Train inspection and shipment integrated preparation method and device

Abstract

The present application relates to the field of rail vehicle technology, and discloses a train inspection and transportation integrated scheduling method and device, the method comprising: generating a feasible path set of each train in a planning time period according to a preset constraint condition and an inspection and transportation task in the planning time period; generating an initial solution of a decision variable related to a feasible path of a train, a transportation task and an inspection task performed on each planning day according to the feasible path set of each train; calculating optimal decision variables of each train in the planning time period to satisfy a preset objective function according to the initial solution of the decision variable of each train, thereby obtaining an optimal inspection and transportation scheduling scheme of each train. Compared with the existing manual scheduling of train inspection and transportation plans, the present application can formulate an optimal inspection and transportation scheduling scheme of the train according to the preset constraint condition and the inspection and transportation tasks of the train in the future period of time, avoid possible errors caused by manual operation, and improve work efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail vehicle technology, in particular to a train inspection and operation integrated scheduling method and device. BACKGROUND

[0002] With the increasingly fierce competition in the urban rail market, the information system carrying the functions of train operation and maintenance has become an important indicator to measure the competitiveness. The train maintenance plan and the transportation plan are interrelated and influence each other. The comprehensive optimization of the maintenance plan and the transportation plan to improve the scheduling quality is a necessary measure to meet the needs of the owner and is of great significance to improve the accuracy of the plan and the intelligent maintenance. However, the current train operation plan and maintenance plan are mostly manually scheduled by dispatchers. The manual scheduling method has low correlation with the transportation plan and cannot consider various influencing factors, resulting in low train utilization and low maintenance efficiency. Moreover, with the increase in the number of trains, the manual scheduling method itself has low efficiency and cannot meet the actual production needs. When the number of trains increases and the number of maintenance operation packages increases, the operation cost of the enterprise will increase and the economic benefit will decrease. SUMMARY

[0003] The present application provides a train inspection and operation integrated scheduling method and device to solve the problem of low correlation between the maintenance plan and the transportation plan in the manual scheduling method of the prior art, which cannot consider various influencing factors and results in low train utilization and low maintenance efficiency.

[0004] The present application provides a train inspection and operation integrated scheduling method, comprising:

[0005] According to the preset constraint condition and the inspection and operation task in the planning time period, a feasible path set of each train in the planning time period is generated, wherein the inspection and operation task includes a transportation task set and a maintenance task set;

[0006] According to the feasible path set of each train, an initial solution of decision variables related to the feasible path of the train, the transportation task and the maintenance task performed in each planning day is generated;

[0007] According to the initial solution of the decision variables of each train, the optimal decision variables of each train in the planning time period that satisfy the preset objective function are calculated, so as to obtain the optimal inspection and operation scheduling scheme of each train.

[0008] According to the train inspection and operation integrated scheduling method provided by the present application, the feasible path set of each train in the planning time period is generated according to the preset constraint condition and the inspection and operation task in the planning time period, which comprises:

[0009] determining, according to the transport task set and the repair task set, a feasible path set of each train in the planning time period.

[0010] determining, according to the transport task set and the repair task set, a feasible path set of each train in the planning time period.

[0011] According to the train repair and transport integrated scheduling method provided by the application, the constraint conditions include:

[0012]

[0013]

[0014]

[0015] ∑ r∈R y gdr ≤1 (4)

[0016] ∑ g∈G ∑ b∈B x gdb ≤N (5)

[0017]

[0018] wherein, l gb denotes the cumulative running mileage between the time when the train g completes the repair work package b for the last time and the time when the train g performs the repair work package b again, l r denotes the mileage of the train performing the transport task r, denotes the upper limit of the repair cycle of the repair work package b with mileage as the repair cycle, t gb denotes the cumulative running time between the time when the train g completes the repair work package b for the last time and the time when the train g performs the repair work package b again, t r denotes the running time of the train performing the transport task r, denotes the upper limit of the repair cycle of the repair work package b with time as the repair cycle, formula (1) and formula (2) indicate that the repair work should be performed within the specified time cycle range and the mileage cycle range, G denotes the train set, R denotes the transport task set, B denotes the repair work package set, denotes the decision variable of the feasible path, and is 1 when the train g performs the feasible path p g , formula (3) indicates that in the feasible path set P g of the train g, at most one feasible path p g can be performed, y gdr is the decision variable of the transport task, and is 1 when the train g performs the transport task r on the planning day d, formula (4) indicates that at most one transport task can be performed by one train in one planning day, x gdbdecision variable representing the maintenance task, 1 if train g performs maintenance work package b on plan day d, formula (5) represents that the maintenance workload of maintenance point per day cannot exceed the upper limit of maintenance capacity N, in formula (6), Set d s1 represents the set of transportation tasks that need to be performed when plan day d is a holiday, and s2 represents the set of transportation tasks that need to be performed when plan day d is a weekday.

[0019] According to the train inspection and transportation integrated scheduling method provided by the application, the initial solution of the decision variable related to the feasible path of the train, the transportation task and the maintenance task performed on each plan day is generated according to the feasible path set of each train, which comprises the following steps:

[0020] For each train, a feasible path is selected from the feasible path set of the train, and the decision variable of the selected feasible path is determined as 1 under the condition that all transportation tasks in the selected feasible path do not appear in the feasible path selected by other trains.

[0021] In the feasible path with the decision variable of 1, the decision variable corresponding to the maintenance task and the decision variable corresponding to the transportation task are determined as 1 according to the maintenance task and the transportation task.

[0022] According to the train inspection and transportation integrated scheduling method provided by the application, the preset objective function is:

[0023] minZ=w1β1Z1+w2β2Z2+w3β3Z3+a1W1+a2W2

[0024]

[0025]

[0026]

[0027] Wherein, Z1, Z2 and Z3 represent the train maintenance cost index, the number of train online operation index and the train operation balance index respectively, w1, w2 and w3 represent the weight value of the three indexes, β1, β2 and β3 represent the normalization processing index of the train maintenance cost index, the number of train online operation index and the train operation balance index respectively, a1 represents the number of transportation tasks not included, W1 represents the penalty value of the transportation task not included in the scheduling scheme, a2 represents the number of maintenance tasks exceeding the maintenance capacity of the depot in the scheduling scheme, W2 represents the penalty value of the total number of maintenance tasks exceeding the maintenance capacity of the depot in a single plan day in the scheduling scheme, T b C represents the maintenance time required by maintenance work package b; t max represents the labor cost per unit time required for maintenance work, and gd and mingd Let x represent the maximum and minimum train mileage within the planned time period, respectively. gdb Let represent the decision variable for the maintenance task, and let be the value 1 when train g performs maintenance work package b on planned day d. The decision variable represents a feasible path, when train g follows feasible path p. g Take 1 at the time.

[0028] According to a train inspection and operation integrated scheduling method provided by the present invention, the step of calculating the optimal decision variables of each train satisfying a preset objective function within the planned time period based on the initial solutions of the decision variables of each train, thereby obtaining the optimal inspection and operation scheduling scheme for each train, includes:

[0029] Calculate multiple neighborhood solutions based on the initial solution of the decision variable, so that the total number of the initial solution and neighborhood solutions reaches the number of individuals in the initial population of the Grey Wolf Optimization Algorithm.

[0030] The initial solution of the decision variable and the initial scheduling scheme corresponding to multiple neighborhood solutions are used as individual inputs to the Grey Wolf optimization algorithm;

[0031] The optimal maintenance and scheduling scheme for each train is obtained iteratively using the Grey Wolf optimization algorithm based on the preset objective function.

[0032] According to the integrated train inspection and operation planning method provided by the present invention, in the iterative process of the Grey Wolf optimization algorithm, the control parameter k is a nonlinear control parameter with gen and MAXGEN as independent variables, and the closer gen is to MAXGEN, the greater the deceleration rate of the control parameter k. Here, gen and MAXGEN represent the current iteration number and the total iteration number, respectively.

[0033] The present invention also provides an integrated train inspection and transportation organization device, comprising:

[0034] The feasible path set generation module is used to generate a set of feasible paths for each train within a planned time period based on preset constraints and maintenance tasks within the planned time period. The maintenance tasks include: a set of transportation tasks and a set of maintenance tasks.

[0035] The initial solution generation module is used to generate initial solutions for decision variables related to the feasible paths of each train, the transportation tasks to be performed on each planned day, and the maintenance tasks, based on the set of feasible paths for each train.

[0036] The optimal solution calculation module is used to calculate the optimal decision variables of each train that satisfy the preset objective function within the planned time period based on the initial solutions of the decision variables of each train, thereby obtaining the optimal maintenance and operation scheduling scheme for each train.

[0037] According to the train inspection and transportation integrated scheduling device provided by the application, the feasible path set generating module comprises:

[0038] An executable transportation task determining module is configured to determine a set of executable transportation tasks of each train on each planning day to meet the constraint condition;

[0039] A feasible path determining module is configured to determine a set of feasible paths of each train in the planning time period according to the set of executable transportation tasks and the set of inspection tasks.

[0040] According to the train inspection and transportation integrated scheduling device provided by the application, the initial solution generating module comprises:

[0041] A first decision variable determining module is configured to determine the decision variable of a selected feasible path as 1 when the selected feasible path is selected from the set of feasible paths of each train and all the transportation tasks in the selected feasible path do not appear in the selected feasible paths of other trains;

[0042] A second decision variable determining module is configured to determine the decision variable corresponding to the inspection task and the decision variable corresponding to the transportation task as 1 according to the inspection task and the transportation task in the feasible path with the decision variable being 1.

[0043] Compared with the existing manual train inspection and transportation scheduling mode, the train inspection and transportation integrated scheduling method and device can formulate an optimal train inspection and transportation scheduling scheme (i.e., an optimal train inspection and transportation scheduling scheme) according to the preset constraint condition and the inspection and transportation tasks of the train in the future period of time, avoid possible errors caused by manual operation, and improve the work efficiency. Moreover, the inspection and transportation tasks are comprehensively considered, so that the whole planning process is more reasonable and convenient, the formulated inspection and transportation scheduling scheme can effectively improve the operation efficiency of the train, reduce the number of trains in use, reduce the inspection cost, and improve the inspection efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0045] Figure 1 is a flowchart of the train inspection and transportation integrated scheduling method provided by the application;

[0046] Figure 2is a flowchart of generating a feasible path set in the train inspection and operation integrated scheduling method provided by the present application.

[0047] Figure 3 is a flowchart of generating an initial solution of decision variables in the train inspection and operation integrated scheduling method provided by the present application.

[0048] Figure 4 is a flowchart of calculating an optimal scheduling scheme in the train inspection and operation integrated scheduling method provided by the present application.

[0049] Figure 5 is a structural schematic diagram of an electronic device provided by the present application. DETAILED DESCRIPTION

[0050] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described below in connection with the drawings in the present application. Obviously, the described embodiments are only some 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 of ordinary skill in the art without any creative work fall within the protection scope of the present application.

[0051] The train inspection and operation integrated scheduling method of the present embodiment, as shown in Figure 1 , includes:

[0052] In step S110, a feasible path set of each train in the planning time period is generated according to the preset constraint condition and the inspection and operation tasks in the planning time period. The feasible path set includes the transport tasks or inspection tasks that can be executed by the train in each planning day. The inspection and operation tasks include a transport task set and an inspection task set. The transport task set refers to the transport task set of a single planning day (i.e., the circuit task to be executed in a single planning day), which may differ according to the type of planning day (holiday, weekday). The inspection task set refers to the inspection task type and the corresponding inspection period of a line, as well as the inspection time of the corresponding inspection task, etc. The daily inspection workload of the inspection point in the inspection task set cannot exceed the upper limit of the inspection capacity.

[0053] Step S120, generating an initial solution of decision variables related to the feasible paths of the trains, the transportation tasks and the maintenance tasks performed on each planning day according to the feasible path set of each train. The decision variables are used to represent whether the feasible paths of the trains, the transportation tasks and the maintenance tasks performed on each planning day are selected to be executed. If the decision variable is 1, it means that the corresponding feasible path, the transportation task and the maintenance task performed on each planning day are selected to be executed. For all trains, the selected feasible paths, the transportation tasks and the maintenance tasks performed on each planning day form a set of inspection and transportation scheduling scheme, and therefore a plurality of decision variables with a value of 1 can be used to represent the inspection and transportation scheduling scheme.

[0054] Step S130, calculating the optimal decision variables of each train in the planning time period to meet the preset objective function, thereby obtaining the optimal inspection and transportation scheduling scheme of each train according to the initial solution of the decision variables of each train. The preset objective function takes the train maintenance cost, the number of train online operation and the train operation balance as the target. A better inspection and transportation scheduling scheme requires relatively low train maintenance cost, relatively small number of train online operation and good train operation balance. The train operation balance is the uniformity of the running mileage of each train in the planning time period, which can be judged by the extreme value of the running mileage of each train in the planning time period.

[0055] Compared with the existing manual preparation of train maintenance plan and transportation plan, the train inspection and transportation integrated preparation method of the embodiment can formulate the optimal inspection and transportation scheduling scheme (i.e. the optimal inspection and transportation plan scheme) of the train according to the preset constraint condition and the future maintenance tasks and transportation tasks of the train for a period of time, avoid possible errors caused by manual operation, and improve the work efficiency. Moreover, the invention comprehensively considers the double factors of maintenance tasks and transportation tasks, so that the whole inspection and transportation plan preparation process is more reasonable and convenient, and the prepared inspection and transportation plan can effectively improve the operation efficiency of the train, reduce the number of train use, reduce the maintenance cost and improve the maintenance efficiency.

[0056] In the embodiment, step S110 includes:

[0057] According to the transportation task set, the executable transportation task set of each train in each planning day that meets the constraint condition is determined, that is, the transportation tasks meeting the constraint condition are selected from the transportation task set to form the executable transportation task set.

[0058] According to the executable transportation task set and the maintenance task set, the feasible path set of each train in the planning time period is determined.

[0059] The constraint condition includes:

[0060]

[0061]

[0062]

[0063] ∑ r∈R y gdr ≤1 (4)

[0064] ∑ g∈G ∑ b∈B x gdb ≤N (5)

[0065]

[0066] wherein, l gb denotes the cumulative running distance between the time when the train g finishes the maintenance work package b last time and the time when the train g performs the maintenance work package b again, l r denotes the running distance of the train g performing the transportation task r, denotes the upper limit of the maintenance cycle of the maintenance work package b with the running distance as the maintenance cycle, t gb denotes the cumulative running time between the time when the train g finishes the maintenance work package b last time and the time when the train g performs the maintenance work package b again, t r denotes the running time of the train g performing the transportation task r, denotes the upper limit of the maintenance cycle of the maintenance work package b with the time as the maintenance cycle, the equations (1) and (2) indicate that the maintenance work should be performed within the specified time cycle range and the running distance cycle range, G denotes the train set, R denotes the transportation task set, B denotes the maintenance work package set, is the decision variable of the feasible path, and is 1 when the train g performs the feasible path p g , the equation (3) indicates that the train g can perform at most one feasible path p g in the feasible path set P g of the train g, y gdr is the decision variable of the transportation task, and is 1 when the train g performs the transportation task r on the planning day d, the equation (4) indicates that a train can perform at most one transportation task in a planning day, x gdb denotes the decision variable of the maintenance task, and is 1 when the train g performs the maintenance work package b on the planning day d, the equation (5) indicates that the maintenance work amount of the maintenance point per day cannot exceed the upper limit of the maintenance capacity N, in the equation (6), Set d denotes the transportation task set that needs to be performed when the planning day d is a holiday, and denotes the transportation task set that needs to be performed when the planning day d is a working day. The feasible path set of the train obtained under the above various constraint conditions is more actual transportation situation and scheduling requirement.

[0067] Specifically, in step S110, the inputs include: the time and mileage of the train's last maintenance work package, the train set G, the planned time period set D, the maintenance work package set B, the transportation task set R, and the aforementioned constraints. Where G = {g | g = g1, g2, ..., g} i} represents the set of trains, where g is the train number; D = {d | d = d1, d2, ..., d} i} represents the set of planned time periods, where d is the planned day number; B = {b | b = b1, b2, ..., b} i} represents the set of maintenance work packages, and b represents the category number of the maintenance work package.

[0068] Algorithm output: The set of feasible routes for each train within the planned time period.

[0069] Specific steps are as follows: Figure 2 As shown, it includes:

[0070] Step 1: Initialization, set the current planned day d = 1, and create a set of feasible paths P. g and set the current path p g Empty.

[0071] Step 2: Solve for feasible transportation tasks on the planned day d. Determine the set of transportation tasks for the day based on whether the current planned day d is a holiday or a workday. Then, for each train g, iterate through the set of transportation tasks and determine which tasks train g can perform on that day. The basis for the determination is whether the following conditions are met: (1) l r l represents the mileage of the train performing the transportation task r. gb This represents the cumulative mileage traveled by train g between the completion of maintenance work package b and the commencement of another maintenance work package b. This indicates the upper limit of the maintenance cycle for maintenance operation b, which uses mileage as the maintenance cycle; (2) t r t represents the travel time of the train performing the transportation task r. gb This represents the cumulative travel time from the last completion of maintenance work package b by train g to the start of the next maintenance work package b. This indicates the upper limit of the maintenance cycle for maintenance operation b, using time as the maintenance cycle. That is, it cannot exceed the upper limit of the maintenance cycle for any maintenance operation package, whether the maintenance cycle is based on running time or mileage. Finally, the set of executable transportation tasks for the train on day d is obtained, proceed to Step 4; if for train g∈G, all transportation tasks on planned day d cannot be executed, it means that one or more maintenance tasks of the train have reached their maintenance cycle and require maintenance, then proceed to Step 3.

[0072] wherein the set of transport tasks is different on weekdays and on holidays.

[0073]

[0074] Set d s1 if the planning day d is a holiday, and s2 if the planning day d is a weekday.

[0075] Step 3: Solve the set of maintenance tasks to be performed on the planning day d. Traverse the set of maintenance work packages of the train to determine the set of maintenance tasks that can be performed by the train on the planning day, and go to Step 4.

[0076] Step 4: Expand the current feasible path. If the train has multiple feasible transport tasks on the planning day d, but the train can only perform one transport task on a single planning day, expand the current path into the same number of feasible paths as the number of executable tasks; if the train cannot perform a transport task on the planning day d, perform a maintenance task, add the maintenance task to the end of the current feasible path, and add the expanded current feasible path p g to the set of feasible paths P g .

[0077] Step 5: Determine whether the current planning day d is the last day of the planning time period D, if so, go to Step 6; otherwise, d is incremented by 1, and go to Step 2.

[0078] Step 6: Output the set of feasible paths P g of the train.

[0079] For example, there are five planned days in the planned time period set D, for train g, according to the transport task set and the above (1) and (2) conditions, it is judged that the train g can execute the transport task in the first day, assuming that two executable transport tasks r1 and r2 are selected, at this time the current feasible path is expanded into two feasible paths, respectively: [r1] and [r2], the second day has no transport task, then the repair task job package executed by the train g in the second day is found by traversing the repair task set, assuming that b is found, then b is added to the two feasible paths, obtaining: [r1, b] and [r2, b]. In the third day, according to the transport task set and the above (1) and (2) conditions, it is judged that the train g can execute the transport task in the third day, assuming that three executable transport tasks r3, r4 and r5 are selected, then the two feasible paths before are expanded into six feasible paths, respectively: [r1, b, r3], [r1, b, r4], [r1, b, r5], [r2, b, r3], [r2, b, r4] and [r2, b, r5]. Similarly, the executable transport task r6 is selected in the fourth day, and the executable transport task r7 is selected in the fifth day, then for the train g, in the planned time period set D, the final feasible path set P g g g P g = { [r1, b, r3, r6, r7], [r1, b, r4, r6, r7], [r1, b, r5, r6, r7], [r2, b, r3, r6, r7], [r2, b, r4, r6, r7], [r2, b, r5, r6, r7]}.

[0080] The above steps Step1-Step6 quickly and effectively filter out the feasible path set of the train g in the planned time period by traversing the transport task set of each planned day for each train g.

[0081] In this embodiment, step S120 includes:

[0082] For each train, a feasible path is selected from the feasible path set of the train, and in the case that all transport tasks in the selected feasible path do not appear in the feasible path selected by other trains, the decision variable of the selected feasible path is determined to be 1.

[0083] In the feasible path with the decision variable being 1, the decision variable corresponding to the repair task and the decision variable corresponding to the transport task are determined to be 1 according to the repair task and the transport task, respectively.

[0084] Algorithm input: planned time period D, transport task R, train set G, feasible path set P of the train in the planned time period g .

[0085] Algorithm output: Initial solution for the decision variables.

[0086] The specific steps of the algorithm are as follows: Figure 3 As shown:

[0087] Step 1: Initialization. Create set H to store the selected feasible paths, create set F to record the first transportation task appearing in the feasible paths, and initialize the decision variables. and Initialize to 0. Continuing with the example above, F records transportation task r1 and transportation task r2.

[0088] Step 2: Iterate through the train set G. Check if G is empty. If it is, go to Step 6; otherwise, set the retrieved train as g.

[0089] Step 3: Traverse the set of feasible paths P for train g. g The variable j is used to record the selected feasible path number, if set P g If not empty, take the current path as p. g If yes, proceed to Step 4; otherwise, proceed to Step 5.

[0090] Step 4: If p g If none of the transportation tasks in p appear in set F, then p g Add to set H, and set p g The transportation tasks in [r1, b, r3, r6, r7] are added to set F, g+1, and then proceed to Step 3. Continuing the example above, since F is initially empty, it's clear that transportation tasks r1, r3, r6, and r7 in [r1, b, r3, r6, r7] are not present in set F. Therefore, [r1, b, r3, r6, r7] are added to set H, and r1, r3, r6, and r7 are added to set F. For each train g, at most one p will be selected. g It's possible that none of them will be selected, and it will be from P. g p selected g Recorded as That is, set H contains the values ​​corresponding to each train g.

[0091] Step 5: Traverse the feasible paths in H. Because train g can only execute at most one feasible path from its own set of feasible paths, when train g selects a feasible path... At that time, Corresponding decision variables Set to 1.

[0092] Step 6: Traverse feasible paths The maintenance and transportation tasks will be carried out by the company. Decision variables for each transportation task is set to 1; and the value of the decision variable corresponding to the feasible path of the train g is set to 1. is set to 1.

[0093] Step 7: If the set H is traversed, output the initial solution of the decision variable, otherwise go to Step 5.

[0094] The above steps Step 1-Step 7 can quickly and effectively obtain the initial solution of the decision variable through the auxiliary sets H and F.

[0095] The initial solution of the decision variable obtained by the step S120 is decoded to obtain the feasible path selected by the train, and the transportation task and the maintenance task performed on each planning day, which correspond to three decision variables, so the initial solution of the decision variable generated by the step S120 corresponds to a train inspection and transportation scheduling scheme, for example: [r1, b, r3, r6, r7] as the feasible path of the selected train g is stored in the set H, and the corresponding decision variable (initial solution of the transportation task decision variable), (initial solution of the maintenance task decision variable), and (initial solution of the feasible path decision variable) are assigned a value of 1, i.e. (indicating that the train g performs the transportation task r1 on the first day), (indicating that the train g performs the maintenance task b on the second day). Wherein the decision variable value of 1 indicates that the train needs to perform the corresponding maintenance task and transportation task, and all the decision variable values of 1 represent a feasible train inspection and transportation scheduling scheme. According to the scheduling scheme, the number and type of maintenance tasks of the train in the planning period can be obtained to further obtain the maintenance task cost, the number of trains used according to the task performed by the train in the scheduling scheme, and the cumulative operating mileage of each train in the planning period according to the operation task arranged in the scheduling scheme to further obtain the train operation balance index. However, considering that the initial solution may not include all the transportation tasks in the planning period, an additional value W1 is added to this solution, and the value is large enough to reflect the penalty value for the transportation tasks not included in the scheme, wherein a1 represents the number of transportation tasks not included. Because the algorithm considers the first and second level maintenance tasks, the depot has different maintenance capabilities for each level of maintenance task, and if the total number of maintenance tasks in a single planning day in the generated scheme exceeds the maintenance capability of the depot, a penalty value W2 based on the maintenance task is applied to the planning, and a2 is the number of maintenance tasks exceeded by the scheduling scheme. Therefore, the objective function is as follows:

[0096] ​minZ = w1β1Z1 + w2β2Z2 + w3β3Z3 + a1W1 + a2W2

[0097]

[0098]

[0099]

[0100] wherein Z1, Z2 and Z3 represent the train maintenance cost index, the number of train on-line operation index and the train operation balance index respectively, w1, w2 and w3 represent the weight values of the three indexes, β1, β2 and β3 represent the normalization processing indexes of the train maintenance cost index, the number of train on-line operation index and the train operation balance index respectively, T b represents the maintenance time required for maintenance operation package b; C t represents the labor cost per unit time required for maintenance operation. Z3 represents the extreme value of the running mileage of all trains in the non-on-line operation time within the planning time period (the operation balance is that the running mileage of each train within the planning time period is uniform, and in the embodiment, the extreme value of the running mileage of each train within the planning time period is used for evaluation); max gd and min gd respectively represent the maximum and minimum values of the train running mileage within the planning time period, x gdb represents the decision variable of the maintenance task, and represents 1 when the train g performs the maintenance operation package b on the planning day d, represents the decision variable of the feasible path, and represents 1 when the train g performs the feasible path

[0101] p g . It should be noted that although Z3 is not directly applied to the transportation task decision variable y gdr , max gd and min gd are decision variables y gdr , which are converted into actual train execution schemes, and the mileage statistics of each transportation task are obtained.

[0102] Step S130 is based on the train feasible path set P g obtained in step S110 and the initial solution of the decision variable obtained in step S120, and applies the grey wolf algorithm to solve the optimal transportation scheduling plan and maintenance scheduling plan of the train within the planning time period that satisfies the above objective function. Step S130 includes:

[0103] According to the initial solution of the decision variable, a plurality of neighborhood solutions are calculated, so that the total number of the initial solution and the neighborhood solution reaches the number of individuals in the initial population of the grey wolf optimization algorithm.

[0104] The initial solution of the decision variable and the initial scheduling scheme corresponding to the plurality of neighborhood solutions are input into the grey wolf optimization algorithm as individuals.

[0105] The optimal inspection and transportation scheduling scheme of each train is obtained by the grey wolf optimization algorithm according to the preset target function.

[0106] The grey wolf optimization algorithm (GWO) is a swarm intelligence optimization algorithm proposed by Mirjalili et al. of Griffith University in Australia. The algorithm is an optimization search method inspired by the hunting activities of grey wolf populations. The algorithm has strong convergence ability, fewer parameters, and is easy to implement. In recent years, it has been successfully applied to job shop scheduling, parameter optimization, and image classification. Based on the above characteristics, the grey wolf optimization algorithm is selected to solve the problem. However, the traditional grey wolf algorithm has the defects of poor population diversity, slow convergence speed in the later stage, easy to fall into local optimal solution, and poor solving efficiency. In view of these problems, the traditional grey wolf optimization algorithm is optimized.

[0107] The optimized part is as follows:

[0108] The linear control parameter is improved. In the traditional grey wolf algorithm, the control parameter k is generally taken as:

[0109]

[0110] where gen and MAXGEN represent the current iteration number and the total iteration number, respectively, that is, k is a linear control parameter with gen and MAXGEN as independent variables. However, such a linear control parameter cannot completely balance the search ability of global and local optimal solutions. Therefore, in this embodiment, a nonlinear control parameter with gen and MAXGEN as independent variables is selected. The closer gen is to MAXGEN, the greater the k decrement rate is. The purpose is to increase the global search ability by decreasing the k decrement rate in the early stage of algorithm search, and to increase the convergence speed of local optimization by increasing the k decrement rate in the later stage of algorithm search.

[0111] The specific process of the grey wolf optimization algorithm in step S130 is as shown in Figure 4 .

[0112] Step 1: initialization. Set the population size as pompsize, the iteration number as MAXGEN, and the control parameter k, the coefficient vector A and C. According to the initial solution and generate neighborhood solutions y′ gdr , x′ gdb and Y′ pgThe way to generate neighborhood solutions is to arbitrarily select one or more trains from the initial solution to form a set G1, and then select the remaining unassigned trains (from the initial solution) to form a neighborhood solution. For each train in the set of feasible paths where all possible paths are 0 (i.e., trains whose feasible paths were not selected into set H in step S120 above), one or more trains are selected to form set G2. Then, for each train in G2, a feasible path is selected from its corresponding feasible path set, such that the transport task in the selected feasible path replaces the transport task of the trains in set G1. This generates a neighborhood solution. Both the initial solution and the neighborhood solution are individuals in the initial population. A population containing pompsize individuals is then generated using this method. For example, if a train g selects feasible paths [r1, r2, refueling, r3, r4] within the planned time period, then the decision variable corresponding to this feasible path is... The decision variable is 1, which represents the transportation tasks included in the feasible path. The decision variables corresponding to the maintenance tasks included in the feasible path are 1. The value is 1, and an individual consists of a set of multiple trains whose respective feasible paths have been selected, which also represents an integrated inspection and transportation scheduling scheme within the planned time period.

[0113] Step 2: Calculate the fitness value of individuals in the gray wolf population according to the above objective function, and set the three individuals with the best fitness values ​​as α wolf, β wolf and δ wolf, and save them. The fitness value represents the way to measure the quality of the generated scheme and reflects the degree to which the generated scheme achieves the above objective function.

[0114] Step 3: Based on the gray wolf individual position update strategy, update the position of the ordinary ω wolf according to the positions of the α wolf, β wolf and δ wolf.

[0115] Step 4: Update the control parameter k, and then update the coefficient vectors A and C.

[0116] Step 5: Calculate the fitness value of all individual gray wolves based on the objective function.

[0117] Step 6: Update the fitness values ​​and positions of α wolves, β wolves, and δ wolves again based on the individual gray wolf positions.

[0118] Step 7: Determine if the current iteration number is equal to MAXGEN. If it is, the algorithm ends; otherwise, return to Step 3.

[0119] The algorithm finally outputs the optimal solution as y″. gdr 、x″ gdb and Y″ pg .

[0120] The improved grey wolf algorithm is used for solving the train inspection and transportation integrated scheduling model in the embodiment. An initial population generation method and a nonlinear control parameter improvement method are designed, so that the solution quality is improved, and compared with existing optimization methods, the improved grey wolf algorithm can achieve a higher optimal solution hit rate and improve the algorithm operation performance.

[0121] The train inspection and transportation integrated scheduling device provided by the application is described below, and the train inspection and transportation integrated scheduling device described below can be correspondingly referred to the train inspection and transportation integrated scheduling method described above.

[0122] As shown in Figure 5 The train inspection and transportation integrated scheduling device provided by the application comprises:

[0123] The feasible path set generation module 510 generates a feasible path set of each train in a planning time period according to a preset constraint condition and an inspection and transportation task in the planning time period, wherein the inspection and transportation task comprises a transportation task set and an inspection task set.

[0124] The initial solution generation module 520 is configured to generate an initial solution of a decision variable related to a feasible path of a train, a transportation task and an inspection task performed on each planning day according to the feasible path set of each train.

[0125] The optimal solution calculation module 530 is configured to calculate optimal decision variables of each train in the planning time period that satisfy a preset objective function according to the initial solution of the decision variable of each train, so as to obtain an optimal inspection and transportation scheduling scheme of each train.

[0126] Compared with the existing manual train inspection and transportation planning method, the train inspection and transportation integrated scheduling device provided by the application can formulate an optimal inspection and transportation scheduling scheme (i.e., an optimal inspection and transportation planning scheme) of a train according to a preset constraint condition and an inspection and transportation task of the train in a future period of time, avoid possible errors caused by manual operation, and improve work efficiency. Moreover, the application comprehensively considers the double factors of the inspection and transportation tasks, so that the entire inspection and transportation planning process is more reasonable and convenient, and the formulated inspection and transportation plan can effectively improve the utilization efficiency of the train, reduce the number of trains used, reduce the inspection cost, and improve the inspection efficiency.

[0127] Optionally, the feasible path set generation module 510 comprises:

[0128] The executable transportation task determination module is configured to determine a set of executable transportation tasks of each train in each planning day that satisfy the constraint condition according to the transportation task set.

[0129] The feasible path determination module is configured to determine a feasible path set of each train in the planning time period according to the executable transport task set and the overhaul task set.

[0130] Optionally, the initial solution generation module 520 comprises:

[0131] The first decision variable determination module is configured to determine a decision variable of a selected feasible path as 1 for each train, if all transport tasks in the selected feasible path do not appear in the selected feasible paths of other trains.

[0132] The second decision variable determination module is configured to determine a decision variable corresponding to an overhaul task and a decision variable corresponding to a transport task as 1 according to the overhaul task and the transport task in the feasible path with the decision variable as 1.

[0133] The above-described apparatus embodiments are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed 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 purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0134] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software plus necessary universal hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, which can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and include a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0135] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A train inspection and dispatch integrated scheduling method, characterized in that, The method comprises the steps of: According to the preset constraint condition and the inspection and transportation task in the planning time period, a feasible path set of each train in the planning time period is generated, wherein the inspection and transportation task comprises a transportation task set and a maintenance task set; According to the feasible path set of each train, an initial solution of decision variables related to the feasible path of the train, the transportation task and the maintenance task performed on each planning day is generated; According to the initial solution of the decision variables of each train, the optimal decision variables of each train in the planning time period satisfying the preset objective function are calculated, so as to obtain the optimal inspection and transportation scheduling scheme of each train; The constraint condition comprises: wherein, l gb denotes the cumulative running distance between the time when the train g last completed the maintenance work of the maintenance work package b and the time when the train g will perform the maintenance work of the maintenance work package b again, l r denotes the distance of the train performing the transportation task r, denotes the upper limit of the maintenance cycle of the maintenance work package b with distance as the maintenance cycle, t gb denotes the cumulative running time between the time when the train g last completed the maintenance work of the maintenance work package b and the time when the train g will perform the maintenance work of the maintenance work package b again, t r denotes the running time of the train performing the transportation task r, denotes the upper limit of the maintenance cycle of the maintenance work package b with time as the maintenance cycle, formula (1) and (2) indicate that the maintenance work should be performed within the specified time cycle range and distance cycle range, G denotes a set of trains, R denotes a set of transportation tasks, B denotes a set of maintenance work packages, Y p g denotes the decision variable of the feasible path, and is 1 when the train g performs the feasible path p g , formula (3) indicates that in the feasible path set P g of the train g, at most one feasible path p g can be performed, y gdr is the decision variable of the transportation task, and is 1 when the train g performs the transportation task r on the planning day d, formula (4) indicates that a train can perform at most one transportation task in a planning day, x gdb denotes the decision variable of the maintenance task, and is 1 when the train g performs the maintenance work package b on the planning day d, formula (5) indicates that the daily maintenance workload of the maintenance point cannot exceed its upper limit of maintenance capacity N, in formula (6), Set d denotes the set of transportation tasks that need to be performed when the planning day d is a holiday, and the set of transportation tasks that need to be performed when the planning day d is a working day is s2.

2. The train inspection and dispatch integrated scheduling method according to claim 1, wherein, According to the preset constraint condition and the inspection and transportation task in the planning time period, a feasible path set of each train in the planning time period is generated, wherein the inspection and transportation task comprises a transportation task set and a maintenance task set; According to the transportation task set, a set of executable transportation tasks of each train satisfying the constraint condition on each planning day is determined; According to the set of executable transportation tasks and the set of maintenance tasks, a set of feasible paths of each train in the planning time period is determined.

3. The train inspection and dispatch integrated scheduling method according to claim 1, wherein, According to the feasible path set of each train, an initial solution of decision variables related to the feasible path of the train, the transportation task and the maintenance task performed on each planning day is generated; For each train, a feasible path is selected from the feasible path set of the train, and in the case that all transportation tasks in the selected feasible path do not appear in the feasible paths selected by other trains, the decision variable of the selected feasible path is determined as 1; In the feasible path with the decision variable of 1, the decision variable corresponding to the maintenance task and the decision variable corresponding to the transportation task are determined as 1 according to the maintenance task and the transportation task.

4. The train inspection and dispatch integrated scheduling method of claim 1, wherein, The preset objective function is: minZ=w1β1Z1+w2β2Z2+w3β3Z3+a1W1+a2W2 wherein Z1, Z2 and Z3 represent the train maintenance cost index, the number of train on-line operation index and the train operation balance index respectively, w1, w2 and w3 represent the weight values of the three indexes, β1, β2 and β3 represent the normalization processing indexes of the train maintenance cost index, the number of train on-line operation index and the train operation balance index respectively, a1 represents the number of excluded transportation tasks, W1 represents the penalty value of the transportation tasks not included in the scheduling scheme, a2 represents the number of maintenance tasks exceeding the maintenance capacity of the depot in the scheduling scheme, W2 represents the penalty value of the total maintenance task quantity of a single planning day exceeding the maintenance capacity of the depot in the scheduling scheme, T b represents the maintenance time required by the maintenance operation package b; C t represents the labor cost per unit time required when performing the maintenance operation, max gd and min gd respectively represent the maximum and minimum values of the train operation mileage in the planning time period, x gdb represents the decision variable of the maintenance task, and represents 1 when the train g performs the maintenance operation package b on the planning day d, represents the decision variable of the feasible path, and represents 1 when the train g performs the feasible path p g .

5. The train inspection and transportation integrated scheduling method according to any one of claims 1 to 4, characterized in that, According to the initial solution of the decision variables of each train, the optimal decision variables of each train in the planning time period satisfying the preset objective function are calculated, so as to obtain the optimal inspection and transportation scheduling scheme of each train, which comprises the steps of: According to the initial solution of the decision variables, a plurality of neighborhood solutions are calculated, so that the total number of the initial solution and the neighborhood solution reaches the number of individuals in the initial population of the grey wolf optimization algorithm; The initial scheduling scheme corresponding to the initial solution and the plurality of neighborhood solutions of the decision variables is input into the grey wolf optimization algorithm as an individual; According to the preset objective function, the optimal inspection and transportation scheduling scheme of each train is obtained by iteration of the grey wolf optimization algorithm.

6. The train inspection and dispatch integrated scheduling method according to claim 5, wherein, In the iteration process of the grey wolf optimization algorithm, the control parameter k is a nonlinear control parameter with gen and MAXGEN as independent variables, and the closer gen is to MAXGEN, the greater the decreasing rate of the control parameter k is, wherein gen and MAXGEN respectively represent the current iteration number and the total iteration number.

7. A train inspection and dispatch integrated preparation device, characterized in that, The method comprises the steps of: A feasible path set generation module is configured to generate a feasible path set of each train in a planning time period according to a preset constraint condition and an inspection and transportation task in the planning time period, wherein the inspection and transportation task comprises a transportation task set and a maintenance task set; The initial solution generation module is configured to generate, according to the feasible path set of each train, an initial solution of a decision variable related to the feasible path of the train, a transportation task performed on each planning day, and a maintenance task; The optimal solution calculation module is configured to calculate, according to the initial solution of the decision variable of each train, an optimal decision variable of each train in the planning time period to meet a preset objective function, thereby obtaining an optimal inspection and transportation scheduling scheme of each train; The constraint conditions include: wherein, l gb denotes the cumulative running distance between the time when the train g last completed the maintenance work of the maintenance work package b and the time when the train g will perform the maintenance work of the maintenance work package b again, l r denotes the distance of the train g performing the transportation task r, denotes the upper limit of the maintenance cycle of the maintenance work package b with distance as the maintenance cycle, t gb denotes the cumulative running time between the time when the train g last completed the maintenance work of the maintenance work package b and the time when the train g will perform the maintenance work of the maintenance work package b again, t r denotes the running time of the train g performing the transportation task r, denotes the upper limit of the maintenance cycle of the maintenance work package b with time as the maintenance cycle, formula (1) and (2) indicate that the maintenance work should be performed within the specified time cycle range and distance cycle range, G represents a set of trains, R represents a set of transportation tasks, and B represents a set of maintenance work packages, denotes the decision variable of the feasible path, and is 1 when the train g performs the feasible path p g , formula (3) indicates that in the feasible path set P g of the train g, at most one feasible path p g can be performed, y gdr is the decision variable of the transportation task, and is 1 when the train g performs the transportation task r on the planning day d, formula (4) indicates that a train can perform at most one transportation task in a planning day, x gdb denotes the decision variable of the maintenance task, and is 1 when the train g performs the maintenance work package b on the planning day d, formula (5) indicates that the daily maintenance workload of the maintenance point cannot exceed its upper limit of maintenance capacity N, in formula (6), Set d denotes the set of transportation tasks that need to be performed when the planning day d is a holiday, and the set of transportation tasks that need to be performed when the planning day d is a working day is s2.

8. The train inspection and dispatch integrated system according to claim 7, wherein The feasible path set generation module includes: The executable transportation task determination module is configured to determine, according to the transportation task set, an executable transportation task set of each train on each planning day to meet the constraint conditions; The feasible path determination module is configured to determine, according to the executable transportation task set and the maintenance task set, a feasible path set of each train in the planning time period.

9. The train inspection and dispatch integrated scheduling apparatus according to claim 7, wherein The initial solution generation module includes: The first decision variable determination module is configured to determine, for each train, a decision variable of a selected feasible path as 1, in a case that all transportation tasks in the selected feasible path do not appear in a selected feasible path of another train. The second decision variable determination module is configured to determine, in the feasible path with the decision variable as 1, a decision variable corresponding to the maintenance task and a decision variable corresponding to the transportation task as 1, respectively.

Images