force.

3. The train diagram optimization method of claim 1, wherein, The construction of the train diagram optimization model with the objective of minimizing the expected value of the total net traction energy consumed by all power supply partitions under the actual perturbation stay scenario comprises: obtaining decision variables of the train diagram optimization model; wherein the decision variables comprise a departure interval duration of the train and a next adjacent train at a departure station, and a predicted running speed level of the train in each interval; Obtaining constraint conditions of a train diagram optimization model; wherein the constraint conditions include: a constraint of the departure interval duration, an integer constraint of the departure interval duration, a constraint of an operation service duration of the train, a constraint of a total travel duration of the train from a starting station to a terminal station, a constraint of an interval duration between a departure of the train from a station and an arrival of a next train at the station, and a selection constraint of the predicted travel speed level; According to the decision variables and the constraint conditions, a train diagram optimization model is constructed, with an objective of minimizing an expected value of total net traction energy consumed by all power supply partitions under the actual disturbance stay scenario.

4. The train diagram optimization method of claim 1, wherein, Solving the train diagram optimization model to obtain the departure interval duration of the train and the next train at the starting station, and the predicted travel speed level of the train in each section, including: Solving the train diagram optimization model according to the adaptive large neighborhood search algorithm to generate the departure interval duration of the train and the next train at the starting station, and the predicted travel speed level of the train in each section. Obtaining initial parameters of the adaptive large neighborhood search algorithm, wherein the initial parameters include: rules of a destruction operator, rules of a repair operator, scores of operators, weights of operators, an initial temperature of simulated annealing, a termination temperature of simulated annealing, a cooling speed of simulated annealing, an iteration number of the adaptive large neighborhood search algorithm, and an iteration number of the simulated annealing algorithm. Assigning the initial solution to a current solution and an optimal solution of the train diagram optimization model, and then optimizing the current solution according to the initial parameters to obtain the optimal solution of the train diagram optimization model, and taking the optimal solution as the departure interval duration of the train and the next train at the starting station, and the predicted travel speed level of the train in each section.

5. The method for optimizing a train working diagram according to claim 4, characterized in that, The optimization of the current solution according to the initial parameters to obtain the optimal solution of the train diagram optimization model includes: Performing an optimization operation on the current solution according to the initial parameters until a preset termination condition is met to obtain the optimal solution of the train diagram optimization model; The optimization operation includes: According to the weights of the operators, using a roulette method to select a corresponding operator as a disturbance operator, and then disturbing the current solution of the train diagram optimization model according to the disturbance operator to obtain a new solution of the train diagram optimization model; Judging the superiority-inferiority relationship between the new solution and the current solution and the optimal solution, and updating the current solution, the optimal solution, and the operator scores according to the superiority-inferiority relationship; When it is detected that the updated operator scores meet a preset weight update condition, updating the operator weights and resetting the operator scores; Updating the initial temperature and the iteration number of simulated annealing according to the new solution.

6. An apparatus for optimizing a train working diagram, characterized by comprising: It includes: An initial train diagram obtaining module, an actual disturbance stay scenario generating module, a train diagram optimization model constructing module, an initial train diagram optimization module, and a net traction energy calculating module; The initial train diagram obtaining module is configured to obtain an initial train diagram of a train; wherein the initial train diagram includes a planned stay duration of each train at each station. The actual disturbance stay scene generation module is configured to, for each train, obtain a preset stay duration interval, randomly extract a stay duration from the stay duration interval as a disturbance stay duration, obtain a predicted stay duration of the train at each station according to the disturbance stay duration and the planned stay duration, obtain a historical actual stay duration of the train at each station, calculate an occurrence probability of the predicted stay duration according to the historical actual stay duration, and generate an actual disturbance stay scene of the train at each station according to the predicted stay duration of the train at each station and the occurrence probability of the predicted stay duration. The train working diagram optimization model construction module is configured to construct a train working diagram optimization model with a minimum expected value of total net traction energy consumed by all power supply partitions under the actual disturbance stay scene as a target. The initial working diagram optimization module is configured to solve the train working diagram optimization model to obtain a departure interval duration of the train and a next adjacent train at a departure station and a predicted running speed level of the train at each section, and then optimize the initial working diagram according to the departure interval duration and the predicted running speed level to obtain an optimized train working diagram. The net traction energy calculation module is configured to, for each power supply partition, calculate the net traction energy consumed by the power supply partition under the actual disturbance stay scene by the following manner: calculating traction energy consumed and braking energy generated by the train at each moment during running according to a running speed level of the train between adjacent stations, calculating total traction energy consumed and total braking energy generated by the power supply partition at each moment according to the traction energy consumed and the braking energy generated by the train at each moment during running, and then calculating the net traction energy consumed by the power supply partition at each moment according to the total traction energy and the total braking energy. The power supply partition corresponds to a plurality of stations.

7. A train diagram optimization device characterized by comprising: The storage medium includes a stored computer program, wherein the computer program controls a device where the computer readable storage medium is located to execute the optimization method of the train working diagram according to any one of claims 1 to 5 when the computer program is running.

8. A storage medium, characterized by The storage medium includes a stored computer program, wherein the computer program controls a device where the computer readable storage medium is located to execute the optimization method of the train working diagram according to any one of claims 1 to 5 when the computer program is running.

Images