Large airport runway scheduling method fusing rule triggering and genetic algorithm
Through the method of fusion rule triggering and genetic algorithms, the aircraft's initial taxiing time and runway allocation are optimized, and the problems of aircraft taxiing conflicts and delays in the prior art are solved, and the safety and resource utilization efficiency of the airport are improved.
Patent Information
- Application Number
- CN202411093314.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-09
AI Technical Summary
The prior art fails to effectively combine the aircraft's starting taxiing time and runway scheduling, which makes it difficult to effectively reduce the conflict points and delayed number during the aircraft's taxiing, affecting the efficiency and safety of airport resource utilization.
Large airport runway scheduling methods using fusion rule triggering and genetic algorithms, obtain and preprocess aircraft data, sort out outbound aircraft using heuristic rule algorithms, and iterative calculations combined with genetic algorithms to finally generate the optimal taxi scheduling scheme to optimize the starting of the aircraft Start taxiing time and runway allocation.
It effectively reduces the conflict points and delays during aircraft taxiing, improves the safety of the airport and the on-time flight rate, and optimizes resource utilization efficiency.
Smart Images

Figure CN119049347A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of runway scheduling, and in particular to a large airport runway scheduling method integrating rule triggering and genetic algorithm. Background Art
[0002] At present, my country's economy is booming, the demand for aviation transportation is growing, and the number of aircraft continues to increase. Aircraft activities are mainly concentrated in two links: air flight and ground taxiing. It is worth noting that the use of runways is exclusive and can only be used by one aircraft at a time; the taxiway system is different. It can accommodate multiple aircraft in parallel at the same time. However, with the increasing number of taxiways, the ground layout of the airport has become more and more complex, providing aircraft with a large number of taxiing routes. However, the taxiing path of aircraft on the airport ground is generally selected by dispatchers and pilots based on historical work experience, which makes the existing resource allocation of the airport unable to make full use of the existing resources of the airport, which not only increases the cost in the operation process, but also increases the risk of aircraft conflicts during taxiing. Compared with economic factors such as taxiing distance, airports pay more attention to the number of aircraft conflict points and the number of delayed flights. Airports would rather increase a small amount of cost during taxiing, and hope that aircraft can reduce more conflicts and reduce the number of delayed aircraft during taxiing. Therefore, how to formulate appropriate taxiing plans to ensure safe and efficient taxiing of aircraft has become a crucial task in airport management.
[0003] 1. Existing research mainly focuses on changing the factors affecting the aircraft taxiing cost, such as path length, taxiing speed, etc. to solve the conflict avoidance problem, and based on this, solve the optimal path. Less consideration is given to the impact of the initial taxiing time when the aircraft enters the taxiway system on the taxiing process, that is, aircraft entering the taxiway at different initial taxiing times will lead to different potential conflict situations in the scene, which in turn affects the final number of aircraft conflicts, number of delayed aircraft and total taxiing time. The waiting time of aircraft in the process of avoiding conflicts is also a component of taxiing cost. Separating the initial taxiing time from the runway scheduling setting cannot effectively reduce the aircraft's taxiing cost, nor can it effectively reduce the risk of conflict during the aircraft's taxiing process.
[0004] Therefore, when analyzing the path conflict avoidance of aircraft, the impact of the delayed change in the initial taxiing time on the potential conflicts in the future path should be considered at the same time. However, the current modeling research that comprehensively considers the trade-off between taxiing cost and delay cost is still insufficient. Summary of the invention
[0005] The purpose of the present invention is to provide a large airport runway scheduling method that integrates rule triggering and genetic algorithm, with the goal of reducing the number of conflict points and delayed aircraft during the taxiing process of aircraft. By reasonably arranging the starting taxiing time of the aircraft and the corresponding runway selected, the algorithm's solution efficiency is improved, thereby reducing the number of conflicts and delays during the operation of aircraft.
[0006] In order to achieve the above tasks, the present invention adopts the following technical solutions:
[0007] A large airport runway scheduling method integrating rule triggering and genetic algorithm, comprising:
[0008] Obtain the airport's standard taxi path diagram and all data on aircraft arriving and departing from the airport during the same period and perform pre-processing;
[0009] Use the heuristic rule algorithm to sort the priority sequence of outbound aircraft for each set of data, and obtain the priority ranking of each outbound aircraft;
[0010] Prioritize each outbound aircraft, assign runways to them according to the priority, and set the corresponding start taxi time. The situation after the assignment is used as the scheduling plan;
[0011] The generated scheduling plan is used as the initial solution, a population is generated based on the initial solution, and iterative calculations are performed through the genetic algorithm. Finally, the optimal individual is obtained through the genetic algorithm, which is the final scheduling plan.
[0012] Furthermore, the preprocessing is an operation of removing irrelevant data; the data after preprocessing includes:
[0013] Departure aircraft data: parking position, assigned runway, aircraft type, aircraft take-off time, ground taxiing time; arrival aircraft data: parking position, landing runway, landing time.
[0014] Furthermore, the aircraft type, aircraft take-off time, and ground taxiing time of each aircraft in the outbound aircraft data are taken as a set of data as the input of the heuristic rule algorithm; the parking position, landing runway entrance, and landing time corresponding to each aircraft in the inbound aircraft data are taken as a set of data as the input of conflict judgment.
[0015] Furthermore, the heuristic rule algorithm obtains the priority ranking of the departing aircraft through the following formula:
[0016]
[0017]
[0018] Among them, I j(t,l) indicates the priority index of the outbound aircraft operation j after the completion of operation l; j is the operation sequence number, which refers to the process of the outbound aircraft with the corresponding sequence number starting from the start taxiing time, following the standard taxiing path to the corresponding runway gate and then taking off; l indicates the index of the completed operation; w j represents the weight of job j, p j represents the ground taxiing time of the outbound aircraft corresponding to operation j;
[0019] d j represents the take-off time of the outbound aircraft corresponding to operation j, t represents the current time, represents the average ground taxiing time of departing aircraft, represents the average safety time interval between departing aircraft, s lj represents the safe time interval for job j to follow l, k1 and k2 are the look-ahead parameter and scaling parameter respectively; τ is the tightness factor, R is the range factor, η is the setting time severity factor, n is the number of jobs, is the average take-off time of outbound aircraft, β is the time-varying correlation coefficient, d max d min Respectively represent the aircraft take-off times of the latest and earliest departing aircraft;
[0020] The priority index I j (t,l) are sorted from large to small, which is the priority sorting of the corresponding departing aircraft.
[0021] Furthermore, the step of allocating runways to outbound aircraft in order includes:
[0022] First, create a departure aircraft set, which contains the relevant corresponding information of all departure aircraft to be sorted; obtain the priority index of each departure aircraft operation based on the heuristic rule algorithm, and then sort them; put the departure aircraft with the highest priority index into the planned aircraft set, then delete it from the original departure aircraft set, and change the existing time to the existing time plus the ground taxi time of the aircraft; recalculate the priority index of the remaining departure aircraft in the departure aircraft set until all departure aircraft are placed in the planned aircraft set in priority order, and then start to allocate runways for takeoff: first allocate a runway to the departure aircraft with the highest priority index. If there is an idle runway, that is, there is no aircraft ready to take off on the runway, it will be allocated to the nearest idle runway according to the path of the standard taxi path diagram, and then the runway will be allocated to the departure aircraft with the next highest priority index; if there is no idle runway, wait until the time interval with the aircraft on the non-idle runway is greater than the safety time interval before allocating it; repeat the above process until all departure aircraft are allocated the corresponding allocated runways and the generation of the scheduling plan is completed.
[0023] Furthermore, the population initialization process of the genetic algorithm includes:
[0024] Initialize an aircraft taxiing plan set according to the initial solution as the population of the genetic algorithm; the initial solution is in the form of {starting taxiing time 1, the runway assigned to outbound aircraft 1; starting taxiing time 2, the runway assigned to outbound aircraft 2...starting taxiing time n, the runway assigned to outbound aircraft n};
[0025] Based on the initial solution, randomly change the starting taxi time and the assigned runway of any number of outbound aircraft, and generate a new scheduling plan after each change. After a certain number of changes, all new scheduling plans and the initial solution are used as individuals in the initial population.
[0026] The method for each change is:
[0027] Any time within 15 minutes before the departure time of the outbound aircraft will be randomly selected as its new starting taxi time; one of the remaining runways in the same direction as the runway assigned to the outbound aircraft will be randomly selected as its reallocated runway.
[0028] Furthermore, the selection operation of the genetic algorithm includes:
[0029] The selection operation adopts the roulette selection method. The roulette wheel is divided into corresponding population sizes according to the proportion of individual fitness values. The size of the roulette wheel area is proportional to the individual fitness. The fitness calculation method is:
[0030] F=ω1g1(c1,c2,…,c 2i )+ω2g2(c1,c2,…,c 2i ) (7)
[0031] Among them, F represents the fitness, g1 represents the total number of conflict points during the taxiing process of outbound aircraft, g2 represents the number of delays of outbound aircraft, and I represents the total number of outbound aircraft; (c1, c2, …, c 2I ) represents an individual in the population, that is, the scheduling plan; where (c1, c2, …, c I ) represents the starting taxiing time of each departing aircraft in the scheduling plan, (c I+1 ,c I+2 ,…,c 2I ) represents the runway assigned to each departing aircraft in the scheduling plan; c i With c I+i is the value of the i-th position of the individual, c i is the initial taxiing time of outbound aircraft i, c I+i The runway assigned to the departing aircraft i; i = 1, 2, ... I; ω1 and ω2 represent the weights of the number of conflict points and the number of delayed aircraft respectively;
[0032] After the runway gate set e and parking stand set s of the incoming aircraft are known, the Dijkstra algorithm is used to solve the path, and then it is merged with the standard taxiing path of the outgoing aircraft. The above-mentioned method of determining whether there is a conflict is used to perform conflict avoidance, and finally the number of conflict points and the waiting time of each aircraft are obtained; then the ground taxiing time of the outgoing aircraft is taken out, and the time for each outgoing aircraft to arrive at the runway gate of the assigned runway is calculated. The time to arrive at the runway gate = c i +s ij / v+ taxi waiting time; when the departure aircraft arrives at the runway gate and the corresponding runway gate c I+i , and without changing the landing time and runway gate of the inbound aircraft, the outbound aircraft are scheduled on a first-come-first-served basis. After the scheduling is completed, the queue time of the outbound aircraft can be obtained, and the actual take-off time of the outbound aircraft can be obtained as follows = the time to arrive at the runway gate + the queue time at the runway gate + t setup , and then compare the actual take-off time with T pi In contrast, when greater than T pi Account for aircraft take-off delays and calculate the total number of departing aircraft delays;
[0033] The fitness of each individual is calculated, and the roulette wheel selection method is used for the entire population to select a certain number of new individuals to form a new population.
[0034] Furthermore, the method for determining the conflict point is:
[0035] Obtain the nodes passed by each inbound and outbound aircraft during taxiing and obtain the data of incoming aircraft; obtain a set of potential conflict points by finding duplicate points, and when there is a potential conflict point in the set, determine whether it is a conflict point; among them, the nodes passed by the outbound aircraft are determined by the nodes passed by all the standard taxiing paths of the corresponding runway; the nodes passed by the inbound aircraft are determined by the nodes passed by the shortest path from the landing runway to the parking position; the judgment method is: first, through the landing time and taxiing path of the inbound aircraft, the starting taxiing time and the standard taxiing path of the outbound aircraft, calculate the time for the inbound and outbound aircraft passing through the point to arrive at the point, and determine whether the time difference is greater than the minimum safe time interval. If it is less than the minimum safe interval, it is a conflict point; when the point is a conflict point, let the aircraft that arrives later wait to avoid it, and the time interval until the time of waiting to taxi through the conflict point is not less than the safe time interval.
[0036] Furthermore, the crossover operation of the genetic algorithm is:
[0037] Using the single-point crossover method, a point is randomly selected in the individual's coding string as the crossover point, and then the chromosome is split into two; the left genetic information of the newly generated daughter chromosome comes from one parent chromosome, and the right genetic information comes from the other parent chromosome;
[0038] The chromosome coding method adopts a double-layer coding structure. The first layer of coding is the starting taxi time of each aircraft, and the second layer of coding is the runway code assigned to each aircraft. The chromosome has a total of 2I digits, representing an outbound aircraft. The first I digits of the chromosome represent the starting taxi time of the outbound aircraft, and the last I digits represent the runway gate code of the aircraft.
[0039] Furthermore, the mutation operation of the genetic algorithm is:
[0040] Single-point mutation means that after confirming that a mutation operation is to be performed, a mutation point is randomly set and a random number within a specified range is generated to replace the value at the mutation point.
[0041] Compared with the prior art, the present invention has the following technical features:
[0042] The main innovation of the present invention is to perform runway scheduling for aircraft at large airports; the present invention can further reduce the number of conflict points and the number of flight delays during aircraft taxiing at the airport, further improve the safety of the airport and the punctuality of flights, and enable the airport to operate better. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 To generate a flow chart for the initial solution;
[0044] Figure 2 It is the flow chart of genetic algorithm;
[0045] Figure 3 is the chromosome coding map;
[0046] Figure 4 Display diagram for sliding conflict;
[0047] Figure 5 Calculation flow chart for the fitness function. DETAILED DESCRIPTION
[0048] The present invention provides a feasible runway selection scheme for outbound aircraft for the airport, that is, to solve the runway scheduling problem. The corresponding application background is: in terms of operation, the airport needs to arrange the departure of aircraft as well as the arrival of aircraft, and the inbound and outbound aircraft will have mutual influence during the taxiing process. During the taxiing process of inbound and outbound aircraft, the airport pays more attention to whether the number of conflict points in the taxiing process and the number of delayed outbound aircraft can be reduced. Therefore, the present invention focuses on reducing the number of conflict points and the number of delayed aircraft during the taxiing of inbound and outbound aircraft. The problems addressed by this scheme have the following characteristics:
[0049] 1) Departing aircraft can delay their departure, but the landing time of arriving aircraft cannot be adjusted; 2) Departing aircraft have a clear departure taxiing path, and by default they taxi along the standard taxiing path under normal circumstances; 3) Incoming aircraft can choose the fastest route to taxi back to the parking position after passing the center point of the runway; 4) The runway selection for arriving and departing aircraft must be strictly based on the wind direction of the day. Even if the road to take off to the north is free, the runway entrance to the north cannot be selected in the wind direction for takeoff to the south; 5) According to relevant literature, the wake turbulence safety interval between large aircraft in front and large aircraft in the back is set to 96s; large aircraft in front and small aircraft in the back The wake turbulence safety interval between the aircraft behind is 196s; the wake turbulence safety interval between the small aircraft in front and the large aircraft behind is 60s; the wake turbulence safety interval between the small aircraft in front and the small aircraft behind is 82s, which is used as the safety time interval; 6) When the actual take-off time of the aircraft is greater than the expected departure time, it is a delay; 7) Runway dispatch only dispatches the aircraft from the parking position to the runway entrance; 8) The default aircraft taxiing speed is constant and the same; 9) It is defaulted that there are no other obstacles to the aircraft's taxiing process except for taxiing conflicts; 10) The default preparation time for take-off of departing aircraft (such as straightening the fuselage) is constant at 2 minutes.
[0050] Since the runway scheduling problem is too complicated, it is difficult to find the optimal solution at present, and we can only try to solve it to obtain a better optimal solution. Therefore, the present invention combines the heuristic rule algorithm with the genetic algorithm, takes the optimal solution obtained by the heuristic rule algorithm as the initial solution, generates the parent population, and then iterates the solution through the genetic algorithm. Experiments have proved that this method can effectively reduce the occurrence of conflicts between aircraft during taxiing, further ensure the safety of aircraft during taxiing, and the reasonable outbound flight sequence and the initial taxiing time planning can reduce the waiting time caused by the busy runway and the inability to start taxiing in time, or the waiting time generated to avoid conflicts during taxiing, thereby reducing delays.
[0051] Referring to the accompanying drawings, a large airport runway scheduling method integrating rule triggering and genetic algorithm provided by the present invention comprises the following steps:
[0052] Step 1: Obtain the standard taxiing path diagram of the airport and all the data of aircraft arriving and departing from the airport in the same period of time, and perform preprocessing to remove irrelevant data.
[0053] In step 1, all relevant data for inbound and outbound aircraft include:
[0054] Departure aircraft data: parking position, assigned runway, aircraft type, aircraft take-off time, ground taxiing time; arrival aircraft data: parking position, landing runway, landing time.
[0055] For the pre-processed data, the aircraft type, take-off time, and ground taxiing time of each aircraft in the outbound aircraft data are taken as a set of data as the input of the heuristic rule algorithm in step 2; the parking position, landing runway gate, and landing time corresponding to each aircraft in the inbound aircraft data are taken as a set of data as one of the inputs of the conflict judgment in step 4. Among them, the runway gate is used for take-off when outbound, and the runway gate is used for landing when inbound.
[0056] Each of the above data sets is data corresponding to one aircraft, that is, within the selected time period, for the outbound aircraft data, there are as many data sets as there are aircraft; the same is true for the inbound aircraft data.
[0057] Step 2, using a heuristic rule algorithm to sort each set of data of the outbound aircraft obtained in step 1 into a priority sequence of the outbound aircraft, and obtain the priority ranking of each outbound aircraft, as shown below.
[0058] The safe time interval between departing aircraft is used as the setting time (for calculating the average setting time), the average value of the original aircraft ground taxiing time provided by the airport is used as the average processing time, and the expiration time is set to the aircraft take-off time set by the airport. Since the relevant calculation formula needs to take into account the importance of each departing aircraft, that is, the weight to calculate the priority of each aircraft, and the airport has no special instructions for this, the weight of each aircraft is set to 1 in this plan, and this parameter is used to calculate the aircraft priority.
[0059] The heuristic rule algorithm obtains the priority ranking of departing aircraft through the following formula:
[0060]
[0061] The variables in the above formula are interpreted as follows:
[0062]
[0063]
[0064]
[0065] Among them, due to the variable τ, C is used in R max variables, and because C max Difficult to estimate, so use an estimator To express.
[0066] The above formula is used to calculate the priority index I of the next outbound aircraft operation j after the completion of operation l. j (t,l), sort the indexes from large to small, which is the priority sorting of the corresponding departing aircraft.
[0067] Step 3: sort the priorities of the outbound aircraft, and assign idle runways to them according to the order. The corresponding starting taxi time is set to 15 minutes before the aircraft takes off. The situation after the allocation is used as the scheduling plan, which includes the runways and starting taxi times assigned to each outbound aircraft, as the initial plan for the subsequent algorithm; the corresponding process is as follows: Figure 1 The specific allocation algorithm is described as follows:
[0068] First, create a departure aircraft set, which contains the relevant corresponding information of all departure aircraft to be sorted; obtain the priority index of each departure aircraft operation based on the heuristic rule algorithm, and then sort them; put the departure aircraft with the highest priority index into the planned aircraft set, and then delete it from the original departure aircraft set, and change the existing time to the existing time plus the ground taxiing time of the aircraft; recalculate the priority index of the remaining departure aircraft in the departure aircraft set until all departure aircraft are placed in the planned aircraft set in priority order, and then start allocating runways for takeoff: first allocate a runway to the departure aircraft with the highest priority index. If there is an idle runway, that is, there is no aircraft ready to take off on the runway, then according to The path of the standard taxi path diagram is assigned to the nearest idle runway, and then the runway is assigned to the outbound aircraft with the next highest priority index; if there is no idle runway, it will be assigned until the time interval with the aircraft on the non-idle runway is greater than the safety time interval. This safety time interval is related to the size of the two aircraft, that is, the setting time between them is obtained by comparing the size of the aircraft type on the current runway with the size of the aircraft type to be departed. When the starting taxi time interval of the two aircraft is greater than the setting time, it is considered that the two aircraft will not affect each other. At this time, the runway corresponding to the aircraft to be departed can be considered as an idle runway for runway allocation; repeat the above process until all outbound aircraft are assigned the corresponding allocated runways and the generation of the scheduling plan is completed.
[0069] Step 4: Use the generated scheduling plan as the initial solution, generate a certain number of populations, and put them into the genetic algorithm for iterative calculation. Finally, the optimal individual is obtained through the genetic algorithm, which is the final scheduling plan.
[0070] Step 4.1, initialize the population
[0071] An aircraft taxiing plan set is initialized according to the initial solution as the population of the genetic algorithm; the initial solution is in the form of {starting taxiing time 1, the runway assigned to outbound aircraft 1; starting taxiing time 2, the runway assigned to outbound aircraft 2…starting taxiing time n, the runway assigned to outbound aircraft n}; considering that the starting taxiing time cannot be later than the take-off time, the starting taxiing time is obtained by randomly selecting a value from 15 minutes before the aircraft's take-off time to the aircraft's take-off time, and based on airport restrictions, outbound aircraft in each direction have only a few fixed runways for take-off and departure.
[0072] Therefore, the operation of initializing the population is to randomly change the starting taxi time and the assigned runway of any number of outbound aircraft based on the original initial solution, and generate a new scheduling plan after each change. After a certain number of changes, all new scheduling plans and the initial solution are used as individuals of the initial population.
[0073] The method for each change is:
[0074] Any time within 15 minutes before the departure time of the outbound aircraft will be randomly selected as its new starting taxi time; one of the remaining runways in the same direction as the runway assigned to the outbound aircraft will be randomly selected as its reallocated runway.
[0075] Step 4.2, selection, mutation, crossover, iteration
[0076] Based on the initial population obtained in step 4.1, perform selection, mutation, crossover, and iterative genetic operations.
[0077] Step 4.2.1, select
[0078] The selection operation adopts the roulette selection method. The roulette wheel divides the roulette wheel into the corresponding population size according to the proportion of individual fitness values. The size of the roulette wheel area is proportional to the individual fitness value, that is, the higher the individual fitness, the larger the roulette wheel area it occupies, and the probability of being selected increases accordingly. Before introducing fitness, we must first introduce conflict judgment, which plays a vital role in the fitness calculation of the present invention. The introduction to conflict judgment is as follows:
[0079] Conflicts on airport taxiways are mainly manifested in conflicts between the taxi paths of aircraft, such as overlapping or crossing taxi paths. Figure 4 shown.
[0080] Among them, overtaking conflict refers to the possibility that the following aircraft overtakes the preceding aircraft when they pass the same taxiway; head-on conflict refers to the conflict caused by two aircraft driving from opposite directions on the same taxiway. Since the airport adopts a one-way taxiing method on the taxiway, this type of conflict is not considered. In the calculation, each aircraft is controlled to taxi in the direction allowed by the taxiway. Crossing conflict refers to the conflict that may be caused when the taxiing paths of two aircraft have intersecting nodes.
[0081] If a conflict occurs, avoidance should be carried out in the following ways:
[0082] Obtain the nodes passed by each inbound and outbound aircraft during taxiing and obtain the data of the inbound aircraft; obtain a set of potential conflict points by finding duplicate points, and when there is a potential conflict point in the set, determine whether it is a conflict point; among them, the nodes passed by the outbound aircraft are determined by the nodes passed by all the standard taxiing paths of the corresponding runway; the nodes passed by the inbound aircraft are determined by the nodes passed by the shortest path from the landing runway to the parking position; the judgment method is: first, through the landing time and taxiing path of the inbound aircraft, the starting taxiing time and standard taxiing path of the outbound aircraft, calculate the time when the inbound and outbound aircraft passing through the point arrive at the point respectively, and determine whether the time difference is greater than the minimum safe time interval. If it is less than the minimum safe interval, it is a conflict point; when the point is a conflict point, let the aircraft that arrives later wait to avoid it, and the time interval until the time of waiting to taxi through the conflict point is not less than the safe time interval; if the waiting time is too long and exceeds the preset time threshold, resulting in the aircraft taking off later than the aircraft take-off time, it is also considered a delay.
[0083] After clarifying the conflict and the conflict point, the fitness can be calculated. The fitness calculation method is as follows:
[0084] The fitness function can be used to estimate each individual in the population. The individual fitness function defined in this scheme is shown in formula (7):
[0085] F=ω1g1(c1,c2,…,c 2I )+ω2g2(c1,c2,…,c 2I ) (7)
[0086] Among them, F represents the fitness, g1 represents the total number of conflict points during the taxiing process of outbound aircraft, g2 represents the number of delays of outbound aircraft, and I represents the total number of outbound aircraft; (c1, c2, …, c 2I ) represents an individual in the population, that is, the scheduling plan; where (c1, c2, …, c I ) represents the starting taxiing time of each departing aircraft in the scheduling plan, (c I+1 ,c I+2 ,…,c 2I ) represents the runway assigned to each departing aircraft in the scheduling plan; that is, c i With c I+i is the value of the individual ith position, c i is the initial taxiing time of outbound aircraft i, c I+i The runway allocated to the departing aircraft i; i = 1, 2, … I; ω1 and ω2 represent the weights of the number of conflict points and the number of delayed aircraft, respectively.
[0087] According to the known data of inbound and outbound aircraft, the standard taxiing path of outbound aircraft i to runway j can be known from the parking positions and assigned runways of outbound aircraft as r ij and the corresponding distance s ij , the departure aircraft is scheduled to take off at T pi , the taxiing speed of the glide vehicle is v, and the runway preparation time for takeoff is t setup , the maximum safe wake interval is t wake , the runway set of the incoming aircraft is e, the landing time set is t, the parking position set is s, and the taxiing time of the incoming aircraft on the runway is t land .
[0088] The process of calculating the fitness function is as follows Figure 5 As shown in FIG. 1 , in order to consider the impact of the incoming aircraft on the outgoing aircraft during the ground taxiing stage, after the runway gate set e and the parking stand set s of the incoming aircraft are known, the Dijkstra algorithm is used to solve the path, which is then merged with the standard taxiing path of the outgoing aircraft. The above-mentioned method of determining whether there is a conflict is used to perform conflict avoidance, and finally the number of conflict points and the waiting time of each aircraft are obtained; then the ground taxiing time of the outgoing aircraft is taken to calculate the time for each outgoing aircraft to arrive at the runway gate of the assigned runway, that is, the time to arrive at the runway gate = c i +s ij / v+ taxi waiting time; when the departure aircraft arrives at the runway gate and the corresponding runway gate c I+i , and without changing the landing time and runway gate of the inbound aircraft, the outbound aircraft are scheduled on a first-come-first-served basis. The aircraft wake safety interval time needs to be considered during the scheduling process. After the scheduling is completed, the queuing time of the outbound aircraft can be obtained. Finally, the actual take-off time of the outbound aircraft = the time to arrive at the runway gate + the queuing time at the runway gate + t setup , and then compare the actual take-off time with T pi In contrast, when greater than T pi Explain the aircraft take-off delay, calculate the total departure aircraft delay, and finally add the weighted number of conflict points and flight delay to get the fitness value.
[0089] Through the above method, the fitness of each individual can be obtained. By using the roulette selection method for the entire population, a certain number of new individuals can be selected to form a new population.
[0090] Step 4.2.2, Cross
[0091] The crossover operation is a specific method that exchanges some genes (the starting taxi time of an outbound aircraft, the assigned runway) between two individuals to create two new individuals.
[0092] In this scheme, a single-point crossover method is used, that is, a point is randomly selected in the individual's coding string as the crossover point, and then the chromosome is split into two; the left genetic information of the newly generated offspring chromosome comes from one parent chromosome, and the right genetic information comes from another parent chromosome. The chromosome coding information of this scheme is as follows:
[0093] The chromosome coding method adopts a double-layer coding structure. The first layer of coding is the starting taxi time of each aircraft, and the second layer of coding is the runway code of each aircraft, such as Figure 3 As shown, the chromosome has 2I digits, representing one outbound aircraft. The first I digits of the chromosome represent the start taxiing time of the outbound aircraft, and the last I digits represent the runway gate code of the aircraft.
[0094] Step 4.2.3, mutation
[0095] The mutation operation is a process in which a new individual is formed after the crossover operation, and there is a certain probability that genetic mutation will occur. The present invention adopts single-point mutation, which means that after confirming that a mutation operation is to be performed, a mutation point is randomly set and a random number within a specified range is generated to replace the value at the mutation point.
[0096] Step 4: Specific flow chart as follows Figure 2 As shown; except for the above-mentioned special instructions, the rest of the genetic algorithm in this scheme is prior art and will not be described in detail.
[0097] Example:
[0098] The scheme of the present invention compares the existing runway scheduling with the new runway scheduling algorithm used in the invention for the conflict points, number of delayed aircraft and taxiing time of two sets of standardized flight paths of the airport.
[0099] The number of conflict points, delayed aircraft and taxiing time have all been significantly reduced, which not only effectively meets the airport's requirements for safety and reduced delays, but also improves airport efficiency and reduces the time required for taxiing. The specific comparison is shown below:
[0100] Table 1 Comparison of conflict points
[0101] Existing runway scheduling plan This program Road Network 1 0.296 0.259 Road Network 2 0.333 0.296
[0102] Table 2 Comparison of delayed flights
[0103] Existing runway scheduling plan This program Road Network 1 1.037 0.222 Road Network 2 0.963 0.296
[0104] Table 3 Comparison of taxiing time
[0105]
[0106]
[0107] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A large airport runway scheduling method integrating rule triggering and genetic algorithm, characterized in that: include: Obtain the airport's standard taxi path diagram and all data on aircraft arriving and departing from the airport during the same period and perform pre-processing; Use the heuristic rule algorithm to sort the priority sequence of outbound aircraft for each set of data, and obtain the priority ranking of each outbound aircraft; Prioritize each outbound aircraft, assign runways to them according to the priority, and set the corresponding start taxi time. The situation after the assignment is used as the scheduling plan; The generated scheduling plan is used as the initial solution, a population is generated based on the initial solution, and iterative calculations are performed through the genetic algorithm. Finally, the optimal individual is obtained through the genetic algorithm, which is the final scheduling plan.
2. The large airport runway scheduling method based on the fusion rule triggering and genetic algorithm according to claim 1 is characterized in that: The preprocessing is an irrelevant data removal operation; The preprocessed data includes: Departure aircraft data: parking position, assigned runway, aircraft type, aircraft take-off time, ground taxiing time; Arrival aircraft data: parking position, landing runway, landing time.
3. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 1 is characterized in that: The aircraft type, take-off time, and ground taxiing time of each aircraft in the outbound aircraft data are taken as a set of data as the input of the heuristic rule algorithm; the parking position, landing runway entrance, and landing time corresponding to each aircraft in the inbound aircraft data are taken as a set of data as the input of conflict judgment.
4. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 1 is characterized in that: The heuristic rule algorithm obtains the priority ranking of departing aircraft through the following formula: Among them, I j (t,l) indicates the priority index of the outbound aircraft operation j after the completion of operation l; j is the operation sequence number, which refers to the process of the outbound aircraft with the corresponding sequence number starting from the start taxiing time, following the standard taxiing path to the corresponding runway gate and then taking off; l indicates the index of the completed operation; w j represents the weight of job j, p j represents the ground taxiing time of the outbound aircraft corresponding to operation j; d j represents the take-off time of the outbound aircraft corresponding to operation j, t represents the current time, represents the average ground taxiing time of departing aircraft, represents the average safety time interval between departing aircraft, s lj represents the safe time interval of operation j following l, k1 and k2 are the look-ahead parameter and the scaling parameter respectively; τ is the tightness factor, R is the range factor, η is the setting time severity factor, n is the number of operations, d is the average take-off time of outbound aircraft, β is the variation time correlation coefficient, d max ,d min Respectively represent the aircraft take-off times of the latest and earliest departing aircraft; The priority index I j (t,l) are sorted from large to small, which is the priority sorting of the corresponding departing aircraft.
5. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 1 is characterized in that: The runway allocation to departing aircraft according to the sequence includes: First, create a departure aircraft set, which contains the relevant corresponding information of all departure aircraft to be sorted; obtain the priority index of each departure aircraft operation based on the heuristic rule algorithm, and then sort them; put the departure aircraft with the highest priority index into the planned aircraft set, then delete it from the original departure aircraft set, and change the existing time to the existing time plus the ground taxi time of the aircraft; recalculate the priority index of the remaining departure aircraft in the departure aircraft set until all departure aircraft are placed in the planned aircraft set in priority order, and then start to allocate runways for takeoff: first allocate a runway to the departure aircraft with the highest priority index. If there is an idle runway, that is, there is no aircraft ready to take off on the runway, it will be allocated to the nearest idle runway according to the path of the standard taxi path diagram, and then the runway will be allocated to the departure aircraft with the next highest priority index; if there is no idle runway, wait until the time interval with the aircraft on the non-idle runway is greater than the safety time interval before allocating it; repeat the above process until all departure aircraft are allocated the corresponding allocated runways and the generation of the scheduling plan is completed.
6. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 1 is characterized in that: The population initialization process of the genetic algorithm includes: Initialize an aircraft taxiing plan set according to the initial solution as the population of the genetic algorithm; the initial solution is in the form of {starting taxiing time 1, the runway assigned to outbound aircraft 1; starting taxiing time 2, the runway assigned to outbound aircraft 2...starting taxiing time n, the runway assigned to outbound aircraft n}; Based on the initial solution, randomly change the starting taxi time and the assigned runway of any number of outbound aircraft, and generate a new scheduling plan after each change. After a certain number of changes, all new scheduling plans and the initial solution are used as individuals in the initial population. The method for each change is: Any time within 15 minutes before the departure time of the outbound aircraft will be randomly selected as its new starting taxi time; one of the remaining runways in the same direction as the runway assigned to the outbound aircraft will be randomly selected as its reallocated runway.
7. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 1 is characterized in that: The selection operation of the genetic algorithm includes: The selection operation adopts the roulette selection method. The roulette wheel is divided into corresponding population sizes according to the proportion of individual fitness values. The size of the roulette wheel area is proportional to the individual fitness. The fitness calculation method is: F=ω1g1(c1,c2,…,c 2I )+ω2g2(c1,c2,…,c 2I ) (7) Among them, F represents the fitness, g1 represents the total number of conflict points during the taxiing process of outbound aircraft, g2 represents the number of delays of outbound aircraft, and I represents the total number of outbound aircraft; (c1, c2, …, c 2I ) represents an individual in the population, that is, the scheduling plan; where (c1, c2, …, c I ) represents the starting taxiing time of each departing aircraft in the scheduling plan, (c I+1 ,c I+2 ,…,c 2I ) represents the runway assigned to each departing aircraft in the scheduling plan; c i With c I+i is the value of the individual ith position, c i is the initial taxiing time of outbound aircraft i, c I+i The runway assigned to the departing aircraft i; i = 1, 2, ... I; ω1 and ω2 represent the weights of the number of conflict points and the number of delayed aircraft respectively; After knowing the runway gate set e and the parking stand set s of the incoming aircraft, use the Dijkstra algorithm to solve the path, and then merge it with the standard taxiing path of the outgoing aircraft. Use the above-mentioned method to determine whether there is a conflict to avoid the conflict, and finally get the number of conflict points and the waiting time of each aircraft; then take the ground taxiing time of the outgoing aircraft, calculate the time for each outgoing aircraft to arrive at the runway gate of the assigned runway, and the time to arrive at the runway gate = c i +s ij / v+ taxi waiting time; when the departure aircraft arrives at the runway gate and the corresponding runway gate c I+i , and without changing the landing time and runway gate of the inbound aircraft, the outbound aircraft are scheduled on a first-come-first-served basis. After the scheduling is completed, the queue time of the outbound aircraft can be obtained, and the actual take-off time of the outbound aircraft can be obtained as follows = the time to arrive at the runway gate + the queue time at the runway gate + t setup , and then compare the actual take-off time with T pi In contrast, when greater than T pi Account for aircraft take-off delays and calculate the total number of departing aircraft delays; The fitness of each individual is calculated, and the roulette wheel selection method is used for the entire population to select a certain number of new individuals to form a new population.
8. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 7 is characterized in that: The method to determine the conflict point is: Obtain the nodes passed by each inbound and outbound aircraft during taxiing and obtain the data of incoming aircraft; obtain a set of potential conflict points by finding duplicate points, and when there is a potential conflict point in the set, determine whether it is a conflict point; among them, the nodes passed by the outbound aircraft are determined by the nodes passed by all the standard taxiing paths of the corresponding runway; the nodes passed by the inbound aircraft are determined by the nodes passed by the shortest path from the landing runway to the parking position; the judgment method is: first, through the landing time and taxiing path of the inbound aircraft, the starting taxiing time and the standard taxiing path of the outbound aircraft, calculate the time for the inbound and outbound aircraft passing through the point to arrive at the point, and determine whether the time difference is greater than the minimum safe time interval. If it is less than the minimum safe interval, it is a conflict point; when the point is a conflict point, let the aircraft that arrives later wait to avoid it, and the time interval until the time of waiting to taxi through the conflict point is not less than the safe time interval.
9. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 1 is characterized in that: The crossover operation of the genetic algorithm is: Using the single-point crossover method, a point is randomly selected in the individual's coding string as the crossover point, and then the chromosome is split into two; the left genetic information of the newly generated daughter chromosome comes from one parent chromosome, and the right genetic information comes from the other parent chromosome; The chromosome coding method adopts a double-layer coding structure. The first layer of coding is the starting taxi time of each aircraft, and the second layer of coding is the runway code assigned to each aircraft. The chromosome has a total of 2I digits, representing an outbound aircraft. The first I digits of the chromosome represent the starting taxi time of the outbound aircraft, and the last I digits represent the runway gate code of the aircraft.
10. The large airport runway scheduling method integrating rule triggering and genetic algorithm according to claim 1 is characterized in that: The mutation operation of the genetic algorithm is: Single-point mutation means that after confirming that a mutation operation is to be performed, a mutation point is randomly set and a random number within a specified range is generated to replace the value at the mutation point.
Citation Information
Patent Citations
Aircraft scene joint scheduling method, device and equipment and storage medium
CN116168566A
Method and device for determining taxiing path of aircraft and storage medium
CN117848343A
Aircraft taxiing scheduling method and device considering runway crossing
CN117852829A
Device for allotting aircraft takeoff and landing order
JP2002197599A
Cited By
Method and system for dispatching multi-machine collaborative electromagnetic ejection in take-off and landing field
CN122116697A
An electromagnetic catapult scheduling method and system for multiple aircrafts on a runway
CN122116697B
