Method and apparatus for processing flight safety mission data

Abstract

The embodiment of the application provides a flight safety task data processing method and device, the method comprises the following steps: creating a flight training plan by a training personnel, obtaining flight preparation professional information by filling in professional information by each flight preparation professional, obtaining flight task training information by filling in training information by a trainee pilot according to the flight preparation professional information, inputting the flight preparation professional information and the flight task training information into a trained risk list generation model to generate a flight safety task risk list, wherein the risk list generation model is to perform feature extraction on historical flight preparation professional information and historical flight task training information, to perform counterfactual analysis on the features after the feature extraction operation according to a preset counterfactual scene to obtain flight safety features, and to input the flight safety features and risk list labels into an initial deep learning model for model training to obtain the risk list generation model, and the application can improve the efficiency and accuracy of flight preparation work.

Description

Technical Field

[0001] This application relates to the field of data processing, specifically to a method and apparatus for processing flight safety mission data. Background Technology

[0002] In the field of flight safety, current flight preparation faces two key problems. First, the data and information that flight preparation personnel from various specialties need to acquire during the flight preparation phase often exist in multiple formats, such as images, paper documents, or electronic documents. The formats submitted to training personnel are inconsistent and relatively fragmented. Training personnel must then compile this information and submit it to their respective flight squadrons, which in turn organize flight personnel to review and study it, conduct flight risk assessments, and have the squadron leader review the risks before finally executing the flight plan. This inconsistent data processing and risk assessment method not only increases the difficulty for training personnel in compiling and organizing data but also reduces the efficiency of flight preparation work.

[0003] Secondly, although the flight preparation process itself is relatively complete, there is a lack of unified management methods for controlling and managing each stage during implementation. This results in blind spots in the execution and monitoring of flight preparation work, making it difficult for managers to have a comprehensive and real-time understanding of the progress of flight preparation, and thus making it impossible to identify and resolve potential problems in a timely manner.

[0004] Therefore, there is an urgent need for a flight safety mission data processing method to improve the efficiency and accuracy of flight preparation work and achieve closed-loop management of the entire flight preparation process. Summary of the Invention

[0005] To address the problems in the prior art, this application provides a flight safety mission data processing method and apparatus, which can improve the efficiency and accuracy of flight preparation work.

[0006] To solve at least one of the above problems, this application provides the following technical solution:

[0007] Firstly, this application provides a method for processing flight safety mission data, including:

[0008] The system receives flight training plans created by flight mission training personnel, sends professional information reporting notifications to each flight preparation professional, enabling each flight preparation professional to report professional information according to the notifications, and receives flight preparation professional information returned by each flight preparation professional. The system then sends the flight preparation professional information to the preset training pilot's terminal, enabling the training pilot to report training information according to the flight preparation professional information, and receives flight mission training information returned by the training pilot.

[0009] Collect historical flight preparation professional information and historical flight mission training information, perform feature extraction on the historical flight preparation professional information and historical flight mission training information, perform counterfactual analysis on the features after feature extraction based on a preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, input the flight safety features and risk list labels into an initial deep learning model for model training to obtain the corresponding risk list generation model;

[0010] The flight preparation professional information and the flight mission training information are input into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

[0011] Furthermore, the process of receiving the flight training plan created by the flight mission training personnel, sending professional information reporting notifications to each flight preparation professional, enabling each of the flight preparation professionals to report professional information according to the notifications, and receiving the flight preparation professional information returned by each of the flight preparation professionals, includes:

[0012] The system receives flight training plans created by flight mission training personnel and sends professional information reporting notices to various flight preparation professionals, enabling them to report professional information based on their respective knowledge of flight safety. These flight preparation professionals include political workers, aviation medical personnel, meteorologists, birdwatchers, aircraft maintenance personnel, and air traffic controllers.

[0013] The system receives professional information returned by each of the flight preparation professionals and sends the professional information to the flight mission training personnel. The flight mission training personnel review the professional information, and if it passes, they determine the corresponding flight preparation professional information.

[0014] Furthermore, before sending the flight preparation professional information to the preset training pilot's terminal, the following steps are included:

[0015] Determine whether the pilot meets the training standards based on the aforementioned flight preparation professional information;

[0016] If the criteria are met, the corresponding training pilot's terminal will be determined.

[0017] Further, the process involves collecting historical flight preparation professional information and historical flight mission training information, performing feature extraction on the historical flight preparation professional information and the historical flight mission training information, conducting counterfactual analysis on the features after the feature extraction operation according to a preset counterfactual scenario, determining the flight risk probability distribution under the counterfactual scenario, comparing the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, constructing a flight safety causal graph, and extracting the causal relationships in the flight safety causal graph to construct corresponding flight safety features, including:

[0018] Collect historical flight preparation professional information and historical flight mission training information. Among them, flight preparation professional information is professional information used to characterize flight safety-related factors, and flight mission training information is information used to characterize the pilot's own training arrangements.

[0019] Multi-source feature extraction and risk correlation analysis are performed on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding highly correlated risk features.

[0020] Based on a preset counterfactual scenario, counterfactual analysis is performed on the highly relevant risk features to determine the flight risk probability distribution under the counterfactual scenario. After comparing the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, a flight safety causal graph is constructed, and the causal relationships in the flight safety causal graph are extracted to construct the corresponding flight safety features.

[0021] Furthermore, the multi-source feature extraction and risk correlation analysis operations performed on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding highly correlated risk features include:

[0022] Multi-source feature extraction is performed on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding multi-source flight safety features. The multi-source flight safety feature data is used as nodes, and feature aggregation is performed on the multi-source flight safety features according to the graph neural network to determine the corresponding global flight safety features.

[0023] Based on the Pearson correlation coefficient, risk correlation analysis is performed on the global flight safety characteristics to determine the corresponding highly correlated risk characteristics.

[0024] Furthermore, the step of performing risk correlation analysis on the global flight safety characteristics based on the Pearson correlation coefficient to determine the corresponding highly correlated risk characteristics includes:

[0025] The global flight safety characteristics are assessed for flight risk correlation based on the Pearson correlation coefficient to determine the corresponding correlation values.

[0026] The correlation value is compared with a preset high correlation threshold to determine the corresponding high correlation risk characteristic.

[0027] Furthermore, after inputting the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model, the method further includes:

[0028] The flight safety mission risk list is sent to the trainee pilot's terminal so that the trainee pilot can learn about the risks based on the flight safety mission risk list;

[0029] The system receives the risk learning results returned by the participating pilots, sends risk confirmation notifications to the preset squadron leader and preset commander terminals, so that the squadron leader and commander can perform risk confirmation operations according to the risk confirmation notifications, and accepts the risk confirmation opinions returned by the squadron leader and commander terminals, and sends the risk confirmation opinions to the participating pilots' terminals, so that the participating pilots can conduct flight training according to the risk confirmation opinions.

[0030] Secondly, this application provides a flight safety mission data processing device, comprising:

[0031] The flight information confirmation module is used to receive flight training plans created by flight mission training personnel, send professional information filling notifications to each flight preparation professional, so that each flight preparation professional can fill in professional information according to the professional information filling notifications, receive flight preparation professional information returned by each flight preparation professional, send the flight preparation professional information to the preset training pilot terminal, so that the training pilot can fill in training information according to the flight preparation professional information, and receive flight mission training information returned by the training pilot.

[0032] The risk list generation model training module is used to collect historical flight preparation professional information and historical flight mission training information, perform feature extraction operations on the historical flight preparation professional information and the historical flight mission training information, perform counterfactual analysis on the features after the feature extraction operation according to the preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, and input the flight safety features and risk list labels into the initial deep learning model for model training to obtain the corresponding risk list generation model;

[0033] The flight safety mission risk list determination module is used to input the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

[0034] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the flight safety mission data processing method.

[0035] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the flight safety mission data processing method described above.

[0036] Fifthly, this application provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the flight safety mission data processing method.

[0037] As can be seen from the above technical solution, this application provides a flight safety mission data processing method and apparatus. The method involves training personnel creating flight training plans, flight preparation professionals filling in professional information to obtain flight preparation professional information, and participating pilots filling in training information based on the flight preparation professional information to obtain flight mission training information. The flight preparation professional information and flight mission training information are then input into a trained risk list generation model to generate a flight safety mission risk list. The risk list generation model extracts features from historical flight preparation professional information and historical flight mission training information. Based on a preset counterfactual scenario, counterfactual analysis is performed on the features after the feature extraction operation to obtain flight safety features. These flight safety features and risk list labels are then input into an initial deep learning model for model training. This improves the efficiency and accuracy of flight preparation work. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is one of the flowcharts illustrating the flight safety mission data processing method in the embodiments of this application;

[0040] Figure 2 This is a second flowchart illustrating the flight safety mission data processing method in the embodiments of this application;

[0041] Figure 3 This is the third flowchart illustrating the flight safety mission data processing method in this application embodiment;

[0042] Figure 4 This is the fourth flowchart illustrating the flight safety mission data processing method in the embodiments of this application;

[0043] Figure 5 This is the fifth flowchart illustrating the flight safety mission data processing method in the embodiments of this application;

[0044] Figure 6 This is the sixth flowchart illustrating the flight safety mission data processing method in the embodiments of this application;

[0045] Figure 7 This is the seventh flowchart illustrating the flight safety mission data processing method in this application embodiment;

[0046] Figure 8 This is a structural diagram of the flight safety mission data processing device in the embodiments of this application;

[0047] Figure 9 This is a schematic diagram of the flight safety mission function module in the embodiments of this application;

[0048] Figure 10 This is a schematic diagram of the structure of the electronic device in the embodiments of this application.

[0049] Figure label:

[0050] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver storage unit 9144, antenna 9111, speaker 9131, microphone 9132. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0052] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.

[0053] In the field of flight safety, current flight preparation information is fragmented and lacks unified management methods, resulting in low efficiency in flight preparation and risk assessment management. This application provides a flight safety mission data processing method and apparatus. The method involves trainees creating flight training plans, flight preparation professionals filling in professional information to obtain flight preparation professional information, and participating pilots filling in training information based on the flight preparation professional information to obtain flight mission training information. The flight preparation professional information and flight mission training information are then input into a trained risk list generation model to generate a flight safety mission risk list. The risk list generation model extracts features from historical flight preparation professional information and historical flight mission training information. Based on a preset counterfactual scenario, counterfactual analysis is performed on the features extracted to obtain flight safety features. These flight safety features, along with risk list labels, are then input into an initial deep learning model for training. This improves the efficiency and accuracy of flight preparation.

[0054] To improve the efficiency and accuracy of flight preparation, this application provides an embodiment of a flight safety mission data processing method, see [link to embodiment]. Figure 1 and Figure 9 The flight safety mission data processing method specifically includes the following:

[0055] Step S101: Receive the flight training plan created by the flight mission training personnel, send a professional information filling notification to each flight preparation professional, so that each flight preparation professional can fill in the professional information according to the professional information filling notification, and receive the flight preparation professional information returned by each flight preparation professional, and send the flight preparation professional information to the preset training pilot terminal, so that the training pilot can fill in the training information according to the flight preparation professional information, and receive the flight mission training information returned by the training pilot.

[0056] Optionally, in this embodiment, this step addresses the technical problem that current flight preparation information is too fragmented and lacks a unified management method.

[0057] Specifically, such as Figure 9 The functional modules implemented by this method include basic information maintenance, task information maintenance, task information entry, task status management, task information retrieval, message push, task risk assessment, risk situation feedback, and risk details retrieval.

[0058] The following is an explanation of the function of each module:

[0059] Basic information maintenance: This includes authorizing secondary administrators to existing ordinary users and maintaining organizational information, such as organization name and members; maintaining all user information within the system, including personnel positions, roles, and basic information; and maintaining risk information, including risk classification, importance, probability of occurrence, cause analysis, and improvement measures. Secondary administrators can maintain user information for their respective organizations, while ordinary users can maintain their own basic information.

[0060] Task Information Maintenance: For users with Level 2 administrator privileges or users with the role of trainer, the system has the function of maintaining task information, including task creation, training day creation, selection of trainees, end of training day, and end of task, as well as selecting trainees and controlling the task execution status.

[0061] Task Information Submission: Users with roles such as training, political work, aviation medicine, meteorology, bird monitoring, aircraft maintenance, air traffic control, and pilots can submit task information for the tasks they are currently performing. They can fill in the professional information corresponding to their role and can make multiple modifications.

[0062] Task Status Management: Users with roles such as trainers, pilots, squadron leaders, and commanders can view the status of the data entered during the current task. Trainers can send pending messages to personnel who have not yet entered data for this task, urging them to complete the data entry as soon as possible. Squadron leaders or commanders can supervise pilots who have not yet conducted risk assessments and confirm risk management for pilots who have completed risk assessments.

[0063] Mission Information Inquiry: Users with roles such as trainers, pilots, squadron leaders, and commanders can access the professional information reported by the mission information reporting module to learn about the details of this mission.

[0064] Message push: For users whose role is training, after creating a training day, a push notification will be sent to users currently participating in this training day whose roles are training, political work, aviation medicine, meteorology, bird monitoring, aircraft maintenance, and air traffic control, reminding them that this training day has been created and to fill in the relevant professional information as soon as possible; for users whose role is training, after each professional information is filled in, a notification will be sent to the training personnel that a certain professional information has been filled in; for users whose roles are squadron leader or commander, after the pilot completes the mission risk assessment, a notification will be sent to the squadron leader and commander that a certain pilot has completed the risk assessment.

[0065] Mission Risk Assessment: After the professional information of training, political work, aviation medicine, meteorology, bird situation, aircraft maintenance, and air traffic control is filled in, the user with the role of pilot can fill in the course information, intermittent information, flight scenario, and personal energy status according to their own training schedule. The system will calculate a list of risks that may occur based on the information filled in by each professional and the information filled in by the pilot, so that the pilot can view and learn from it.

[0066] Risk Feedback: During the debriefing and analysis of training days, users with the role of pilots can use this function to report the risks encountered during the actual training process.

[0067] Risk details viewing: Open to all users, allowing fuzzy searches by risk name or combined queries based on probability of occurrence and importance, enabling users to view and learn about risks.

[0068] Specifically, with the cooperation of the above functional modules, the specific process implemented in this step is as follows:

[0069] First, the trainees receive the training plan and then create training days.

[0070] After the training day is created, trainees, political workers, aviation medical personnel, meteorologists, birdwatchers, aircraft maintenance personnel, and air traffic controllers will be notified via message reminders to fill in professional information.

[0071] After each major is filled out, the training staff will review it. If the review fails, the information will be modified. Once all major information is approved, the major information will be obtained.

[0072] Based on the information submitted by the professional, it is determined whether the pilot meets the training standards. If the standards are not met, the pilot cannot participate in this training. If the standards are met, the pilot is allowed to enter the flight preparation stage to fill in training information and view professional information.

[0073] The system will conduct a risk assessment based on the professional information and pilot training information provided, and determine the potential risks that the pilot may face based on this training mission.

[0074] After the pilots complete the risk learning, they notify their squadron leader and the commander of the training session via message reminders to confirm the risks.

[0075] After the squadron leader and commander confirmed the risks, the pilot was allowed to conduct this flight training.

[0076] After the flight training, a debriefing and explanation were conducted to understand the actual risks that occurred during the training. The risk feedback function was then used to report the actual risks encountered.

[0077] After completing the training for this training day, the trainees will set the training day status to "end" to conclude the training session.

[0078] Through the above steps, the problem of inconsistent and fragmented data, information, and file formats that required professionals to obtain from various business systems after developing training plans during the flight preparation phase was successfully solved, thus improving the management efficiency of the flight preparation phase.

[0079] Step S102: Collect historical flight preparation professional information and historical flight mission training information, perform feature extraction on the historical flight preparation professional information and the historical flight mission training information, perform counterfactual analysis on the features after feature extraction based on a preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, input the flight safety features and risk list labels into the initial deep learning model for model training to obtain the corresponding risk list generation model;

[0080] Optionally, in this embodiment, this step is a specific implementation of the mission risk assessment function in step S101 above. The risk assessment function automatically generates a risk list based on the flight preparation professional information submitted by professionals and the flight mission training information submitted by pilots, effectively improving the efficiency of manual risk assessment.

[0081] Specifically, the professional information on flight preparation reported by professionals includes, but is not limited to:

[0082] Political personnel: the pilots' psychological state, teamwork, etc.

[0083] Flight medical personnel: the pilot's health condition, fatigue level, etc.

[0084] Meteorologists: Weather conditions, wind speed, visibility, etc.

[0085] Bird monitoring personnel: bird activity status, bird strike risk, etc.

[0086] Aircraft maintenance personnel: aircraft maintenance status, equipment malfunction records, etc.

[0087] Air traffic controllers: airspace usage, route planning, etc.

[0088] Specifically, the flight mission training information filled out by pilots includes, but is not limited to:

[0089] Pilot: Specific information about the flight mission, such as flight course, flight scenario, intermittent information, and personal energy level.

[0090] Optionally, in this embodiment, the aforementioned historical information is collected and cleaned, standardized, and formatted so that subsequent models can process it effectively.

[0091] Optionally, in this embodiment, feature engineering is performed to extract features related to flight safety from the collected data. Feature engineering is a crucial step, as it determines whether the model can extract useful information from the data for risk assessment.

[0092] In feature engineering, the first step is to extract flight safety-related features for each type of professional information.

[0093] From the professional information of historical political workers, we extract characteristics such as pilots' psychological stress, emotional stability, anxiety level, cooperation between pilots and crew members, communication efficiency, pilots' flight hours, and mission experience.

[0094] Extract information such as pilots' physical health indicators, pilots' fatigue index, and whether pilots have recent illnesses or injury records from historical aviation medical personnel's professional information;

[0095] Extract information such as wind speed, visibility, precipitation probability, temperature, weather changes in the next few hours, and whether there will be extreme weather such as thunderstorms, strong winds, and dense fog from historical meteorological personnel's professional information;

[0096] Bird information is extracted from historical bird information personnel, including the frequency of bird activity in the flight area, the presence of large or high-risk birds, and whether bird strikes have occurred in the area.

[0097] Extract information such as the operating status of each aircraft system, whether maintenance has been performed recently, whether the maintenance was qualified, and whether the aircraft has any fault records from the historical professional information of aircraft maintenance personnel.

[0098] Extract information from historical air traffic controllers' professional information, such as the current airspace's busyness, whether there are other aircraft activities, whether the flight path passes through complex terrain or dangerous areas, and whether there are temporary traffic controls or flight path adjustments.

[0099] Extract information from historical flight mission training data, such as the type of flight mission (e.g., takeoff and landing training, formation flight, night flight), the scenario of the flight mission (e.g., mountainous areas, ocean, cities), whether the pilot has not flown for a long time, and the ratio of personal energy level.

[0100] The second step involves extracting information separately and then modeling the flight safety characteristics of various professionals (political workers, aviation medical personnel, meteorologists, birdwatchers, aircraft maintenance personnel, and air traffic controllers) and pilots into a graph structure. Nodes in the graph represent entities (such as pilots, meteorological data, aircraft maintenance data, etc.), and edges represent the relationships between entities (such as the relationship between pilots and meteorological conditions, the relationship between aircraft maintenance personnel and aircraft status, etc.). Weights for each edge are set according to the strength of the association between features.

[0101] Then, for each node, the features of its neighboring nodes are aggregated. For example, a pilot's neighboring nodes include meteorological data, aircraft maintenance data, and aviation medical data. By updating the features of each node through a graph neural network, and combining its own features with those of its neighbors, the graph neural network can extract global features through multi-layer information transmission and aggregation.

[0102] By leveraging the information transfer and aggregation capabilities of multi-layer graph neural networks, global multi-source data fusion features can be extracted. These features can capture the complex relationships between different data sources, providing more comprehensive information for flight safety mission risk assessment.

[0103] For example: Suppose that the initial feature of each node is its original data.

[0104] For node 1 (pilot):

[0105] By aggregating features from neighboring nodes 2 (meteorological data), 3 (aircraft maintenance data), and 4 (aviation medical data), the updated flight safety features may include: flight experience, psychological stress score, wind speed, visibility, engine status, fatigue index, etc. The updated features further aggregate global information from other nodes, such as the interaction between meteorological and aircraft maintenance data.

[0106] Ultimately, global multi-source data fusion features can be extracted:

[0107] Global Feature 1: Pilot operational risks under complex weather conditions (combining pilot experience, psychological state, wind speed, visibility, etc.).

[0108] Global Feature 2: Pilot operational risks when the aircraft is in poor condition (combining pilot experience, engine condition, maintenance records, etc.).

[0109] Global Feature 3: Operational risks of pilots under fatigue (combining pilot experience, fatigue index, heart rate, etc.).

[0110] The third step is to extract the global features and then use the Pearson correlation coefficient to analyze the correlation between the extracted global features and the target variable, selecting the features that are highly correlated with the target variable.

[0111] For example, suppose we have a global feature: "the operational risk of a pilot in a state of fatigue."

[0112] According to Pearson correlation coefficient analysis, if the correlation between the feature "pilot's operational risk under fatigue" and "flight risk" is >0.5 (correlation threshold), then this feature is selected.

[0113] The fourth step is to construct a counterfactual scenario after screening out highly relevant global features, and further screen out highly relevant global features that have a direct causal relationship with flight risks through counterfactual analysis.

[0114] Specifically, all "global characteristic variables" highly correlated with flight safety have now been identified, along with the target variable "flight risk." Counterfactual analysis is then performed on each "global characteristic variable" to assess how flight risk would change if a particular characteristic were to change.

[0115] For example:

[0116] Assuming a real-world scenario: the pilot's fatigue index is 6, and the probability of the flight risk being "high" is 20%.

[0117] Define a counterfactual scenario: Suppose that the "pilot fatigue index" decreases from 6 to 3.

[0118] Calculate the probability distribution of "flight risk" under counterfactual scenarios.

[0119] Counterfactual scenario: With a pilot fatigue index of 3, the probability of a "high" flight risk decreases to 10%.

[0120] Conclusion: "Pilot fatigue index" has a direct causal impact on "flight risk".

[0121] Optionally, after identifying features that have a direct causal impact on flight risk through causal reasoning, features that are only correlated with flight risk but not causally related (such as airspace congestion) are excluded, and features that have a direct causal impact on flight risk (such as wind speed, pilot fatigue index, etc.) are selected to construct a flight safety causal graph.

[0122] Finally, the final set of flight safety features is constructed through the causal relationships in the causal graph, which is used for training and prediction of the flight safety mission risk assessment model.

[0123] Understandably, the Pearson correlation coefficient is primarily used to measure the linear correlation between features and flight risk, but it cannot directly identify causal relationships. Therefore, subsequent counterfactual analysis and causal graph construction are necessary to ensure that the selected features are not only related to flight risk but also have a direct causal impact on it. Counterfactual analysis is one of the core techniques of causal reasoning; it helps identify causal relationships by simulating the impact of feature changes on flight risk and ultimately constructs a causal graph to represent the causal relationship between features and flight risk in an intuitive way.

[0124] Optionally, in this embodiment, after feature engineering is completed, model training is performed based on the obtained flight safety features and risk list labels.

[0125] Specifically, risk list tags are used to enable the model to automatically generate a detailed risk list, listing potential risks and their probability of occurrence, severity, etc. Risk list tags include:

[0126] Risk categories include: weather risk, equipment risk, personnel risk, bird strike risk, etc.

[0127] Risk Description: Describe the specific circumstances of each risk, such as "excessive wind speed may lead to flight instability" or "pilot fatigue may lead to operational errors".

[0128] Risk level: The risk level is marked according to the model prediction, including minor, moderate, significant, major, and extremely significant.

[0129] Risk probability: Outputs the probability of this risk occurring, such as very low, low, high, or very high.

[0130] Specifically, for each flight mission, an input flight safety feature and a corresponding output risk list are constructed.

[0131] For example,

[0132] Input characteristics: wind speed 15 knots, visibility 5 km, pilot fatigue index 7, aircraft maintenance time 7 days ago, frequent bird activity.

[0133] Output labels (risk list):

[0134] Risk 1: High wind speeds may cause flight instability. The risk level is moderate and the probability is low.

[0135] Risk 2: Pilot fatigue may lead to operational errors, which is a significant risk with a high probability.

[0136] Risk 3: Bird activity may lead to bird strikes during takeoff; the risk level is moderate and the probability is low.

[0137] Then, the model builds a prediction model by learning the mapping relationship between input features and output labels in historical data.

[0138] When the input features are "wind speed 15 knots, visibility 5 km, pilot fatigue index 7", the model learns the association between these features and the risks of "excessive wind speed" and "pilot fatigue". By adjusting the internal parameters, the predicted risk list is made as close as possible to the true labels in historical data.

[0139] Once the model is trained, a risk list can be generated in real time for new flight missions, helping pilots and commanders to better assess and manage flight risks.

[0140] Step S103: Input the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

[0141] Optionally, in this embodiment, this step is the model application process.

[0142] A specific example is provided for illustration:

[0143] Assuming feature engineering is performed on the flight preparation professional information submitted by each flight preparation professional, the final input features obtained are:

[0144] Meteorological data: Wind speed: 20 knots; Visibility: 3 km; Probability of precipitation: 40%

[0145] Pilot data: Fatigue index: 8 (out of 10); Flight experience: 300 hours

[0146] Aircraft status data: Maintenance time: 10 days ago; Fuel status: 70%

[0147] Bird activity data: Frequent; Bird strike risk: Moderate

[0148] Air traffic control data: Airspace traffic density: High; Route conflict points: 2

[0149] Mission Information: Flight Subject: Nighttime Low-Altitude Flight; Flight Scenario: Complex Terrain

[0150] The model's final output risk list is:

[0151] Risk 1: High wind speeds may cause flight instability.

[0152] Risk level: High

[0153] Probability of occurrence: High

[0154] Risk 2: Pilot fatigue may lead to operational errors.

[0155] Risk level: High

[0156] Probability of occurrence: Very high

[0157] Risk 3: Bird strike risk, which may cause damage to the aircraft.

[0158] Risk level: Moderate

[0159] Probability of occurrence: Low

[0160] Risk 4: Low visibility may cause navigation difficulties.

[0161] Risk level: Moderate

[0162] Probability of occurrence: High

[0163] Risk 5: The aircraft maintenance time is long, and there is a risk of equipment failure.

[0164] Risk level: Moderate

[0165] Probability of occurrence: Low

[0166] Finally, based on the rules, risk response strategies are generated for the aforementioned risks.

[0167] Specific countermeasures and recommendations:

[0168] If high wind speeds may cause flight instability, it is recommended to adjust the flight altitude or postpone the flight.

[0169] Regarding pilot fatigue, which may lead to operational errors, it is recommended to increase rest time or replace the pilot.

[0170] To mitigate the risk of bird strikes, which could damage the aircraft, it is recommended to avoid areas with high bird activity.

[0171] Given the low visibility, which may cause navigation difficulties, it is recommended to use auxiliary navigation devices.

[0172] Regarding aircraft maintenance issues, there may be a risk of equipment failure: it is recommended to conduct an equipment inspection.

[0173] This example demonstrates how this embodiment unifies the flight preparation management process and improves flight preparation efficiency based on a risk prediction model.

[0174] As described above, the flight safety mission data processing method provided in this application embodiment can generate a flight safety mission risk list by having trainees create flight training plans, flight preparation professionals fill in professional information to obtain flight preparation professional information, and participating pilots fill in training information based on the flight preparation professional information to obtain flight mission training information. The flight preparation professional information and flight mission training information are then input into a trained risk list generation model to generate a flight safety mission risk list. The risk list generation model extracts features from historical flight preparation professional information and historical flight mission training information, performs counterfactual analysis on the features after the feature extraction operation based on a preset counterfactual scenario to obtain flight safety features, and inputs the flight safety features and risk list labels into an initial deep learning model for model training. This improves the efficiency and accuracy of flight preparation work.

[0175] In one embodiment of the flight safety mission data processing method of this application, see [link to relevant documentation]. Figure 2 It can also specifically include the following:

[0176] Step S201: Receive the flight training plan created by the flight mission training personnel, and send a professional information filling notice to each flight preparation professional so that each flight preparation professional can fill in the professional information according to their respective flight safety knowledge. Among them, flight preparation professionals include political workers, aviation medical personnel, meteorological personnel, bird monitoring personnel, aircraft maintenance personnel, and air traffic control personnel.

[0177] Step S202: Receive the professional information returned by each of the flight preparation professionals and send the professional information to the flight mission training personnel. The flight mission training personnel review the professional information. If it passes, the corresponding flight preparation professional information is confirmed.

[0178] Optionally, in this embodiment, the trainees first create training days after receiving the training plan;

[0179] Then, after the training day is created, the training personnel, political workers, aviation medical personnel, meteorological personnel, bird monitoring personnel, aircraft maintenance personnel, and air traffic control personnel will be notified by message reminder to fill in the professional information in their respective fields.

[0180] Finally, after each major is filled out, the training staff will review it. If the review fails, the information will be modified. Once all major information is approved, the major information will be obtained.

[0181] The pilot can then determine whether he or she is eligible to participate in the training based on the information provided in the professional report.

[0182] Through step S202, this embodiment obtains professional reporting information, which reflects all the professional data required for flight preparation before a flight training exercise, laying the foundation for subsequent flight risk assessment.

[0183] In one embodiment of the flight safety mission data processing method of this application, see [link to relevant documentation]. Figure 3 It can also specifically include the following:

[0184] Step S301: Determine whether the pilot meets the training standards based on the flight preparation professional information;

[0185] Step S302: If the conditions are met, determine the corresponding training pilot's terminal.

[0186] Optionally, after obtaining the professional field information in step S202, the system determines whether the pilot meets the training standards based on the professional information submitted. If the pilot does not meet the standards, he / she cannot participate in this training. If the pilot meets the standards, he / she is a trainee pilot and can proceed to the flight preparation stage to submit training information and view professional information.

[0187] Through step S302, this embodiment first makes a preliminary judgment on whether the pilot is suitable to participate in this training based on the professional information filled in, and selects the pilot who meets the standard as the training pilot to reduce the probability of flight risk.

[0188] In one embodiment of the flight safety mission data processing method of this application, see [link to relevant documentation]. Figure 4 It can also specifically include the following:

[0189] Step S401: Collect historical flight preparation professional information and historical flight mission training information. Among them, flight preparation professional information is professional information used to characterize flight safety-related factors, and flight mission training information is information used to characterize the pilot's own training arrangements.

[0190] Step S402: Perform multi-source feature extraction and risk correlation analysis on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding highly correlated risk features;

[0191] Step S403: Perform counterfactual analysis on the highly relevant risk features according to the preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, and extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features.

[0192] Optionally, in this embodiment, this step is feature engineering during model training.

[0193] Step S402 extracts historical flight information separately and constructs a graph structure to extract global multi-source data fusion features. Then, through risk correlation analysis, highly correlated global features are selected.

[0194] After selecting highly relevant global features, step S403 constructs a counterfactual scenario and further selects highly relevant global features that have a direct causal relationship with flight risks through counterfactual analysis.

[0195] Specifically, all "global characteristic variables" highly correlated with flight safety have now been identified, along with the target variable "flight risk." Counterfactual analysis is then performed on each "global characteristic variable" to assess how flight risk would change if a particular characteristic were to change.

[0196] For example:

[0197] Assuming a real-world scenario: the pilot's fatigue index is 6, and the probability of the flight risk being "high" is 20%.

[0198] Define a counterfactual scenario: Suppose that the "pilot fatigue index" decreases from 6 to 3.

[0199] Calculate the probability distribution of "flight risk" under counterfactual scenarios.

[0200] Counterfactual scenario: With a pilot fatigue index of 3, the probability of a "high" flight risk decreases to 10%.

[0201] Conclusion: "Pilot fatigue index" has a direct causal impact on "flight risk".

[0202] Optionally, after identifying features that have a direct causal impact on flight risk through causal reasoning, features that are only correlated with flight risk but not causally related (such as airspace congestion) are excluded, and features that have a direct causal impact on flight risk (such as wind speed, pilot fatigue index, etc.) are selected to construct a flight safety causal graph.

[0203] Finally, the final set of flight safety features is constructed through the causal relationships in the causal graph, which is used for training and prediction of the flight safety mission risk assessment model.

[0204] Through step S403, this embodiment successfully extracted flight safety features with direct causal relationships through feature engineering, laying a solid data foundation for subsequent model training.

[0205] In one embodiment of the flight safety mission data processing method of this application, see [link to relevant documentation]. Figure 5 It can also specifically include the following:

[0206] Step S501: Perform multi-source feature extraction on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding multi-source flight safety features. Use the multi-source flight safety feature data as nodes and perform feature aggregation on the multi-source flight safety features according to the graph neural network to determine the corresponding global flight safety features.

[0207] Step S502: Perform risk correlation analysis on the global flight safety features based on the Pearson correlation coefficient to determine the corresponding highly correlated risk features.

[0208] Optionally, in this embodiment, this step is the process of extracting multi-source data and risk analysis in the feature extraction step.

[0209] Specifically, firstly, flight safety-related features are extracted for each type of professional information.

[0210] From the professional information of historical political workers, we extract characteristics such as pilots' psychological stress, emotional stability, anxiety level, cooperation between pilots and crew members, communication efficiency, pilots' flight hours, and mission experience.

[0211] Extract information such as pilots' physical health indicators, pilots' fatigue index, and whether pilots have recent illnesses or injury records from historical aviation medical personnel's professional information;

[0212] Extract information such as wind speed, visibility, precipitation probability, temperature, weather changes in the next few hours, and whether there will be extreme weather such as thunderstorms, strong winds, and dense fog from historical meteorological personnel's professional information;

[0213] Bird information is extracted from historical bird information personnel, including the frequency of bird activity in the flight area, the presence of large or high-risk birds, and whether bird strikes have occurred in the area.

[0214] Extract information such as the operating status of each aircraft system, whether maintenance has been performed recently, whether the maintenance was qualified, and whether the aircraft has any fault records from the historical professional information of aircraft maintenance personnel.

[0215] Extract information from historical air traffic controllers' professional information, such as the current airspace's busyness, whether there are other aircraft activities, whether the flight path passes through complex terrain or dangerous areas, and whether there are temporary traffic controls or flight path adjustments.

[0216] Extract information from historical flight mission training data, such as the type of flight mission (e.g., takeoff and landing training, formation flight, night flight), the scenario of the flight mission (e.g., mountainous areas, ocean, cities), whether the pilot has not flown for a long time, and the ratio of personal energy level.

[0217] Secondly, after extracting the information, the flight safety characteristic data of various professionals (political workers, aviation medical personnel, meteorological personnel, bird monitoring personnel, aircraft maintenance personnel, and air traffic controllers) and pilots are modeled into a graph structure. The nodes in the graph represent entities (such as pilots, meteorological data, aircraft maintenance data, etc.), and the edges represent the relationships between entities (such as the relationship between pilots and meteorological conditions, the relationship between aircraft maintenance personnel and aircraft status, etc.).

[0218] Then, for each node, the features of its neighboring nodes are aggregated. For example, a pilot's neighboring nodes include meteorological data, aircraft maintenance data, and aviation medical data. By updating the features of each node through a graph neural network, and combining its own features with those of its neighbors, the graph neural network can extract global features through multi-layer information transmission and aggregation.

[0219] By leveraging the information transfer and aggregation capabilities of multi-layer graph neural networks, global multi-source data fusion features can be extracted. These features can capture the complex relationships between different data sources, providing more comprehensive information for flight safety mission risk assessment.

[0220] For example: Suppose that the initial feature of each node is its original data.

[0221] For node 1 (pilot):

[0222] By aggregating features from neighboring nodes 2 (meteorological data), 3 (aircraft maintenance data), and 4 (aviation medical data), the updated flight safety features may include: flight experience, psychological stress score, wind speed, visibility, engine status, fatigue index, etc. The updated features further aggregate global information from other nodes, such as the interaction between meteorological and aircraft maintenance data.

[0223] Ultimately, global multi-source data fusion features can be extracted:

[0224] Global Feature 1: Pilot operational risks under complex weather conditions (combining pilot experience, psychological state, wind speed, visibility, etc.).

[0225] Global Feature 2: Pilot operational risks when the aircraft is in poor condition (combining pilot experience, engine condition, maintenance records, etc.).

[0226] Global Feature 3: Operational risks of pilots under fatigue (combining pilot experience, fatigue index, heart rate, etc.).

[0227] Finally, after extracting the global features, the correlation between the extracted global features and the target variable is analyzed using the Pearson correlation coefficient, and the features with a high correlation with the target variable are selected.

[0228] Through step S502, this embodiment successfully extracted global features through multi-source data feature fusion. These features contain more complex feature representations, which can make the model learn more accurately. Then, highly correlated global features are selected through correlation analysis to reduce the learning pressure on the model.

[0229] In one embodiment of the flight safety mission data processing method of this application, see [link to relevant documentation]. Figure 6 It can also specifically include the following:

[0230] Step S601: Assess the flight risk correlation of the global flight safety characteristics based on the Pearson correlation coefficient and determine the corresponding correlation value;

[0231] Step S602: Compare the correlation value with a preset high correlation threshold to determine the corresponding high correlation risk characteristic.

[0232] Optionally, this step enables the application of feature correlation analysis.

[0233] Specifically, after extracting the global features, the correlation between the extracted global features and the target variable is analyzed using the Pearson correlation coefficient, and features with a high correlation to the target variable are selected.

[0234] For example, suppose we have a global feature: "the operational risk of a pilot in a state of fatigue."

[0235] According to Pearson correlation coefficient analysis, if the correlation between the feature "pilot's operational risk under fatigue" and "flight risk" is >0.5 (correlation threshold), then this feature is selected.

[0236] Through step S602, this embodiment successfully obtained global features that are highly correlated with flight risk, laying a solid data foundation for subsequent model training.

[0237] In one embodiment of the flight safety mission data processing method of this application, see [link to relevant documentation]. Figure 7 It can also specifically include the following:

[0238] Step S701: Send the flight safety mission risk list to the trainee pilot's terminal so that the trainee pilot can learn about risks based on the flight safety mission risk list;

[0239] Step S702: Receive the risk learning results returned by the trainee pilot's terminal, send risk confirmation notifications to the preset squadron leader terminal and the preset commander terminal, so that the squadron leader and commander can perform risk confirmation operations according to the risk confirmation notifications, and accept the risk confirmation opinions returned by the squadron leader terminal and the commander terminal, and send the risk confirmation opinions to the trainee pilot's terminal, so that the trainee pilot can conduct flight training according to the risk confirmation opinions.

[0240] Optionally, this step is performed after obtaining the risk list.

[0241] Specifically, a risk list will be sent to the participating pilots for risk warnings and training;

[0242] After the pilots complete the risk learning, they notify their squadron leader and the commander of the training session via message reminders to confirm the risks.

[0243] After the squadron leader and commanders confirmed the risks, the flight training exercise was conducted.

[0244] After the flight training, a debriefing and explanation were conducted to understand the actual risks that occurred during the training. The risk feedback function was then used to report the actual risks encountered.

[0245] After completing the training for this training day, the trainees will set the training day status to "end" to conclude the training session.

[0246] Understandably, feedback data from actual risks will serve as data material for the model to continue learning, enabling the model to continuously respond to new risk scenarios.

[0247] Through step S702, this embodiment successfully applies the risk list generated by the model for flight preparation and formal flight training, improving the efficiency and accuracy of flight preparation work.

[0248] To improve the efficiency and accuracy of flight preparation, this application provides an embodiment of a flight safety mission data processing apparatus for implementing all or part of the aforementioned flight safety mission data processing method. See [link to embodiment]. Figure 8 The flight safety mission data processing device specifically includes the following components:

[0249] The flight information confirmation module 10 is used to receive the flight training plan created by the flight mission training personnel, send a professional information filling notification to each flight preparation professional, so that each flight preparation professional can fill in the professional information according to the professional information filling notification, receive the flight preparation professional information returned by each flight preparation professional, send the flight preparation professional information to the preset training pilot terminal, so that the training pilot can fill in the training information according to the flight preparation professional information, and receive the flight mission training information returned by the training pilot.

[0250] The risk list generation model training module 20 is used to collect historical flight preparation professional information and historical flight mission training information, perform feature extraction operations on the historical flight preparation professional information and the historical flight mission training information, perform counterfactual analysis on the features after the feature extraction operation according to the preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, and input the flight safety features and risk list labels into the initial deep learning model for model training to obtain the corresponding risk list generation model;

[0251] The flight safety mission risk list determination module 30 is used to input the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

[0252] As described above, the flight safety mission data processing device provided in this application embodiment can generate a flight safety mission risk list by having trainees create flight training plans, flight preparation professionals fill in professional information to obtain flight preparation professional information, and participating pilots fill in training information based on the flight preparation professional information to obtain flight mission training information. The flight preparation professional information and flight mission training information are then input into a trained risk list generation model to generate a flight safety mission risk list. The risk list generation model extracts features from historical flight preparation professional information and historical flight mission training information, performs counterfactual analysis on the features after the feature extraction operation based on a preset counterfactual scenario to obtain flight safety features, and inputs the flight safety features and risk list labels into an initial deep learning model for model training. This improves the efficiency and accuracy of flight preparation work.

[0253] From a hardware perspective, in order to improve the efficiency and accuracy of flight preparation, this application provides an embodiment of an electronic device for implementing all or part of the flight safety mission data processing method, wherein the electronic device specifically includes the following:

[0254] The system comprises a processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between the flight safety mission data processing method and core business systems, user terminals, and related databases and other related devices; the logic controller can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the logic controller can be implemented with reference to the embodiments of the flight safety mission data processing method in the present embodiment, and the contents of the embodiments of the flight safety mission data processing method are incorporated herein, and repeated details will not be described again.

[0255] It is understood that the user terminal may include smartphones, tablet computers, network set-top boxes, portable computers, desktop computers, personal digital assistants (PDAs), in-vehicle devices, smart wearable devices, etc. Among these, the smart wearable devices may include smart glasses, smartwatches, smart bracelets, etc.

[0256] In practical applications, the flight safety mission data processing method can be partially executed on the electronic device side as described above, or all operations can be completed on the client device. The choice can be made based on the processing power of the client device and the limitations of the user's usage scenario. This application does not impose any limitations on this. If all operations are completed on the client device, the client device may further include a processor.

[0257] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0258] Figure 10 This is a schematic block diagram illustrating the system configuration of the electronic device 9600 according to an embodiment of this application. Figure 10As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that... Figure 10 This is an example; other types of structures can also be used to supplement or replace this structure to achieve telecommunications functions or other functions.

[0259] In one embodiment, the flight safety mission data processing method function can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control:

[0260] Step S101: Receive the flight training plan created by the flight mission training personnel, send a professional information filling notification to each flight preparation professional, so that each flight preparation professional can fill in the professional information according to the professional information filling notification, and receive the flight preparation professional information returned by each flight preparation professional, and send the flight preparation professional information to the preset training pilot terminal, so that the training pilot can fill in the training information according to the flight preparation professional information, and receive the flight mission training information returned by the training pilot.

[0261] Step S102: Collect historical flight preparation professional information and historical flight mission training information, perform feature extraction on the historical flight preparation professional information and the historical flight mission training information, perform counterfactual analysis on the features after feature extraction based on a preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, input the flight safety features and risk list labels into the initial deep learning model for model training to obtain the corresponding risk list generation model;

[0262] Step S103: Input the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

[0263] As described above, the electronic device provided in this application embodiment allows trainees to create flight training plans, and various flight preparation professionals to fill in professional information to obtain flight preparation professional information. Trainees then fill in training information based on the flight preparation professional information to obtain flight mission training information. The flight preparation professional information and flight mission training information are input into a trained risk list generation model to generate a flight safety mission risk list. The risk list generation model extracts features from historical flight preparation professional information and historical flight mission training information. It then performs counterfactual analysis on the features extracted based on a preset counterfactual scenario to obtain flight safety features. These flight safety features, along with risk list labels, are input into an initial deep learning model for training. This process improves the efficiency and accuracy of flight preparation work.

[0264] In another embodiment, the flight safety mission data processing method can be configured separately from the central processing unit 9100. For example, the flight safety mission data processing method can be configured as a chip connected to the central processing unit 9100, and the flight safety mission data processing method function can be implemented through the control of the central processing unit.

[0265] like Figure 10 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily need to include these components. Figure 10 All components shown; in addition, the electronic device 9600 may also include Figure 10 For components not shown, please refer to existing technologies.

[0266] like Figure 10 As shown, the central processing unit 9100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device, which receives inputs and controls the operation of various components of the electronic device 9600.

[0267] The memory 9140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 9100 may execute the program stored in the memory 9140 to perform information storage or processing, etc.

[0268] Input unit 9120 provides input to central processing unit 9100. Input unit 9120 may be, for example, a keypad or touch input device. Power supply 9170 provides power to electronic device 9600. Display 9160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0269] The memory 9140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or processes for executing the operation of the electronic device 9600 via the central processing unit 9100.

[0270] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0271] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which is the same as in a conventional mobile communication terminal.

[0272] Based on different communication technologies, multiple communication modules 9110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby realizing typical telecommunications functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Furthermore, the audio processor 9130 is also coupled to a central processing unit 9100, enabling on-device recording via the microphone 9132 and on-device playback of stored sound via the speaker 9131.

[0273] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps of the flight safety mission data processing method with a server or client as the execution subject in the above embodiments. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the flight safety mission data processing method with a server or client as the execution subject in the above embodiments. For example, when the processor executes the computer program, it implements the following steps:

[0274] Step S101: Receive the flight training plan created by the flight mission training personnel, send a professional information filling notification to each flight preparation professional, so that each flight preparation professional can fill in the professional information according to the professional information filling notification, and receive the flight preparation professional information returned by each flight preparation professional, and send the flight preparation professional information to the preset training pilot terminal, so that the training pilot can fill in the training information according to the flight preparation professional information, and receive the flight mission training information returned by the training pilot.

[0275] Step S102: Collect historical flight preparation professional information and historical flight mission training information, perform feature extraction on the historical flight preparation professional information and the historical flight mission training information, perform counterfactual analysis on the features after feature extraction based on a preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, input the flight safety features and risk list labels into the initial deep learning model for model training to obtain the corresponding risk list generation model;

[0276] Step S103: Input the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

[0277] As described above, the computer-readable storage medium provided in this application embodiment allows trainees to create flight training plans, and various flight preparation professionals to fill in professional information to obtain flight preparation professional information. Trainees then fill in training information based on the flight preparation professional information to obtain flight mission training information. The flight preparation professional information and flight mission training information are input into a trained risk list generation model to generate a flight safety mission risk list. The risk list generation model extracts features from historical flight preparation professional information and historical flight mission training information. It then performs counterfactual analysis on the features extracted based on a preset counterfactual scenario to obtain flight safety features. These flight safety features and risk list labels are then input into an initial deep learning model for model training. This process improves the efficiency and accuracy of flight preparation work.

[0278] Embodiments of this application also provide a computer program product capable of implementing all steps of the flight safety mission data processing method with the execution subject being a server or client in the above embodiments. When executed by a processor, this computer program / instruction implements the steps of the flight safety mission data processing method. For example, the computer program / instruction implements the following steps:

[0279] Step S101: Receive the flight training plan created by the flight mission training personnel, send a professional information filling notification to each flight preparation professional, so that each flight preparation professional can fill in the professional information according to the professional information filling notification, and receive the flight preparation professional information returned by each flight preparation professional, and send the flight preparation professional information to the preset training pilot terminal, so that the training pilot can fill in the training information according to the flight preparation professional information, and receive the flight mission training information returned by the training pilot.

[0280] Step S102: Collect historical flight preparation professional information and historical flight mission training information, perform feature extraction on the historical flight preparation professional information and the historical flight mission training information, perform counterfactual analysis on the features after feature extraction based on a preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, input the flight safety features and risk list labels into the initial deep learning model for model training to obtain the corresponding risk list generation model;

[0281] Step S103: Input the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

[0282] As described above, the computer program product provided in this application embodiment allows trainees to create flight training plans, flight preparation professionals to fill in professional information to obtain flight preparation professional information, and participating pilots to fill in training information based on the flight preparation professional information to obtain flight mission training information. The flight preparation professional information and flight mission training information are then input into a trained risk list generation model to generate a flight safety mission risk list. The risk list generation model extracts features from historical flight preparation professional information and historical flight mission training information, performs counterfactual analysis on the features after the feature extraction operation based on a preset counterfactual scenario to obtain flight safety features, and inputs the flight safety features and risk list labels into an initial deep learning model for model training. This improves the efficiency and accuracy of flight preparation work.

[0283] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0284] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0285] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0286] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0287] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for processing flight safety mission data, characterized in that, The method includes: The system receives flight training plans created by flight mission training personnel, sends professional information reporting notifications to each flight preparation professional, enabling each flight preparation professional to report professional information according to the notifications, and receives flight preparation professional information returned by each flight preparation professional. The system then sends the flight preparation professional information to the preset training pilot's terminal, enabling the training pilot to report training information according to the flight preparation professional information, and receives flight mission training information returned by the training pilot. Collect historical flight preparation professional information and historical flight mission training information, perform feature extraction on the historical flight preparation professional information and historical flight mission training information, perform counterfactual analysis on the features after feature extraction based on a preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, input the flight safety features and risk list labels into an initial deep learning model for model training to obtain the corresponding risk list generation model; The flight preparation professional information and the flight mission training information are input into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

2. The flight safety mission data processing method according to claim 1, characterized in that, The process involves receiving flight training plans created by flight mission training personnel, sending professional information reporting notifications to each flight preparation professional, enabling them to report their professional information according to the notifications, and receiving flight preparation professional information returned by each of the flight preparation professionals, including: The system receives flight training plans created by flight mission training personnel and sends professional information reporting notices to various flight preparation professionals, enabling them to report professional information based on their respective knowledge of flight safety. These flight preparation professionals include political workers, aviation medical personnel, meteorologists, birdwatchers, aircraft maintenance personnel, and air traffic controllers. The system receives professional information returned by each of the flight preparation professionals and sends the professional information to the flight mission training personnel. The flight mission training personnel review the professional information, and if it passes, they determine the corresponding flight preparation professional information.

3. The flight safety mission data processing method according to claim 1, characterized in that, Before sending the flight preparation professional information to the preset training pilot's terminal, the following is included: Determine whether the pilot meets the training standards based on the aforementioned flight preparation professional information; If the criteria are met, the corresponding training pilot's terminal will be determined.

4. The flight safety mission data processing method according to claim 1, characterized in that, The process involves collecting historical flight preparation professional information and historical flight mission training information, performing feature extraction on the historical flight preparation professional information and historical flight mission training information, conducting counterfactual analysis on the features after feature extraction based on a preset counterfactual scenario, determining the flight risk probability distribution under the counterfactual scenario, comparing the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under a preset actual scenario, constructing a flight safety causal graph, and extracting the causal relationships in the flight safety causal graph to construct corresponding flight safety features, including: Collect historical flight preparation professional information and historical flight mission training information. Among them, flight preparation professional information is professional information used to characterize flight safety-related factors, and flight mission training information is information used to characterize the pilot's own training arrangements. Multi-source feature extraction and risk correlation analysis are performed on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding highly correlated risk features. Based on a preset counterfactual scenario, counterfactual analysis is performed on the highly relevant risk features to determine the flight risk probability distribution under the counterfactual scenario. After comparing the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, a flight safety causal graph is constructed, and the causal relationships in the flight safety causal graph are extracted to construct the corresponding flight safety features.

5. The flight safety mission data processing method according to claim 4, characterized in that, The process of performing multi-source feature extraction and risk correlation analysis on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding highly correlated risk features includes: Multi-source feature extraction is performed on the historical flight preparation professional information and the historical flight mission training information to determine the corresponding multi-source flight safety features. The multi-source flight safety features are used as nodes, and feature aggregation is performed on the multi-source flight safety features according to the graph neural network to determine the corresponding global flight safety features. Based on the Pearson correlation coefficient, risk correlation analysis is performed on the global flight safety characteristics to determine the corresponding highly correlated risk characteristics.

6. The flight safety mission data processing method according to claim 5, characterized in that, The step of performing risk correlation analysis on the global flight safety characteristics based on the Pearson correlation coefficient to determine the corresponding highly correlated risk characteristics includes: The global flight safety characteristics are assessed for flight risk correlation based on the Pearson correlation coefficient to determine the corresponding correlation values. The correlation value is compared with a preset high correlation threshold to determine the corresponding high correlation risk characteristic.

7. The flight safety mission data processing method according to claim 1, characterized in that, After inputting the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model, the method further includes: The flight safety mission risk list is sent to the trainee pilot's terminal so that the trainee pilot can learn about the risks based on the flight safety mission risk list; The system receives the risk learning results returned by the participating pilots, sends risk confirmation notifications to the preset squadron leader and preset commander terminals, so that the squadron leader and commander can perform risk confirmation operations according to the risk confirmation notifications, and accepts the risk confirmation opinions returned by the squadron leader and commander terminals, and sends the risk confirmation opinions to the participating pilots' terminals, so that the participating pilots can conduct flight training according to the risk confirmation opinions.

8. A flight safety mission data processing device, characterized in that, The device includes: The flight information confirmation module is used to receive flight training plans created by flight mission training personnel, send professional information filling notifications to each flight preparation professional, so that each flight preparation professional can fill in professional information according to the professional information filling notifications, receive flight preparation professional information returned by each flight preparation professional, send the flight preparation professional information to the preset training pilot terminal, so that the training pilot can fill in training information according to the flight preparation professional information, and receive flight mission training information returned by the training pilot. The risk list generation model training module is used to collect historical flight preparation professional information and historical flight mission training information, perform feature extraction operations on the historical flight preparation professional information and the historical flight mission training information, perform counterfactual analysis on the features after the feature extraction operation according to the preset counterfactual scenario, determine the flight risk probability distribution under the counterfactual scenario, compare the flight risk probability distribution under the counterfactual scenario with the flight risk probability distribution under the preset actual scenario, construct a flight safety causal graph, extract the causal relationships in the flight safety causal graph to construct the corresponding flight safety features, and input the flight safety features and risk list labels into the initial deep learning model for model training to obtain the corresponding risk list generation model; The flight safety mission risk list determination module is used to input the flight preparation professional information and the flight mission training information into the risk list generation model to obtain the flight safety mission risk list output by the risk list generation model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the flight safety mission data processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the flight safety mission data processing method according to any one of claims 1 to 7.

Images