An airport network time configuration method, system, device and storage medium based on cluster compression

Abstract

The application discloses an airport network time configuration method, system and device based on clustering compression and a storage medium, and the method comprises the following steps: constructing a time application operation date matrix according to time application data of a future planning period, and performing clustering compression on the dates in the matrix to obtain a plurality of date clusters; globally selecting a key operation date to represent other dates in the date cluster; constructing an airport network time configuration model, and solving the model based on the selected key operation date to obtain an initial time configuration scheme; dynamically adjusting the priority weight of the time application, and updating the selection of the key operation date based on the adjusted weight to update the time configuration scheme of the future period. The application optimizes the solving speed of the flight time configuration related problems, and ensures that the time configuration is efficiently and fairly performed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of air traffic control, and particularly relates to an airport network time configuration method, system, device and storage medium based on clustering compression. BACKGROUND

[0002] The airport time configuration in the air transportation system is of great significance to ensure the efficient and fair use of limited airport infrastructure.

[0003] The time configuration problem aims to configure the take-off and landing time to the appropriate request when the runway capacity of one or more airports and the request of the airline for a specific time are given, and various priorities, conflicts and capacity requirements are considered. In practical applications, a large number of time applications, complex time configuration rules and the necessity of seamless coordination among airport networks together lead to the high complexity of the time configuration problem, and the balancing of various conflicting interests and priorities further aggravates the complexity of the problem.

[0004] The prior art has the following two defects when dealing with the highly complex problem of airport time configuration:

[0005] On the one hand, the time configuration problem, especially the airport network time configuration problem, is a highly complex problem, and it is difficult to obtain an optimal solution in a short time;

[0006] On the other hand, most time configuration models do not consider the condition that the time application exceeds the airport capacity, i.e. it may not be possible to allocate all applications to time even if the allocation is done in any way. Therefore, the existing time configuration model is almost impossible to solve this application over-capacity situation. SUMMARY

[0007] The purpose of the present application is to provide a method for quickly configuring airport network time, which supports efficient and fair configuration of multiple airports, ensures airport capacity balance and airline satisfaction. Another purpose of the present application is to provide an airport network time configuration system, electronic device and computer readable storage medium corresponding to the aforementioned method.

[0008] Technical solution: According to a first aspect of the present application, a method for configuring airport network time based on clustering compression is provided, comprising the following steps:

[0009] S1: Based on the time application data of the future planning period, a time application operation date matrix is constructed, and clustering analysis is performed on the date column to classify multiple future operation dates into several date clusters;

[0010] S2: selecting at least one key operation date from each of the date clusters to represent other dates in the date cluster; recovering part of the to-be-eliminated application and constructing a corresponding virtual operation date, expanding the key operation date as the selected key operation date;

[0011] S3: constructing an airport network time configuration model for the compressed time configuration problem, and solving based on the selected key operation date to obtain a time configuration initial scheme;

[0012] S4: dynamically adjusting the priority weight of the time application according to the result of the time configuration initial scheme, and updating the key operation date selection in step S2 based on the adjusted weight to update the future period time configuration scheme.

[0013] The application adopts a date clustering compression method, reduces the solution size of a large-scale combinatorial optimization problem, and ensures the accuracy of the time configuration scheme through dynamic adjustment.

[0014] Optionally, the time application data based on the future planning period is used to construct a time application operation date matrix, including:

[0015] Obtaining time application data submitted by each airline, including but not limited to: flying airline, take-off and landing airport, application take-off time, application landing time, application period, operation week, application flight type, flying aircraft type, and connecting flight;

[0016] Integrating the time application date, time application time, time application priority type, and connecting flight to obtain a time application data table;

[0017] Constructing a time application operation date matrix according to the time application data table.

[0018] Further, according to the time application data integration, a time application data table is obtained,

[0019] Optionally, the distance measurement for the clustering analysis is based on the uncovered degree of the flight application operation between the date columns; wherein the uncovered degree of the first date column to the second date column is defined as: the number or weighted sum of time applications that run in the second date column but do not run in the first date column.

[0020] Optionally, the calculation of the uncovered degree can also add a corresponding weight to the application, and generally uses the operation days of the application.

[0021] Further, the uncovered degree between the date columns is calculated to obtain an asymmetric matrix, and the matrix is used as a distance matrix.

[0022] Further, the matrix is symmetrically processed, including but not limited to: taking the minimum, taking the average, and taking the maximum of two symmetric points.

[0023] In addition, distance matrices can also be constructed using Jaccard similarity, cosine similarity, and other methods.

[0024] Based on the distance matrix, a clustering algorithm is used to classify the running dates, resulting in multiple date clusters.

[0025] Optionally, the clustering algorithm includes, but is not limited to, k-means, k-mediods, and DBSCAN algorithms.

[0026] Optionally, to resolve time configuration conflicts caused by the difference between the key running date and other represented running days, conflicting applications are removed after the key running date is selected. To minimize the number of applications removed, a greedy algorithm is used to select the key running date after clustering.

[0027] Furthermore, the greedy algorithm includes:

[0028] Calculate the number of operating days requested for each flight;

[0029] Based on the results of the cluster analysis, each candidate date is calculated as a set of conflict time requests that need to be removed after the key date;

[0030] All time requests in the set are scored, and their scores are dynamically updated based on the number of times they are selected for removal in the current group category: each time a time request is selected for removal, its score increases by an increment related to the current number of selections;

[0031] Calculate the total score for each candidate date, which is the sum of the latest scores of all applications in the corresponding conflict time application set;

[0032] The key run date is selected based on the total score of the candidate dates. A lower total score for a candidate date indicates a smaller impact from the resulting elimination set when that run date is selected.

[0033] Optionally, dynamically updating its score based on the number of times it has been selected for removal in the current grouping category includes: calculating the number of running days for each application. And record each application according to the classification. The set of time slots that need to be removed after being selected as critical execution dates is determined. After obtaining the set, all time slot applications are scored, with the score increasing each time a time slot is selected for removal. , Select application for current classification group The number of times. The score reduction mechanism within the same group can effectively reduce the impact of the same elimination preference caused by similarity within the same group on the overall score.

[0034] Preferably, selecting the key operation date according to the total score of the candidate dates comprises: calculating the difference between the lowest score and the second lowest score in each group, and preferentially selecting the lowest score date in the group with a larger difference as the key operation date. If the difference between the highest score and the lowest score in multiple groups is equal, preferentially selecting the group with a lower score.

[0035] Optionally, the airport capacity information includes but is not limited to: 60 min take-off and landing capacity, 15 min take-off and landing capacity.

[0036] Further, the 60 min take-off and landing capacity and the 15 min take-off and landing capacity each further include: runway capacity, apron capacity.

[0037] Optionally, the airport network time configuration model includes:

[0038] Setting the remaining applications after the elimination as base applications, and setting the selected key operation date as a base operation date, and constructing an airport network time configuration model based on the base applications and the base operation date;

[0039] Setting the objective function in the airport network time configuration model, including: maximum time offset , total time offset , and unallocated flight frequency ; wherein, is an index of a time application, is a date index, is a date set, is a time positive offset of an application , is a time negative offset of an application , indicates whether an application runs on a day or not; indicates the maximum value in or , indicates whether a time application is allocated or not; The constraint conditions include capacity constraints, turnaround constraints, and priority constraints; wherein, the capacity constraints include: airport take-off and landing constraints, airport landing constraints, and airport total take-off and landing constraints; the turnaround constraints include: maximum turnaround time constraints of connecting flights and minimum turnaround time constraints of connecting flights; the priority constraints use time offset to constrain time applications of different priorities;

[0040]

[0041] ​​​The application at each time is sorted based on preset influence factor priorities, and each time application is sequentially inserted into a pre-constructed time slot allocation scheme according to the sorting result to obtain an initial solution of the airport network time slot configuration model;

[0042] The initial solution includes an allocated flight set obtained by taking a position with a minimum insertion penalty value as a target insertion position and an unallocated flight set composed of time slot application groups that cannot be inserted; and the penalty value is determined by the number of capacity conflicts, the total offset and the number of connection flight violation times.

[0043] The approximate integer solution is taken as a target solution of the airport network time slot configuration model, and a flight configuration result and a flight plan are obtained.

[0044] Further, in the turn-around constraint,

[0045] The maximum turn-around time constraint of the connection flight is:

[0046] ;

[0047] The minimum turn-around time constraint of the connection flight is:

[0048] ;

[0049] Wherein, denotes the allocated time slot of the application ; is 0 when the allocated time slot of the time slot application is earlier than the time slot ; denotes the flight time of the application ; Similarly, denotes the allocated time slot of the application ; and are the maximum turn-around time and the minimum turn-around time of the connection flight, respectively, is a maximum value for removing the turn-around limit of the unallocated flight; is the earliest allocation time of the time slot application ; is the earliest allocation time of the time slot application ; is the latest allocation time of the time slot application ; is the latest allocation time of the time slot application ; denotes whether the time slot application is allocated;

[0050] The priority constraint is:

[0051] ;

[0052] ;

[0053] in, , ; , ;

[0054] in, Apply for time The application time and the specific time Related, if If it was applied for earlier than the deadline, then =1, otherwise =0; Apply for time Historical operating time, and specific moments Related, if If it was applied for earlier than the deadline, then =1, otherwise =0; A collection of all historical moment requests; Request a set for each moment.

[0055] Optionally, the unconfigured time points are sorted by the number of running dates and inserted sequentially into the initial solution of the model.

[0056] Optionally, the dynamic adjustment of the priority weight of the time slot applications includes: readjusting the scoring weight of the time slot applications based on their offset relative to the time slots in the initial time configuration scheme, and adjusting the original scores of each time slot application from... Adjusted to ;in, These are the original weights; This is the offset; It is a constant, determined based on the number of applications and airport capacity; The preset reduction factor, For the first The number of times an application can be selected at any given moment.

[0057] According to a second aspect of the present invention, an airport network time slot configuration system is provided, comprising:

[0058] The date clustering module is used to construct a matrix of time application execution dates based on time application data for future planning periods, and to perform cluster analysis on the date columns to classify multiple future execution dates into several date clusters;

[0059] A key day selection module is used to select at least one key running date from each of the date clusters;

[0060] A model solving module is configured to construct an airport network time configuration model and solve the model based on the selected key operation date to obtain a time configuration initial scheme;

[0061] A dynamic optimization module is configured to dynamically adjust priority weight of the time configuration based on the result of the time configuration initial scheme, and trigger the key date selection module to reselect the key operation date for iterative optimization.

[0062] According to a third aspect of the present application, an electronic device is provided, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements part or all steps of the airport network time configuration method based on cluster compression when executing the program.

[0063] According to a fourth aspect of the present application, a computer readable storage medium is provided, which stores a computer program, wherein the program is executed by a processor to implement part or all steps of the airport network time configuration method based on cluster compression.

[0064] Advantages: Compared with the prior art, the present application has the following advantages:

[0065] 1. The present application improves the technical defects of the prior art that the solving speed is slow when performing time configuration, only a single airport is considered, and the request over-capacity situation cannot be solved;

[0066] 2. By constructing a time application operation date matrix, extracting operation day features, and then combining a density clustering and a partition clustering hybrid strategy, high-dimensional date data is compressed into typical date clusters, and the solving time and iteration speed are significantly shortened by using a small-scale solving method, thereby improving the solving efficiency;

[0067] 3. A weighted conflict network is constructed based on the conflict intensity within the cluster, a bidirectional greedy strategy is used to dynamically select a representative key operation date, an airport network space-time coupling model is established based on the key operation date, an initial feasible solution is generated by conflict constraint relaxation, and gradient insertion optimization is performed based on the date distribution characteristics of the unconfigured request; The conflict network construction method in this process can be based on different weight determination methods, and the key date selection algorithm can be based on different greedy principles, which provides good expandability for subsequent algorithms to adapt to different characteristics of the airport network model.

[0068] 4. The priority of the airline and the time value weight are dynamically corrected according to the time offset, and closed-loop optimization is achieved through cluster center relocation and key date rescreening until the airport capacity balance and airline satisfaction constraints are met. This method uses an iterative method to continuously improve the solution to achieve the possibility of optimality, reducing the local optimal disadvantages caused by the lack of global vision. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 Fig. 1 is a schematic diagram of the overall process of the present application;

[0070] Figure 2 Fig. 2 is a schematic diagram of the basic structure of cluster compression in an embodiment;

[0071] Figure 3 Fig. 3 is a flowchart of the compressed greedy algorithm in an embodiment;

[0072] Figure 4 Fig. 4 is a structural block diagram of an electronic device. DETAILED DESCRIPTION

[0073] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application are further described below with reference to the drawings.

[0074] In the prior art, when dealing with the highly complex problem of airport schedule configuration, on the one hand, only single-airport problems are considered, although the problem is simplified, but the strong complementarity of configuring the schedule for a single flight at the departure airport and the destination airport is ignored, thus there is a defect that cannot meet the actual demand. On the other hand, the condition that the schedule application exceeds the airport capacity is not considered, thus it is impossible to solve the application exceeding the capacity. Based on this, the present embodiment provides an airport schedule configuration method based on large neighborhood search, which seeks the optimal way of schedule configuration from a global perspective, and ensures the efficient and fair use of limited airport infrastructure.

[0075] As shown in Fig. 1, a cluster compression-based airport network schedule configuration method comprises the following steps: Figure 1

[0076] S1: Based on the schedule application data of the future planning period, a schedule application operation date matrix is constructed, and cluster analysis is performed on the date column to classify multiple future operation dates into several date clusters.

[0077] S11: Obtain the schedule application data submitted by each airline, including but not limited to: operating airline, departure and arrival airport, application departure time, application landing time, application period, operation day, application flight type, operating aircraft type, and connecting flight;

[0078] Integrate the schedule application date, schedule application time, schedule application priority type, and connecting flight to obtain a schedule application data table;

[0079] Construct a schedule application operation date matrix according to the schedule application data table.

[0080] As shown in Fig. 2, a cluster compression-based airport network schedule configuration method comprises the following steps: Figure 2 ​As shown, in one embodiment, the sequence number 1 is the original time application running date matrix, wherein the horizontal axis represents the date, and the vertical axis represents different applications, 1 represents that the corresponding application runs on that day, which is marked in green in the figure; 0 represents that the corresponding application does not run on that day. The colored numbers above the sequence number 1 represent different dates, and different colors represent classification results.

[0081] S12: Using a distance calculation method based on one-way distance, the date column is classified.

[0082] First, define the uncovered degree, and after obtaining the running date matrix, the running date uncovered degree of column a to column b is calculated as follows: the number of applications that run in column b but do not run in column a is called the uncovered degree of a to b. The calculation of this uncovered degree can also add a corresponding weight to the application, and the running days of the application are generally used. The uncovered degree between the date columns is calculated, and an asymmetric matrix is obtained, which is used as a distance matrix, or the matrix is symmetrically processed, including but not limited to taking the minimum, taking the average, and taking the maximum of two symmetric points. In addition, the distance matrix can also be constructed by Jaccard similarity, cosine similarity, etc.

[0083] The classification method includes but is not limited to k-means; k-mediods; DBSCAN algorithm, etc.

[0084] Figure 2 The sequence number 2 shows the results after clustering classification, and dates with high similarity are classified into a class.

[0085] S2: Select at least one key running date from each of the date clusters to represent other dates in the date cluster, thereby reducing the problem solving scale.

[0086] To solve the time configuration conflict caused by the difference between the key running day and the other represented running day, the conflict application is removed after the key running date is selected. In order to make the number of removed applications less, a greedy algorithm is used to select the key running date after clustering.

[0087] As Figure 3 shown, the steps of the greedy algorithm are as follows: the running days of each application are calculated and the time set that needs to be removed after each application is selected as a key running date is recorded according to the classification, and after obtaining the set, all time applications are scored, and the score increases , for each time selected and removed. The score of the application selected by the current classification group is counted. The score decreasing mechanism of the same group can effectively reduce the influence of the same removal preference on the whole due to the similarity of the same group.

[0088] After scoring the time point application, the corresponding operation date is scored, and the score of each operation day is calculated as the sum of the time point scores corresponding to the time point application exclusion set generated by the current operation day. The lower the score, the smaller the impact of selecting the operation day to generate the exclusion set. When selecting the key operation day, calculate the difference between the lowest score and the second lowest score in each group, and preferentially select the lowest score date in the group with a large difference as the key operation day. Then, update the exclusion application list and recalculate the date score of each group. The applications included in the exclusion list are not considered.

[0089] After clustering, one or several key operation days are selected from each class to represent all operation days in the class to reduce the problem solving scale.

[0090] Figure 2 Figure 3 shows the selection of a key date from each class as a representative of the class after clustering (the time point application exclusion process is not shown in the figure).

[0091] Restore part of the applications to be excluded and construct the corresponding virtual operation date; thereafter, model and solve the compressed time point configuration problem, which can reduce the model scale.

[0092] As Figure 2 As shown in Figure 4, the key date is expanded to all operation dates after the result, and the result is used for solving, sacrificing part of the accuracy to reduce the solving scale.

[0093] S3: Build an airport network time point configuration model and solve it based on the selected key operation date to obtain an initial time point configuration scheme. Due to the pre-processing of step S2, the model is significantly smaller in scale compared to the time point configuration model constructed by directly solving the original problem.

[0094] S31: Build an airport network time point configuration model according to the time point application data table and airport capacity information.

[0095] The airport capacity information includes: 60min take-off and landing capacity, 15min take-off and landing capacity; wherein the 60min take-off and landing capacity and the 15min take-off and landing capacity each include: runway capacity, apron capacity.

[0096] Set the remaining applications after exclusion as the basic applications, and set the selected key operation date as the basic operation date. Build an airport network time point configuration model based on the basic applications and the basic operation date.

[0097] The objective function in the model includes: maximum time point offset , total time point offset Dpmax and unallocated flight frequency ; wherein i is the index of the time point application, d is the date index, for a date set, for a positive offset of time for application i, for a negative offset of time for application i, denotes whether d days for application i are running; denotes or the maximum value in, denotes whether time application is assigned;

[0098] The constraint conditions include capacity constraints, turnaround constraints and priority constraints; wherein the capacity constraints include airport take-off slot constraints, airport landing slot constraints and airport total take-off and landing slot constraints; the turnaround constraints include maximum turnaround time constraints of connecting flights and minimum turnaround time constraints of connecting flights; the priority constraints use time offset to constrain time applications of different priorities;

[0099] Based on the preset priority of influencing factors, each time application is sorted, and each time application is sequentially inserted into a pre-constructed empty time assignment scheme according to the sorting result to obtain an initial solution of the airport network time configuration model;

[0100] The initial solution includes an assigned flight set obtained by taking the position with the minimum insertion penalty value as the target insertion position, and an unassigned flight set composed of time applications that cannot be inserted; the penalty value is determined by the number of capacity conflicts, the total offset and the number of connecting flight violations;

[0101] The approximate integer solution is taken as the target solution of the airport network time configuration model, and then the flight configuration result and the flight plan are obtained.

[0102] In the turnaround constraints,

[0103] The maximum turnaround time constraint of the connecting flight is:

[0104] ;

[0105] The minimum turnaround time constraint of the connecting flight is:

[0106] ;

[0107] wherein, denotes the assigned time of application , and when the assigned time of time application is earlier than the time , it is 0, otherwise it is 1; denotes the flight time of application , which can be obtained according to the expected landing time and the expected take-off time; and respectively the maximum and minimum turnaround time of the connecting flight, is a maximum value for removing the turnaround limit for unassigned flights; is the earliest assignment time of the time application is the earliest assignment time of the time application is the latest assignment time of the time application is the latest assignment time of the time application is the latest assignment time of the time application is the latest assignment time of the time application is the latest assignment time of the time application denotes whether the time application is assigned;

[0108] The priority constraint is:

[0109] ;

[0110] ;

[0111] wherein , ; , ;

[0112] wherein is the application time of the time application , which is relevant to the specific time , and is 1 if is earlier than the application time, and 0 otherwise; is the historical running time of the time application , which is relevant to the specific time , and is 1 if is earlier than the application time, and 0 otherwise; is the set of time applications of all historical time changes; is the set of time applications. S32: solve the model, sort the unassigned applications according to the running date, and insert each time application into the pre-constructed empty time assignment scheme in turn to obtain the initial solution of the airport network time configuration model.

[0113] S4: after solving the above model, re-sort the unassigned time for insertion to obtain the initial solution, adjust the scoring weight of the time application according to the time offset generated by the initial solution, re-select the key running date, update the list of applications to be removed, re-construct the virtual running date and re-sort the insertion, and iterate in turn until the target value is unchanged.

[0114] S4: after solving the above model, re-sort the unassigned time for insertion to obtain the initial solution, adjust the scoring weight of the time application according to the time offset generated by the initial solution, re-select the key running date, update the list of applications to be removed, re-construct the virtual running date and re-sort the insertion, and iterate in turn until the target value is unchanged.

[0115] ​In summary, the airport network time configuration model is provided with multiple constraints, three operator structures and large neighborhood algorithm are combined in the solving process, so that the configuration result is effective for multi-airport network. Meanwhile, the non-allocated option is introduced, so that the problem solving is more in line with the actual situation, and the high dimension of the solution space of the planning is also considered, so that the problem of high complexity in optimization solving is solved.

[0116] The application further provides an airport network time configuration system, comprising:

[0117] A date clustering module is configured to construct a time application operation date matrix based on time application data of a future planning period, and perform clustering analysis on a date column to classify multiple future operation dates into several date clusters.

[0118] A key day selection module is configured to select at least one key operation date from each date cluster.

[0119] A model solving module is configured to construct an airport network time configuration model and solve the model based on the selected key operation date to obtain a time configuration initial scheme.

[0120] A dynamic optimization module is configured to dynamically adjust the priority weight of the time application according to the result of the time configuration initial scheme, trigger the key day selection module to reselect a key operation date, and perform iterative optimization.

[0121] To implement the above-mentioned airport network time configuration method based on clustering compression, the application further provides an electronic device and a computer readable storage medium.

[0122] As shown in Figure 4 The electronic device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements part or all of the steps of the above-mentioned airport network time configuration method based on clustering compression when executing the program.

[0123] The computer readable storage medium has a computer program stored thereon, and the program is executed by the processor to implement part or all of the steps of the above-mentioned airport network time configuration method based on clustering compression.

[0124] The above is only some embodiments of the application, not all embodiments. Based on the described embodiments of the application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the application. Unless otherwise defined, the technical terms or scientific terms used herein should have their usual meanings understood by those skilled in the art in the field of the application.

[0125] As used in the specification and claims, the words "first", "second", and similar words of distinction do not imply any order, quantity, or importance, but are used to distinguish one element from another. Also, the words "a" and "an" and "the" and similar words of distinction do not imply a quantity of one, but instead mean that at least one of something is present. The words "comprise" and "comprising" and similar words of conformance mean that the elements or objects preceding the word "comprise" or "comprising" encompass the features, integers, steps, operations, elements, and / or components listed after the word "comprise" or "comprising" and do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0126] The above programs can be run in a processor, or can also be stored in a memory (or called computer readable storage medium), the computer readable medium includes permanent and non-permanent, removable and non-removable media, which can be realized by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape disk storage or other magnetic storage device, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. According to the definition herein, computer readable medium does not include temporary computer readable medium, such as modulated data signals and carriers.

[0127] These computer programs can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate computer implemented processes, so that the instructions executed on the computer or other programmable device provide a process for implementing the flow Figure 1 The steps of the flow or the functions specified in the block Figure 1 corresponding to different steps can be implemented by different modules.

Claims

1. A method for configuring airport network time slots based on clustering compression, characterized in that, Includes the following steps: Based on the time slot application data of future planning cycles, a time slot application operation date matrix is ​​constructed, and cluster analysis is performed on the date columns to classify multiple future operation dates into several date clusters; Select at least one key running date from each of the date clusters to represent the other dates within that date cluster; The selection of the critical operating date uses a greedy algorithm, including: calculating the number of operating days for each flight application; calculating each candidate date as a set of conflicting time slot applications to be removed after the critical date, based on the results of the cluster analysis; scoring all time slot applications in the set, and dynamically updating their scores based on the number of times they are selected for removal in the current group category: each time a time slot application is selected for removal, its score increases by an increment related to the current selection count; after scoring the time slot applications, calculating the total score for each candidate date, which is the sum of the latest scores of all applications in the corresponding set of conflicting time slot applications to be removed; selecting the critical operating date based on the total score of the candidate dates, including: calculating the difference between the lowest and second lowest scores in each group, prioritizing the lowest score date in the group with the largest difference as the critical operating date, then updating the list of applications to be removed, recalculating the date score for each group, and excluding applications already included in the removal list; Recover some of the applications to be removed and construct corresponding virtual running dates, then expand the key running dates as the selected key running dates; To address the compressed time slot configuration problem, an airport network time slot configuration model is constructed and solved based on selected key operating dates to obtain an initial time slot configuration scheme. Based on the results of the initial time-configuration scheme, the priority weights of time slot applications are dynamically adjusted, and the selection of key running dates is updated based on the adjusted weights, thus updating the time-configuration scheme for future periods. The dynamic adjustment of the priority weights of the time slot applications includes: readjusting the scoring weights of the time slot applications based on their offset from the times in the initial time-configuration scheme, changing the original scores of each time slot application from... Adjusted to ;in, These are the original weights; This is the offset; It is a constant, determined based on the number of applications and airport capacity; The preset reduction factor, For the first The number of times an application can be selected at any given moment.

2. The method according to claim 1, characterized in that, The clustering analysis is based on the distance metric of the non-coverage of flight application operations between date columns; wherein, the non-coverage of the first date column to the second date column is defined as: the number or weighted sum of time slot applications that operate in the second date column but not in the first date column.

3. The method according to claim 2, characterized in that, In the calculation of the coverage, the weight assigned to the time slot application is the number of days it is planned to run.

4. The method according to claim 1, characterized in that, The airport network time slot configuration model includes: The remaining applications after elimination are set as basic applications, and the selected key operating dates are set as basic operating dates. An airport network slot configuration model is then built based on the basic applications and basic operating dates. The objective function in the airport network time configuration model is set, including: maximum time offset. Total time offset and unallocated flight frequencies ;in, For indexes requested at any time, For date indexing, For a set of dates, For application The time is shifting. The negative offset of the time of application i. Indicates application of Is the heavens functioning? express or The maximum value in, Indicates time application Whether it has been assigned; The constraints include capacity constraints, turnaround constraints, and priority constraints; wherein, the capacity constraints include: airport takeoff number constraints, airport landing number constraints, and total airport takeoff and landing number constraints; the turnaround constraints include: maximum turnaround time constraints for connecting flights and minimum turnaround time constraints for connecting flights; the priority constraints use time offsets to constrain time requests of different priorities. Based on the preset priority of influencing factors, each time slot application is sorted, and then inserted into a pre-built empty time slot allocation scheme in sequence according to the sorting result to obtain the initial solution of the airport network time slot configuration model; wherein, the initial solution includes the set of allocated flights obtained by taking the position with the minimum penalty value at the time of insertion as the target insertion position, and the set of unallocated flights consisting of time slot applications that cannot be inserted; the penalty value is determined by the number of capacity conflicts, the total offset, and the number of connecting flight violations; The approximate integer solution is used as the target solution of the airport network time slot configuration model, thereby obtaining the flight configuration results and flight plan.

5. The method according to claim 4, characterized in that, In the aforementioned turnover constraints, The maximum turnaround time constraint for connecting flights is: ； The minimum turnaround time constraint for connecting flights is: ； in, Indicates application The time slot is allocated when the time slot is requested. The time of allocation is earlier than the time of allocation. The value is 0 if the condition is met, and 1 otherwise. Indicates application Flight time; and These refer to the maximum and minimum turnaround times for connecting flights. This is a maximum value used to lift turnaround restrictions on unassigned flights; Apply for time The earliest allocation time, Apply for time The earliest allocation time, Apply for time The latest allocation time, Apply for time The latest allocation time, Indicates time application Whether it has been assigned; The priority constraint is as follows: ； ； in, , ; , ; Apply for time The application time and the specific time Related, if If it was applied for earlier than the deadline, then =1, otherwise =0; Apply for time Historical operating time, and specific moments Related, if If it was applied for earlier than the deadline, then =1, otherwise =0; A collection of all historical moment requests; Request a set for each moment.

6. An airport network time slot configuration system based on the method of any one of claims 1-5, characterized in that, include: The date clustering module is used to construct a matrix of time slot application execution dates based on time slot application data for future planning periods, and to perform cluster analysis on the date columns to classify multiple future execution dates into several date clusters; A key day selection module is used to select at least one key running date from each of the date clusters; The model solving module is used to construct the airport network time slot configuration model and solve it based on the selected key operating dates to obtain the initial time slot configuration scheme; The dynamic optimization module is used to dynamically adjust the priority weight of time application based on the result of the initial scheme configured at the time, and trigger the key day selection module to reselect the key running date for iterative optimization.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps of the method as described in any one of claims 1-5.

Images