Real-time analysis method and system for pilot control qualities under wind shear conditions

Abstract

This invention provides a method and system for real-time analysis of pilot control quality under wind shear conditions, relating to the field of aviation flight technology. First, it captures flight control and environmental interaction data, including flight status feedback data and pilot control execution data. Next, it establishes a set of response correlation links between control actions and environmental changes, using wind shear environmental changes as trigger nodes, pilot control execution as response nodes, and flight status feedback as correlation nodes. Then, it traces the temporal trajectory of control actions and their compatibility with environmental changes, outputting a set of temporal compatibility relationships. Based on the evolutionary analysis of the temporal compatibility relationship set, it analyzes the adaptive dynamic changes of control actions and outputs the adaptive dynamic change results. Finally, it generates control adjustment guidance based on the adaptive dynamic change results to assist pilots in optimizing control actions in real time, improving flight safety and control quality.

Description

Technical Field

[0001] This invention relates to the field of aviation flight technology, and more specifically, to a method and system for real-time analysis of pilot control qualities under wind shear conditions. Background Technology

[0002] In the field of aviation, wind shear is an extremely complex and dangerous meteorological phenomenon. Wind shear refers to a sudden change in wind direction and speed over a short distance. These changes can have a drastic impact on the flight status of an aircraft, causing it to suddenly ascend, descend, or yaw, greatly increasing flight safety risks.

[0003] Currently, when dealing with wind shear environments, pilots primarily rely on their experience, training, and information provided by flight instruments to make control decisions. However, most existing flight instruments and auxiliary systems can only provide basic flight status data, such as altitude, speed, and attitude, lacking in-depth analysis and real-time feedback on the complex relationship between pilot actions and environmental changes in wind shear environments. Pilots struggle to accurately determine whether their actions are compatible with changes in wind shear and flight status, and cannot promptly understand the dynamic changes in the adaptability of their actions. This limits the accuracy and timeliness of pilot decisions in complex and ever-changing wind shear environments, hindering real-time optimization of control actions and ultimately impacting flight safety. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, the present invention provides a method for real-time analysis of pilot handling qualities under wind shear conditions, the method comprising:

[0005] The system captures flight control and environmental interaction data under wind shear conditions. The flight control and environmental interaction data includes flight status feedback data and pilot control execution data. The flight status feedback data reflects the state changes of the aircraft after the wind shear environment acts on it, and the pilot control execution data reflects the control actions taken by the pilot in response to the wind shear environment and changes in flight status.

[0006] Based on the interaction data between flight control and environment, a response correlation link between control actions and environmental changes is established, and a set of response correlation links between control actions and environmental changes is output. The response correlation link takes wind shear environmental change as the trigger node, pilot control execution as the response node, and flight status feedback as the correlation node.

[0007] Based on the set of response association links, the temporal trajectory of the control action and the fit relationship with environmental changes are traced, and a set of temporal fit relationships between the control action and environmental changes is output. The fit relationship reflects the correspondence between the timing of the control action, the type of action and the magnitude of wind shear environmental changes and the trend of flight state changes.

[0008] Based on the evolutionary analysis of the time-series fit relationship set, the adaptive dynamic changes of the control actions are analyzed, and the adaptive dynamic changes of the control actions are output. The adaptive dynamic changes reflect the adjustment effect of the control actions as the wind shear environment changes continuously and the flight status is continuously fed back.

[0009] Based on the adaptive dynamic change results, control adjustment guidelines are generated. These guidelines provide adjustment directions for control actions that do not conform to environmental changes and flight status feedback in the adaptive dynamic change results, thereby assisting pilots in optimizing control actions in real time.

[0010] Furthermore, the present invention also provides a real-time analysis system for pilot control qualities under wind shear conditions, comprising:

[0011] A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to perform the above-described real-time analysis method for pilot control qualities under wind shear conditions by executing the machine-executable instructions.

[0012] In another aspect, the present invention also provides a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, the processor of a computer device reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the computer device to execute the above-described method for real-time analysis of pilot handling qualities under wind shear conditions.

[0013] Based on the above, by capturing flight control and environmental interaction data under wind shear conditions, including flight status feedback data and pilot control execution data, a set of response correlation links between control actions and environmental changes is established based on this data. This reveals the intrinsic connection between wind shear environment changes, pilot control execution, and flight status feedback. By tracing the temporal trajectory of control actions and their fit with environmental changes, a set of temporal fit relationships is output, revealing the correspondence between the timing and type of control actions and the magnitude of wind shear environment changes and flight status change trends. Furthermore, based on the evolutionary analysis of the temporal fit relationship set, the adaptability dynamic changes of control actions are analyzed, and the results of adaptability dynamic changes are output. This reflects the adjustment effect of control actions as the wind shear environment continuously changes and the flight status continuously responds, allowing pilots to understand the adaptability changes of control actions in a timely manner. Finally, based on the adaptability dynamic change results, control adjustment guidance is generated, providing precise adjustment directions for control actions that do not fit environmental changes and flight status feedback. This effectively assists pilots in optimizing control actions in real time, significantly improving pilot control quality and response capabilities in wind shear environments, and thus greatly enhancing flight safety. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the execution flow of the real-time analysis method for pilot control qualities under wind shear environment provided in an embodiment of the present invention.

[0015] Figure 2 This is a schematic diagram of exemplary hardware and software components of the real-time analysis system for pilot handling qualities under wind shear conditions provided in an embodiment of the present invention. Detailed Implementation

[0016] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a real-time analysis method for pilot control qualities under wind shear conditions, provided in one embodiment of the present invention. The following is a detailed description of this real-time analysis method for pilot control qualities under wind shear conditions.

[0017] Step S110: Capture flight control and environmental interaction data under wind shear conditions. The flight control and environmental interaction data includes flight status feedback data and pilot control execution data. The flight status feedback data reflects the state changes of the aircraft after the wind shear environment acts on it. The pilot control execution data reflects the control actions implemented by the pilot in response to the wind shear environment and changes in flight status.

[0018] In this embodiment, this step is achieved through multiple sensors and data acquisition devices mounted on the aircraft. Specifically, the sensors include atmospheric data sensors, inertial measurement units, attitude sensors, engine parameter sensors, etc., which continuously collect various parameters of the aircraft during flight. Atmospheric data sensors acquire environmental data such as air pressure, temperature, wind speed, and wind direction around the aircraft, directly reflecting changes in wind shear. The inertial measurement unit collects motion parameters such as acceleration and angular velocity of the aircraft. Attitude sensors provide attitude information such as pitch angle, roll angle, and yaw angle. Engine parameter sensors record power-related data such as engine thrust, speed, and fuel flow. Pilot control execution data is collected through devices such as control stick displacement sensors, throttle position sensors, and pedal position sensors in the cockpit. This data accurately reflects the pilot's control actions on the control stick, throttle, pedals, and other control components.

[0019] During data acquisition, all sensors and data acquisition devices are synchronously sampled according to a unified time base to ensure temporal consistency between the acquired flight status feedback data and pilot control execution data. The sampling frequency is set according to the characteristics of different parameters. For rapidly changing parameters such as acceleration and angular velocity, a higher sampling frequency is set to ensure that instantaneous state changes can be captured; for relatively slowly changing parameters such as fuel flow, the sampling frequency can be appropriately reduced. The acquired data first undergoes preliminary format conversion and verification, converting the raw data output from different sensors into a unified data format. Outliers and missing values ​​are preliminarily processed. Outliers are identified by comparison with preset normal ranges, and missing values ​​are filled using interpolation to ensure the accuracy of subsequent data processing.

[0020] Step S120: Establish a response correlation link between control actions and environmental changes based on flight control and environmental interaction data, and output a set of response correlation links between control actions and environmental changes. The response correlation link takes wind shear environmental changes as the trigger node, pilot control execution as the response node, and flight status feedback as the correlation node.

[0021] After acquiring flight control and environmental interaction data, in-depth analysis and processing are required to establish a response correlation link between control actions and environmental changes. First, the raw data is preprocessed, including data denoising and smoothing, to reduce noise interference in subsequent analysis. Data denoising employs a moving average filtering method, smoothing the data curve by calculating the average value of data within a certain time window. Smoothing uses a low-pass filtering algorithm to remove high-frequency noise components while retaining the main trends in data variation. The preprocessed data is stored in a dedicated database for retrieval and analysis in subsequent steps.

[0022] Step S121: Separate the flight status feedback data and pilot control execution data in the flight control and environment interaction data, distinguish between flight status feedback data reflecting changes in wind shear environment and flight status feedback data reflecting the aircraft's own state, and distinguish between pilot control execution data for attitude adjustment and pilot control execution data for power control.

[0023] Step S1211: Traverse the flight control and environmental interaction data, perform preliminary classification according to the source and representation object of the data, and separate the flight status feedback data and pilot control execution data.

[0024] When performing data segmentation, a comprehensive review of flight control and environmental interaction data is first conducted. During this review, the data type is determined based on its source identifier and the meaning of its fields. For example, data from atmospheric data sensors and inertial measurement units, reflecting the aircraft's external environment and its own motion status, are categorized as flight status feedback data; while data from cockpit control component sensors are categorized as pilot control execution data. In the initial classification process, a correspondence table between data source and data type is established. This table is consulted to quickly determine the data type, ensuring the accuracy and efficiency of the classification.

[0025] Step S1212: Perform data cleaning on the initially separated flight status feedback data, perform feature recognition on the cleaned flight status feedback data, and extract feature information related to the external environment. The feature information related to the external environment includes sudden state changes and uncontrolled state fluctuations reflected in the data.

[0026] The data cleaning process includes removing duplicate data and correcting outliers. For flight status feedback data, duplicate data is identified by comparing the same parameter data collected by different sensors; for duplicate data, the most recent timestamp is retained. Correction of outliers is based on the physical characteristics of the parameters and historical data distribution. For outliers exceeding reasonable ranges, statistical models based on historical data are used for prediction and correction. After cleaning, feature identification is performed on the data, and features related to external environmental influences are extracted by analyzing data change patterns. Sudden state changes are identified by calculating the first and second derivatives of the data; when the absolute value of the derivative exceeds a preset threshold, it is considered a sudden state change. Uncontrolled state fluctuations are identified by comparing flight status feedback data with pilot control execution data over time; when flight status feedback data fluctuates but there is no pilot control execution data within the corresponding time period, it is considered an uncontrolled state fluctuation.

[0027] Step S1213: Based on the feature information related to the external environment, establish a differentiation model to divide the flight status feedback data into flight status feedback data reflecting changes in the wind shear environment and flight status feedback data reflecting the aircraft's own state. The differentiation model is based on preset judgment rules, which include: if the time of occurrence of the state change is not temporally related to the pilot's control execution data, and the magnitude of the state change exceeds a first preset threshold of the corresponding flight status parameter, then it is determined to be flight status feedback data reflecting changes in the wind shear environment; otherwise, it is determined to be flight status feedback data reflecting the aircraft's own state.

[0028] The model is established based on the feature information related to the external environment extracted above. First, a first preset threshold is set for each flight state parameter. This first preset threshold is determined based on the aircraft's design performance, data in the flight manual, and a large amount of flight test data; different flight state parameters correspond to different first preset thresholds. During the segmentation process, for each state change event, it is first checked whether its occurrence time is temporally correlated with pilot control execution data. Temporal correlation is determined by judging whether pilot control execution data exists within a certain time range before and after the state change occurrence time. This time range is set based on the delay in the impact of control actions on the flight state. If the occurrence time of the state change is not temporally correlated with pilot control execution data, and the magnitude of the state change exceeds the first preset threshold of the corresponding flight state parameter, then the flight state feedback data corresponding to the state change is determined to reflect changes in the wind shear environment; otherwise, it is determined to reflect the aircraft's own state.

[0029] Step S1214: Perform data cleaning on the initially separated pilot control execution data, identify the action attributes of the cleaned pilot control execution data, and extract relevant information about the target of the control action. The relevant information about the target includes the aircraft system targeted by the control action and the flight parameters that the control action is expected to change.

[0030] The process of cleaning pilot control data is similar to that of cleaning flight status feedback data, including removing duplicate data and correcting abnormal data. Action attribute recognition is achieved by analyzing the characteristics of pilot control data. For example, the direction and magnitude of stick displacement reflect the pilot's intention to adjust attitude parameters such as pitch and roll angles; the throttle position reflects the pilot's intention to control engine thrust. By establishing a mapping relationship between control actions and their objects, the aircraft system targeted by the control action and the desired flight parameters to be changed are extracted. For example, stick manipulation corresponds to the aircraft's attitude control system, with the desired change being attitude parameters such as pitch and roll angles; throttle manipulation corresponds to the aircraft's propulsion system, with the desired change being power parameters such as thrust and speed.

[0031] Step S1215: Based on the relevant information of the object of the control action, establish a classification model to divide the pilot control execution data into pilot control execution data for attitude adjustment and pilot control execution data for power control. The object of the pilot control execution data for attitude adjustment is the attitude control system of the aircraft, and the desired change is the attitude parameters of the aircraft. The object of the pilot control execution data for power control is the power system of the aircraft, and the desired change is the power parameters of the aircraft.

[0032] The classification model is built based on information related to the object of the control actions. According to the aircraft system targeted by the control actions and the flight parameters to be changed, pilot control execution data is divided into two categories. Control action data targeting the attitude control system, with the aim of changing attitude parameters, is classified as pilot control execution data for attitude adjustment; control action data targeting the power system, with the aim of changing power parameters, is classified as pilot control execution data for power control. In the classification process, a decision tree algorithm is used to construct the classification model, using information related to the object of the control actions as input features. The model is trained using training samples to improve classification accuracy.

[0033] Step S1216: Label the different types of data, organize the labeled data by category to form a classified flight status feedback data set and a pilot control execution data set. The labeling information includes the data category and the basis for differentiation.

[0034] Different labels were added to the differentiated flight status feedback data reflecting wind shear environmental changes, flight status feedback data reflecting the aircraft's own state, pilot control execution data for attitude adjustment, and pilot control execution data for power control. The labeling information used a unified encoding method, with data categories represented by different code values. The differentiation criteria were recorded in the data's attribute fields for subsequent traceability and verification. After labeling, the data of each category was sorted chronologically and stored in different data files, forming a categorized dataset for easy retrieval and processing in subsequent steps.

[0035] Step S122: Extract features from the split flight state feedback data reflecting changes in the wind shear environment, identify state fluctuation features directly related to the wind shear environment in the data, and further confirm the attribute attribution of the data based on the state fluctuation features. The state fluctuation features reflect the state changes caused by the force exerted by the wind shear environment on the aircraft.

[0036] After data splitting, feature extraction was performed on flight status feedback data reflecting changes in the wind shear environment. First, time-domain and frequency-domain analyses were conducted. Time-domain analysis included calculating statistical characteristics such as the mean, variance, peak value, and trough value to describe the data's distribution characteristics in the time domain. Frequency-domain analysis converted the time-domain data to frequency-domain data using Fourier transform and calculated features such as power spectral density to identify the frequency components contained in the data. State fluctuation characteristics directly related to the wind shear environment mainly include the frequency, amplitude, and duration of the fluctuations. State fluctuations caused by the wind shear environment typically have specific frequency ranges and amplitude variation patterns. By comparing the extracted features with a preset wind shear feature template, state fluctuation characteristics directly related to the wind shear environment in the data were identified.

[0037] Based on the identified state fluctuation characteristics, the attribute attribution of the data is further confirmed. Data with a high degree of matching between its features and the wind shear feature template is confirmed as flight state feedback data reflecting changes in the wind shear environment. For data with a low degree of matching, it is necessary to re-examine the data splitting process for errors or whether other external environmental factors have affected the flight state. During the confirmation process, a comprehensive judgment is made by combining flight trajectory data and meteorological data. Flight trajectory data can provide information on the geographical location of the aircraft, while meteorological data can provide information on wind shear occurrence in the region. Through the fusion analysis of multi-source data, the accuracy of data attribute attribution confirmation is improved.

[0038] Step S123: Using the confirmed flight status feedback data reflecting changes in the wind shear environment as the trigger node data, the trigger node data are arranged sequentially according to the time dimension to form a continuous trigger node sequence. The adjacent trigger node data in the trigger node sequence maintain temporal continuity, and each trigger node data is accompanied by its corresponding acquisition time information.

[0039] After confirming the flight status feedback data reflecting changes in the wind shear environment, this data is used as trigger node data. The trigger node data is arranged according to the time sequence of data acquisition, with each trigger node containing its corresponding acquisition time information accurate to the millisecond level to ensure the accuracy of the time sequence. During the arrangement process, the time interval between adjacent trigger node data needs to be checked to ensure temporal continuity. If the time interval between adjacent data exceeds the preset maximum allowable interval, it is determined that data is missing. In this case, null nodes need to be inserted within the missing time period, and these null nodes will be specially processed in subsequent analysis to avoid affecting the continuity of the entire trigger node sequence.

[0040] After the trigger node sequence is formed, detailed attribute records are made for each trigger node data, including the data acquisition time, corresponding flight status parameter values, and status fluctuation characteristics. This information is stored in a structured data table, where each row represents a trigger node data point and each column represents an attribute field. Simultaneously, an index is created for the trigger node sequence, using the acquisition time as the key, to facilitate quick querying and access to trigger node data at specific times.

[0041] Step S124: Perform motion feature recognition on the split pilot control execution data, distinguish the execution features of different control actions, and define the target of the control actions based on the execution features.

[0042] The identification of pilot control execution data based on motion characteristics is primarily based on the dynamic changes in control actions. First, time-series analysis is performed on the pilot control execution data to extract characteristic parameters such as the start and end times, control amplitude, and control speed of the control actions. Control amplitude is determined by calculating the displacement of the control components, while control speed is calculated as the ratio of displacement to control time. Different control actions have different execution characteristics. For example, pushing or pulling the control stick corresponds to attitude adjustment, characterized by a large displacement and a relatively fast control speed; pushing or pulling the throttle lever corresponds to power control, characterized by a relatively slow control speed and a large range of control amplitude variations.

[0043] Based on the extracted execution features, a pattern recognition algorithm is used to distinguish different control actions. The pattern recognition algorithm employs a support vector machine (SVM) model, which learns the execution feature patterns of different control actions through training samples. The feature parameters of the control action to be recognized are then input into the model, and the model outputs the category of the control action. During model training, a large amount of control data from different pilots is collected as training samples. These samples are labeled and then input into the model for training. The recognition accuracy is improved by adjusting the model's parameters.

[0044] When defining the objective of a control action, it is determined based on the type and characteristics of the action. For example, the objective of attitude adjustment control actions is to change the aircraft's attitude parameters such as pitch, roll, and yaw angles; the objective of power control control actions is to change the engine thrust, thereby affecting the aircraft's speed, altitude, and other parameters. The definition of the objective is recorded in the attribute fields of the control action data for use in establishing subsequent response correlation links.

[0045] Step S125: Using the pilot's control execution data after feature recognition as response node data, the response node data are arranged sequentially according to the time dimension to form a continuous response node sequence. Each response node data in the response node sequence is associated with the corresponding control action acquisition time information.

[0046] Similar to the arrangement of trigger node data, response node data is also arranged according to the time dimension, forming a response node sequence. Each response node contains the type of manipulation action, execution characteristics, target, and corresponding acquisition time information. During the arrangement process, temporal continuity must also be ensured. For missing data, null nodes are inserted using the same processing method as for the trigger node sequence. The data structure of the response node sequence is similar to that of the trigger node sequence, using a structured data table for storage and establishing an index with the acquisition time as the key.

[0047] Step S126: Extract the flight status feedback data between the trigger node sequence and the response node sequence as associated node data, confirm that the timestamp of the associated node data is between the timestamps of the corresponding trigger node data and the response node data, arrange the associated node data according to the time dimension to form an associated node sequence, and each data in the associated node sequence is accompanied by the corresponding timestamp information.

[0048] Correlation node data serves as a bridge connecting trigger node data and response node data. It reflects the changes in flight status after a change in wind shear environment and before the pilot's maneuvers. When extracting correlation node data, first, each trigger node and its corresponding response node are identified. By querying the indices of the trigger node sequence and response node sequence, temporally adjacent pairs of trigger and response node data are found. Then, the flight status feedback data between these two pairs of node data is extracted as correlation node data. It is necessary to ensure that the timestamp of the correlation node data lies between the timestamps of the trigger and response node data.

[0049] The extracted associated node data is arranged chronologically to form an associated node sequence. Each associated node also includes a corresponding timestamp and the specific values ​​of the flight status parameters. During the arrangement process, the associated node data undergoes quality checks to ensure its integrity and accuracy. Abnormal data is corrected or removed using the same methods as before. The associated node sequence is stored in the same way as the trigger node sequence and response node sequence for convenient subsequent analysis and processing.

[0050] Step S127: Connect each trigger node data in the trigger node sequence with the associated node data in the corresponding associated node sequence and the response node data in the response node sequence in chronological order to form a single control action and environmental change response association link.

[0051] When forming a response association link, a unique link identifier is first assigned to each trigger node data. Then, based on the timestamp correspondence, the trigger node data is matched with subsequent association node data and response node data. Specifically, for each trigger node data, the association node data whose timestamp immediately follows the trigger node data is found in the association node sequence, and the response node data whose timestamp follows the association node data is found in the response node sequence. These three node data are then concatenated in the order of trigger node data -> association node data -> response node data to form a single response association link.

[0052] During the chaining process, it is necessary to ensure the correct temporal order of data between each node, and that the chain contains complete trigger, association, and response information. In cases with multiple associated or response node data, the data is chained sequentially according to the timestamps to form a multi-node response association chain. Each response association chain is stored in an independent data structure, containing information such as the chain identifier, trigger node data, a list of associated node data, and response node data.

[0053] Step S128: For response node data that does not correspond to trigger node data, query the trigger node data that is adjacent to it, extract the associated node data corresponding to the adjacent trigger node data, and concatenate the response node data with the extracted associated node data.

[0054] In practice, some response node data may not have corresponding trigger node data. This could be due to the relatively weak intensity of wind shear changes, which prevented it from being identified as trigger node data, or it could be due to omissions during data acquisition. For these response node data without corresponding trigger node data, it is necessary to query their preceding and following trigger node data. By traversing the trigger node sequence, the largest trigger node data with a timestamp smaller than the response node data is identified as the preceding adjacent trigger node data, and the smallest trigger node data with a timestamp larger than the response node data is identified as the following adjacent trigger node data.

[0055] Then, the associated node data corresponding to the preceding adjacent trigger node data and the following adjacent trigger node data are extracted. This associated node data is then concatenated with the response node data that does not correspond to a trigger node data. The concatenation order is: associated node data corresponding to the preceding adjacent trigger node data -> response node data -> associated node data corresponding to the following adjacent trigger node data, forming a response association chain that includes the association information between the preceding and following trigger nodes. This ensures that all response node data is included in the response association chain, improving data integrity and the comprehensiveness of the analysis.

[0056] Step S129: For trigger node data that does not have corresponding response node data, query its most recent response node data, extract the associated node data corresponding to the response node data, connect the trigger node data and the extracted associated node data, and integrate all the connected single manipulation actions and environmental change response association links to form a set of manipulation actions and environmental change response association links.

[0057] For trigger node data without corresponding response node data, it indicates that after the wind shear environmental change occurred, the pilot did not take timely control actions, or the control actions were too small to be identified as response node data. In this case, the nearest response node data following the trigger node data is queried. By traversing the response node sequence, the smallest response node data with a timestamp greater than the trigger node data is found as the nearest subsequent response node data. Then, the associated node data corresponding to the response node data is extracted, and the trigger node data and associated node data are concatenated to form a response association link containing both trigger node data and associated node data.

[0058] After connecting all individual response-related links, these links are integrated. The integration process includes removing duplicate links and merging similar links. Duplicate links are identified by comparing node data within the links; for identical links, only one is retained. Similar links are determined by calculating their similarity; links with high similarity are merged, and the merged links retain key node data and common features. After integration, a set of response-related links relating manipulatory actions to environmental changes is formed. This set contains all possible response relationships between manipulatory actions and environmental changes.

[0059] Step S130: Based on the response association link set, trace the temporal trajectory of the control action and the fit relationship with environmental changes, and output the temporal fit relationship set of the control action and environmental changes, wherein the fit relationship reflects the correspondence between the timing of the control action, the type of action and the magnitude of wind shear environmental changes and the trend of flight state changes.

[0060] After obtaining the set of response correlation links, it is necessary to conduct a source analysis on the correlation between the temporal trajectory of the maneuvering actions and environmental changes. First, each response correlation link is analyzed in depth, extracting detailed information on the trigger node data, correlation node data, and response node data. The magnitude of wind shear environmental changes in the trigger node data is determined by the amplitude parameter of the state fluctuation characteristics; the timing of the maneuvering actions in the response node data is determined by the data acquisition time information, and the action type is determined based on the maneuvering action category identified in the previous steps; the flight state change trend is analyzed by the direction and rate of change of flight state parameters in the correlation node data. The direction of change is determined by comparing the parameter values ​​at adjacent time points, and the rate of change is calculated by the ratio of the change in parameter value to the time interval.

[0061] By correlating the timing and type of control actions with the magnitude of wind shear changes and the trend of flight status changes, the correlation between these factors is analyzed. For example, when the wind shear changes significantly, did the pilot execute the appropriate control actions at the right time, and were the types of actions effective in responding to the changing trends in flight status? The results of the correlation analysis are recorded in the form of data pairs, each containing information such as the timing of the control action, the type of action, the magnitude of the wind shear change, and the trend of flight status changes.

[0062] Step S131: Extract the trigger node data, response node data and associated node data of each response-related link from the set of response-related links one by one. During the extraction process, record the unique identification information of each link so that each type of node data forms a unique mapping relationship with the corresponding link.

[0063] When extracting node data, each link in the response-related link set is traversed, and the corresponding trigger node data, response node data, and associated node data are obtained through the unique identifier information of the link. For each link, information such as the collection time and state fluctuation characteristics of the trigger node data, the manipulation action type, implementation time, and execution characteristics of the response node data, and the flight status parameter values ​​and timestamps of the associated node data are extracted and stored in a temporary data structure. Simultaneously, the unique identifier information of the link is added as an attribute to each node data to ensure a unique mapping relationship between various types of node data and their corresponding links, facilitating subsequent data association and analysis.

[0064] Step S132: Analyze the extracted trigger node data, identify the key features in the trigger node data that reflect changes in the wind shear environment, and classify the types of wind shear environment changes based on the key features. Each type corresponds to a combination of key features.

[0065] The analysis of trigger node data primarily focuses on key features reflecting changes in the wind shear environment. These key features include the frequency, amplitude, duration, and trend of state fluctuations, reflecting the type and intensity of the wind shear environment. For example, wind shear caused by microbursts typically exhibits large amplitude variations and specific frequency components over a short period; while frontal wind shear is characterized by longer durations and a relatively gentler trend. Based on these key features, the trigger node data is classified using a clustering algorithm. The K-means clustering algorithm is employed, dividing the trigger node data into different categories according to a preset number of clusters. Each category corresponds to a type of wind shear environment change. The key feature combinations for each type are recorded in a type definition table for subsequent type identification and matching.

[0066] Step S133: Perform action feature analysis on the extracted response node data, extract the implementation time information of the manipulation action, identify the specific execution method and the part of action to be affected by the manipulation action, and determine the action type information of each manipulation action.

[0067] The motion feature analysis of response node data first extracts the execution time information of the control action, which is directly obtained from the acquisition time attribute of the response node data. Then, the specific execution method of the control action is analyzed, including the motion trajectory of the control components and changes in control force. This information is reflected through the dynamic characteristics of the control execution data. The point of application is determined according to the type of control component; for example, the control stick corresponds to the attitude control component of the aircraft, and the throttle lever corresponds to the power control component of the engine.

[0068] When determining the action type information, a comprehensive judgment is made by combining the execution method of the control action, the part of the body affected, and the category of control action identified in the previous steps. For example, the left and right swing of the control stick corresponds to roll attitude adjustment, the forward and backward push and pull of the control stick corresponds to pitch attitude adjustment, and the push and pull of the throttle stick corresponds to power increase or decrease. The action type information is encoded as a specific identifier and stored in the attribute field of the response node data for subsequent matching relationship analysis.

[0069] Step S134: Analyze the state changes of the extracted associated node data, track the continuous changes of flight state parameters after the occurrence of trigger node data, extract the change rate and change trend direction information of different flight state parameters, and integrate them to form flight state change trend information. The flight state change trend information is used to reflect the change trend of flight state from the occurrence of trigger node data to the implementation of the corresponding control action of response node data.

[0070] Analyzing the state changes of associated node data requires tracking the continuous changes in flight state parameters after the trigger node data occurs. First, a time series analysis is performed on each flight state parameter in the associated node data to calculate the rate of change and direction of change at different time points. The rate of change is calculated as the ratio of the difference between parameter values ​​at adjacent time points to the time interval. The direction of change is determined by comparing the signs of the rates of change across multiple consecutive time points: a positive rate of change indicates an increasing trend, while a negative rate of change indicates a decreasing trend.

[0071] For multiple flight status parameters, their changes need to be considered comprehensively to form overall flight status trend information. For example, a decreasing trend in altitude and a decreasing trend in speed may indicate that the aircraft is in a dangerous stall state; while stable altitude and increasing speed indicate that the aircraft is recovering to normal. During the integration process, the rate of change and direction of change of different flight status parameters are weighted and synthesized. The weights are determined based on the importance of the parameter to flight safety, with higher weights assigned to parameters of higher importance, such as altitude and speed, thus forming flight status trend information that comprehensively reflects the direction of flight status changes.

[0072] Step S135: Associate the type of wind shear environment change, the timing and type of the control action, and the flight status change trend information in the same response association link according to the link's unique identifier information to form a multi-dimensional association information group for a single response association link.

[0073] When forming multi-dimensional association information groups, the unique identifier of the link is used as the key to associate the wind shear environment change type, timing of the control action, action type, and flight status change trend information within the same response association link. This information is integrated into a structured data object, with each data object representing multi-dimensional association information for one response association link. The data object contains fields such as link identifier, wind shear type, timing of implementation, action type, and status change trend, with each field storing the corresponding information content.

[0074] Step S136: For each multi-dimensional related information group, analyze the timing of the implementation of the control action and the occurrence of the wind shear environment change, determine whether the time sequence meets the preset condition that the wind shear environment change precedes the control action, and determine whether the interval length is within the preset time range determined according to the type of wind shear environment change. At the same time, analyze the corresponding adaptation information between the action type of the control action and the flight state change trend, and determine whether the action type can produce a preset adjustment effect in response to the flight state change trend.

[0075] The time sequence analysis first compares the occurrence time of the wind shear environmental change (i.e., the time of data acquisition at the trigger node) with the implementation time of the control action (i.e., the time of data acquisition at the response node) to ensure that the time sequence conforms to the preset condition that the wind shear environmental change precedes the control action. If the implementation time of the control action is earlier than the occurrence time of the wind shear environmental change, the time sequence is considered abnormal. The determination of the interval length is based on the type of wind shear environmental change. Different types of wind shear environmental changes correspond to different preset time ranges, which are derived from flight manuals and historical flight data. For example, for micro-downburst type wind shear, the preset time range is shorter, requiring the pilot to respond within a shorter time; for frontal wind shear, the preset time range is longer. The actual time interval is compared with the preset time range. If it is within the range, the time sequence is considered normal; otherwise, it is considered abnormal.

[0076] The analysis of the correspondence between action types and flight status change trends is based on the function of the action type and the requirements of the flight status change trend. Each action type has its preset adjustment effect. For example, pitch attitude adjustment actions can change the aircraft's altitude change trend, and power increase actions can increase the aircraft's speed. The preset adjustment effect of the action type is matched with the flight status change trend. If the adjustment effect of the action type can offset or correct the unfavorable change trend of the flight status, the action type is judged to be compatible; otherwise, it is judged to be incompatible.

[0077] Step S137: Classify and organize the time-series connection analysis results and action type adaptation information of all multi-dimensional related information groups, group them according to the type of wind shear environment change and the action type of the control action, arrange the analysis results in chronological order within each group, integrate all the organized analysis results to form a set of time-series matching relationships between control actions and environmental changes. Each time-series matching relationship in the set contains the corresponding unique link identifier information, the type of wind shear environment change, the timing and action type of the control action, the flight status change trend information, the time-series connection analysis results, and the action type adaptation information.

[0078] During the classification and organization process, the multi-dimensional related information groups were first divided into different major categories according to the type of wind shear environment change. Then, within each major category, they were further divided into different subcategories according to the type of maneuvering action. Within each subcategory, the analysis results were arranged in chronological order to form an ordered sequence of analysis results. During the integration process, the information in each analysis result sequence was extracted and organized according to the format requirements of time-series alignment relationships. Each time-series alignment relationship includes fields such as unique link identifier information, type of wind shear environment change, timing and type of maneuvering action, flight status change trend information, time connection analysis results, and action type adaptation information. The integrated set of time-series alignment relationships is stored in a database for subsequent evolution analysis and generation of maneuvering adjustment guidelines.

[0079] Step S140: Based on the set of temporal matching relationships, trace the temporal trajectory of the control action and the matching relationship with environmental changes, and output the set of temporal matching relationships between the control action and environmental changes. The matching relationship reflects the correspondence between the timing of the control action, the type of action, the magnitude of wind shear environmental changes, and the trend of flight state changes.

[0080] Step S141: Sort each temporal matching relationship in the set of temporal matching relationships in chronological order to form a continuous temporal matching relationship sequence. Divide the temporal matching relationship sequence into continuous time periods, each time period containing multiple temporal matching relationships.

[0081] When sorting the set of temporal fit relationships, the execution time information of the maneuver in each relationship is used as the key, and the relationships are arranged in chronological order to form a continuous sequence of temporal fit relationships. A sliding window method is used to divide the time periods, with the window size determined based on the rate of change of flight status and the frequency of maneuver actions. The window size must be set to ensure that each time period contains a sufficient number of temporal fit relationships to reflect the overall fit between maneuver actions and environmental changes within that time period. The step size of the sliding window can be adjusted as needed; a smaller step size improves temporal resolution but increases data processing workload, while a larger step size reduces data processing workload but decreases temporal resolution. Experiments are conducted to determine an appropriate window size and step size so that the divided time periods accurately reflect the dynamic relationship between maneuver actions and environmental changes.

[0082] Step S142: Extract the type change information of wind shear environment from the temporal fit relationship within each time period, track the duration and intensity of the same type of wind shear environment change, and record the conversion sequence and conversion nodes of different types of wind shear environment changes to form the continuous change of wind shear environment within that time period.

[0083] Within each time period, the temporal correlations are traversed to extract the type information of wind shear environment changes. By comparing the types of wind shear environment changes in adjacent temporal correlations, the transition between types is identified, and the transition sequence and timestamps of the transition nodes are recorded. The duration of the same type of wind shear environment change is determined by calculating the time interval between the emergence of that type and its transition to another type. The intensity of change is reflected by statistical measures such as the average and maximum values ​​of the state fluctuation characteristics of that type of wind shear environment change within that time period. All of the above information is integrated to form a description of the continuous changes in the wind shear environment within that time period, including the type sequence, the duration of each type, the intensity of change, and the transition nodes.

[0084] Step S143: Extract the timing adjustment information and action type conversion information of the control action from the timing fit relationship within each time period, record the continuous adjustment range and adjustment interval of the timing of the same control action, and track the conversion logic and conversion frequency between different action types to form the dynamic adjustment process of the pilot's control actions within that time period.

[0085] Information on the timing of pilot maneuvers is extracted by comparing the time interval between the implementation time of the same type of maneuver in different temporal fits and the time of wind shear environmental changes. The adjustment magnitude is the difference in time interval between two adjacent maneuvers, and the adjustment interval is the time difference between two adjustments. Maneuver type conversion information is extracted by recording the order of occurrence and number of conversions of different maneuver types in temporal fits; the conversion frequency is the ratio of the number of conversions to the length of the time interval. The conversion logic between different maneuver types is determined by analyzing the flight state change trends and maneuver type compatibility before and after the conversion. For example, when pitch attitude adjustment maneuvers cannot effectively correct altitude change trends, the pilot may switch to power control maneuvers. Integrating the above information forms a description of the dynamic adjustment process of pilot pilot maneuvers, including the magnitude and interval of timing adjustments, the order, frequency, and logic of maneuver type conversions.

[0086] Step S144: Extract feedback information on the flight status change trend from the temporal fit relationship within each time period, record the magnitude and direction of change of flight status parameters before and after the implementation of control actions, analyze the time lag relationship between flight status changes and the implementation of control actions, and form the change of flight status with control actions within that time period.

[0087] The extraction of feedback information on flight status change trends primarily focuses on the changes in flight status parameters before and after the implementation of control actions. For each temporal fit, the magnitude and direction of change of flight status parameters within a certain time window before and after the implementation of the control action are compared. The magnitude of change is calculated by the difference between the maximum and minimum values ​​of the parameters, while the direction of change is determined by the overall trend of parameter change. The time lag relationship is determined by calculating the time interval between the implementation of the control action and the moment when the flight status parameters begin to show significant changes. The above information is arranged in chronological order to analyze the impact of different control actions on flight status changes, forming a description of how flight status changes with control action adjustments, including information such as the magnitude, direction, and time lag relationship.

[0088] Step S145: Synchronously correlate the continuous changes in wind shear environment, the dynamic adjustment process of pilot control actions, and the feedback changes in flight status within each time period to form a three-dimensional correlated information group for that time period.

[0089] Step S1451: Assign time markers to the continuous changes in the wind shear environment, with each time marker corresponding to a fixed time segment.

[0090] The time stamps are divided into equally spaced time segments, the length of which is determined by the size of the time period and the density of the data. For example, a time period can be divided into several time segments of 0.1 seconds each, with each segment corresponding to a unique time stamp. The time stamp format consists of the start time of the time period plus the time segment number for easy identification and sorting.

[0091] Step S1452: Mark each change event in the continuous change of the wind shear environment, and associate each change event with the corresponding time marker. The change events include the conversion of the wind shear environment change type and the adjustment of the change intensity exceeding the preset fluctuation range.

[0092] Change events in the continuously changing wind shear environment are identified by monitoring the type and intensity parameters of these changes. A change event is recorded when the type changes or the intensity exceeds a preset fluctuation range. A timestamp is added to each change event, and the time segment to which it belongs is determined based on the timestamp, thus associating the change event with the corresponding time marker. Detailed information about the change event, such as the type before and after the change, and the magnitude of the intensity change, is also recorded in the event description.

[0093] Step S1453: Assign the same time stamp to the dynamic adjustment process of the pilot's control actions. For each control action adjustment event, determine the time segment in which it occurs and associate the control action adjustment event with the corresponding time stamp. Control action adjustment events include adjustments to the timing of implementation and changes in the type of action.

[0094] The time stamping of the pilot's dynamic adjustment process of control actions is consistent with the time stamping of the continuous changes in the wind shear environment, using the same time segment division. Control action adjustment events are identified by monitoring changes in the timing and type of control actions. When the adjustment magnitude of the timing exceeds a preset threshold or the type of action changes, it is recorded as a control action adjustment event. The time segment to which the event belongs is determined based on its occurrence time, the event is associated with the corresponding time stamp, and detailed event information is recorded, such as the timing before the adjustment, the timing after the adjustment, and the type of action before and after the change.

[0095] Step S1454: Record each adjustment detail feature in the dynamic adjustment process of the pilot's control actions, associate the adjustment detail feature with the corresponding control action adjustment event, and then establish an indirect association between the control action adjustment event and the time stamp.

[0096] Adjustment details include the adjustment range, speed, and force of the manipulation action. These features are extracted through dynamic analysis of the manipulation execution data. Each adjustment detail is associated with a corresponding manipulation action adjustment event, and a reference to the adjustment detail is added to the attributes of the manipulation action adjustment event. Since the manipulation action adjustment event is already associated with a time stamp, the adjustment detail features are indirectly associated with the time stamp through the event.

[0097] Step S1455: Assign the same time stamp to the feedback changes in flight status. For each flight status change event, determine the time segment in which it occurs and associate the flight status change event with the corresponding time stamp to keep the flight status change event consistent with the corresponding time segment. Flight status change events include significant fluctuations in flight status parameters and changes in the trend of change.

[0098] The time stamping for flight status feedback changes remains consistent with the previous method. Flight status change events are identified by monitoring changes in flight status parameters. When a parameter fluctuation exceeds a preset threshold or its trend changes, it is recorded as a flight status change event. The time segment to which the event belongs is determined based on its occurrence time, associated with a time stamp, and detailed event information is recorded, such as parameter name, fluctuation amplitude, and direction before and after the trend change.

[0099] Step S1456: Record each change detail in the feedback changes of the flight status, associate the change details with the corresponding flight status change events, and then establish an indirect association between the flight status change events and the time stamp.

[0100] The details of the changes include the specific numerical sequence of flight status parameters during the change event, and the rate of change. These details are then associated with the corresponding flight status change events, and references to these details are added to the event's attributes, establishing an indirect link between the event and the time stamp.

[0101] Step S1457: Integrate wind shear environment change information, control action adjustment information and flight status feedback information under the same time mark in the order of time mark, and establish an association index between time mark and various types of information. The association index contains the correspondence between time mark, wind shear environment change events and details, control action adjustment events and details, and flight status change events and details.

[0102] Following the order of the time stamps, each time stamp is traversed, and the wind shear environment change events and details, control action adjustment events and details, and flight status change events and details under that time stamp are integrated to form the comprehensive information corresponding to that time stamp. Then, a relational index table is created, using the time stamp as the key, to record the storage location or reference pointers of various events and details corresponding to each time stamp. Through the relational index, various information under any time stamp can be quickly retrieved, facilitating subsequent analysis and processing.

[0103] Step S1458: Arrange the integrated information corresponding to different time markers in chronological order to form a continuous sequence of related information. The dynamic correlation process of wind shear environment, control actions, and flight status over time is presented through the sequence of related information.

[0104] The integrated information corresponding to all time markers is arranged in the order of the time markers, forming a continuous sequence of related information. Each element in this sequence represents integrated information for a given time marker, containing detailed information about the wind shear environment, maneuvering actions, and flight status within that time segment. By traversing this sequence, the dynamic changes and interrelationships of these three factors over time can be observed.

[0105] Step S146: For each three-dimensional associated information group, analyze the degree of matching between the adjustment process of the control action and the changes in the wind shear environment, determine whether the frequency and amplitude of the control action adjustment form a preset adaptation relationship with the frequency and intensity of the wind shear environment changes, and at the same time analyze the trend of the flight status feedback after the control action adjustment to determine whether the flight status changes in the preset stable direction.

[0106] Step S1461: Set the matching analysis dimensions for control actions and wind shear environment changes. The matching analysis dimensions include adjustment timing matching dimension, adjustment frequency matching dimension, adjustment amplitude matching dimension, and action type matching dimension. Each matching analysis dimension defines specific analysis points. The analysis points for the adjustment timing matching dimension include the time difference between the timing of control action adjustment and the occurrence of wind shear environment changes, and the degree of fit between the implementation sequence of adjustment actions and the evolution sequence of wind shear environment changes. The analysis points for the adjustment frequency matching dimension include the consistency between the frequency of control action adjustment and the frequency of wind shear environment changes, and the correlation between the change in adjustment frequency and the intensity of wind shear environment changes. The analysis points for the adjustment amplitude matching dimension include the adaptability of the amplitude of control action adjustment and the intensity of wind shear environment changes, and the synergy between the change in adjustment amplitude and the rate of wind shear environment change. The analysis points for the action type matching dimension include the correspondence between the type of control action and the type of wind shear environment change, and the relevance of the function of the action type to the flight status problems caused by wind shear environment changes.

[0107] Detailed evaluation criteria and quantification methods have been established for the key points of each matching analysis dimension. For example, the time difference in the timing matching dimension is quantified by calculating the difference between the timing of the maneuver adjustment and the timing of the wind shear environment change; the fit is evaluated by comparing the consistency between the sequence of maneuver implementation and the sequence of wind shear environment change evolution, using a sequence alignment algorithm to calculate the similarity score. Frequency consistency in the frequency matching dimension is measured by calculating the ratio of the maneuver adjustment frequency to the wind shear environment change frequency; a ratio close to 1 indicates high consistency; correlation is evaluated by calculating the correlation coefficient between the change in adjustment frequency and the intensity of wind shear environment change. Adaptability in the adjustment amplitude matching dimension is judged by comparing the ratio of the maneuver adjustment amplitude to the intensity of wind shear environment change with a preset adaptation range; synergy is evaluated by analyzing the synchronization between the rate of change of adjustment amplitude and the rate of change of wind shear environment. Correspondence in the action type matching dimension is determined by querying a preset correspondence table between action types and wind shear environment change types; relevance is evaluated based on the functional description of the action type and the requirements of the flight status problem.

[0108] Step S1462: Based on the specific analysis points of each matching analysis dimension, establish a matching degree evaluation space. The matching degree evaluation space includes the evaluation criteria and weight allocation for each analysis point. The evaluation criteria define the basis for judging different matching degrees, and the weight allocation is determined according to the importance of each analysis point to the overall matching degree.

[0109] The matching degree assessment space is a multi-dimensional assessment model, with each analytical point corresponding to one dimension. The assessment criteria set multiple matching degree levels for each dimension, such as excellent, good, average, and poor, with each level corresponding to specific judgment criteria. For example, in the time difference assessment criteria for the timing matching dimension, a time difference within the range of 0-0.5 seconds is excellent, 0.5-1 second is good, 1-2 seconds is average, and more than 2 seconds is poor. Weight allocation is determined using the analytic hierarchy process (AHP). Flight experts are invited to score the importance of each analytical point, and then consistency checks and weight calculations are performed to obtain the weight value of each analytical point in the overall matching degree assessment. Highly important analytical points, such as the relevance of the action type matching dimension, are assigned higher weights; relatively less important analytical points, such as the relevance of the adjustment frequency matching dimension, are assigned lower weights.

[0110] Step S1463: For each three-dimensional associated information group, based on the matching degree evaluation space, analyze the matching status of each matching analysis dimension one by one, compare the degree of fit between the adjustment process of the manipulation action and the changes in the wind shear environment at each analysis point, and output the matching level of each analysis point according to the evaluation criteria.

[0111] For each three-dimensional related information group, each analysis point in the matching analysis dimension is traversed. The actual value is calculated according to the quantification method of the analysis point, and then compared with the level range in the evaluation criteria to determine the matching level of that analysis point. For example, in the adaptability analysis point of the adjustment amplitude matching dimension, the ratio of the adjustment amplitude of the manipulation action to the intensity of wind shear environmental change is calculated. If this ratio is within the preset adaptability range, the matching level is good; if it exceeds the range, it is determined to be average or poor based on the degree of exceedance. The matching level of each analysis point is recorded as the basis for subsequent calculation of the comprehensive matching score.

[0112] Step S1464: Based on the weight allocation of each analysis point, calculate the comprehensive matching score of each matching analysis dimension, and then combine the weight allocation of each matching analysis dimension to calculate the overall matching degree score between the maneuver adjustment process and the wind shear environment change. Summarize to obtain the quantitative result of the degree of fit. The overall matching degree score is used to quantitatively reflect the matching situation between the two.

[0113] First, for each matching analysis dimension, the matching level of each analysis point under that dimension is converted into a corresponding score. For example, excellent corresponds to 5 points, good to 4 points, average to 3 points, poor to 2 points, and very poor to 1 point. Then, based on the weight of each analysis point, the weighted average score of that dimension is calculated as the comprehensive matching score for that matching analysis dimension. Next, a weight is assigned to each matching analysis dimension. This weight is determined based on the importance of each dimension in the overall matching degree assessment and is calculated using the analytic hierarchy process (AHP). The comprehensive matching score of each matching analysis dimension is multiplied by its corresponding weight and then summed to obtain the overall matching degree score between the maneuver adjustment process and the wind shear environment change. The higher the overall matching degree score, the better the matching between the two.

[0114] Step S1465: Track the continuous changes in flight status feedback data after the control action is adjusted, record the specific numerical changes of flight status parameters in chronological order, and mark the key change nodes of flight status. Key change nodes include the moment when the parameter reaches the extreme value, the moment when the parameter change direction changes, and the moment when the parameter change rate changes significantly.

[0115] After adjusting the control actions, the flight status feedback data is continuously tracked, and the values ​​of the flight status parameters are recorded in chronological order. By calculating the first and second derivatives of the data, extreme points, points of change in direction, and points of significant change in rate of change of the parameters are identified; these points are the key change nodes. When recording key change nodes, not only the timestamp is recorded, but also the parameter value, direction of change, and rate of change at that moment are also recorded to analyze the changing trends of the flight status.

[0116] Step S1466: Analyze the change patterns between flight status change nodes, calculate the change characteristics between adjacent key change nodes, and determine the change trend information of flight status feedback based on the change characteristics. The change characteristics include parameter change amount, change rate, and change duration.

[0117] After identifying key change nodes, the changes in flight status parameters between adjacent nodes are analyzed. The parameter change is calculated as the difference between parameter values ​​between nodes; the rate of change is the ratio of the parameter change to the duration of change; and the duration of change is the time interval between adjacent nodes. Based on these characteristics, the trend of flight status feedback can be determined. For example, a positive parameter change with a gradually increasing rate of change indicates that the flight status is rapidly changing in a favorable direction; a negative parameter change with a gradually decreasing rate of change indicates that the unfavorable trend of flight status is slowing down.

[0118] Step S1467: Associate the quantification result of the degree of fit with the information on the trend of change, analyze the intrinsic relationship between the two, determine whether the overall matching degree score meets the preset threshold and whether the flight status feedback shows a stable convergence trend, and whether the overall matching degree score does not reach the preset threshold and whether the flight status feedback shows an unstable divergence trend, and obtain the correlation result.

[0119] The preset threshold is determined based on flight safety standards and historical data statistics. When the overall matching score is higher than this threshold, the control maneuver adjustment process is considered to be well matched with the changes in wind shear environment; otherwise, the matching is considered poor. The quantitative results of the matching degree are correlated with the information on the trend of change to compare the trends of flight status feedback under different matching scores. If, when the overall matching score is higher than the preset threshold, the flight status feedback shows a stable convergence trend (i.e., parameters gradually tend to stable values ​​and the rate of change gradually decreases), and when the score is lower than the preset threshold, the flight status feedback shows an unstable divergence trend (i.e., parameter fluctuations increase and the rate of change is unstable), then there is a positive correlation between the two. The correlation results are recorded as a table showing the correspondence between matching score and flight status change trends.

[0120] Step S1468: Integrate the results of the fit degree analysis, the trend of flight status feedback changes, and the correlation between the two to form an analysis conclusion. The analysis conclusion includes the matching status of the control action adjustment process with the wind shear environment changes in various dimensions, the overall matching degree score, the specific trend of flight status feedback changes, and the correlation between the matching degree and the flight status changes.

[0121] The analysis integrates the matching results across various dimensions, the overall matching score, the trends in flight status feedback, and the correlation results to form a comprehensive analytical conclusion. This conclusion is presented using a combination of natural language descriptions and data tables. It includes qualitative descriptions, such as "the type of maneuver adjustment process matches well with the changes in wind shear environment, but the timing of the adjustment is only moderately matched," as well as quantitative data, such as an overall matching score of 85 points and a decrease in the rate of change of flight status parameters from 2 m / s² to 0.5 m / s². The correlation relationships are displayed in tabular form, showing the flight status change trends corresponding to different matching score ranges.

[0122] Step S147: Summarize the matching degree analysis results and flight status feedback change trends corresponding to the three-dimensional correlation information groups for all time periods, arrange the analysis results in the order of time periods, identify the common characteristics and differences of adaptability changes in different time periods, supplement the adaptability change trend analysis across time periods, and form the dynamic change results of the adaptability of control actions.

[0123] During the aggregation process, the analysis conclusions of the three-dimensional correlation information groups for each time period are arranged in chronological order to form a continuous sequence of analysis results. By comparing the analysis results of different time periods, common characteristics and differences in adaptability changes are identified. Common characteristics include, for example, that the adjustment range of maneuvering actions is generally larger during periods of greater wind shear intensity; differences include, for example, significant differences in the adaptability scores of the action type matching dimension across different time periods. The analysis of adaptability change trends across time periods is conducted by calculating the curves of changes in indicators such as the overall matching degree score and the matching scores of each dimension over time, analyzing the upward or downward trends of these indicators, as well as the inflection points and causes of these trends. The above analysis results are integrated to form a dynamic change result of maneuvering action adaptability. This dynamic change result details the changes in maneuvering action adaptability over time, including the overall trend, the characteristics of changes in each dimension, and influencing factors.

[0124] Step S150: Generate control adjustment guidelines based on the adaptive dynamic change results. The control adjustment guidelines provide adjustment directions for control actions that do not conform to environmental changes and flight status feedback in the adaptive dynamic change results, in order to assist the pilot in optimizing control actions in real time.

[0125] Step S151: Filter out control actions that do not match the wind shear environment changes and flight status feedback from the adaptive dynamic change results. During the filtering process, set the judgment criteria for non-matching based on the matching degree analysis results and the flight status feedback change trend. Non-matching control actions include control actions whose timing is not properly connected with the wind shear environment changes, and control actions whose action type does not meet the preset requirements for adaptability to the flight status change trend.

[0126] The screening of mismatched control actions first establishes judgment criteria. For the criterion of inappropriate timing, if the timing analysis result is abnormal or the matching level of the timing matching dimension is poor, it is judged as inappropriate timing. For the criterion of action type adaptability not meeting preset requirements, if the action type adaptability information is incompatible or the matching level of the action type matching dimension is poor, it is judged as incompatible action type. Each temporal fit relationship in the dynamic adaptability change results is traversed, and information related to mismatched control actions is filtered out according to the judgment criteria. This includes link unique identifier information, the timing of the control action implementation, the action type, the type of wind shear environment change, and the flight status change trend.

[0127] Step S152: Classify and organize the information related to the selected mismatched manipulation actions, group them according to the type of mismatch, arrange the relevant information of the manipulation actions in each group in chronological order, extract the common features of each group of mismatched manipulation actions, and locate the typical problems of each group of mismatched manipulation actions in terms of implementation timing or action type.

[0128] The information related to mismatched maneuvers was categorized into two groups based on the type of mismatch: improper timing and incompatible action type. Within each group, the maneuver information was arranged chronologically to observe the development trend of the problem. Analysis of each group revealed common characteristics. For example, in the improper timing group, most maneuvers were implemented after the wind shear environmental change occurred, and the lag time was relatively close. In the incompatible action type group, a specific action type frequently failed to match the specific wind shear environmental change. Typical problems were identified based on these common characteristics, such as delayed implementation timing and incorrect selection of specific action types.

[0129] Step S153: For each group of non-matching control actions, compare the actual implementation time of the non-matching control actions with the ideal time interval between the time when the wind shear environment changes, define the difference information between the implementation timing and the optimal timing, and obtain the implementation timing deviation information.

[0130] The ideal time interval is determined based on the type of wind shear environmental change and the recommended response time in the flight manual. Different types of wind shear environmental changes correspond to different ideal time intervals. For each set of mismatched control actions, the time interval between the actual execution time of each control action and the time when the wind shear environmental change occurs is calculated. This time interval is compared with the ideal time interval to obtain the difference in time interval, i.e., the timing deviation information. The deviation information includes the direction of the deviation (advanced or delayed) and the magnitude of the deviation, for example, delayed by 0.8 seconds, advanced by 0.3 seconds, etc.

[0131] Step S154: Combining the type of wind shear environment change and the flight state change trend, compare the difference between the actual function and expected requirements of the group of non-matching control actions, locate the inconsistencies between the action type and the environmental and state requirements, and obtain the action type mismatch information for each group of non-matching control actions.

[0132] The actual function of a maneuver type is determined based on its preset adjustment effect, while the expected requirement is determined based on the type of wind shear environment change and the flight state change trend. For example, when facing a micro-downburst type of wind shear and the flight state change trend is a rapid descent in altitude, the expected requirement is to increase pitch attitude and power; if the actual maneuver type is to decrease pitch attitude, then the actual function differs from the expected requirement. By comparing the actual function with the expected requirement, discrepancies are identified, such as incorrect adjustment direction of the maneuver type or insufficient adjustment intensity, forming maneuver type mismatch information.

[0133] Step S155: Real-time acquisition of changes in the current wind shear environment, including the type, intensity, and trend of changes in the current wind shear environment, and simultaneously acquisition of feedback information on the current flight status, including the specific values, rate of change, and direction of change of the current flight status parameters.

[0134] Real-time data acquisition is conducted through sensors and data acquisition systems on the aircraft, collecting data including current atmospheric data, inertial measurement data, engine parameters, etc. The type of current wind shear environment change is determined using the same wind shear environment change type identification method as in the previous steps; the current intensity is calculated using the amplitude parameters of the state fluctuation characteristics; and the current change trend is determined by analyzing the changing trends of the state fluctuation characteristics. The acquisition of current flight status feedback information directly obtains the values ​​of current flight status parameters, calculates the rate of change by the first derivative of the parameters, and determines the direction of change by the sign of the derivative.

[0135] Step S156: Combining the current changes in the wind shear environment and the current feedback information of the flight status, determine the adjustment direction of the control actions for each set of non-matching control action timing deviation information and action type mismatch information. The adjustment direction includes advancing or delaying the implementation timing, adjusting the implementation interval, changing the action type, adjusting the action intensity, and optimizing the action combination.

[0136] Regarding timing deviation information, if the actual implementation time lags behind the ideal time interval, the adjustment direction is to advance the implementation time; if it is advanced, the adjustment direction is to delay the implementation time. The adjustment of the implementation interval is determined based on the frequency of wind shear environment changes and the preset time range. If the current wind shear environment changes frequently, it is recommended to reduce the implementation interval; conversely, increase the implementation interval. For information regarding incompatible maneuver types, replace them with appropriate maneuver types according to expected needs; for example, replace the pitch decrease maneuver with the pitch increase maneuver. The adjustment of maneuver intensity is determined based on the intensity of wind shear environment changes and the magnitude of flight state change trends. When the intensity is high, it is recommended to increase the maneuver intensity; conversely, decrease the maneuver intensity. Optimization of maneuver combinations is recommended based on multiple wind shear environment change types and complex flight state change trends, suggesting the use of combinations of multiple maneuver types to achieve better adjustment results.

[0137] Step S157: Based on the adjustment direction, refine the specific adjustment content of the control actions, define the specific implementation time range after the adjustment for the timing of implementation; define the specific action type after the replacement for the action type; define the range of action intensity after the adjustment for the action intensity; define the optimized action combination sequence and coordination method for the action combination optimization, and define the adjustment range and rhythm of the control actions. The adjustment range and rhythm adapt to the current wind shear environment changes and flight status feedback to obtain preliminary control adjustment guidance.

[0138] The timing range for implementation is determined based on the timing deviation information and the type of current wind shear environment change. For example, for a timing deviation of 0.8 seconds, if the current wind shear environment change type is a micro-downburst and the ideal time interval is 0.5 seconds, then the adjusted timing range is set to 0.3-0.7 seconds after the occurrence of the wind shear environment change. The specific action type after replacement is determined based on action type incompatibility information and expected requirements, and is directly designated as the preset compatible action type. The action intensity range is defined based on the intensity of the current wind shear environment change and the magnitude of the flight status change trend, combined with the optimal action intensity range from historical data. The action combination sequence and coordination method are determined based on the function and implementation effect of the action type, determining which actions to implement first, which actions to implement later, and the time interval and intensity coordination between actions. The adjustment range and adjustment rhythm are determined based on the intensity and rate of change of the current wind shear environment and the changes in flight status feedback, ensuring that the adjustment range can effectively respond to environmental changes and that the adjustment rhythm matches the frequency of environmental changes. Integrating the above specific adjustment contents forms a preliminary control adjustment guide.

[0139] Step S158: Retrieve scene records from historical flight data that are similar to the current wind shear environment and flight status. Select control action cases from the retrieved scene records that meet the preset standards for control action adaptability. Combine the control action cases that meet the preset standards to improve the preliminary control adjustment guidelines and generate the final control adjustment guidelines.

[0140] For example, step S1581: Establish a historical flight data retrieval system, which contains multiple historical flight data, and the historical flight data is classified and stored according to wind shear environment type, flight stage and aircraft model.

[0141] The historical flight data retrieval system is built using a relational database. Historical flight data is categorized and stored according to key fields such as wind shear environment type, flight stage, and aircraft model. Each historical flight data record contains detailed information on wind shear environment changes, pilot control execution data, flight status feedback data, and control action adaptability analysis results. To improve retrieval efficiency, indexes are created on key fields, such as wind shear environment type indexes and flight stage indexes.

[0142] Step S1582: Set search conditions, which include the type, intensity, and trend of the current wind shear environment change, the initial parameters, rate of change, and direction of change of the current flight state, and the current flight stage.

[0143] The search criteria are set based on the current actual situation, taking into account the type, intensity, and trend of the current wind shear environment changes, the initial parameters of the current flight state (such as altitude, speed, attitude angle, etc.), the rate of change, the direction of change, and the current flight stage (such as takeoff, climb, cruise, descent, landing, etc.). Each search criterion has a certain matching tolerance range set to improve the flexibility and accuracy of the search.

[0144] Step S1583: Perform a search operation through the historical flight data retrieval system to obtain scene records that meet the search conditions, and sort the search results by relevance. The sorting is based on the similarity score between the scene record and the current scene. The similarity score is calculated based on the matching degree of the search conditions. The similarity score is obtained by weighted calculation of the matching degree of each search condition. The search conditions include the type, intensity, and trend of wind shear environment changes, as well as parameters of the flight state.

[0145] During the retrieval process, the set search criteria are input into the historical flight data retrieval system. The system then queries the database based on these criteria and returns scene records that match the conditions. The similarity score for the search results is calculated as follows: First, the matching degree of each search criterion is calculated. A perfect match for the type scores 1 point, otherwise 0 points. The intensity matching degree is calculated by dividing the current intensity by the intensity of the historical scene by the ratio of the difference between the current intensity and the intensity of the historical scene; a smaller ratio indicates a higher matching degree. The trend matching degree is calculated by comparing the similarity of the direction and rate of change trends. The flight state parameter matching degree is measured by calculating the Euclidean distance between the current parameters and the parameters of the historical scene; a smaller distance indicates a higher matching degree. Then, weights are assigned according to the importance of each search criterion: type and flight stage have higher weights, intensity and trend have lower weights, and flight state parameters have relatively lower weights. The matching degree of each search criterion is multiplied by its corresponding weight and summed to obtain the similarity score of the scene record. The search results are then sorted from highest to lowest similarity score.

[0146] Step S1584: Select control action cases from the sorted scene records that meet the preset standards for control action adaptability. During the selection process, refer to the analysis results of the control action matching degree in historical scenes and the trend of flight status feedback changes, and select cases whose matching degree meets the preset threshold and whose flight status feedback shows a stable convergence trend.

[0147] The preset threshold is determined based on the overall matching score of cases with good control action adaptability in historical data, typically set at an overall matching score of 80 or above (out of 100). A stable convergence trend in flight status feedback is determined by the changing trends of flight status parameters in historical scenarios. When parameters gradually approach stable values ​​and the rate of change gradually decreases, it is considered a stable convergence trend. The sorted scenario records are iterated through, and the control action matching analysis results for each scenario are checked to see if they meet the preset threshold, and whether the flight status feedback trend is a stable convergence trend. Cases of control actions that meet the criteria are then selected.

[0148] Step S1585: Analyze the selected manipulation action cases and extract the key information of the manipulation actions in the manipulation action cases. The key information of the manipulation actions includes the timing of the implementation of the manipulation action, the type of action, the adjustment range, the adjustment rhythm, the combination of actions, and the detailed control during the implementation process.

[0149] The analysis of control maneuver cases begins with obtaining the pilot's control execution data and the results of control maneuver suitability analysis. From the pilot's control execution data, information such as the timing of the control maneuvers, the type of maneuver, the adjustment range (displacement of control components), the adjustment rhythm (control speed and time intervals), and the combination of maneuvers (the sequence and coordination of different maneuver types) is extracted. Detailed control during implementation includes changes in control force and fine-tuning during the control process; this information is extracted by analyzing the dynamic characteristics of the control execution data. The above key information is organized into a structured data format to facilitate comparative analysis with currently unsuitable control maneuvers.

[0150] Step S1586: Compare and analyze the key information of the extracted case manipulation actions with the relevant information of the current non-matching manipulation actions, define the differences between the manipulation actions in the case and the current non-matching manipulation actions in terms of implementation timing, action type, and adjustment range, analyze the reasons for the differences, and obtain the comparative analysis results.

[0151] The comparative analysis was conducted on aspects such as timing of implementation, type of action, and magnitude of adjustment. Differences in timing of implementation were defined by calculating the difference between the timing of the case study's maneuver and the timing of the currently mismatched maneuver; differences in action type were defined by comparing whether the type of action in the case study was consistent with the type of action currently mismatched; differences in magnitude of adjustment were defined by calculating the ratio of the magnitude of the adjustment in the case study to the magnitude of the adjustment in the currently mismatched action. The analysis of the causes of these differences considered the differences between the current wind shear environment and flight status and those in the case study. For example, if the intensity of the current wind shear environment change is greater than that in the case study, it may lead to insufficient adjustment of the current maneuver. The differences and their causes were recorded in the comparative analysis results.

[0152] Step S1587: Based on the comparative analysis results, extract the corresponding control adaptation technology solutions. The control adaptation technology solutions include the optimal control action implementation timing selection method for specific wind shear environment change types, the most suitable action type recommendation for specific flight state change trends, applicable scenarios for different adjustment amplitudes and adjustment rhythms, and effective action combination strategies.

[0153] Based on the comparative analysis results, when the timing of the unsuitable control maneuver is delayed, the optimal timing selection method for this type of wind shear environment change is extracted from the case studies, such as implementing the control maneuver 0.5-1 second in advance; when the maneuver type is incompatible, the most suitable maneuver type recommendation for the current flight state change trend is extracted; when the adjustment range is insufficient, suggestions for the adjustment range and adjustment rhythm applicable to the current intensity are extracted; when maneuver combination optimization is required, effective maneuver combination strategies from the case studies are extracted, such as implementing pitch attitude adjustment first, followed by power control. The above control adaptation technology solutions are then compiled into specific guiding principles and methods.

[0154] Step S1588: Integrate the control adaptation technology solution with the adjustment direction in the preliminary control adjustment guide. For each adjustment direction, modify the adjustment content in conjunction with the control adaptation technology solution, and perform preliminary optimization on the control adjustment guide after integrating the control adaptation technology solution. Input the optimized control adjustment guide into the control adjustment guide feasibility verification model, perform simulation verification operation, and obtain simulation verification results. The simulation verification results include the flight state parameter change curve after implementing the adjustment guide, flight state stability assessment, and potential risk point identification.

[0155] During the integration process, specific methods and suggestions from the control adaptation technology solution are applied to the corresponding adjustment directions of the initial control adjustment guidelines. For example, in the direction of adjusting the implementation timing, the optimal implementation timing selection method is used to correct the implementation time range; in the direction of adjusting the action type, the most suitable action type recommendation is adopted to replace the action type in the initial guidelines. Preliminary optimization includes refining and quantifying the adjustment content, such as specifying the adjustment range to the displacement range of control components, and specifying the adjustment rhythm to the numerical range of control speed and time interval. The feasibility verification model for the control adjustment guidelines is built using flight simulator software. The optimized control adjustment guidelines are input into the simulator, initial conditions similar to the current wind shear environment and flight state are set, the simulator is run, and the flight state parameter change curves after implementing the adjustment guidelines are obtained. By analyzing the change curves, the stability of the flight state is evaluated, such as whether the parameter fluctuation range is within the allowable range and whether it can quickly converge to a stable value; potential risk points are identified, such as whether there is a stall risk or an overload exceeding the limit risk.

[0156] Step S1589: Analyze the simulation verification results, determine the feasibility and effectiveness of the optimized operation adjustment guidelines in the current scenario, and obtain the final operation adjustment guidelines based on the analysis results.

[0157] The analysis of simulation results first assesses whether the flight state parameter variation curves meet preset stability standards, such as parameter fluctuation amplitude within preset ranges and convergence time within preset timeframes. Control adjustment guidelines with good flight state stability and no potential risks are deemed feasible and effective. If potential risks exist or the stability assessment is poor, the control adjustment guidelines are revised based on the issues identified in the simulation results, such as adjusting the intensity of actions or changing the sequence of action combinations. The simulation is then repeated until feasible and effective control adjustment guidelines are obtained. The final control adjustment guidelines are described in clear and concise language, including specific guidance on the type of control action, timing of implementation, adjustment amplitude, adjustment rhythm, and action combination method, so that pilots can quickly understand and execute them.

[0158] Based on the same inventive concept, please refer to Figure 2 This paper shows a schematic block diagram of a real-time analysis system 100 for pilot handling quality under wind shear environment, provided in an embodiment of this application, for performing the above-described real-time analysis method for pilot handling quality under wind shear environment. The real-time analysis system 100 for pilot handling quality under wind shear environment may include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0159] In this embodiment, both the machine-readable storage medium 120 and the processor 130 are located in the real-time pilot handling quality analysis system 100 under wind shear conditions and are separately configured. Alternatively, the machine-readable storage medium 120 can also be integrated into the processor 130 and can communicate and interact with external systems through the communication unit 110. The machine-readable storage medium 120 is used to store machine-executable instructions for executing the scheme of this application, and the processor 130 is used to execute the machine-executable instructions stored in the machine-readable storage medium 120 to implement the real-time pilot handling quality analysis method under wind shear conditions provided in the aforementioned method embodiments.

[0160] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A method for real-time analysis of pilot control qualities under wind shear conditions, characterized in that, The method includes: The system captures flight control and environmental interaction data under wind shear conditions. The flight control and environmental interaction data includes flight status feedback data and pilot control execution data. The flight status feedback data reflects the state changes of the aircraft after the wind shear environment acts on it, and the pilot control execution data reflects the control actions taken by the pilot in response to the wind shear environment and changes in flight status. Based on the interaction data between flight control and environment, a response correlation link between control actions and environmental changes is established, and a set of response correlation links between control actions and environmental changes is output. The response correlation link takes wind shear environmental change as the trigger node, pilot control execution as the response node, and flight status feedback as the correlation node. Based on the set of response association links, the temporal trajectory of the control action and the fit relationship with environmental changes are traced, and a set of temporal fit relationships between the control action and environmental changes is output. The fit relationship reflects the correspondence between the timing of the control action, the type of action and the magnitude of wind shear environmental changes and the trend of flight state changes. Based on the evolutionary analysis of the time-series fit relationship set, the adaptive dynamic changes of the control actions are analyzed, and the adaptive dynamic changes of the control actions are output. The adaptive dynamic changes reflect the adjustment effect of the control actions as the wind shear environment changes continuously and the flight status is continuously fed back. Based on the adaptive dynamic change results, control adjustment guidelines are generated. These guidelines provide adjustment directions for control actions that do not conform to environmental changes and flight status feedback in the adaptive dynamic change results, thereby assisting pilots in optimizing control actions in real time.

2. The real-time analysis method for pilot control qualities under wind shear environment according to claim 1, characterized in that, The process establishes a response correlation link between control actions and environmental changes based on flight control and environmental interaction data, and outputs a set of response correlation links between control actions and environmental changes, including: The flight status feedback data and pilot control execution data in the flight control and environment interaction data are separated. The flight status feedback data reflecting changes in wind shear environment is distinguished from the flight status feedback data reflecting the aircraft's own state. The pilot control execution data for attitude adjustment is distinguished from the pilot control execution data for power control. Feature extraction is performed on the split flight state feedback data reflecting changes in the wind shear environment. The state fluctuation features directly related to the wind shear environment in the data are identified. Based on the state fluctuation features, the attribute of the data is further confirmed. The state fluctuation features reflect the state changes caused by the force exerted by the wind shear environment on the aircraft. The confirmed flight status feedback data reflecting changes in the wind shear environment is used as the trigger node data. The trigger node data is arranged sequentially according to the time dimension to form a continuous trigger node sequence. The adjacent trigger node data in the trigger node sequence maintains temporal continuity, and each trigger node data is accompanied by its corresponding acquisition time information. The split pilot control execution data is subjected to motion feature identification to distinguish the execution characteristics of different control actions, and the target of the control actions is defined based on the execution characteristics. The pilot's control execution data after feature recognition is used as response node data. The response node data is arranged sequentially according to the time dimension to form a continuous response node sequence. Each response node data in the response node sequence is associated with the corresponding control action acquisition time information. Extract flight status feedback data between the trigger node sequence and the response node sequence as associated node data, confirm that the timestamp of the associated node data is between the timestamps of the corresponding trigger node data and the response node data, arrange the associated node data according to the time dimension to form an associated node sequence, and each data in the associated node sequence is accompanied by the corresponding timestamp information. The trigger node data in the trigger node sequence is sequentially linked with the associated node data in the corresponding associated node sequence and the response node data in the response node sequence in chronological order to form a single control action and response association link to environmental changes. For response node data that does not correspond to trigger node data, query the trigger node data that is adjacent to it, extract the associated node data corresponding to the adjacent trigger node data, and concatenate the response node data with the extracted associated node data. For trigger node data that does not have a corresponding response node data, query its most recent response node data, extract the associated node data corresponding to the response node data, connect the trigger node data and the extracted associated node data, and integrate all the connected single manipulation actions and environmental change response association links to form a set of manipulation actions and environmental change response association links.

3. The real-time analysis method for pilot control qualities under wind shear environment according to claim 1, characterized in that, The method of tracing the temporal trajectory of the manipulation action and the fit between it and environmental changes based on the response association link set, and outputting a set of temporal fit relationships between the manipulation action and environmental changes, includes: The trigger node data, response node data, and associated node data of each response-related link are extracted one by one from the set of response-related links. During the extraction process, the unique identification information of each link is recorded so that each type of node data and the corresponding link form a unique mapping relationship. The extracted trigger node data is parsed to identify key features reflecting wind shear environment changes. Based on these key features, the types of wind shear environment changes are classified, and each type corresponds to a combination of key features. The extracted response node data is analyzed for action features to extract the implementation time information of the manipulation action, and at the same time, the specific execution method and the part of action to be applied are identified to determine the action type information of each manipulation action. The extracted associated node data is analyzed for state changes. The continuous changes of flight state parameters are tracked after the occurrence of trigger node data. The rate of change and direction of change of different flight state parameters are extracted and integrated to form flight state change trend information. The flight state change trend information is used to reflect the change trend of flight state from the occurrence of trigger node data to the implementation of the corresponding control action of response node data. The types of wind shear environment changes, timing and type of control actions, and flight status change trends in the same response association link are associated with each other according to the link's unique identifier information, forming a multi-dimensional association information group for a single response association link. For each multi-dimensional related information group, analyze the timing of the implementation of the control action and the occurrence of the wind shear environment change, determine whether the time sequence meets the preset condition that the wind shear environment change precedes the control action, and determine whether the interval length is within the preset time range determined according to the type of wind shear environment change. At the same time, analyze the corresponding adaptation information between the action type of the control action and the flight state change trend, and determine whether the action type can produce a preset adjustment effect in response to the flight state change trend. The time-series connection analysis results and action type adaptation information of all multi-dimensional related information groups are classified and organized. They are grouped according to the type of wind shear environment change and the action type of the control action. The analysis results within each group are arranged in chronological order. All the organized analysis results are integrated to form a set of time-series matching relationships between control actions and environmental changes. Each time-series matching relationship in the set contains the corresponding unique link identifier information, the type of wind shear environment change, the timing and type of the control action, the flight status change trend information, the time-series connection analysis results, and the action type adaptation information.

4. The real-time analysis method for pilot handling qualities under wind shear environment according to claim 1, characterized in that, The evolutionary analysis of the adaptive dynamic changes of the manipulation actions based on the set of temporal fit relationships, and the output of the adaptive dynamic change results of the manipulation actions, include: The temporal matching relationships in the set are sorted in chronological order to form a continuous temporal matching relationship sequence. The temporal matching relationship sequence is then divided into continuous time periods, each time period containing multiple temporal matching relationships. Extract the type change information of wind shear environment from the temporal fit relationship within each time period, track the duration and intensity of the same type of wind shear environment change, and record the conversion sequence and conversion nodes of different types of wind shear environment changes to form the continuous change of wind shear environment within that time period. Extract the timing adjustment information and action type conversion information of the pilot's control actions from the timing fit relationship within each time period, record the continuous adjustment range and adjustment interval of the timing of the same control action, and track the conversion logic and conversion frequency between different action types to form the dynamic adjustment process of the pilot's control actions within that time period. The feedback information on the trend of flight status change is extracted from the temporal fit relationship within each time period. The magnitude and direction of the change of flight status parameters before and after the implementation of the control action are recorded. The time lag relationship between the change of flight status and the implementation of the control action is analyzed to form the change of flight status with the control action within that time period. The continuous changes in wind shear environment, the dynamic adjustment process of pilot control actions, and the feedback changes in flight status within each time period are synchronously correlated to form a three-dimensional correlated information group within that time period. For each three-dimensional associated information group, analyze the degree of matching between the adjustment process of the control action and the changes in the wind shear environment, determine whether the frequency and amplitude of the control action adjustment form a preset adaptation relationship with the frequency and intensity of the wind shear environment changes, and at the same time analyze the trend of the flight status feedback after the control action adjustment to determine whether the flight status changes in the preset stable direction. The matching degree analysis results and flight status feedback change trends of the three-dimensional correlation information groups for all time periods are summarized, and the analysis results are arranged in order of time period. The common characteristics and differences of adaptability changes in different time periods are identified, and the trend analysis of adaptability changes across time periods is supplemented to form the dynamic change results of the adaptability of control actions.

5. The real-time analysis method for pilot handling qualities under wind shear environment according to claim 4, characterized in that, The continuous changes in wind shear environment, the dynamic adjustment process of pilot control actions, and the feedback changes in flight status within each time period are synchronously correlated to form a three-dimensional correlated information group for that time period, including: Time stamps are assigned to the continuous changes in the wind shear environment, with each time stamp corresponding to a fixed time segment; Each change event in the continuous change of the wind shear environment is marked, and each change event is associated with a corresponding time mark. The change events include the conversion of the wind shear environment change type and the adjustment when the change intensity exceeds the preset fluctuation range. Assign the same time stamp to the dynamic adjustment process of pilot control actions. For each control action adjustment event, determine the time segment in which it occurs and associate the control action adjustment event with the corresponding time stamp. Control action adjustment events include adjustments to the timing of implementation and changes in the type of action. Each adjustment detail feature in the dynamic adjustment process of the pilot's control actions is recorded, and the adjustment detail feature is associated with the corresponding control action adjustment event. Then, an indirect association is established between the control action adjustment event and the time stamp. Assign the same time stamp to the feedback changes in flight status. For each flight status change event, determine the time segment in which it occurs and associate the flight status change event with the corresponding time stamp to keep the flight status change event consistent with the corresponding time segment. Flight status change events include significant fluctuations in flight status parameters and changes in the trend of change. Record every detail of the changes in the flight status feedback, associate the details of the changes with the corresponding flight status change events, and then establish an indirect relationship between the flight status change events and time stamps. Information on wind shear environment changes, control action adjustments, and flight status feedback under the same time mark is integrated sequentially according to the time mark order, and an association index between time marks and various types of information is established. The association index contains the correspondence between time marks, wind shear environment change events and details, control action adjustment events and details, and flight status change events and details. The integrated information corresponding to different time markers is arranged in chronological order to form a continuous sequence of related information. The dynamic correlation process of wind shear environment, control actions, and flight status over time is presented through the sequence of related information.

6. The real-time analysis method for pilot control qualities under wind shear environment according to claim 1, characterized in that, The process of generating control adjustment guidelines based on the adaptive dynamic change results provides adjustment directions for control actions that do not align with environmental changes and flight status feedback, assisting pilots in optimizing control actions in real time, including: From the results of the adaptive dynamic changes, filter out control actions that do not match the changes in wind shear environment and flight status feedback. During the screening process, based on the matching degree analysis results and the trend of flight status feedback changes, set the judgment criteria for non-matching. Non-matching control actions include control actions whose timing is not properly connected with the changes in wind shear environment, and control actions whose action type does not match the flight status change trend in a preset manner. The relevant information of the selected mismatched manipulation actions is classified and organized, and grouped according to the type of mismatch. Within each group, the relevant information of the manipulation actions is arranged in chronological order. The common features of the mismatched manipulation actions in each group are extracted, and the typical problems of the timing or type of the mismatched manipulation actions in each group are identified. For each set of mismatched control actions, compare the actual implementation time of the mismatched control actions with the ideal time interval between the time when the wind shear environment changes, define the difference between the implementation timing and the optimal timing, and obtain the implementation timing deviation information; By combining the types of wind shear environment changes and the trends of flight status changes, the differences between the actual functions and expected requirements of the group of mismatched control actions are compared, the discrepancies between the action types and the environmental and status requirements are located, and the mismatch information of the action types of each group of mismatched control actions is obtained. Real-time data collection of changes in the current wind shear environment, including the type, intensity, and trend of changes in the current wind shear environment; and collection of feedback information on the current flight status, including the specific values, rate of change, and direction of change of the current flight status parameters. Based on the current changes in the wind shear environment and the current feedback information of the flight status, for each set of mismatched control actions, the adjustment direction of the control actions is determined. The adjustment direction includes advancing or delaying the implementation timing, adjusting the implementation interval, changing the action type, adjusting the action intensity, and optimizing the action combination. Based on the adjustment direction, the specific adjustment content of the control actions is refined. For the timing of implementation, the specific implementation time range after the adjustment is defined; for the action type, the specific action type after replacement is defined; for the action intensity, the range of action intensity after the adjustment is defined; for the action combination optimization, the sequence and coordination method of the optimized action combination are defined. At the same time, the adjustment range and rhythm of the control actions are defined. The adjustment range and rhythm adapt to the current wind shear environment changes and flight status feedback to obtain preliminary control adjustment guidance. Retrieve historical flight data of scenarios similar to the current wind shear environment and flight status. Select control action cases that meet the preset standards from the retrieved scenario records. Combine the control action cases that meet the preset standards to improve the preliminary control adjustment guidelines and generate the final control adjustment guidelines.

7. The real-time analysis method for pilot control qualities under wind shear environment according to claim 2, characterized in that, The separation of flight status feedback data and pilot control execution data in the flight control and environmental interaction data, distinguishing between flight status feedback data reflecting wind shear environmental changes and flight status feedback data reflecting the aircraft's own state, and distinguishing between pilot control execution data for attitude adjustment and pilot control execution data for power control, includes: By traversing the flight control and environmental interaction data, preliminary classification is performed according to the data source and the object represented, separating flight status feedback data and pilot control execution data; The initially separated flight status feedback data is cleaned, and the cleaned flight status feedback data is feature identified to extract feature information related to the external environment. The feature information related to the external environment includes sudden state changes and uncontrolled state fluctuations reflected in the data. Based on the characteristic information related to the effects of the external environment, a differentiation model is established to divide flight status feedback data into flight status feedback data reflecting changes in the wind shear environment and flight status feedback data reflecting the aircraft's own state. The differentiation model is based on preset judgment rules, which include: if the time of occurrence of the state change is not temporally related to the pilot's control execution data, and the magnitude of the state change exceeds a first preset threshold of the corresponding flight status parameter, then it is determined to be flight status feedback data reflecting changes in the wind shear environment; otherwise, it is determined to be flight status feedback data reflecting the aircraft's own state. The pilot control execution data that has been initially separated is cleaned, and the cleaned pilot control execution data is subjected to action attribute identification. The relevant information of the object of the control action is extracted, including the aircraft system targeted by the control action and the flight parameters that the control action is expected to change. Based on the relevant information of the object of the control action, a classification model is established to divide the pilot control execution data into pilot control execution data for attitude adjustment and pilot control execution data for power control. The object of the pilot control execution data for attitude adjustment is the attitude control system of the aircraft, and the desired change is the attitude parameters of the aircraft. The object of the pilot control execution data for power control is the power system of the aircraft, and the desired change is the power parameters of the aircraft. The various types of data after differentiation are labeled, and the labeled data are organized by category to form a classified flight status feedback data set and a pilot control execution data set. The labeling information includes the data category and the basis for differentiation.

8. The real-time analysis method for pilot handling qualities under wind shear environment according to claim 3, characterized in that, The process involves parsing the state changes of the extracted associated node data, tracking the continuous changes in flight state parameters after the occurrence of trigger node data, extracting the rate of change and direction of change of different flight state parameters, and integrating them to form flight state change trend information, including: The flight status parameter change information in the associated node data is analyzed. The flight status parameters include the aircraft's altitude, speed, attitude, and heading parameters. The records of changes in each flight status parameter are arranged in chronological order to form a sequence of changes for each parameter. Trend analysis is performed on the sequence of changes for each flight status parameter, the amount of parameter change between adjacent data points is calculated, the direction of increase or decrease of the parameter is determined based on the sign of the change, the time interval between adjacent data points is calculated, and the rate of change of the parameter is calculated based on the amount of parameter change and the time interval. By analyzing the direction and rate of change of multiple consecutive data points, the trend of increase or decrease of parameter change is determined. Identify the correlation between multiple related flight state parameters, which include parameter groups that influence and restrict each other. Analyze the synergistic relationship between the increase and decrease trends of the change sequence of each flight state parameter, which includes synchronous increase and decrease, reverse increase and decrease, and sequential increase and decrease. Based on the synergistic relationship, construct an overall trend of flight state changes, which is used to reflect the changes in the aircraft state under the combined effect of multiple related parameters. Based on the overall situation, the subsequent direction of flight status changes is predicted. Combined with the subsequent development of similar historical flight status change cases, and taking into account the continuous impact of wind shear environmental changes and the expected effect of pilot control actions, a first flight status change trend information is formed. The first flight status change trend information includes the possible subsequent change direction, change rate range, and change duration estimate. Analyze the impact characteristics of the trigger node data, combine these impact characteristics to determine its potential effect on the flight state change trend, correct the content in the first flight state change trend information that does not conform to the impact characteristics, and obtain the second flight state change trend information. The wind shear environment change type corresponding to the trigger node data is different, and its degree of influence, range of influence, and duration of influence on the flight state parameters are also different. Short-term flight status feedback data is collected after collecting data from related nodes. It is compared with the second flight status change trend information, the reasons for the differences are analyzed, the second flight status change trend information is further corrected, and the third flight status change trend information is obtained. This short-term flight status feedback data is used to reflect the actual subsequent trend of flight status changes. Integrate the third flight status change trend information to form flight status change trend information.

9. The real-time analysis method for pilot control qualities under wind shear environment according to claim 4, characterized in that, For each group of three-dimensional associated information, the degree of matching between the adjustment process of the control actions and the changes in the wind shear environment is analyzed. It is determined whether the frequency and amplitude of the control action adjustments form a preset adaptation relationship with the frequency and intensity of the wind shear environment changes. Simultaneously, the trend of changes in the flight state feedback after the control action adjustments is analyzed to determine whether the flight state is changing towards a preset stable direction, including: The analysis establishes matching dimensions between control actions and wind shear environmental changes. These dimensions include adjustment timing, adjustment frequency, adjustment magnitude, and action type. Each dimension defines specific analysis points. For example, the adjustment timing dimension considers the time difference between the control action adjustment and the occurrence of wind shear environmental changes, and the degree of fit between the implementation sequence of the adjustment actions and the evolution sequence of the wind shear environmental changes. The adjustment frequency dimension considers the consistency between the frequency of control action adjustments and the frequency of wind shear environmental changes, and the correlation between changes in adjustment frequency and the intensity of wind shear environmental changes. The adjustment magnitude dimension considers the adaptability between the magnitude of control action adjustments and the intensity of wind shear environmental changes, and the synergy between changes in adjustment magnitude and the rate of wind shear environmental changes. The action type dimension considers the correspondence between the type of control action and the type of wind shear environmental change, and the relevance of the action type's function to the flight status problems caused by wind shear environmental changes. Based on the specific analysis points of each matching analysis dimension, a matching degree evaluation space is established. The matching degree evaluation space includes the evaluation criteria and weight allocation for each analysis point. The evaluation criteria define the basis for judging different matching degrees, and the weight allocation is determined according to the importance of each analysis point to the overall matching degree. For each three-dimensional related information group, based on the matching degree evaluation space, the matching situation of each matching analysis dimension is analyzed one by one, and the degree of fit between the adjustment process of the manipulation action and the change of wind shear environment at each analysis point is compared. The matching level of each analysis point is output according to the evaluation criteria. Based on the weight allocation of each analysis point, the comprehensive matching score of each matching analysis dimension is calculated. Then, combined with the weight allocation of each matching analysis dimension, the overall matching degree score between the maneuver adjustment process and the wind shear environment change is calculated. The results of the overall matching degree are summarized to obtain the quantitative result of the degree of fit. The overall matching degree score is used to quantitatively reflect the matching situation between the two. Track the continuous changes in flight status feedback after control adjustments, record the specific numerical changes of flight status parameters in chronological order, and mark key change nodes in flight status. Key change nodes include the moment when the parameter reaches its extreme value, the moment when the direction of parameter change changes, and the moment when the rate of parameter change changes significantly. Analyze the change patterns between flight status change nodes, calculate the change characteristics between adjacent key change nodes, and determine the change trend information of flight status feedback based on the change characteristics, including parameter change amount, change rate, and change duration. By correlating the quantitative results of the degree of fit with the information on the trend of change, the intrinsic relationship between the two is analyzed, and it is determined whether the overall degree of fit score meets the preset threshold and whether the flight status feedback shows a stable convergence trend, or whether the overall degree of fit score does not reach the preset threshold and whether the flight status feedback shows an unstable divergence trend, the correlation results are obtained. The results of the compatibility analysis, the trend of flight status feedback changes, and the correlation between the two are integrated to form an analysis conclusion. The analysis conclusion includes the matching status of the control action adjustment process with the wind shear environment changes in various dimensions, the overall compatibility score, the specific trend of flight status feedback changes, and the correlation between the compatibility and flight status changes.

10. A real-time analysis system for pilot control qualities under wind shear conditions, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the real-time analysis method for pilot handling qualities under wind shear conditions as described in any one of claims 1 to 9 by executing the machine-executable instructions.

Images