Airport scene operation scene interpretable similarity measurement method and system

By collecting and processing flight and meteorological data during airport scene operation and calculating similarity using neural networks, the problem of difficulty in measuring the similarity between target scenarios and similar scenarios in the prior art is solved, and higher prediction accuracy and trust are achieved. New methods are provided for refined scene control.

CN119005484AActive Publication Date: 2024-11-22NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF 11 Cites 0 Cited by

Patent Information

Application Number
CN202410928838.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-11
Publication Date
2024-11-22
Estimated Expiration
2044-07-11

AI Technical Summary

Technical Problem

It is difficult for the prior art to effectively measure the similarity between the target scenario and similar scenarios during airport scene operation, resulting in insufficient prediction accuracy and affecting the controller's trust in the prediction model and the promotion of the model.

Method used

A similarity measurement method that can explain the operation scenario of the airport scene is adopted. By collecting flight data and meteorological data, preprocessing and secondary feature extraction are performed, and divided into flight static attribute characteristics and environmental dynamic attribute characteristics, constructing data structures and applying neural networks Calculate the similarity and calculate the scene similarity using multiple time scale comparison.

Benefits of technology

A similarity measurement with strong interpretability of airport scene operation scenarios is realized, prediction accuracy and trust are improved, and a new method is provided for refined scene control, filling the technical gap in dynamic interpretable similarity calculation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119005484A_ABST
    Figure CN119005484A_ABST
Patent Text Reader

Abstract

The invention discloses an airport scene operation scene interpretable similarity measurement method and system, and the method comprises the following steps: 1, collecting flight data and meteorological data from different data channels, taking the flight data and meteorological data as scene data, carrying out the preprocessing of the scene data, and constructing a scene operation data set under a departure flight perspective; step 2, performing secondary feature processing on the data in the scene operation data set obtained in the step 1, and dividing the data into flight static attribute features and environment dynamic attribute features according to feature characteristics; 3, constructing a data structure for explaining similarity data by using the two types of features obtained in the step 2; and 4, applying the data obtained in the step 3 to the network for calculating the interpretable similarity, and calculating the interpretable similarity of the scene surface operation scene based on the two types of data.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of airport scene situation measurement and recognition, and specifically relates to an interpretable similarity measurement method and system, in particular to an interpretable similarity measurement method and system for airport scene operation scenarios. Background Art

[0002] In the study of airport surfaces, as the design of prediction models becomes more and more complex, the prediction accuracy of various indicators is also getting higher and higher. However, in actual surface control work, due to the opacity of the prediction process, controllers cannot fully understand the working principle of the model, so that controllers are more inclined to their own experience in order to ensure operational safety, which hinders the promotion of related technologies. In order to improve the trust of controllers in the prediction model, a feasible solution is to adopt a prediction model based on similar scenarios, that is, by comparing the actual operation information differences between historical scenarios and current scenarios, to provide reference historical statistical results for the prediction of indicators in the target scenario. In the field of indicator prediction, multiple historical data that are most similar to the current scenario are used to improve the prediction accuracy of the indicators of departing aircraft in the target scenario. Considering that the scenarios obtained by this method are based on actual historical operation data and are completely visible, the prediction results will be more easily accepted by front-line operators, so the process is considered to be explainable.

[0003] In the past, some studies have been conducted on similar scenes, which are mainly divided into two types of methods: "scene division, identification and prediction" based on clustering and indicator prediction based on improved similarity measurement. However, it is obvious that the result obtained by the clustering-based method is an inaccurate division result. The prediction method of the scene generated by this method has weak generalization performance and limited improvement in overall prediction performance, which is obviously difficult to be promoted in refined scene operation management. In the method based on constructing similarity measurement, by calculating the deep features between scenes and performing similarity measurement, the average of the prediction indicators under the closest multiple scenes is taken as the prediction indicator of the target scene. This method can alleviate the situation that the traditional similarity measurement formula is difficult to apply to the airport scene operation environment with more classification features, and improve the prediction effect to a certain extent. Therefore, in the current prediction research, there is an urgent need for a method that can effectively measure the similarity between the target scene and the similar scene, which is used to screen similar scenes and assist in subsequent prediction or classification content. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an interpretable similarity measurement method and system for airport surface operation scenarios in view of the deficiencies in the prior art, and to calculate the quantitative calculation results of the scenario similarities under different application backgrounds.

[0005] To achieve the above object, the present invention adopts the following technical solution:

[0006] A method for measuring the interpretable similarity of airport surface operation scenarios comprises the following steps:

[0007] Step 1: collect flight data and meteorological data from different data channels as scenario data, pre-process them, and construct a scenario operation data set from the perspective of departing flights;

[0008] Step 2, processing the data in the scene operation data set obtained in step 1 with secondary features, and dividing them into two types of features: flight static attribute features and environment dynamic attribute features according to their characteristics;

[0009] Step 3, constructing a data structure that can explain similarity data using the two types of features obtained in step 2;

[0010] Step 4, apply the data obtained in step 3 to the network for calculating interpretable similarity, and calculate the interpretable similarity of the field scene operation scenarios based on the two types of data.

[0011] The step 1 specifically comprises the following steps:

[0012] Step 1-1, collecting original flight and weather data from different data channels, including at least: airport collaborative management system data, airline operation control system data, and relevant scene data involving the target airport in airport weather message data;

[0013] Step 1-2, preprocessing the raw data collected in step 1-1, including screening out abnormal data and completing missing data.

[0014] The step 2 specifically includes the following steps:

[0015] Step 2-1, extract secondary features from the data processed in step 1, integrate all features, and perform normalization and encoding processing;

[0016] Step 2-2, according to the characteristics of the extracted features, the features are divided into two parts, namely, flight static attribute features and environmental dynamic attribute features; the processed and statistically processed data are stored, and a flight static attribute feature database and an environmental dynamic attribute feature database are constructed.

[0017] The step 3 specifically comprises the following steps:

[0018] Step 3-1: Construct the input structure of flight static attributes. Most of the flight static attribute features are categorical features and require additional encoding processing. The processed vector is called the static attribute vector θ i, all sample target scenes and their respective candidate similar scene sets are stacked into an input structure, and the final input data format is s×n×2×a cat , where s represents the number of sample scenes, α cat is the dimension of the processed static attribute vector, and n is the number of candidate scenes. Under this construction method, the static attributes of the flights of all sample target scenes and each candidate similar scene are compared.

[0019] Step 3-2: construct the input structure of the environment dynamic attribute. The environment dynamic attribute features of all scenes at the same time in the previous few days are spliced ​​to form a multi-time scale environment dynamic attribute input vector for one day, which is called the dynamic attribute vector θ e , process the vectors of all scene data according to the above steps; combine all scenes with their respective candidate similar scene sets to finally form the input data.

[0020] The step 4 specifically comprises the following steps:

[0021] Step 4-1: For the main supervision index T(n) under the nth candidate similar scene, use the supervision index T of the target scene ξ ξ The deviation δ between the nth candidate similar scene and the target scene n Therefore, the following method is used to model and analyze the composition of the candidate similar scene departure supervision indicators:

[0022] T(n)=T ξ +δ n

[0023] Assume that the supervision index deviation in the nth candidate scenario follows a normal distribution with a mean of 0 such that in The variance is generated under the condition of this scene due to the difference in key features between it and the target scene;

[0024] Step 4-2: Assume that there are only two different departure flight operation scenarios i and j in a scenario category, and rewrite the supervision index as Similarly, rewrite another sample as

[0025] Combine multiple samples with the same mean but different variances and apply weights to each sample to minimize the overall variance and obtain a more accurate T S Value:

[0026]

[0027] Among them, ω is the scene similarity, is the estimated value of the supervision index of the target scene, limiting the sum of similarities ω i +ωj =1;

[0028] Step 4-3: Get T through statistical knowledge S The variance value is

[0029]

[0030] Derivative the similarity in the formula:

[0031]

[0032]

[0033] Obviously, the second-order derivative is greater than zero, so the minimum variance value is obtained, and the corresponding similarity values ​​are:

[0034]

[0035] If the corresponding similarity of multiple scenes is determined, the expression is as follows:

[0036]

[0037] Where S is a scene set, which contains n different scenes, so the following results are obtained:

[0038]

[0039] The sum of all similarities is 1, that is,

[0040] Step 4-4: Considering that the sample variance is the minimum value, combined with the final prediction result, it is approximately considered that the supervision index of the target scene is the weighted sum of the samples, which is expressed as:

[0041]

[0042] Similarly, the sum of similarities is 1, that is, Where each weight ω i That is the explainable similarity of the scene;

[0043] Based on the characteristics of explainable similarity, the processed two types of data are respectively input into two neural networks NET1 and NET2. The original data input shapes of the two neural networks are n×1×a cat and n×2×40, where n is the number of candidate similar scenes, α cat is the processed classification feature dimension, and the calculation process is shown in steps 4-5 to 4-8:

[0044] Step 4-5: Downsample the data and visualize it. Downsample the data with a 1×2 convolution kernel. Expand the simplified data to form a two-dimensional tensor, which is similar in size and shape to common pictures.

[0045] Step 4-6: Image-like convolution, using the LeNet-based network framework structure to perform convolution pooling on the tensor, access the fully connected layer at the end of the two neural networks to obtain their respective similarity vectors, and combine the weights to obtain an n×1 similarity fitting result, the number of which is consistent with the number of candidate similar scenes, where the loss function is calculated as:

[0046]

[0047] Where ω is the similarity set of candidate similar scenes under all target scenes, ω mn is the similarity of the nth scene under the mth target scene, The meaning of is the mth target scene feature, x n The meaning is the nth historical scene feature, α is the target scene set, β is the candidate similar scene set, λ is the deviation importance parameter, K is the number of target operation scenes in which all departing aircraft are located during the training process, T(x n ) is the supervision index of the nth candidate similar scene, For the target scene The actual supervision indicator under the target scenario; the loss function consists of two parts, which minimizes the prediction error of the supervision indicator on the training set while reducing the similarity value of the high-deviation scenario. The actual role of the neural network in this process is to calculate the similarity between different scenarios and the target scenario and use it to linearly generate the supervision indicator of the departure flight under the target scenario;

[0048] Step 4-7: Priori processing. In order to ensure that the sum of similarities is 1, the following conditions are required: Based on this restriction, the similarities output from the two neural networks need to be additionally processed a priori: first, all negative similarities of the output need to be mapped to 0, that is, an additional ReLU layer is added at the end of the two neural networks; secondly, all weights need to be normalized, that is, a normalization layer is added after the ReLU layer of the two neural networks; the above-processed similarities are combined and weighted to generate the overall similarity between the target scene and the candidate similar scenes, which is calculated as follows:

[0049]

[0050] Where μ is the combination similarity distribution coefficient, and are the static and dynamic vector similarities of the candidate similar scenes output by NET1 and NET2 under the nth sample target scene;

[0051] Step 4-8: Iterative training and similarity extraction. By linearly generating supervision indicators, the predicted combined similarity is linearly weighted summed with the historical supervision indicators in the candidate similar scenes, and iterative training is performed according to the loss function to reduce the error between the two.

[0052] The airport surface operation scenario interpretable similarity measurement system is characterized by being used in the above method, including:

[0053] Airport surface operation scene data processing module, used to collect data from multiple data sources and perform pre-processing;

[0054] The surface operation scene feature extraction and grouping module is used to extract secondary features according to data types, and divide all features into flight static attribute features and environment dynamic attribute features according to feature characteristics;

[0055] The scene dynamic interpretable similarity calculation module integrates the input data and calculates the dynamic interpretable similarity between the target scene and the historical operation scene.

[0056] Beneficial effects: The airport surface operation scenario interpretable similarity measurement method and system of the present invention designs a neural network model with strong interpretability based on similar scenarios to realize the scenario similarity calculation process. The present invention proposes an airport surface operation scenario interpretable similarity measurement method, which calculates the scenario interpretable similarity by multi-time scale comparison based on heterogeneous feature types, and provides a new method for realizing precise scene control. It can be widely used in the prediction and classification of key indicators of subsequent systems, filling the technical gap in the dynamic interpretable similarity calculation method of airport surface operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A hierarchical partition diagram of the interpretable similarity measurement method for airport surface operation scenarios;

[0058] Figure 2 The core flow chart of the interpretable similarity measurement method for airport surface operation scenarios;

[0059] Figure 3 The flight static attribute feature data structure for interpretable similarity measurement method of airport surface operation scenarios;

[0060] Figure 4 The environmental dynamic attribute feature data structure for interpretable similarity measurement methods of airport surface operation scenarios;

[0061] Figure 5The weight comparison results of one day obtained by interpretable similarity measurement method for airport surface operation scenarios;

[0062] Figure 6 The weight comparison results of different dates obtained by the interpretable similarity measurement method for airport surface operation scenarios;

[0063] Figure 7 Flow chart of an interpretable similarity measurement method for airport surface operation scenarios in an embodiment. DETAILED DESCRIPTION

[0064] In order to make the design purpose, improvement measures and combined advantages of the present invention easier to understand, the present invention is further explained in detail in combination with the following figures and embodiments. It should also be noted that the scene classification samples used in this section are only used to illustrate the feasibility of the classification method and are not exclusively applicable to the processes included in the present invention.

[0065] Figure 1 This is a hierarchical division diagram of the interpretable similarity measurement method for airport surface operation scenarios, which describes the hierarchical division of the interpretable similarity measurement method for airport surface operation scenarios.

[0066] Figure 2 The figure is a flow chart of the interpretable similarity measurement method for airport surface operation scenarios, which describes the process of the method.

[0067] We selected the scenarios faced by departure flights at a certain hub airport and their accompanying data for summary, and set the monitoring object as delay time. The analysis results are as follows:

[0068] Step 1: Collect and process multi-source data. The essence is to collect data from various available data sources from airlines, airports and meteorological departments, integrate and pre-process the data, and build a surface operation scenario feature system from the perspective of preliminary departure flights.

[0069] It includes the following sub-steps:

[0070] (1.1) Collect original flight and weather data from different data channels, including at least: airport collaborative management system data, airline operation control system data, airport weather message data, and relevant scene data involving the target airport;

[0071] (1.2) Preprocessing the raw data obtained from multiple data sources, including screening for abnormal data and filling in missing data. The preprocessing method is to delete all missing data that cannot be filled, and extract flight operation scenario data within the target time period based on the features selected above. The present invention eliminates all flight entries whose taxiing time exceeds three times the standard deviation of the mean based on the Laida criterion.

[0072] Statistics of various flight-related data at the busy airport include but are not limited to airport collaborative decision-making data, flight plan data submitted by airlines, airport weather message time, etc. The original data is processed by means of difference completion or deletion of abnormal data. The cleaned original data is integrated to form the original departure flight scenario database.

[0073] Step 2: Feature extraction and grouping in the field operation scenario is essentially to process the secondary features based on the extracted scene data. According to the characteristics of the features, they are divided into two types of features: flight static attribute features and environmental dynamic attribute features to provide conditions for subsequent network input and data construction. The specific process is as follows:

[0074] (2.1) Perform secondary feature extraction on the processed data, integrate all features, and perform normalization and encoding processing;

[0075] (2.2) According to the characteristics of the extracted features, the features are divided into two parts, namely, flight static attribute features and environmental dynamic attribute features; the processed and statistically processed data are stored, and a flight static attribute feature database and an environmental dynamic attribute feature database are established. The flight static attribute refers to the features related to the flight status itself or the content specified in the flight plan, and the environmental dynamic attribute features refer to the features related to the dynamic attributes such as the current traffic and weather at the airport.

[0076] According to the characteristics of the supervision target, secondary indicators are extracted, such as flight number, aircraft type, scene congestion evaluation index, airport weather status and other indicators. The categorical variables are processed by unique hot encoding. The processed similar scene data are classified according to their characteristics. The characteristics unique to the current flight are classified as flight static attribute characteristics, and the characteristics that have periodic changes over time are classified as environmental dynamic attribute characteristics. The two characteristics are matched and stored separately. The construction result is as follows Figure 3 and Figure 4 shown.

[0077] Step 3: The data construction method of interpretable similarity data is essentially to compare the similarity according to the characteristics of the two types of data. The two types of data are constructed separately according to the repeatability characteristics of the data. The specific construction process includes the following steps:

[0078] (3.1) Construction of flight static attribute input structure. Most of the flight static attribute features are categorical features and require additional encoding processing. The processed vector is called the static attribute vector θ i , all sample target scenes and their respective candidate similar scene sets are stacked into an input structure, and the final input data format is s×n×2×a cat , where s represents the number of sample scenes, αcat is the dimension of the processed static attribute vector, and n is the number of candidate scenes. Under this construction method, the static attributes of all sample target scenes and each candidate similar scene are compared.

[0079] (3.2) The input structure of environmental dynamic attributes is constructed. The environmental dynamic attribute features of all scenes at the same time in the previous few days are spliced ​​to form a multi-time scale environmental dynamic attribute input vector for one day, which is called the dynamic attribute vector θ e , process all scene data according to the above steps; combine all scenes with their respective candidate similar scene sets to finally form input data.

[0080] The data obtained in step 2 are further processed. For data without time period characteristics, i.e., flight static attribute features and their corresponding similar data models in candidate similar scenes are stacked to form a flight static attribute input structure. For environmental dynamic attribute features with time period properties, several periodic data are extracted to construct the environmental dynamic attribute input structure. Finally, two feature sets for training dynamic similar scenes are formed.

[0081] Step 4: The data construction method of dynamic interpretable similarity is essentially a dynamic interpretable similarity calculation method based on the static attribute feature data of the flight and the dynamic attribute data of the environment to evaluate the similarity, which specifically includes the following steps:

[0082] (4.1) Scene index decomposition. For the main supervisory index T(n) under the nth candidate similar scene, the present invention uses the supervisory index T of the target scene ξ ξ The deviation δ between the nth candidate similar scene and the target scene n Therefore, the following method is used to model and analyze the composition of the candidate similar scene departure supervision indicators:

[0083] T(n)=T ξ +δ n

[0084] Various influencing factors and uncertainties involved in the scene will cause the supervision index of the current scene to be shortened or extended compared with the target scene. Therefore, for the convenience of analysis, it is assumed that the supervision index deviation in the nth candidate scene follows a normal distribution with a mean of 0. in The variance is generated under the condition of this scene due to the difference in key features between it and the target scene.

[0085] (4.2) Assuming that there are only two different departure flight operation scenarios i and j in a scenario class, the supervision index can be rewritten as Another example can be rewritten as

[0086] Combine multiple samples with the same mean but different variances, and apply weights to each sample to minimize the overall variance and obtain a more accurate T S Value:

[0087]

[0088] Among them, ω is the scene similarity, is the estimated value of the supervision index of the target scene, limiting the sum of similarities ω i +ω j =1.

[0089] (4.3) Through statistical knowledge, we can get T S The variance value is

[0090]

[0091] By taking the derivative of the similarity in the formula, we can get:

[0092]

[0093]

[0094] Obviously, the second-order derivative is greater than zero, so the minimum variance value can be obtained, and the corresponding similarity values ​​are:

[0095]

[0096] If the corresponding similarity of multiple scenes is determined, the expression is as follows:

[0097]

[0098] Where S is a scene set, including n different scenes. Therefore, the following results can be obtained:

[0099]

[0100] The sum of all similarities is 1, that is, (4.4) At the same time, considering that the sample variance is the minimum value, combined with the final prediction result, it can be approximately considered that the supervision index of the target scene is the weighted sum of the samples. The experimental results show that this approximation method can meet the actual calculation requirements, and its expression is:

[0101]

[0102] Similarly, the sum of similarities is 1, that is, Where each weight ω i This is the explainable similarity of the scene.

[0103] Based on the characteristics of explainable similarity, the processed two types of data are respectively input into two neural networks NET1 and NET2. The original data input shapes of the two neural networks are n×1×a cat and n×2×40, where n is the number of candidate similar scenes, α cat is the processed classification feature dimension, and the calculation process is shown in steps (4-5) to (4-8):

[0104] (4.5) Data downsampling and visualization: downsampling the data with a 1×2 convolution kernel; the simplified data is expanded to form a two-dimensional tensor, whose size and shape are similar to common pictures;

[0105] (4.6) Image convolution, using the LeNet-based network framework structure to perform convolution pooling on the tensor. At the end of the two neural networks, the fully connected layer is connected to obtain the respective similarity vectors, and the weights are combined to obtain an n×1 similarity fitting result, the number of which is consistent with the number of candidate similar scenes used in the present invention. The loss function is calculated as follows:

[0106]

[0107] Where ω is the similarity set of candidate similar scenes under all target scenes, ω mn is the similarity of the nth scene under the mth target scene, The meaning of is the mth target scene feature, x n The meaning is the nth historical scene feature, α is the target scene set, β is the candidate similar scene set, λ is the deviation importance parameter, K is the number of target operation scenes in which all departing aircraft are located during the training process, T(x n ) is the supervision index of the nth candidate similar scene, For the target scene The actual supervision indicator under the target scenario. The loss function consists of two parts. It minimizes the prediction error of the supervision indicator on the training set while reducing the similarity value of the high-deviation scenario. The actual role of the neural network in this process is to calculate the similarity between different scenarios and the target scenario and use it to linearly generate the supervision indicator of the departure flight under the target scenario;

[0108] (4.7) Priori processing, in order to ensure that the similarity sum is 1, the following conditions are required: Based on this constraint, the similarity output from the two neural networks needs to be processed a priori: first, all negative similarities of the output need to be mapped to 0, that is, an additional ReLU layer is added at the end of the two neural networks; second, all weights need to be normalized, that is, a normalization layer is added after the ReLU layer of the two neural networks. The above processed similarities are combined and weighted to produce the overall similarity between the target scene and the candidate similar scenes, which is calculated as follows:

[0109]

[0110] Where μ is the combination similarity distribution coefficient, and They are the static and dynamic vector similarities of the candidate similar scenes output by NET1 and NET2 under the nth sample target scene. This method can significantly reduce the parameter search space and accelerate the convergence of the neural network. (4.8) Iterative training and similarity extraction, by linearly generating supervision indicators, the combined similarity generated by the prediction is linearly weighted summed with the historical supervision indicators in the candidate similar scenes, and iterative training is performed according to the loss function to reduce the error between the two.

[0111] The flight data in step 3 is extracted and trained using the convolutional neural network proposed in step 4 of the present invention. The parameter settings are shown in Table 1. The scene similarities in the static attribute features and dynamic environment attribute features of the flight are compared respectively. Finally, the similarity between the target scene and all candidate similar scenes is evaluated based on the scene metric vector after the full connection layer and prior processing. The similarity evaluation results of one day are shown in Figure 5 As shown in Figure 2, the similarity comparison results for multiple days are as follows: Figure 6 The resulting scene dynamic similarity can be used in subsequent scene indicator prediction or scene category clustering analysis.

[0112] Table 1. Neural network parameter diagram in the interpretable similarity measurement method for airport surface operation scenarios

[0113]

[0114]

[0115] The present invention also provides an interpretable similarity measurement system for airport surface operation scenarios, which is used to implement the above method, specifically comprising:

[0116] Airport surface operation scene data processing module, used to collect data from multiple data sources and perform pre-processing;

[0117] The surface operation scene feature extraction and grouping module is used to extract secondary features according to data types, and divide all features into flight static attribute features and environment dynamic attribute features according to feature characteristics;

[0118] The scene dynamic interpretable similarity calculation module integrates the input data and calculates the dynamic interpretable similarity between the target scene and the historical operation scene.

[0119] The present invention provides an interpretable similarity measurement method and system for airport surface operation scenarios. There are many methods and ways to implement the technical solution. The above is only a preferred implementation of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. A method for measuring the interpretable similarity of airport surface operation scenarios, characterized by: The following steps are involved: Step 1: collect flight data and meteorological data from different data channels as scenario data, pre-process them, and construct a scenario operation data set from the perspective of departing flights; Step 2, processing the data in the scene operation data set obtained in step 1 with secondary features, and dividing them into two types of features: flight static attribute features and environment dynamic attribute features according to their characteristics; Step 3, constructing a data structure that can explain similarity data using the two types of features obtained in step 2; Step 4, apply the data obtained in step 3 to the network for calculating interpretable similarity, and calculate the interpretable similarity of the field scene operation scenarios based on the two types of data.

2. The interpretable similarity measurement method for airport surface operation scenarios according to claim 1 is characterized by: The step 1 specifically comprises the following steps: Step 1-1, collecting original flight and weather data from different data channels, including at least: airport collaborative management system data, airline operation control system data, and relevant scene data involving the target airport in airport weather message data; Step 1-2, preprocessing the raw data collected in step 1-1, including screening out abnormal data and completing missing data.

3. The interpretable similarity measurement method for airport surface operation scenarios according to claim 1 is characterized by: The step 2 specifically includes the following steps: Step 2-1, extract secondary features from the data processed in step 1, integrate all features, and perform normalization and encoding processing; Step 2-2, according to the characteristics of the extracted features, the features are divided into two parts, namely, flight static attribute features and environmental dynamic attribute features; the processed and statistically processed data are stored, and a flight static attribute feature database and an environmental dynamic attribute feature database are constructed.

4. The interpretable similarity measurement method for airport surface operation scenarios according to claim 1 is characterized by: The step 3 specifically comprises the following steps: Step 3-1: Construct the input structure of flight static attributes. Most of the flight static attribute features are categorical features and require additional encoding processing. The processed vector is called the static attribute vector θ i , all sample target scenes and their respective candidate similar scene sets are stacked into an input structure, and the final input data format is s×n×2×a cat , where s represents the number of sample scenes, α cat is the dimension of the processed static attribute vector, and n is the number of candidate scenes. Under this construction method, the static attributes of the flights of all sample target scenes and each candidate similar scene are compared. Step 3-2: construct the input structure of the environment dynamic attribute. The environment dynamic attribute features of all scenes at the same time in the previous few days are spliced ​​to form a multi-time scale environment dynamic attribute input vector for one day, which is called the dynamic attribute vector θ e , process the vectors of all scene data according to the above steps; combine all scenes with their respective candidate similar scene sets to finally form the input data.

5. The interpretable similarity measurement method for airport surface operation scenarios according to claim 1 is characterized by: The step 4 specifically comprises the following steps: Step 4-1: For the main supervision index T(n) under the nth candidate similar scene, use the supervision index T of the target scene ξ ξ The deviation δ between the nth candidate similar scene and the target scene n Therefore, the following method is used to model and analyze the composition of the candidate similar scene departure supervision indicators: T(n)=T ξ +δ n Assume that the supervision index deviation in the nth candidate scenario follows a normal distribution with a mean of 0 such that in The variance is generated under the condition of this scene due to the difference in key features between it and the target scene; Step 4-2: Assume that there are only two different departure flight operation scenarios i and j in a scenario category, and rewrite the supervision index as Similarly, rewrite another sample as Combine multiple samples with the same mean but different variances, and apply weights to each sample to minimize the overall variance and obtain a more accurate T S Value: Among them, ω is the scene similarity, is the estimated value of the supervision index of the target scene, limiting the sum of similarities ω i +ω j =1; Step 4-3: Get T through statistical knowledge S The variance value is Derivative the similarity in the formula: Obviously, the second-order derivative is greater than zero, so the minimum variance value is obtained, and the corresponding similarity values ​​are: If the corresponding similarity of multiple scenes is determined, the expression is as follows: Where S is a scene set, which contains n different scenes, so the following results are obtained: The sum of all similarities is 1, that is, Step 4-4: Considering that the sample variance is the minimum value, combined with the final prediction result, it is approximately considered that the supervision index of the target scene is the weighted sum of the samples, which is expressed as: Similarly, the sum of similarities is 1, that is, Where each weight ω i That is the explainable similarity of the scene; Based on the characteristics of explainable similarity, the processed two types of data are respectively input into two neural networks NET1 and NET2. The original data input shapes of the two neural networks are n×1×a cat and n×2×40, where n is the number of candidate similar scenes, α cat is the processed classification feature dimension, and the calculation process is shown in steps 4-5 to 4-8: Step 4-5: Downsample the data and visualize it. Downsample the data with a 1×2 convolution kernel. Expand the simplified data to form a two-dimensional tensor, which is similar in size and shape to common pictures. Step 4-6: Image-like convolution, using the LeNet-based network framework structure to perform convolution pooling on the tensor, access the fully connected layer at the end of the two neural networks to obtain their respective similarity vectors, and combine the weights to obtain an n×1 similarity fitting result, the number of which is consistent with the number of candidate similar scenes, where the loss function is calculated as: Where ω is the similarity set of candidate similar scenes under all target scenes, ω mn is the similarity of the nth scene under the mth target scene, The meaning of is the mth target scene feature, x n The meaning is the nth historical scene feature, α is the target scene set, β is the candidate similar scene set, λ is the deviation importance parameter, K is the number of target operation scenes in which all departing aircraft are located during the training process, T(x n ) is the supervision index of the nth candidate similar scene, For the target scene The actual supervision indicator under the target scenario; the loss function consists of two parts, which minimizes the prediction error of the supervision indicator on the training set while reducing the similarity value of the high-deviation scenario. The actual role of the neural network in this process is to calculate the similarity between different scenarios and the target scenario and use it to linearly generate the supervision indicator of the departure flight under the target scenario; Step 4-7: Priori processing. In order to ensure that the sum of similarities is 1, the following conditions are required: Based on this restriction, the similarities output from the two neural networks need to be additionally processed a priori: first, all negative similarities of the output need to be mapped to 0, that is, an additional ReLU layer is added at the end of the two neural networks; secondly, all weights need to be normalized, that is, a normalization layer is added after the ReLU layer of the two neural networks; the above-processed similarities are combined and weighted to generate the overall similarity between the target scene and the candidate similar scenes, which is calculated as follows: Where μ is the combination similarity distribution coefficient, and are the static and dynamic vector similarities of the candidate similar scenes output by NET1 and NET2 under the nth sample target scene; Step 4-8: Iterative training and similarity extraction. By linearly generating supervision indicators, the predicted combined similarity is linearly weighted summed with the historical supervision indicators in the candidate similar scenes, and iterative training is performed according to the loss function to reduce the error between the two.

6. An interpretable similarity measurement system for airport surface operation scenarios, characterized by: The method for implementing any one of claims 1 to 5 comprises: Airport surface operation scene data processing module, used to collect data from multiple data sources and perform pre-processing; The surface operation scene feature extraction and grouping module is used to extract secondary features according to data types, and divide all features into flight static attribute features and environment dynamic attribute features according to feature characteristics; The scene dynamic interpretable similarity calculation module integrates the input data and calculates the dynamic interpretable similarity between the target scene and the historical operation scene.

Citation Information

Patent Citations

  • A data preprocessing method based on 1 / 2 similarity deviation

    CN109902762A

  • Airport surface variable slide-out time prediction method based on big data deep learning

    CN110826788A

  • Scene variable slide-out time prediction system based on big data deep learning

    CN110852497A

  • Semi-supervised measurement, adjustment measurement matrix and terminal area operation scene identification method

    CN114529179A

  • Inplausible SAR target identification method and system based on VIT model and prototype network

    CN115775332A