CN110807525A - Neural network flight guarantee service time estimation method based on improved genetic algorithm - Google Patents
Neural network flight guarantee service time estimation method based on improved genetic algorithm Download PDFInfo
- Publication number
- CN110807525A CN110807525A CN201911037529.XA CN201911037529A CN110807525A CN 110807525 A CN110807525 A CN 110807525A CN 201911037529 A CN201911037529 A CN 201911037529A CN 110807525 A CN110807525 A CN 110807525A
- Authority
- CN
- China
- Prior art keywords
- neural network
- service time
- time estimation
- flight
- genetic algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 106
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 78
- 230000002068 genetic effect Effects 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000012549 training Methods 0.000 claims abstract description 26
- 238000012360 testing method Methods 0.000 claims abstract description 24
- 230000009471 action Effects 0.000 claims abstract description 21
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 230000008569 process Effects 0.000 claims description 18
- 210000000349 chromosome Anatomy 0.000 claims description 16
- 238000004140 cleaning Methods 0.000 claims description 11
- 230000009467 reduction Effects 0.000 claims description 9
- 239000010813 municipal solid waste Substances 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 239000010865 sewage Substances 0.000 claims description 7
- 238000013461 design Methods 0.000 claims description 6
- 210000001503 joint Anatomy 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000007689 inspection Methods 0.000 claims description 4
- 238000012423 maintenance Methods 0.000 claims description 3
- 230000002028 premature Effects 0.000 claims description 3
- 238000004321 preservation Methods 0.000 claims description 2
- 238000012847 principal component analysis method Methods 0.000 claims description 2
- 230000035772 mutation Effects 0.000 abstract description 18
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 210000001726 chromosome structure Anatomy 0.000 abstract description 2
- 108090000623 proteins and genes Proteins 0.000 description 26
- 230000006870 function Effects 0.000 description 8
- 230000003044 adaptive effect Effects 0.000 description 7
- 239000011159 matrix material Substances 0.000 description 7
- 238000000513 principal component analysis Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 239000000446 fuel Substances 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000003032 molecular docking Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000010006 flight Effects 0.000 description 1
- 239000010794 food waste Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002779 inactivation Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 238000011012 sanitization Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G06Q50/40—
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
A neural network flight guarantee service time estimation method based on an improved genetic algorithm is disclosed. The method comprises the steps of dividing flight guarantee services into three types; selecting a standard action node, reducing the dimension of an original variable, and constructing a BP neural network flight guarantee service time estimation model; preprocessing and normalizing the data, and training and testing the model; optimizing the model using an improved genetic algorithm; and actually applying the model and the like. The neural network flight guarantee service time estimation method based on the improved genetic algorithm has the following beneficial effects: on the basis of the traditional genetic algorithm, a chromosome structure, a fitness function, a selection operator, a crossover operator, a mutation operator and a crossover mutation probability are respectively designed, so that the flight guarantee service time is accurately estimated, and the flight guarantee service efficiency can be improved.
Description
Technical Field
The invention belongs to the technical field of civil aviation, and particularly relates to a neural network flight guarantee service time estimation method based on an improved genetic algorithm (AMGA).
Background
The manager of the airport support service not only considers the constraint of the flight support service operation time, but also considers the exclusivity of the aircraft to the equipment occupation. Factors such as aircraft type, flight time, vehicle driving route and the like need to be considered when special vehicle equipment is scheduled, so that the flight guarantee process has the characteristics of nonlinearity, non-stationarity and dynamics. Once a certain link in the whole flight support service process is interfered and the operation is not completed within the time specified by the standard, the subsequent flight support service link is directly influenced to form a sweep effect, and even serious consequences can be caused. Therefore, a mathematical model and a computer simulation technology of flight support service need to be established, flight operation efficiency is improved by describing the ground operation state in detail, and resource utilization and equipment use maximization are ensured.
When the BP neural network solves the complex nonlinear problem, one part is forward propagation of samples, the other part is backward propagation of errors, the connection weight and the threshold of each layer are adjusted to reduce the actual errors, and the process is repeatedly circulated until the output error of the model is smaller than the precision requirement, namely the output result of the model is infinitely close to the actual result. But currently, methods for estimating flight guarantee service time by using the BP neural network are few.
Disclosure of Invention
In order to solve the above problems, the present invention provides a neural network flight support service time estimation method based on an improved genetic algorithm.
In order to achieve the above purpose, the neural network flight support service time estimation method based on the improved genetic algorithm provided by the invention comprises the following steps in sequence:
step 1: analyzing flight support service flow of airport scene operation, dividing flight support service into three types according to different controlled objects, and respectively serving an aircraft, passengers and luggage, goods and mails;
step 2: selecting n standard action nodes from the three types of flight guarantee services, taking the time variable of each standard action node as an original variable, reducing the dimension of the original variable to obtain m main components, taking the m main components as input variables of a BP neural network flight guarantee service time estimation model, and then constructing the BP neural network flight guarantee service time estimation model according to the number of the main components;
and step 3: preprocessing and normalizing the data related to the input variables, randomly dividing all the processed data into training samples and testing samples, inputting the training samples into a BP neural network flight guarantee service time estimation model to train the model, and finally inputting the testing samples into the trained BP neural network flight guarantee service time estimation model;
and 4, step 4: optimizing the trained BP neural network flight guarantee service time estimation model by using an improved genetic algorithm to obtain an optimized neural network flight guarantee service time estimation model of the improved genetic algorithm;
and 5: and (3) inputting the flight guarantee service data of any airport into the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm after being processed according to the steps 1 and 2, wherein the output of the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm is the estimated value of the flight guarantee service time of the airport.
In step 1), the method for analyzing the flight support service flow of the airport scene operation, and dividing the flight support service into three types according to the difference of controlled objects, wherein the three types are respectively the service of an aircraft, the service of a passenger and the service of luggage and luggage postal service, and the method comprises the following steps:
the aircraft service mainly comprises: the method comprises the following steps of wheel block mounting, gallery bridge butt joint, passenger elevator car butt joint, cabin cleaning, garbage treatment, sewage cleaning, food adding, aviation oil filling, gallery bridge evacuation, passenger elevator car evacuation, maintenance inspection and wheel block removal; the service of the passenger mainly comprises the operations of opening a passenger cabin door, taking off the passenger, boarding the passenger and closing the passenger cabin door; the service of the luggage goods and the mails mainly comprises the operations of opening a goods cabin door, unloading the mail luggage, loading the mail luggage and closing the goods cabin door.
In step 2), the method for selecting n standard action nodes from the three types of flight support services, using the time variable of each standard action node as an original variable, performing dimensionality reduction on the original variable to obtain m principal components, using the m principal components as input variables of a BP neural network flight support service time estimation model, and then constructing the BP neural network flight support service time estimation model according to the number of the principal components comprises the following steps: firstly, selecting standard action nodes as original variables from flight support services required by analysis according to the step 1), then using a principal component analysis method to reduce the dimensions of the original variables, using the original variables after the dimension reduction as input layer nodes of a BP neural network, calculating the number of nodes of a corresponding hidden layer output layer, and finally constructing a BP neural network flight support service time estimation model.
In step 3), the method for preprocessing and normalizing the data related to the input variables, then randomly dividing all the processed data into training samples and testing samples, then inputting the training samples into the BP neural network flight guarantee service time estimation model to train the model, and finally inputting the testing samples into the trained BP neural network flight guarantee service time estimation model comprises the following steps: firstly, selecting data related to input variables of a BP neural network flight guarantee service time estimation model from actual data of an airport in an actual flight guarantee service process, processing the data by using a normalization method to reduce influence caused by partial data loss, dividing the data after normalization processing into training samples and testing samples, then inputting the training samples and the testing samples into the BP neural network flight guarantee service time estimation model in sequence to train and test the model, and finally obtaining the trained BP neural network flight guarantee service time estimation model.
In step 4), the method for obtaining the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm by optimizing the trained BP neural network flight guarantee service time estimation model by using the improved genetic algorithm is as follows: firstly, improving two-layer structure chromosomes of a genetic algorithm, improving a fitness function of the genetic algorithm, secondly, selecting an operation method of an optimal individual preservation strategy and a random tournament selection strategy to improve a selection operator of the genetic algorithm, avoiding the reduction of the efficiency of the genetic algorithm caused by too high selection probability from damaging excellent individuals in a population, simultaneously avoiding the selection probability from falling into 'premature' due to too low selection probability to design self-adaptive cross probability and variation probability, and finally completing a neural network flight guarantee service time estimation model for improving the genetic algorithm.
The neural network flight guarantee service time estimation method based on the improved genetic algorithm has the following beneficial effects: on the basis of the traditional genetic algorithm, a chromosome structure, a fitness function, a selection operator, a crossover operator, a mutation operator and a crossover mutation probability are respectively designed, so that the flight guarantee service time is accurately estimated, and the flight guarantee service efficiency can be improved.
Drawings
Fig. 1 is a schematic diagram of a specific flight support service.
Fig. 2 is a schematic diagram of a basic structure of a BP neural network provided by the present invention.
FIG. 3 is a flight assurance service workflow of a selected example of the present invention.
FIG. 4 shows a two-layer chromosome with a ladder structure and its coding map.
FIG. 5 is a flow chart of the improved genetic algorithm (AMGA) designed by the present invention.
FIG. 6 is a graph illustrating the number of hidden layer nodes according to the present invention.
FIG. 7 is a comparison graph of the estimation errors of three models of an improved genetic algorithm (AMGA), a traditional Genetic Algorithm (GA) and an unadditized algorithm.
FIG. 8 is a diagram comparing an estimated value and an actual value of flight support service time.
Detailed Description
The following describes the neural network flight support service time estimation method based on the improved genetic algorithm in detail with reference to the accompanying drawings and specific embodiments.
The neural network flight guarantee service time estimation method based on the improved genetic algorithm comprises the following steps in sequence:
step 1: analyzing flight support service flow of airport scene operation, dividing flight support service into three types according to different controlled objects, and respectively serving an aircraft, passengers and luggage, goods and mails;
the service of the aircraft mainly comprises: the method comprises the following steps of wheel block mounting, gallery bridge butt joint, passenger elevator car butt joint, cabin cleaning, garbage treatment, sewage cleaning, food adding, aviation oil filling, gallery bridge evacuation, passenger elevator car evacuation, maintenance inspection and wheel block removal; the service of the passenger mainly comprises the operations of opening a passenger cabin door, taking off the passenger, boarding the passenger and closing the passenger cabin door; the service of the luggage goods and the mails mainly comprises the operations of opening a goods cabin door, unloading the mail luggage, loading the mail luggage and closing the goods cabin door. Therefore, the flight support service is very complex, a specific flight support service schematic diagram is shown in fig. 1, and when an airplane is parked on an apron, various flight support service operations sometimes need to be performed synchronously.
The flight support service includes several jobs, each of which requires the use of several resources and equipment, and the following is an analysis of the equipment involved in the main flight support service job. The passenger boarding and boarding operation mainly comprises a corridor bridge, a ferry vehicle and a passenger ladder vehicle, wherein the corridor bridge is used when the passenger boarding and boarding operation is carried out at a near airport, and the passenger ladder vehicle and the ferry vehicle are used when the passenger boarding and boarding operation is carried out at a far airport; the food adding operation is mainly used for a food car, provides food for the aircraft and is used for serving different types of aircraft; the cabin cleaning and sewage disposal operation is mainly carried out by using a garbage truck for cleaning food waste, a clean water truck for supplementing water consumed in the flight process of the aircraft, and a sewage truck for discharging sewage generated in the flight process; the aviation fuel filling operation is mainly used for aviation fuel trucks which are divided into tank trucks and fuel tanker trucks and are used for supplementing fuel for aircrafts; the operation of loading and unloading the luggage and the goods mail mainly uses a conveyor belt vehicle, a tractor and a platform vehicle so as to facilitate loading and unloading the luggage, the goods mail; aircraft push-out operations involve tractors, also known as trailers, which are used to tow aircraft service equipment.
Through the research on the flight support service, the characteristics of the whole flight support service operation and the characteristics of the single flight support service operation are analyzed. In view of the whole, the flight guarantee operation flow of a single flight can be roughly divided into 4 parallel work flows, namely, a flight inspection service, a cabin service, a cargo hold service and an aviation oil filling service. Part of the parallel work flow also comprises serial flight support service operation, wherein the cargo hold service comprises support operation such as cargo hold opening, baggage cargo mail unloading, baggage cargo mail loading, cargo hold closing and the like; the cabin service includes not only serial operations but also parallel operations. Through deep analysis and research, flight support service processes can be abstracted into a complex network topological graph, the relationship among flight support service operations is complicated, the coupling performance is strong, and the flight support service operations not only have temporal sequence, but also have logical relationship. According to the stipulation of the airport flight operation guarantee standard, each specific flight guarantee service operation has the constraint condition, and the operation service must be completed within the time specified by the standard, wherein the cleaning of the passenger cabin and the addition of the flight food can be started after the passengers and the unit are off-board; the sanitization operation should be completed 15 minutes before the flight plan closes the doors, etc. The flight support service manager considers not only the flight support service operating time constraints, but also the exclusivity of the aircraft in equipment occupancy. Factors such as aircraft type, flight time, vehicle driving route and the like need to be considered when special vehicle equipment is scheduled, so that the flight guarantee service process has the characteristics of nonlinearity, non-stationarity and dynamics. If a certain link in the whole flight support service process is interfered and the operation is not completed within the time specified by the standard, the subsequent flight support service operation is directly influenced, a sweep effect is formed, and even serious consequences can be caused.
Step 2: selecting n standard action nodes from the three types of flight guarantee services, taking the time variable of each standard action node as an original variable, reducing the dimension of the original variable to obtain m main components, taking the m main components as input variables of a BP neural network flight guarantee service time estimation model, and then constructing the BP neural network flight guarantee service time estimation model according to the number of the main components;
from the above analysis, the flight assurance service process is a complex system with nonlinear and dynamic characteristics, and the BP neural network is suitable for solving the problem, and the basic structure of the BP neural network is shown in fig. 2.
And selecting n standard action nodes from the three types of flight support services. In the present invention, as shown in fig. 3, the number of standard action nodes is 21, and they are: the method comprises the following steps of loading a wheel block, butting a gallery bridge, butting a passenger lift car, opening a passenger lift door, opening a cargo hold door, unloading a passenger, unloading postal luggage, cleaning a passenger cabin, performing garbage treatment, performing sewage cleaning operation, adding food, filling aviation oil, polling a flight service, boarding a unit, loading the postal luggage, boarding a passenger, closing the passenger lift door, closing the cargo hold door, evacuating the gallery bridge, evacuating the passenger lift car and removing the wheel block. In the complex mesh topology diagram of fig. 3, the operations are performed sequentially in the direction indicated by the arrow, and the standard action nodes have time sequence and logic sequence, which cannot be reversed, so that the rationality of the flight guarantee service process can be ensured. The time variable of the 21 standard action nodes is changed into (X) by X1,x2,...xn) And (n-1, 2, …,21), and taking the time variable of each standard action node as an original variable.
In order to simplify the complexity of a BP neural network flight support service time estimation model, the invention adopts a Principal Component Analysis (PCA) method to carry out dimension reduction on the original variables, the Principal Component Analysis (PCA) method utilizes the idea of dimension reduction (linear transformation) to convert a plurality of original variables into a plurality of irrelevant principal components under the condition of little information loss, each principal component is a linear combination of the original variables, and all the principal components are not relevant to each other. The method comprises the following specific steps:
(1) normalizing the original variables according to the following formula to obtain normalized variables;
wherein the content of the first and second substances,
n represents the number of flight guarantee service standard action nodes; p is the input times of the time variable of each standard action node of the flight guarantee service; x is the number ofijInputting the j time variable of the ith flight support service standard action node;
the mean value of the time variable of the flight guarantee service input for the j time;
(2) calculating elements of a standardized matrix according to the normalized variation obtained in the last step, and constructing a standardized matrix Z by all the elements;
wherein the content of the first and second substances,
constructing a standardized matrix Z from all elements;
(3) solving a correlation coefficient matrix R by using the standardized matrix Z;
wherein the content of the first and second substances,
ZTa transpose operation representing the normalized matrix Z;
(4) solving the eigen equation | R- λ I of the correlation coefficient matrix RpGet p features | ═ 0Value of lambdaj(j ═ 1,2, …, p) according to the formula
P contribution rates η are calculated and are expressed according to the formula of cumulative contribution rate
Calculating p accumulated contribution rates, sorting the accumulated contribution rates from large to small, and finally selecting m original variables with the accumulated contribution rate more than or equal to 0.85 as principal components X ═ X (X ═ X)1,x2,...xm) And the principal components are used as the input of a BP neural network flight guarantee service time estimation model.
The number m of the principal components calculated by the PCA method is used as the number of nodes of an input layer to construct a BP neural network flight guarantee service time estimation model, in the invention, the number m of the nodes of the input layer is 10, and the number m of the nodes of the input layer is respectively the aviation fuel filling time, the boarding time, the loading and unloading time, the boarding bridge docking time, the boarding bridge gear withdrawing time, the boarding bridge evacuating time, the garbage disposal time and the clean catering time; the number of the hidden layer nodes is assumed according to an empirical formula, and the common empirical formula of the number of the hidden layer nodes includes three types: k is log2m、
α is a constant obtained according to experience, the number k of hidden layer nodes obtained in the invention is 14, and the number q of output layer nodes is 1, namely flight guarantee service time, is determined according to the obtained problem, therefore, the structure of the BP neural network flight guarantee service time estimation model determined in the invention is 10-14-1, and finally, the BP neural network flight guarantee service time estimation model is constructed.
In the BP neural network flight guarantee service time estimation model with the three-layer structure, m input variables of an input layer are m main components selected by the PCA method
The output variable of the input layer isThe input variable of the hidden layer is
The output variable of the hidden layer is
The connection weight from the input layer to the hidden layer is wij(i-1, 2, …, m; j-1, 2, …, k); connection weight v of hidden layer to output layerjt(j ═ 1,2, …, k; t ═ 1,2, …, q); output threshold theta of each unit of hidden layerj(j ═ 1,2, …, q); output threshold value gamma of each unit of output layerj(j ═ 1,2, …, q); p is the number of inputs. In the present invention, x in the input variables1Indicating the time of flight oil filling, x2Indicating gear-on time, x3Indicating passenger departure time, x4Indicating passenger boarding time, x5Indicating the time, x, of loading or unloading the goods and postal baggage6Represents the docking time of the gallery bridge, x7Indicating the time of removing the gear, x8Representing corridor bridge evacuation time, x9Denotes the garbage disposal time, x10Indicating clean catering time. By ci(i ═ 1,2, …,14) denotes each node in the hidden layer, and y denotes flight guarantee service time.
And step 3: preprocessing and normalizing the data related to the input variables, randomly dividing all the processed data into training samples and testing samples, inputting the training samples into a BP neural network flight guarantee service time estimation model to train the model, and finally inputting the testing samples into the trained BP neural network flight guarantee service time estimation model;
the data related to the input variables in the invention is selected from real data recorded in the actual flight guarantee service process in a certain period of an airport. However, in the actual recording process of the airport, the loss of partial data due to special situations is inevitable, and therefore, the data needs to be preprocessed to reduce the influence caused by the loss of partial data. Part of the flight provisioning service data before processing is shown in table 1.
Table 1, partial flight support service data before processing
Firstly, removing the code of the starting time and the ending time of the flight guarantee service operation, such as [06] in the table 1; and then deleting the stop number after the flight arrives, and finally solving the duration of each flight guarantee service operation, namely subtracting the starting time from the ending time of each operation because the relationship between the stop number and the content to be researched is not large. In table 1, the above round-robin operation is taken as an example, the other flight support service operations are processed similarly to the above round-robin operation, and the processed flight support service data are shown in table 2.
TABLE 2 processed Provisioning service data
In order to improve the estimation accuracy of a BP neural network flight guarantee service time estimation model, an output variable y in the model is subjected to1And an input variable x1、x2、x3、x4、x5、x6、x7、x8、x9、x10Normalization is performed, and the formula is as follows:
where u represents the processed input variable data, xiRepresenting input variable data before processing, xmaxRepresenting maximum input variable data; x is the number ofminRepresenting minimum input variable data; u. ofmaxRepresents the processed upper limit data uminRepresents the processed lower limit data, assuming that the processed data range is controlled to [0, 1 ]]Then u ismax＝1，umin＝0。
And (3) carrying out pretreatment and normalization treatment on the data according to the ratio of 3: 1, randomly dividing the training samples into training samples and testing samples, inputting the training samples into a BP neural network flight guarantee service time estimation model to train the model, inputting the testing samples into the trained BP neural network flight guarantee service time estimation model, and determining whether the model meets the precision requirement according to the error.
And 4, step 4: optimizing the trained BP neural network flight guarantee service time estimation model by using an improved genetic algorithm to obtain an optimized neural network flight guarantee service time estimation model of the improved genetic algorithm;
the improved genetic algorithm is characterized in that chromosomes in the traditional genetic algorithm are represented as a two-layer structure, corresponding operators are improved, the traditional mutation probability is designed as the self-adaptive cross mutation probability, and the network structure, the network connection weight and the threshold are optimized. The method comprises the following specific steps:
(1) two-layer structural chromosome design for improved genetic algorithms:
the traditional chromosome is improved, the structure of the chromosome is formed by arranging a plurality of genes according to layers, the chromosome gene design is divided into an upper layer and a lower layer, the upper layer comprises a reference gene and a parameter gene, and the reference gene is positioned at the upper layer and used for controlling the number of nodes of a hidden layer and optimizing the structure of a BP (back propagation) neural network; the parameter genes are arranged at the lower layer and used for optimizing the connection weight and the threshold value of the BP neural network, and the parameter gene strings at the lower layer are controlled by the control genes at the upper layer. Coding the gene, wherein the code of the reference gene is binary, and '1' represents that the corresponding gene is in an activated state, and a lower-layer gene string associated with the gene is effective; "0" represents that the corresponding gene is in an inactivated state, and the lower gene string associated with this gene is not effective; the parameter gene is encoded as a real number. The two-layer structure chromosome and its code map are designed as shown in FIG. 4. The chromosome designed by the invention can be divided into two levels, the coding length of the contrast gene should be equal to the number m of nodes of the hidden layer, and the position of the contrast gene should be at the upper layer of the chromosome; the position of the parameter gene should be in the lower layer of the chromosome, and its coding length should be equal to the total number of connection weights and thresholds in the chromosome (n +1) × (m +1) × p.
(2) Fitness function design for improved genetic algorithms
The improved genetic algorithm not only needs to realize the optimization of the BP neural network structure, but also needs to realize the optimization of the BP neural network connection weight and the threshold value, so that the estimation error of the flight guarantee service time can be minimized, and the complexity of the established model can be optimized, which is a dual-target optimization problem. The designed fitness function can reflect the complexity of the BP neural network structure and the estimation precision of the BP neural network structure. The estimation precision is determined by the total estimation error of actual training sample data of each flight guarantee service operation time, and the network complexity is determined by the number of hidden layer nodes of the designed BP neural network structure. The fitness function is designed as follows:
f＝αfrmse+βf com0＜α,β＜1
in the formula (f)rmse∈[0,1]Is the Root Mean Square Error (RMSE), f, of the actual training sample data for each operating time of the flight assurance servicecomIs the complexity of the network structure and,is an estimated value of flight guarantee service time yiThe flight guarantee service time estimation method is an actual value of flight guarantee service time, and the two values are obtained by a trained BP neural network flight guarantee service time estimation model; n (1) and N (0) represent activation and inactivation, respectively, in the control genesThe number of the live neurons, namely the number of 1 s and 0 s in the control gene, β and α respectively represent the adjustment coefficient of the structural complexity of the BP neural network and the adjustment coefficient of the estimation precision of the BP neural network.
(3) Selection operator for improved genetic algorithms
Conventional roulette wheels and some methods based on fitness scaling often suffer from "premature maturation" or "closed competition" such that there is no feasible way to retrieve, and eventually local extreme points rather than the most extreme points tend to be trapped. To address this limitation, the operating methods of "best individual conservation strategy" and "random tournament selection strategy of size 2" were selected.
1) Optimal individual conservation strategy: the most adaptive individuals in the parent population are selected, and the selected individuals are directly selected into the next generation population, so that the optimal individuals in the previous generation population are stored, and the global convergence of the genetic algorithm is ensured.
2) Tournament selection strategy size 2: for all individuals except the optimal solution in the previous generation population, randomly selecting two individuals to compare the fitness of the individuals, selecting the individuals with better fitness to enter the next generation population, and eliminating the individuals with poor fitness until a complete offspring population is generated, so that the individuals with relatively high quality can be ensured to enter the next generation population.
(4) Crossover and mutation operators for improved genetic algorithms
In the genetic algorithm, a single-point crossover operator and a simple mutation operator are used in a contrast gene layer at the upper layer of a chromosome; the parametric gene layer under the chromosome uses an integral arithmetic crossover operator and a non-uniform mutation operator. The integral arithmetic crossover operator uses the superposition principle of geometric vectors to calculate each component of the intersected previous generation vector, thereby expanding the search range of the algorithm; the non-uniform mutation operator makes the mutation associated with the evolution algebra of the population, and the number of elite individuals is small in the early stage of the evolution process, so that the using range is large. In the later stage of the evolution process, in order to prevent the excellent individuals from being destroyed, the allowable variation range is narrower, so that the local optimum value can be obtained.
(5) Adaptive cross probability and mutation probability of improved genetic algorithm
Because the selection of crossover and mutation probabilities can cause the reduction of the efficiency of genetic algorithms, if the selection probability is too large, the selection probability can easily destroy excellent individuals in the population; if the probability of picking is too small, the individual updating speed is much slower, and the individual is easy to fall into early maturity. The calculation formula of the self-adaptive cross probability is as follows:
in the formula (f)cRepresenting crossing individuals with smaller fitness values, fminRepresenting the minimum fitness value, f, in the current populationaMean value 0 < k representing current population fitness1,k2Less than or equal to 1, and in order to ensure that the algorithm can search the whole situation, a larger self-adaptive cross probability needs to be selected, and k can be selected generally1＝1,k21. The calculation formula of the adaptive mutation probability is as follows:
in the formula (f)mRepresenting the fitness value of the individual to be mutated, fminRepresenting the minimum fitness value, f, in the current populationaMean fitness value representing the current population, 0 < k3,k4Less than or equal to 1. Because the cross mutation operation in the genetic algorithm is the simulation of the mutation of an object with life in nature in the genetic process, the self-adaptive mutation probability is smaller, and generally k can be taken3＝0.5,k4＝0.5。
The improved genetic algorithm design is completed through the steps, the operation parameters of the group are set, the total number is N, the maximum evolution generation G, the number of nodes (usually adopting a larger value) of the initial assumed hidden layer is set, and the like. N individuals were randomly generated to make up the initial population, the population was divided into 2 sub-populations and the chromosomes were encoded into a two-layered structure. Decoding the individual to count the number of 1 in a single gene string in the upper-layer sub-generation group, namely the number of hidden layer nodes of the corresponding BP neural network; decomposing the real parameter string of the lower parameter gene associated with the upper control gene into a value 1 to obtain the initial connection weight and threshold of the hidden layer node; and (3) training the BP neural network according to the calculation formula in the step (2) to calculate an adaptive value. And assigning the obtained connection weight and the threshold value to a BP neural network, training the BP neural network by using a training sample, and testing the BP neural network by using a test sample to obtain a test error. And (4) according to different adaptive values of each individual, entering elite individuals of the next generation according to the selection operator designed in the step (4), and performing mutation operation on the selected individuals by using adaptive probability to generate offspring groups. Decoding upper and lower layer individuals in the offspring population, obtaining the structure, initial connection weight and threshold of the BP neural network, training the network for many times, and calculating the fitness value of the upper and lower layer individuals. And determining whether the optimal individual fitness value can meet a set value or can increase to the maximum evolution number, decoding the individual with the optimal fitness value, and obtaining the optimal number of the hidden layer nodes, the optimal initial connection weight and the threshold value of the nodes. A flow chart of the improved genetic algorithm is shown in fig. 5.
The optimal number of hidden layer nodes obtained by calculation of the improved genetic algorithm, the optimal initial connection weight and the threshold value of the hidden layer nodes are distributed to a BP neural network flight guarantee service time estimation model, and the structural parameters of the improved genetic algorithm after optimization of the BP neural network are determined as follows, wherein the population size N is 100, the maximum evolution algebra G is 200, the fitness function α of the estimation precision adjustment coefficient is 0.9, the network complexity adjustment coefficient β is 0.1, and the adaptive intersection and variation probability Pc、PmMiddle coefficient k1＝1,k2＝1,k3＝0.5,k4＝0.5、k3Finally determining the number m of the nodes of the input layer to be 10, the number of the nodes of the output layer to be 1 and the number m of the nodes of the hidden layer to be 14 when being 0.5; connection weight wij、vjtAnd a threshold value thetajGamma is in the range of [ -3,3 [)]. As shown in fig. 6, through the evolution of 103 generations, the average fitness reaches a minimum, and basically does not change any more as the evolution generation increases. When the evolution algebra is 103 generationsSince the number of corresponding hidden layer nodes is 7, the optimal number of hidden layer nodes of the BP neural network is m-7.
Table 3 optimal initial connection weight, threshold
The improved genetic algorithm neural network flight support service time estimation model has the corresponding optimal initial connection weight and threshold value shown in Table 3, where x1、x2、x3、x4、x5、x6、x7、x8、x9、x10Represents 10 input variables, c1、c2、c3、c4、c5、c6、c7Which represents 7 hidden layer variables, y an output variable, theta a threshold of the hidden layer, and gamma a threshold of the output layer.
Establishing a BP neural network flight guarantee service time estimation model with a network structure of 10-7-1, distributing the optimal initial weight and threshold in the table 3, selecting a Sigmoid function as an excitation function, taking the learning rate of 0.01, taking the step length of 0.9, taking the maximum training frequency of 2000, taking the training expected value of 0.01, and finally forming the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm.
And 5: and (3) inputting the flight guarantee service data of any airport into the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm after being processed according to the steps 1 and 2, wherein the output of the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm is the estimated value of the flight guarantee service time of the airport.
In order to verify the effect of the invention, the inventor estimates the flight guarantee service time by respectively using three algorithms of the optimized neural network flight guarantee service time estimation model (AMGA-BP model) of the improved genetic algorithm, the neural network flight guarantee service time estimation model (GA-BP model) of the traditional genetic algorithm and the neural network flight guarantee service time estimation model (BP model) without the algorithm.
And the absolute value RE of the relative error, the average absolute error MAE and the Hillton coefficient TIC are used as standards for measuring the advantages and the disadvantages of the three algorithms. The absolute value RE is used for calculating the estimation error of each flight guarantee service time; the average absolute error MAE and the Hillton coefficient TIC measure the accuracy of the algorithm by calculating the integral error between the estimated value and the true value of the flight guarantee service time. Assuming that a total of N estimated flight support service times are set in the test set, the actual value of the flight support service time is yiThe estimated value is
The formula is as follows:
TABLE 4 evaluation of Effect evaluation index values
45 groups of samples in the test set are randomly extracted, the evaluation index value of the estimation effect is shown in table 4, and the guarantee service time error pair estimated by different algorithms is shown in fig. 7.
As can be seen from Table 4, the mean absolute error MAE value of the AMGA-BP model is smaller than that of the GA-BP model and the BP model, and the Hillton coefficient TIC value is smaller than that of the GA-BP model and the BP model, which indicates that the AMGA-BP model has higher estimation accuracy. As can be seen in fig. 7: the estimation error curve of the AMGA-BP model is basically lower than the error curves of the GA-BP model and the BP model; the change trends of the error estimation curves of the GA-BP model and the BP model are basically the same. Estimated values of guaranteed service time of the three models are compared with actual values, as shown in fig. 8.
(AMGA-BP model), neural network flight guarantee service time estimation model of traditional genetic algorithm (GA-BP model) and neural network flight guarantee service time estimation model without algorithm (BP model)
In conclusion, compared with the GA-BP model and the BP model, the AMGA-BP model can better estimate the flight guarantee service time and can better process the nonlinear problem. It is worth noting that the AMGA-BP model is not very accurate for estimating the guaranteed service time of each flight, but the overall estimation is more robust. From FIG. 7, it can be seen that for some flights, the accuracy of the AMGA-BP model on the estimation of the flight guarantee service time is much higher than that of the GA-BP model and the BP model, but the error is still not particularly small, which indicates that the AMGA-BP model needs to further improve the adaptability of the flight guarantee service time under different conditions. Fig. 8 compares the estimated values of flight guarantee service time of the AMGA-BP model, the GA-BP model, and the BP model with the actual flight guarantee service time, respectively, and it can be clearly seen that the difference between the estimated value of the BP model and the actual value is the largest, and the GA-BP model is the second to illustrate that aiming at the complex nonlinear problem of flight guarantee service, the expected requirement cannot be met only by using the conventional genetic algorithm, and improvement must be performed on the basis of the neural network flight guarantee service time estimation model. The flight guarantee service time estimated by the AMGA-BP model is closest to the actual flight guarantee service time.
Claims (5)
1. A neural network flight guarantee service time estimation method based on an improved genetic algorithm is characterized by comprising the following steps: the method comprises the following steps which are carried out in sequence:
step 1: analyzing flight support service flow of airport scene operation, dividing flight support service into three types according to different controlled objects, and respectively serving an aircraft, passengers and luggage, goods and mails;
step 2: selecting n standard action nodes from the three types of flight guarantee services, taking the time variable of each standard action node as an original variable, reducing the dimension of the original variable to obtain m main components, taking the m main components as input variables of a BP neural network flight guarantee service time estimation model, and then constructing the BP neural network flight guarantee service time estimation model according to the number of the main components;
and step 3: preprocessing and normalizing the data related to the input variables, randomly dividing all the processed data into training samples and testing samples, inputting the training samples into a BP neural network flight guarantee service time estimation model to train the model, and finally inputting the testing samples into the trained BP neural network flight guarantee service time estimation model;
and 4, step 4: optimizing the trained BP neural network flight guarantee service time estimation model by using an improved genetic algorithm to obtain an optimized neural network flight guarantee service time estimation model of the improved genetic algorithm;
and 5: and (3) inputting the flight guarantee service data of any airport into the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm after being processed according to the steps 1 and 2, wherein the output of the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm is the estimated value of the flight guarantee service time of the airport.
2. The neural network flight support service time estimation method based on the improved genetic algorithm as claimed in claim 1, wherein: in step 1), the method for analyzing the flight support service flow of the airport scene operation, and dividing the flight support service into three types according to the difference of controlled objects, wherein the three types are respectively the service of an aircraft, the service of a passenger and the service of luggage and luggage postal service, and the method comprises the following steps:
the aircraft service mainly comprises: the method comprises the following steps of wheel block mounting, gallery bridge butt joint, passenger elevator car butt joint, cabin cleaning, garbage treatment, sewage cleaning, food adding, aviation oil filling, gallery bridge evacuation, passenger elevator car evacuation, maintenance inspection and wheel block removal; the service of the passenger mainly comprises the operations of opening a passenger cabin door, taking off the passenger, boarding the passenger and closing the passenger cabin door; the service of the luggage goods and the mails mainly comprises the operations of opening a goods cabin door, unloading the mail luggage, loading the mail luggage and closing the goods cabin door.
3. The neural network flight support service time estimation method based on the improved genetic algorithm as claimed in claim 1, wherein: in step 2), the method for selecting n standard action nodes from the three types of flight support services, using the time variable of each standard action node as an original variable, performing dimensionality reduction on the original variable to obtain m principal components, using the m principal components as input variables of a BP neural network flight support service time estimation model, and then constructing the BP neural network flight support service time estimation model according to the number of the principal components comprises the following steps: firstly, selecting standard action nodes as original variables from flight support services required by analysis according to the step 1), then using a principal component analysis method to reduce the dimensions of the original variables, using the original variables after the dimension reduction as input layer nodes of a BP neural network, calculating the number of nodes of a corresponding hidden layer output layer, and finally constructing a BP neural network flight support service time estimation model.
4. The neural network flight support service time estimation method based on the improved genetic algorithm as claimed in claim 1, wherein: in step 3), the method for preprocessing and normalizing the data related to the input variables, then randomly dividing all the processed data into training samples and testing samples, then inputting the training samples into the BP neural network flight guarantee service time estimation model to train the model, and finally inputting the testing samples into the trained BP neural network flight guarantee service time estimation model comprises the following steps: firstly, selecting data related to input variables of a BP neural network flight guarantee service time estimation model from actual data of an airport in an actual flight guarantee service process, processing the data by using a normalization method to reduce influence caused by partial data loss, dividing the data after normalization processing into training samples and testing samples, then inputting the training samples and the testing samples into the BP neural network flight guarantee service time estimation model in sequence to train and test the model, and finally obtaining the trained BP neural network flight guarantee service time estimation model.
5. The neural network flight support service time estimation method based on the improved genetic algorithm as claimed in claim 1, wherein: in step 4), the method for obtaining the optimized neural network flight guarantee service time estimation model of the improved genetic algorithm by optimizing the trained BP neural network flight guarantee service time estimation model by using the improved genetic algorithm is as follows: firstly, improving two-layer structure chromosomes of a genetic algorithm, improving a fitness function of the genetic algorithm, secondly, selecting an operation method of an optimal individual preservation strategy and a random tournament selection strategy to improve a selection operator of the genetic algorithm, avoiding the reduction of the efficiency of the genetic algorithm caused by too high selection probability from damaging excellent individuals in a population, simultaneously avoiding the selection probability from falling into 'premature' due to too low selection probability to design self-adaptive cross probability and variation probability, and finally completing a neural network flight guarantee service time estimation model for improving the genetic algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911037529.XA CN110807525B (en) | 2019-10-29 | 2019-10-29 | Neural network flight guarantee service time estimation method based on improved genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911037529.XA CN110807525B (en) | 2019-10-29 | 2019-10-29 | Neural network flight guarantee service time estimation method based on improved genetic algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110807525A true CN110807525A (en) | 2020-02-18 |
| CN110807525B CN110807525B (en) | 2023-04-14 |
Family
ID=69489427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911037529.XA Active CN110807525B (en) | 2019-10-29 | 2019-10-29 | Neural network flight guarantee service time estimation method based on improved genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110807525B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112036567A (en) * | 2020-09-18 | 2020-12-04 | 北京机电工程研究所 | Genetic programming method, apparatus and computer readable medium |
| CN112163790A (en) * | 2020-10-29 | 2021-01-01 | 广东机场白云信息科技有限公司 | Airport ground resource scheduling method, electronic device and computer readable storage medium |
| CN112330186A (en) * | 2020-11-18 | 2021-02-05 | 杨媛媛 | Method for evaluating ground operation guarantee capability |
| CN113112167A (en) * | 2021-04-21 | 2021-07-13 | 中国民航大学 | Dynamic control method for flight ground support service process |
| CN113378337A (en) * | 2021-06-03 | 2021-09-10 | 安徽富煌科技股份有限公司 | Urban public transport network optimization method based on passenger flow analysis |
| CN113505934A (en) * | 2021-07-19 | 2021-10-15 | 浙江永迅投资管理有限公司 | TMS intelligent logistics time management method and device |
| CN113781820A (en) * | 2021-08-24 | 2021-12-10 | 威海广泰空港设备股份有限公司 | Analysis method for guaranteeing flight efficiency of airport special vehicle |
| CN117314201A (en) * | 2023-11-28 | 2023-12-29 | 中国民用航空总局第二研究所 | Method and system for determining key links of flight guarantee service |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000007113A1 (en) * | 1998-07-31 | 2000-02-10 | Cet Technologies Pte Ltd. | Automatic freeway incident detection system using artificial neural networks and genetic algorithms |
| US20020059154A1 (en) * | 2000-04-24 | 2002-05-16 | Rodvold David M. | Method for simultaneously optimizing artificial neural network inputs and architectures using genetic algorithms |
| CN107241135A (en) * | 2017-06-30 | 2017-10-10 | 北京邮电大学 | A kind of satellite network switching method and device |
| CN108460461A (en) * | 2018-02-06 | 2018-08-28 | 吉林大学 | Mars earth shear parameters prediction technique based on GA-BP neural networks |
-
2019
- 2019-10-29 CN CN201911037529.XA patent/CN110807525B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000007113A1 (en) * | 1998-07-31 | 2000-02-10 | Cet Technologies Pte Ltd. | Automatic freeway incident detection system using artificial neural networks and genetic algorithms |
| US20020059154A1 (en) * | 2000-04-24 | 2002-05-16 | Rodvold David M. | Method for simultaneously optimizing artificial neural network inputs and architectures using genetic algorithms |
| CN107241135A (en) * | 2017-06-30 | 2017-10-10 | 北京邮电大学 | A kind of satellite network switching method and device |
| CN108460461A (en) * | 2018-02-06 | 2018-08-28 | 吉林大学 | Mars earth shear parameters prediction technique based on GA-BP neural networks |
Non-Patent Citations (2)
| Title |
|---|
| 邢志伟;唐云霄;罗谦;: "基于贝叶斯网络的航班保障服务时间动态估计" * |
| 邢志伟;蒋骏贤;罗晓;罗谦;: "基于贝叶斯网的离港航班滑行时间动态估计" * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112036567A (en) * | 2020-09-18 | 2020-12-04 | 北京机电工程研究所 | Genetic programming method, apparatus and computer readable medium |
| CN112036567B (en) * | 2020-09-18 | 2023-10-31 | 北京机电工程研究所 | Genetic programming method, apparatus and computer readable medium |
| CN112163790A (en) * | 2020-10-29 | 2021-01-01 | 广东机场白云信息科技有限公司 | Airport ground resource scheduling method, electronic device and computer readable storage medium |
| CN112330186A (en) * | 2020-11-18 | 2021-02-05 | 杨媛媛 | Method for evaluating ground operation guarantee capability |
| CN113112167A (en) * | 2021-04-21 | 2021-07-13 | 中国民航大学 | Dynamic control method for flight ground support service process |
| CN113378337A (en) * | 2021-06-03 | 2021-09-10 | 安徽富煌科技股份有限公司 | Urban public transport network optimization method based on passenger flow analysis |
| CN113505934A (en) * | 2021-07-19 | 2021-10-15 | 浙江永迅投资管理有限公司 | TMS intelligent logistics time management method and device |
| CN113505934B (en) * | 2021-07-19 | 2022-06-17 | 浙江工企信息技术股份有限公司 | TMS intelligent logistics time management method and device |
| CN113781820A (en) * | 2021-08-24 | 2021-12-10 | 威海广泰空港设备股份有限公司 | Analysis method for guaranteeing flight efficiency of airport special vehicle |
| CN117314201A (en) * | 2023-11-28 | 2023-12-29 | 中国民用航空总局第二研究所 | Method and system for determining key links of flight guarantee service |
| CN117314201B (en) * | 2023-11-28 | 2024-02-06 | 中国民用航空总局第二研究所 | Method and system for determining key links of flight guarantee service |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110807525B (en) | 2023-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110807525B (en) | Neural network flight guarantee service time estimation method based on improved genetic algorithm | |
| CN109919358B (en) | Real-time station flow prediction method based on neural network space-time attention mechanism | |
| CN109785618B (en) | Short-term traffic flow prediction method based on combinational logic | |
| CN109141847B (en) | Aircraft system fault diagnosis method based on MSCNN deep learning | |
| CN107958269A (en) | A kind of driving risk factor Forecasting Methodology based on hidden Markov model | |
| Akgüngör et al. | An artificial intelligent approach to traffic accident estimation: Model development and application | |
| CN111063194A (en) | Traffic flow prediction method | |
| CN111030889B (en) | Network traffic prediction method based on GRU model | |
| CN111931978A (en) | Urban rail transit passenger flow state prediction method based on space-time characteristics | |
| CN109635246B (en) | Multi-attribute data modeling method based on deep learning | |
| Dachyar | Simulation and optimization of services at port in Indonesia | |
| CN111079827B (en) | Railway data state evaluation method and system | |
| CN117056785A (en) | Federal learning classification model training method based on self-adaptive model disturbance | |
| CN112529284A (en) | Private car residence time prediction method, device and medium based on neural network | |
| CN112257914A (en) | Aviation safety cause and effect prediction method based on random forest | |
| CN111967308A (en) | Online road surface unevenness identification method and system | |
| CN113269402B (en) | Flight space control method and device and computer equipment | |
| CN117669837A (en) | Airport land side transfer traffic mode selection prediction method and system | |
| CN111967720B (en) | Scheduling method and system for network taxi taking | |
| CN116303786B (en) | Block chain financial big data management system based on multidimensional data fusion algorithm | |
| CN116739213A (en) | Subway connection bus optimization method based on agent model auxiliary algorithm | |
| Iyigunlu et al. | Agent-based modelling of aircraft boarding methods | |
| CN110222892A (en) | The get-off stop prediction technique and device of passenger | |
| CN112967420B (en) | Interval two-type based heavy-duty train operation process monitoring method and system | |
| CN113687651A (en) | Path planning method and device for delivering vehicles according to needs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |