Airport task allocation and optimization system and method based on Internet of Things

Abstract

The invention relates to the technical field of task allocation, and discloses an airport task allocation and optimization system and method based on the Internet of Things. The method comprises the following steps: firstly, acquiring airport ground support task time information to obtain an airport task time set, and deleting abnormal airport task time data by using an outlier factor detection algorithm; constructing an objective function of an airport task allocation model, establishing the airport task allocation model, and constraining the airport task allocation model; taking the airport task allocation model as a fitness function, and allocating and optimizing airport tasks by using a football team training algorithm under the constraint condition of the airport task allocation model; and finally, according to the flight delay condition, the airport tasks are distributed again, and an airport task distribution scheme is dynamically adjusted. According to the method, the airport tasks are allocated and optimized by constructing the airport task allocation model and using a novel meta-heuristic algorithm, and the method is accurate and novel.

Description

Technical Field

[0001] The present invention relates to the technical field of task allocation, and in particular to an airport task allocation and optimization system and method based on the Internet of Things. Background Art

[0002] Chinese patent CN111126667A discloses a vehicle task unloading and resource allocation method and system based on active topology optimization, characterized in that the steps specifically include obtaining vehicle information of the parking area and business information to be processed, wherein the vehicle information of the parking area includes parked position information and unparked position information; combining all parked position information with each unparked position information respectively, constructing a vehicle cloud model, and obtaining a first adjacency matrix corresponding to each vehicle cloud model; then constructing a business model based on the business information to be processed, and obtaining a second adjacency matrix corresponding to the business model; judging whether the model matches according to the first adjacency matrix and the second adjacency matrix corresponding to each vehicle cloud model; when the model matches, calculating the utility function value corresponding to the vehicle cloud model that matches the business model, obtaining the unparked position information corresponding to the minimum utility function value from all the utility function values, and making the unparked position information correspond to the vehicle task unloading. This invention does not achieve task optimization, and the task execution efficiency is not high.

[0003] Traditional task allocation methods often have problems such as slow task allocation, low staff utilization, and unreasonable allocation methods, and the allocation process is complicated. At the same time, when optimizing task allocation, high-tech technologies such as neural networks and deep learning are not used, the data processing speed is slow, and the optimization strategy cannot reach the optimal level. Summary of the invention

[0004] In response to the problems in the related technology, the present invention provides an airport task allocation and optimization system and method based on the Internet of Things to overcome the above-mentioned technical problems existing in the existing related technology.

[0005] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention is an airport task allocation and optimization method based on the Internet of Things, comprising the following steps:

[0007] S1. Acquire airport ground support mission time information within a period of time, record it as an airport mission time set, delete abnormal data from the airport mission time data in the airport mission time set based on an outlier factor detection algorithm, and obtain a processed airport mission time set;

[0008] S2. Based on the airport service quality, the enthusiasm of airport staff and the fairness of airport task allocation, construct the objective function of the airport task allocation model, establish the airport task allocation model, and constrain the airport task allocation model to obtain the constraint conditions of the airport task allocation model;

[0009] S3, encoding the airport tasks, taking the airport task allocation model as the fitness function, and using the football team training algorithm to allocate the airport tasks under the constraints of the airport task allocation model, finding the best fitness function value, and completing the airport task allocation and optimization;

[0010] S4. Based on the data from the civil aviation route management platform, airport tasks are reallocated in response to flight delays, and the airport task allocation plan is dynamically adjusted.

[0011] The invention obtains airport ground support task time information, generates an airport task time set, uses an outlier factor detection algorithm, detects the local reachable density around the airport task time data, thereby obtaining the local outlier factor, and deletes the airport task time data that does not meet the threshold; the algorithm achieves the purpose of anomaly detection by calculating the density near the data point, can effectively delete abnormal data, and has high detection accuracy; secondly, based on the airport service quality, the enthusiasm of airport staff and the fairness of airport task allocation, respectively establish an airport service quality objective function, an airport staff enthusiasm objective function and an airport task allocation fairness objective function, construct an airport task allocation model, and constrain the airport task allocation model; the task allocation problem can be quantitatively analyzed to facilitate task allocation and optimization; then the airport task allocation model is used as a fitness function, and under the constraints of the airport task allocation model, the football team training algorithm is used to allocate airport tasks, complete airport task allocation and optimization, and optimize the problem by simulating the football training process; the algorithm has a fast search speed and a strong optimization performance; finally, in response to flight delays, the airport tasks are reallocated and the airport task allocation plan is dynamically adjusted; the method records the process of each dynamic adjustment in detail to facilitate subsequent analysis and improvement.

[0012] Preferably, the S1 comprises the following steps:

[0013] S11. During the process from landing to take-off, the airport ground support tasks before, during, after, before and during take-off are obtained. The tasks before landing include allocating aprons, the tasks during landing include guiding the aircraft into position, the tasks after landing include disembarkation guidance, special vehicle support, the tasks before take-off include refueling and power supply, and the tasks during take-off include pushing the aircraft, etc.; the time information of all airport ground support tasks within a period of time is recorded, which is recorded as the airport task time set A = {(a1,b1),(a2,b2),(a3,b3),...,(a m ,b m )}, where (a m ,b m ) represents the time data of the mth airport mission, including the start execution time and the completion time;

[0014] S12, in the airport mission time set, select the m1th airport mission time data, set an arbitrary positive integer k, record the Euclidean distance between the kth nearest neighbor data of the m1th airport mission time data as the k-neighbor distance of m1, and keep the k airport mission time data in the airport mission time set; wherein the Euclidean distance between the k airport mission time data and the m1th airport mission time data is less than or equal to the k-neighbor distance of m1, use the outlier factor detection algorithm to detect abnormal airport mission time data in the airport mission time set, and delete the abnormal airport mission time data to obtain the processed airport mission time set, the specific steps are as follows:

[0015] S121, calculate the Euclidean distance between the airport mission time data in the airport mission time set and the m1th airport mission time data, record it as the k-neighborhood distance, add the airport mission time data corresponding to the k-neighborhood distance less than or equal to m1 to the neighborhood set to obtain the neighborhood set; set the Euclidean distance between the m1th airport mission time data and the i1th airport mission time data in the neighborhood set as the reachable distance The local reachable density calculation formula is as follows:

[0016]

[0017] in, represents the local reachability density of the m1th airport mission time data, b represents the number of airport mission time data in the neighborhood set, i1=1,2,3,...,b;

[0018] S122. Calculate the local reachability density of the mission time data of the i1th airport in the neighborhood set. The calculation formula of the local outlier factor is as follows:

[0019]

[0020] in, represents the local outlier factor of the mission time data of the m1th airport, represents the local reachability density of the mission time data of the i1th airport in the neighborhood set;

[0021] The local outlier factors of all airport mission time data in the neighborhood set are calculated to obtain a local outlier factor set, the local outlier factors in the local outlier factor set are sorted in ascending order, the local outlier factor threshold is set to ω, the local outlier factors in the local outlier factor set that are greater than the local outlier factor threshold are recorded as abnormal local outlier factors, the airport mission time data corresponding to the abnormal local outlier factors in the airport mission time set are deleted, and the processed airport mission time set is obtained.

[0022] The invention uses an outlier factor detection algorithm on the airport mission time set, detects the local reachable density around the airport mission time data, and obtains the local outlier factor. By comparing the threshold, the airport mission time data that does not meet the requirements is deleted. The algorithm achieves the purpose of anomaly detection by calculating the density near the data point, and can effectively delete abnormal data. The algorithm has good performance and high detection accuracy.

[0023] Preferably, S2 comprises the following steps:

[0024] S21, obtaining the airport task execution start time and the airport task completion time according to the processed airport task time set; setting the airport task set to A′={a′1, a′2, a′3, ..., a′ m′}, where a′ m′ represents the m′th airport task, and the airport staff set is A″={c1,c2,c3,...,c n}, where c n represents airport staff member n; assign the airport tasks in the airport task set to the airport staff members in the airport staff set, and set the airport task allocation coefficient to χ. When the airport task allocation coefficient is 1, it means that the m″th airport task is assigned to the airport staff member n′; when the airport task allocation coefficient is 0, it means that the m″th airport task is not assigned to the airport staff member n′; B(m″,n′) represents the ability matching result of the airport staff member n′ to the m″th airport task, and the airport service quality objective function is established as follows:

[0025]

[0026] Among them, F1 represents the airport service quality objective function;

[0027] Assume that B′(m″,n′) represents the interest matching result of airport staff n′ for the m″th airport task, and establish the airport staff motivation objective function as follows:

[0028]

[0029] Among them, F2 represents the airport staff motivation objective function;

[0030] Assume that B″(m″,n′) represents the load of airport worker n′ on the m″th airport task, and the average load of all airport workers in the airport worker set is The objective function of airport task allocation fairness is established as follows:

[0031]

[0032] Among them, F2 represents the fairness objective function of airport task allocation;

[0033] S22, making the airport service quality objective function and the airport staff enthusiasm objective function at the maximum value, giving the airport service quality objective function a weight δ1, giving the airport staff enthusiasm objective function a weight δ2, and δ1+δ2=1; and converting the airport task allocation fairness objective function into a constraint condition, and establishing an airport task allocation model as follows:

[0034]

[0035] Among them, F represents the airport task allocation model;

[0036] S23. Set the number of resources owned by airport staff member n' to d n′ , the resource consumed by the m″th airport task assigned to airport staff n′ is d′(m″,n′), and the airport staff resource constraint is established Set the rated maximum completion time to The airport mission starts at The airport mission will be completed in Establishing time constraints for airport missions Select airport worker n″ from the airport worker set, set the airport task start execution time and airport task completion time of airport worker n′ respectively and The airport task start execution time and airport task completion time of airport staff n″ are and Establishing airport mission time conflict constraints The airport staff resource constraints, airport task time constraints and airport task time conflict constraints constitute the constraints of the airport task allocation model.

[0037] The invention establishes an airport service quality objective function, an airport staff enthusiasm objective function and an airport task allocation fairness objective function based on airport service quality, airport staff enthusiasm and airport task allocation fairness, constructs an airport task allocation model, and sets constraints on the airport task allocation model; mathematical modeling is used to facilitate the solution of the intelligent optimization algorithm.

[0038] Preferably, S3 comprises the following steps:

[0039] S31, encode the airport tasks in the airport task set to obtain an airport task code string A″′={a″1, a″2, a″3, ..., a″ m′}, where a″ m′ Represents the code corresponding to the m′th airport task, combined with the staff set, we get the coding function C(n)=a″ m′ , the encoding function represents assigning the m′th airport task to the airport staff member n; taking the airport task assignment model as the fitness function, and using the football team training algorithm to find the optimal fitness function value under the constraints of the airport task assignment model, the specific steps are as follows:

[0040] S311, generate a random code string, use the coding function to represent the g-th football player in the e-th iteration of the football team, set the number of football players in the football team to be equal to the number of airport tasks, set the dimension of football players in the football team to be equal to the number of staff, the current number of iterations is e, and the g-th football player in the current e-th iteration is Where i represents the football player number, i.e., the code corresponding to the airport task, j represents the football player dimension, corresponding to the airport staff, and the g-th football player in the e-th iteration after training is The best trained football player in the football team is D e (best, j), g1 represents a random number and g1∈[0,1]. The calculation formula for the g-th football player in the e-th iteration after training is as follows:

[0041]

[0042] During the collective training of the football team, the football players in the football team try their best to catch up with the best-trained football players while avoiding becoming the worst-trained football players. The worst-trained football players in the football team are set as D. e (worst, j), g2 represents a random number and g2∈[0,1], g3 represents a random number and g3∈[0,1], then the g-th football player of the e-th iteration after training is updated, and the calculation formula is as follows:

[0043]

[0044] Assume that the state fluctuation of the football players in the football team becomes smaller and smaller as the number of training times increases. Use the normal distribution h(e) to represent the state fluctuation of the football players in the football team as the number of iterations increases. Update the g-th football player in the e-th iteration after training again. The calculation formula is as follows:

[0045]

[0046] In the iterative process, the best fitness function value is continuously sought, and the current football player number and the current football player dimension corresponding to the best fitness function value are found. At this time, the current football player number corresponds to the code corresponding to the current airport task, and the current football player dimension corresponds to the current airport staff. The code corresponding to the current airport task is assigned to the current airport staff.

[0047] S312, after the collective training, the football team enters the collective training process, and the football players in the football team are grouped according to the dimension, and divided into forward group, midfield group, defender group and goalkeeper group. Different groups can communicate and learn, including optimal learning, random learning and random communication; set the learning probability to ε1, the communication probability to ε2, and in the optimal learning stage, set the best football player in group h1 to Any football player in group h1 is Where j represents the best football player dimension, e represents the current iteration number e, and the calculation formula for any football player is as follows:

[0048]

[0049] In the random learning stage, any football player is selected from the h1th group again, and the football player is updated again according to the learning probability and the communication probability, to obtain the updated football player;

[0050] In the random communication stage, the communication probability is set to ε3, g4 represents a random number and g4∈[0,1], when the communication probability is less than or equal to g4, a normal distribution random number is set, and any football player is updated using the normal distribution random number to obtain an updated football player; when the communication probability is greater than g4, any football player is directly updated;

[0051] S313. After the collective training, the football team enters the individual training process, performs additional individual training on the updated football players, sets the additional individual training coefficients to φ1 and φ2, selects the best football player from the updated football players, and records it as In the process of continuous individual training, the best football players are updated and D e(best), the calculation formula is as follows:

[0052]

[0053] Set the maximum number of iterations to E, and continuously iterate the football players in the football team, while continuously searching for the best fitness function value. When the current number of iterations reaches the maximum number of iterations, stop iterating and get the final excellent football player corresponding to the best fitness function value.

[0054] S32. According to the final excellent football player, find the corresponding final excellent football player number and final excellent football player dimension. At this time, the final excellent football player number is the code corresponding to the airport task, and the final excellent football player dimension corresponds to the airport staff. The code corresponding to the airport task is assigned to the airport staff to complete the airport task allocation and optimization.

[0055] The invention uses the airport task allocation model as the fitness function and, under the constraints of the airport task allocation model, uses the football team training algorithm to allocate airport tasks, and completes the airport task allocation and optimization by finding the best fitness function value; the algorithm optimizes by simulating the football training process, has a fast search speed, and has strong optimization performance.

[0056] Preferably, S4 comprises the following steps:

[0057] S41. Collect data from the civil aviation route management platform, including flight status, resource availability, airport task progress, and environmental changes, monitor flight delays in flight status, set a time threshold ξ, and when the difference between the flight rated time and the flight estimated arrival time is greater than the time threshold, the flight is delayed; otherwise, the flight is not delayed, and the civil aviation route management platform is used to verify and identify the delay;

[0058] S42. In response to flight delays, the airport task allocation model and the airport task allocation model constraints are updated, the airport tasks are reallocated, a new airport task allocation plan is generated, and the new airport task allocation plan is evaluated.

[0059] This embodiment also discloses a system for airport task allocation and optimization method based on the Internet of Things, which specifically includes: an airport task data processing module, an airport task allocation model and constraint condition establishment module, an airport task allocation and optimization module, and an airport task allocation dynamic adjustment module;

[0060] The airport mission data processing module is used to delete abnormal data from the airport mission data using an outlier factor detection algorithm;

[0061] The airport task allocation model and constraint condition establishment module is used to establish the airport task allocation model and impose constraints on the airport task allocation model;

[0062] The airport task allocation and optimization module is used to allocate and optimize airport tasks using a football team training algorithm;

[0063] The airport task allocation dynamic adjustment module is used to dynamically readjust the airport tasks according to flight delays.

[0064] The present invention has the following beneficial effects:

[0065] 1. The invention uses an outlier factor detection algorithm to delete abnormal data from the airport mission time data, and achieves the purpose of anomaly detection by calculating the density near the data points. It can effectively delete abnormal data and has high detection accuracy.

[0066] 2. The invention establishes an objective function and constructs an airport task allocation model based on airport service quality, airport staff enthusiasm and fairness of airport task allocation, and uses mathematical modeling to facilitate quantitative analysis of task allocation problems.

[0067] 3. The invention constrains the airport task allocation model to obtain the constraint conditions of the airport task allocation model, which is convenient for the intelligent optimization algorithm to solve.

[0068] 4. The invention allocates airport tasks by using a football team training algorithm and completes airport task allocation and optimization by finding the optimal fitness function value. The algorithm is optimized by simulating the football training process, has a fast search speed, and has strong optimization performance.

[0069] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0071] Figure 1 The present invention provides a flow chart of airport task allocation and optimization by an airport task allocation and optimization system based on the Internet of Things. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0074] Example 1

[0075] See also Figure 1 , this implementation discloses an airport task allocation and optimization method based on the Internet of Things, which specifically includes the following contents:

[0076] S1. Acquire airport ground support mission time information within a period of time, record it as an airport mission time set, delete abnormal data from the airport mission time data in the airport mission time set based on an outlier factor detection algorithm, and obtain a processed airport mission time set;

[0077] The S1 comprises the following steps:

[0078] S11. During the process from landing to take-off, the airport ground support tasks before, during, after, before and during take-off are obtained. The tasks before landing include allocating aprons, the tasks during landing include guiding the aircraft into position, the tasks after landing include disembarkation guidance, special vehicle support, the tasks before take-off include refueling and power supply, and the tasks during take-off include pushing the aircraft, etc.; the time information of all airport ground support tasks within a period of time is recorded, which is recorded as the airport task time set A = {(a1,b1),(a2,b2),(a3,b3),...,(a m ,b m )}, where (a m ,b m ) represents the time data of the mth airport mission, including the start execution time and the completion time;

[0079] S12, in the airport mission time set, select the m1th airport mission time data, set an arbitrary positive integer k, record the Euclidean distance between the kth nearest neighbor data of the m1th airport mission time data as the k-neighbor distance of m1, and keep the k airport mission time data in the airport mission time set; wherein the Euclidean distance between the k airport mission time data and the m1th airport mission time data is less than or equal to the k-neighbor distance of m1, use the outlier factor detection algorithm to detect abnormal airport mission time data in the airport mission time set, and delete the abnormal airport mission time data to obtain the processed airport mission time set, the specific steps are as follows:

[0080] S121, calculate the Euclidean distance between the airport mission time data in the airport mission time set and the m1th airport mission time data, record it as the k-neighborhood distance, add the airport mission time data corresponding to the k-neighborhood distance less than or equal to m1 to the neighborhood set to obtain the neighborhood set; set the Euclidean distance between the m1th airport mission time data and the i1th airport mission time data in the neighborhood set as the reachable distance The local reachable density calculation formula is as follows:

[0081]

[0082] in, represents the local reachability density of the m1th airport mission time data, b represents the number of airport mission time data in the neighborhood set, i1=1,2,3,...,b;

[0083] S122. Calculate the local reachability density of the mission time data of the i1th airport in the neighborhood set. The calculation formula of the local outlier factor is as follows:

[0084]

[0085] in, represents the local outlier factor of the mission time data of the m1th airport, represents the local reachability density of the mission time data of the i1th airport in the neighborhood set;

[0086] Calculate the local outlier factors of all airport mission time data in the neighborhood set to obtain a local outlier factor set, sort the local outlier factors in the local outlier factor set in ascending order, set the local outlier factor threshold as ω, record the local outlier factors in the local outlier factor set that are greater than the local outlier factor threshold as abnormal local outlier factors, delete the airport mission time data corresponding to the abnormal local outlier factors in the airport mission time set, and obtain a processed airport mission time set;

[0087] S2. Based on the airport service quality, the enthusiasm of airport staff and the fairness of airport task allocation, construct the objective function of the airport task allocation model, establish the airport task allocation model, and constrain the airport task allocation model to obtain the constraint conditions of the airport task allocation model;

[0088] The S2 comprises the following steps:

[0089] S21, obtaining the airport task execution start time and the airport task completion time according to the processed airport task time set; setting the airport task set to A′={a′1, a′2, a′3, ..., a′ m′}, where a′ m′ represents the m′th airport task, and the airport staff set is A″={c1,c2,c3,...,c n}, where c n represents airport staff member n; assign the airport tasks in the airport task set to the airport staff members in the airport staff set, and set the airport task allocation coefficient to χ. When the airport task allocation coefficient is 1, it means that the m″th airport task is assigned to the airport staff member n′; when the airport task allocation coefficient is 0, it means that the m″th airport task is not assigned to the airport staff member n′; B(m″,n′) represents the ability matching result of the airport staff member n′ to the m″th airport task, and the airport service quality objective function is established as follows:

[0090]

[0091] Among them, F1 represents the airport service quality objective function;

[0092] Assume that B′(m″,n′) represents the interest matching result of airport staff n′ for the m″th airport task, and establish the airport staff motivation objective function as follows:

[0093]

[0094] Among them, F2 represents the airport staff motivation objective function;

[0095] Assume that B″(m″,n′) represents the load of airport worker n′ on the m″th airport task, and the average load of all airport workers in the airport worker set is The objective function of airport task allocation fairness is established as follows:

[0096]

[0097] Among them, F2 represents the fairness objective function of airport task allocation;

[0098] S22, making the airport service quality objective function and the airport staff enthusiasm objective function at the maximum value, giving the airport service quality objective function a weight δ1, giving the airport staff enthusiasm objective function a weight δ2, and δ1+δ2=1; and converting the airport task allocation fairness objective function into a constraint condition, and establishing an airport task allocation model as follows:

[0099]

[0100] Among them, F represents the airport task allocation model;

[0101] S23. Set the number of resources owned by airport staff member n' to d n′ , the resource consumed by the m″th airport task assigned to airport staff n′ is d′(m″,n′), and the airport staff resource constraint is established Set the rated maximum completion time to The airport mission starts at The airport mission will be completed in Establishing time constraints for airport missions Select airport worker n″ from the airport worker set, set the airport task start execution time and airport task completion time of airport worker n′ respectively and The airport task start execution time and airport task completion time of airport staff n″ are and Establishing airport mission time conflict constraints The airport staff resource constraints, airport task time constraints and airport task time conflict constraints constitute the airport task allocation model constraints;

[0102] S3, encoding the airport tasks, taking the airport task allocation model as the fitness function, and using the football team training algorithm to allocate the airport tasks under the constraints of the airport task allocation model, finding the best fitness function value, and completing the airport task allocation and optimization;

[0103] The S3 comprises the following steps:

[0104] S31, encode the airport tasks in the airport task set to obtain an airport task code string A″′={a″1, a″2, a″3, ..., a″ m′}, where a″ m′ Represents the code corresponding to the m′th airport task, combined with the staff set, we get the coding function C(n)=a″ m′, the encoding function represents assigning the m′th airport task to the airport staff member n; taking the airport task assignment model as the fitness function, and using the football team training algorithm to find the optimal fitness function value under the constraints of the airport task assignment model, the specific steps are as follows:

[0105] S311, generate a random code string, use the coding function to represent the g-th football player in the e-th iteration of the football team, set the number of football players in the football team to be equal to the number of airport tasks, set the dimension of football players in the football team to be equal to the number of staff, the current number of iterations is e, and the g-th football player in the current e-th iteration is Where i represents the football player number, i.e., the code corresponding to the airport task, j represents the football player dimension, corresponding to the airport staff, and the g-th football player in the e-th iteration after training is The best trained football player in the football team is D e (best, j), g1 represents a random number and g1∈[0,1]. The calculation formula for the g-th football player in the e-th iteration after training is as follows:

[0106]

[0107] During the collective training of the football team, the football players in the football team try their best to catch up with the best-trained football players while avoiding becoming the worst-trained football players. The worst-trained football players in the football team are set as D. e (worst, j), g2 represents a random number and g2∈[0,1], g3 represents a random number and g3∈[0,1], then the g-th football player of the e-th iteration after training is updated, and the calculation formula is as follows:

[0108]

[0109] Assume that the state fluctuation of the football players in the football team becomes smaller and smaller as the number of training times increases. Use the normal distribution h(e) to represent the state fluctuation of the football players in the football team as the number of iterations increases. Update the g-th football player in the e-th iteration after training again. The calculation formula is as follows:

[0110]

[0111] In the iterative process, the best fitness function value is continuously sought, and the current football player number and the current football player dimension corresponding to the best fitness function value are found. At this time, the current football player number corresponds to the code corresponding to the current airport task, and the current football player dimension corresponds to the current airport staff. The code corresponding to the current airport task is assigned to the current airport staff.

[0112] S312, after the collective training, the football team enters the collective training process, and the football players in the football team are grouped according to the dimension, and divided into forward group, midfield group, defender group and goalkeeper group. Different groups can communicate and learn, including optimal learning, random learning and random communication; set the learning probability to ε1, the communication probability to ε2, and in the optimal learning stage, set the best football player in group h1 to Any football player in group h1 is Where j represents the best football player dimension, e represents the current iteration number e, and the calculation formula for any football player is as follows:

[0113]

[0114] In the random learning stage, any football player is selected from the h1th group again, and the football player is updated again according to the learning probability and the communication probability, to obtain the updated football player;

[0115] In the random communication stage, the communication probability is set to ε3, g4 represents a random number and g4∈[0,1], when the communication probability is less than or equal to g4, a normal distribution random number is set, and any football player is updated using the normal distribution random number to obtain an updated football player; when the communication probability is greater than g4, any football player is directly updated;

[0116] S313. After the collective training, the football team enters the individual training process, performs additional individual training on the updated football players, sets the additional individual training coefficients to φ1 and φ2, selects the best football player from the updated football players, and records it as In the process of continuous individual training, the best football players are updated and D e (best), the calculation formula is as follows:

[0117]

[0118] Set the maximum number of iterations to E, and continuously iterate the football players in the football team, while continuously searching for the best fitness function value. When the current number of iterations reaches the maximum number of iterations, stop iterating and get the final excellent football player corresponding to the best fitness function value.

[0119] S32. According to the final excellent football player, the corresponding final excellent football player number and the final excellent football player dimension are found. At this time, the final excellent football player number is the code corresponding to the airport task, and the final excellent football player dimension corresponds to the airport staff. The code corresponding to the airport task is assigned to the airport staff, and the airport task assignment and optimization are completed.

[0120] S4. Based on the data from the civil aviation route management platform, airport tasks are reallocated in response to flight delays, and the airport task allocation plan is dynamically adjusted;

[0121] The S4 comprises the following steps:

[0122] S41. Collect data from the civil aviation route management platform, including flight status, resource availability, airport task progress, and environmental changes, monitor flight delays in flight status, set a time threshold ξ, and when the difference between the flight rated time and the flight estimated arrival time is greater than the time threshold, the flight is delayed; otherwise, the flight is not delayed, and the civil aviation route management platform is used to verify and identify the delay;

[0123] S42. In response to flight delays, the airport task allocation model and the airport task allocation model constraints are updated, the airport tasks are reallocated, a new airport task allocation plan is generated, and the new airport task allocation plan is evaluated.

[0124] Example 2

[0125] This embodiment also discloses a system for airport task allocation and optimization method based on the Internet of Things, which specifically includes: an airport task data processing module, an airport task allocation model and constraint condition establishment module, an airport task allocation and optimization module, and an airport task allocation dynamic adjustment module;

[0126] The airport mission data processing module is used to delete abnormal data from the airport mission data using an outlier factor detection algorithm;

[0127] The airport task allocation model and constraint condition establishment module is used to establish the airport task allocation model and impose constraints on the airport task allocation model;

[0128] The airport task allocation and optimization module is used to allocate and optimize airport tasks using a football team training algorithm;

[0129] The airport task allocation dynamic adjustment module is used to dynamically readjust the airport tasks according to flight delays.

[0130] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0131] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. An airport task allocation and optimization method based on the Internet of Things, characterized in that: The steps include: S1. Acquire airport ground support mission time information within a period of time, record it as an airport mission time set, delete abnormal data from the airport mission time data in the airport mission time set based on an outlier factor detection algorithm, and obtain a processed airport mission time set; S2. Based on the airport service quality, the enthusiasm of airport staff and the fairness of airport task allocation, construct the objective function of the airport task allocation model, establish the airport task allocation model, and constrain the airport task allocation model to obtain the constraint conditions of the airport task allocation model; S3, encoding the airport tasks, taking the airport task allocation model as the fitness function, and allocating the airport tasks under the constraints of the airport task allocation model, finding the best fitness function value, and completing the airport task allocation and optimization; S4. Based on the data from the civil aviation route management platform, airport tasks are reallocated in response to flight delays, and the airport task allocation plan is dynamically adjusted.

2. The method for airport task allocation and optimization based on the Internet of Things according to claim 1 is characterized in that: The S1 comprises the following steps: S11, obtaining the time information of all airport ground support tasks within a period of time, and recording it as an airport task time set; S12. Use an outlier factor detection algorithm to detect abnormal airport mission time data in the airport mission time set, and delete the abnormal airport mission time data to obtain a processed airport mission time set.

3. The airport task allocation and optimization method based on the Internet of Things according to claim 2 is characterized in that: The S12 comprises the following steps: S121, selecting the m1th airport mission time data in the airport mission time set, determining the k-neighborhood distance of the m1th airport mission time data, obtaining a neighborhood set, and calculating the local reachability density of the m1th airport mission time data in the neighborhood set; S122. Calculate the local outlier factor of the m1-th airport mission time data according to the local reachable density, obtain a local outlier factor set, delete the airport mission time data corresponding to the abnormal local outlier factor in the local outlier factor set, and obtain a processed airport mission time set.

4. The method for airport task allocation and optimization based on the Internet of Things according to claim 3 is characterized in that: Deleting the airport mission time data corresponding to the abnormal local outlier factor in the local outlier factor set comprises the following steps: The local outlier factors in the local outlier factor set are sorted in ascending order, the local outlier factor threshold is set to ω, the local outlier factors in the local outlier factor set that are greater than the local outlier factor threshold are recorded as abnormal local outlier factors, and the airport mission time data corresponding to the abnormal local outlier factors in the airport mission time set are deleted.

5. The method for airport task allocation and optimization based on the Internet of Things according to claim 4 is characterized in that: The S2 comprises the following steps: S21. Based on the airport service quality, the enthusiasm of airport staff and the fairness of airport task allocation, establish the airport service quality objective function, the airport staff enthusiasm objective function and the airport task allocation fairness objective function; S22, assign weights to the airport service quality objective function and the airport staff enthusiasm objective function, and transform the airport task allocation fairness objective function into constraint conditions to establish an airport task allocation model; S23. Establish airport staff resource constraints, airport task time constraints, and airport task time conflict constraints, and combine the constraints to form the airport task allocation model constraints.

6. The method for airport task allocation and optimization based on the Internet of Things according to claim 5 is characterized in that: The S3 comprises the following steps: S31, encoding the airport tasks in the airport task set to obtain an airport task encoding string, and constructing an encoding function; using the airport task allocation model as a fitness function, and using a football team training algorithm to find the optimal fitness function value under the constraints of the airport task allocation model; S32. According to the final excellent football player, find the corresponding final excellent football player number and final excellent football player dimension. At this time, the final excellent football player number is the code corresponding to the airport task, and the final excellent football player dimension corresponds to the airport staff. The code corresponding to the airport task is assigned to the airport staff to complete the airport task allocation and optimization.

7. The method for airport task allocation and optimization based on the Internet of Things according to claim 6 is characterized in that: The method of using the football team training algorithm to find the best fitness function value includes the following steps: S311, generate a random code string, use the coding function to represent the g-th football player in the e-th iteration of the football team, set the number of football players in the football team to be equal to the number of airport tasks, the dimension of football players in the football team to be equal to the number of staff, the current number of iterations is e, and the g-th football player in the current e-th iteration is Where i represents the football player number, i.e., the code corresponding to the airport task, j represents the football player dimension, corresponding to the airport staff, and the g-th football player in the e-th iteration after training is The best trained football player in the football team is D e (best, j), g1 represents a random number and g1∈[0,1]. The calculation formula for the g-th football player in the e-th iteration after training is as follows: During the collective training of the football team, the football players in the football team try their best to catch up with the best-trained football players while avoiding becoming the worst-trained football players. The worst-trained football players in the football team are set as D. e (worst, j), g2 represents a random number and g2∈[0,1], g3 represents a random number and g3∈[0,1], then the g-th football player of the e-th iteration after training is updated, and the calculation formula is as follows: Assume that the state fluctuation of the football players in the football team becomes smaller and smaller as the number of training times increases. Use the normal distribution h(e) to represent the state fluctuation of the football players in the football team as the number of iterations increases. Update the g-th football player in the e-th iteration after training again. The calculation formula is as follows: In the iterative process, the best fitness function value is continuously sought, and the current football player number and the current football player dimension corresponding to the best fitness function value are found. At this time, the current football player number corresponds to the code corresponding to the current airport task, and the current football player dimension corresponds to the current airport staff. The code corresponding to the current airport task is assigned to the current airport staff. S312, after the collective training, the football team enters the collective training process, and the football players in the football team are grouped according to the dimension, and divided into forward group, midfield group, defender group and goalkeeper group. Different groups can communicate and learn, including optimal learning, random learning and random communication; set the learning probability to ε1, the communication probability to ε2, and in the optimal learning stage, set the best football player in group h1 to Any football player in group h1 is Where j represents the best football player dimension, e represents the current iteration number e, and the calculation formula for any football player is as follows: In the random learning stage, any football player is selected from the h1th group again, and the football player is updated again according to the learning probability and the communication probability, to obtain the updated football player; In the random communication stage, the communication probability is set to ε3, g4 represents a random number and g4∈[0,1], when the communication probability is less than or equal to g4, a normal distribution random number is set, and any football player is updated using the normal distribution random number to obtain an updated football player; when the communication probability is greater than g4, any football player is directly updated; S313. After the collective training, the football team enters the individual training process, performs additional individual training on the updated football players, sets the additional individual training coefficients to φ1 and φ2, selects the best football player from the updated football players, and records it as In the process of continuous individual training, the best football players are updated and D e (best), the calculation formula is as follows: Set the maximum number of iterations to E, and continuously iterate the football players in the football team while continuously searching for the best fitness function value. When the current number of iterations reaches the maximum number of iterations, stop iterating and get the final excellent football player corresponding to the best fitness function value.

8. The method for airport task allocation and optimization based on the Internet of Things according to claim 7 is characterized in that: The S4 comprises the following steps: S41. Monitor flight delays in flight status and use the civil aviation route management platform to verify and identify delays. In case of flight delays, update the airport task allocation model and airport task allocation model constraints, reallocate airport tasks, and generate a new airport task allocation plan.

9. A system for implementing the airport task allocation and optimization method based on the Internet of Things as described in any one of claims 1 to 8, characterized in that: Specifically include: Airport task data processing module, airport task allocation model and constraint condition establishment module, airport task allocation and optimization module and airport task allocation dynamic adjustment module; The airport mission data processing module is used to delete abnormal data from the airport mission data using an outlier factor detection algorithm; The airport task allocation model and constraint condition establishment module is used to establish the airport task allocation model and impose constraints on the airport task allocation model; The airport task allocation and optimization module is used to allocate and optimize airport tasks using a football team training algorithm; The airport task allocation dynamic adjustment module is used to dynamically readjust the airport tasks according to flight delays.

Images