Radio altimeter test data anomaly detection method and system

Abstract

The present invention discloses a method and system for detecting anomalies in radio altimeter test data, which relate to the technical field of altimeter testing. The method comprises the following steps: obtaining historical test data, outputting a first analysis result, collecting real-time flight parameters, calculating dynamic statistical data based on the first analysis result, establishing an adaptive baseline, constructing a first anomaly detection model, obtaining a real-time flight stage and outputting an anomaly prediction result, selecting an alarm level based on the anomaly prediction result and inputting the result into a flight control terminal. The present invention realizes adaptive anomaly detection of radio altimeter test data by constructing a dynamic baseline, thereby improving accuracy, and combines multimodal flight parameters for fusion analysis to enhance the comprehensiveness of detection. A physical verification mechanism is introduced to ensure the reliability of anomaly detection. At the same time, an alarm level is divided according to the anomaly result, thereby realizing real-time response, providing a strong guarantee for flight safety, and significantly improving the intelligence and safety of altimeter anomaly detection.

Description

Technical Field

[0001] The present invention relates to the technical field of altimeter testing, and in particular to a method and system for detecting abnormality in test data of a radio altimeter. Background Art

[0002] As a crucial navigational device for aircraft, the accuracy of the radio altimeter is directly related to flight safety. By measuring the relative altitude between the aircraft and the ground or obstacles, the altimeter provides pilots with critical flight information. However, in actual flight, due to factors such as ambient noise, signal interference, and equipment aging, radio altimeter test data may exhibit anomalies, resulting in pilots receiving erroneous altitude information, potentially leading to flight accidents.

[0003] At present, the Chinese invention patent with application number CN202110868184.3 discloses a method for improving the low precision of a radio altimeter, which mainly includes seven steps: determining whether the low-altitude error of the altimeter is large after installation, detecting and confirming that the altimeter is working properly, testing whether the zero error δ0 on the altimeter is qualified, testing whether the low-altitude error δ1 on the altimeter is qualified, calibrating the zero point of the altimeter, re-adjusting the error of the altimeter and confirming that the performance of the altimeter is qualified, and judging whether the system low-altitude error δ2 on the altimeter is qualified. If qualified, the method ends; if unqualified, the method returns to the step of detecting and confirming that the altimeter is working properly.

[0004] The above technologies make it difficult to achieve accurate anomaly detection of radio altimeter test data through dynamic baseline construction, multimodal fusion detection, physical verification mechanism and alarm level classification, which affects the detection effect. Summary of the Invention

[0005] The technical problem solved by the present invention is that it is difficult for the existing technology to achieve accurate anomaly detection of radio altimeter test data through dynamic baseline construction, multimodal fusion detection, physical verification mechanism and alarm level classification, which affects the detection effect.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] A method for detecting anomalies in radio altimeter test data includes the following steps:

[0008] Step S1, obtaining historical test data and performing spatiotemporal feature extraction and analysis, and outputting a first analysis result;

[0009] Step S2: collecting real-time flight parameters according to the first analysis result and storing the real-time flight parameters in a test database;

[0010] Step S3, calculating dynamic statistical data according to the first analysis result, and establishing an adaptive baseline based on the dynamic statistical data;

[0011] Step S4, constructing a first anomaly detection model, obtaining the real-time flight phase, matching the normal and outputting the anomaly prediction result;

[0012] Step S5: Select an alarm level based on the abnormality prediction result and input it to the flight control terminal.

[0013] Preferably, step S1 includes the following sub-steps:

[0014] Step S101, retrieving historical test data from a test database, wherein the historical test data includes flight parameters during takeoff, cruising, and landing phases, and the flight parameters include historical altitude values, historical altitude change rates, historical signal strengths, and historical ambient noise;

[0015] Step S102, performing quantized wavelet decomposition on the historical test data and outputting a time-frequency energy matrix;

[0016] Step S103 , extracting spatiotemporal features from the time-frequency energy matrix and calculating feature correlation coefficients, and outputting a first analysis result, wherein the first analysis result includes a takeoff phase feature correlation coefficient, a cruise phase feature correlation coefficient, and a landing phase feature correlation coefficient.

[0017] Preferably, step S2 includes the following sub-steps:

[0018] Step S201, parsing the first analysis result:

[0019] If the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient, or the landing phase characteristic correlation coefficient is less than a preset baseline correlation coefficient, it is determined to be an environmentally sensitive phase, and real-time flight parameters corresponding to the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient, or the landing phase characteristic correlation coefficient are collected and a high-frequency sampling mode is set. The real-time flight parameters include real-time altitude value, altitude change rate, signal strength, and environmental noise;

[0020] If the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient, or the landing phase characteristic correlation coefficient is greater than or equal to the preset benchmark correlation coefficient, it is determined to be a stable phase and the preset standard sampling frequency is used;

[0021] Step S202: align the real-time flight parameters by timestamp and store them in the test database.

[0022] Preferably, step S3 includes the following sub-steps:

[0023] Step S301, calculating dynamic statistical data according to the first analysis result, wherein the dynamic statistical data includes a sliding window mean and a standard deviation;

[0024] The mathematical expression of the sliding window mean is:

[0025] ;

[0026] in, is the sliding window mean, is the sliding window length, For the Real-time flight parameters at all times, is the time point index within the window, is the current time point;

[0027] The mathematical expression of the standard deviation is:

[0028] ;

[0029] in, is the standard deviation;

[0030] Step S302: establishing an adaptive baseline based on dynamic statistical data.

[0031] Preferably, step S4 includes the following sub-steps:

[0032] Step S401: construct a first anomaly detection model based on the adaptive baseline, and use a local outlier factor algorithm to calculate the ratio of the local density to the neighboring density of the historical parameters. The mathematical expression of the local outlier factor algorithm is:

[0033] ;

[0034] in, for point The local outlier factor, is any data point in the real-time flight parameters, for point The adjacent collection, for point Neighborhood points, is a natural number greater than 0, for point The local reachable density of for point The local reachable density of

[0035] Step S402: classify the models according to the flight phases, and output a cruise phase model, a landing phase model, and a takeoff phase model;

[0036] Step S403 , calculate the Mahalanobis distance between the current height value and the time series baseline, analyze the Mahalanobis distance based on the abnormal distance prediction classification logic, and mark it as an abnormal prediction result if the distance exceeds a preset Mahalanobis distance threshold.

[0037] Preferably, the abnormal result distance prediction classification logic is:

[0038] If the Mahalanobis distance satisfies the Mahalanobis distance threshold that is greater than or equal to twice, it is marked as confirmed abnormal;

[0039] If the Mahalanobis distance satisfies the condition that the Mahalanobis distance is greater than or equal to the Mahalanobis distance threshold and less than twice the Mahalanobis distance threshold, it is marked as a suspected anomaly;

[0040] If the Mahalanobis distance is less than the Mahalanobis distance threshold, it is marked as normal;

[0041] Confirmed anomalies, suspected anomalies, and normal conditions are combined and output as anomaly prediction results.

[0042] Preferably, step S5 includes the following sub-steps:

[0043] Step S501: Select an alarm level based on the abnormality prediction result:

[0044] If the abnormal prediction result is a confirmed abnormality, an alarm signal with a level 1 alarm is triggered;

[0045] If the abnormal prediction result is a suspected abnormality, the triggering alarm level is a level 2 alarm signal;

[0046] If the abnormal prediction result is normal, no alarm is triggered;

[0047] Step S502: Combine the warning level and the corresponding real-time flight parameters and transmit them to the flight control terminal.

[0048] Preferably, it also includes:

[0049] Step S601: Verify whether the altitude change rate exceeds the performance envelope in combination with the aircraft dynamics model. The constraints are:

[0050] ;

[0051] in, is the height change rate, is the maximum airspeed, is the maximum angle of attack;

[0052] Step S602: If the verification in step S601 fails, the current detection result is overwritten and an alarm signal with an alarm level of triggering a level 1 alarm is triggered.

[0053] Preferably, the mathematical expression of the Mahalanobis distance threshold is:

[0054] ;

[0055] in, is the Mahalanobis distance threshold, is the critical value of the chi-square distribution.

[0056] A radio altimeter test data anomaly detection system, which is applied to the radio altimeter test data anomaly detection method, includes a data analysis module, a real-time acquisition module, a baseline establishment module, a model construction module and an anomaly alarm module;

[0057] The data analysis module is used to obtain historical test data and perform spatiotemporal feature extraction and analysis, and output a first analysis result;

[0058] The real-time acquisition module is used to collect real-time flight parameters according to the first analysis result and store the real-time flight parameters in a test database;

[0059] The baseline establishment module is used to calculate dynamic statistical data according to the first analysis result, and establish an adaptive baseline based on the dynamic statistical data;

[0060] The model building module is used to build a first anomaly detection model, obtain the real-time flight stage, match the normal and output the anomaly prediction result;

[0061] The abnormality warning module is used to select the warning level according to the abnormality prediction result and input it to the flight control end.

[0062] The beneficial effects of the present invention are as follows: the present invention realizes adaptive anomaly detection of radio altimeter test data by constructing a dynamic baseline, thereby improving accuracy; combines multimodal flight parameters for fusion analysis, thereby enhancing the comprehensiveness of detection; introduces a physical verification mechanism, thereby ensuring the reliability of anomaly detection; at the same time, divides alarm levels according to anomaly results, realizes real-time response, provides strong protection for flight safety, and significantly improves the intelligence and safety of altimeter anomaly detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A flowchart of the steps of a method for detecting abnormality in radio altimeter test data provided by one embodiment of the present invention;

[0064] Figure 2 A schematic diagram of the basic flow of a radio altimeter test data anomaly detection system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0066] Example 1, reference Figure 1 , provides a method for detecting anomaly in radio altimeter test data, comprising the following steps:

[0067] Step S1: Acquire historical test data, extract and analyze spatiotemporal features, and output a first analysis result.

[0068] Step S2: collecting real-time flight parameters according to the first analysis result, and storing the real-time flight parameters in a test database.

[0069] Step S3: Calculate dynamic statistical data according to the first analysis result, and establish an adaptive baseline based on the dynamic statistical data.

[0070] Step S4: construct a first anomaly detection model, obtain the real-time flight phase, match the normal and output the anomaly prediction result.

[0071] Step S5: Select an alarm level based on the abnormality prediction result and input it to the flight control terminal.

[0072] Step S1 includes the following sub-steps:

[0073] Step S101, retrieve historical test data from a test database, the historical test data including flight parameters of the takeoff phase, cruise phase and landing phase, the flight parameters including historical altitude values, historical altitude change rates, historical signal strengths and historical ambient noise.

[0074] Step S101 comprehensively retrieves historical test data from the test database, covering flight parameters during takeoff, cruising, and landing phases, providing detailed data support for subsequent analysis.

[0075] Step S102: Perform quantized wavelet decomposition on the historical test data and output a time-frequency energy matrix.

[0076] Step S102 performs quantized wavelet decomposition on the historical test data to effectively extract the time-frequency energy matrix, revealing the energy distribution of flight parameters at different frequencies and time scales, laying the foundation for subsequent feature extraction.

[0077] Step S103 , extracting spatiotemporal features from the time-frequency energy matrix and calculating feature correlation coefficients, and outputting a first analysis result, which includes a takeoff phase feature correlation coefficient, a cruise phase feature correlation coefficient, and a landing phase feature correlation coefficient.

[0078] Step S103 extracts spatiotemporal features from the time-frequency energy matrix, calculates feature correlation coefficients, and outputs a first analysis result closely related to the flight phase, providing a key basis for subsequent anomaly detection and warning level classification.

[0079] Step S1 extracts spatiotemporal features related to flight phases from historical test data, providing a foundation for subsequent anomaly detection. Through quantized wavelet decomposition and spatiotemporal feature extraction, we can deeply analyze the changing patterns of flight parameters at different stages, thereby outputting characteristic correlation coefficients closely related to takeoff, cruise, and landing, providing a strong basis for flight safety assessment.

[0080] Step S2 includes the following sub-steps:

[0081] Step S201, parsing the first analysis result:

[0082] If the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient or the landing phase characteristic correlation coefficient is less than the preset baseline correlation coefficient, it is determined to be an environmentally sensitive phase, and the real-time flight parameters corresponding to the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient or the landing phase characteristic correlation coefficient are collected and a high-frequency sampling mode is set. The real-time flight parameters include real-time altitude value, altitude change rate, signal strength and environmental noise.

[0083] If the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient or the landing phase characteristic correlation coefficient is greater than or equal to the preset reference correlation coefficient, it is determined to be a stable phase and the preset standard sampling frequency is adopted.

[0084] Step S201 intelligently analyzes the first analysis results and determines the environmental sensitivity of the flight phase based on the magnitude of the characteristic correlation coefficient. During environmentally sensitive phases, a high-frequency sampling mode is used to more accurately capture subtle changes in flight parameters and improve the sensitivity of anomaly detection. During stable phases, a preset standard sampling frequency is used to balance data quality and acquisition efficiency.

[0085] Step S202: align the real-time flight parameters by timestamp and store them in the test database.

[0086] Step S202 aligns the real-time flight parameters by timestamp and stores them in the test database, ensuring the data's sequentiality and consistency. This orderly data storage not only facilitates subsequent data analysis and mining, but also helps track anomalies during flight, providing strong support for flight safety assessments.

[0087] Based on the first analysis results, step S2 intelligently adjusts the sampling frequency of real-time flight parameters and sequentially stores the collected data. By identifying environmentally sensitive phases and applying a high-frequency sampling pattern, key changes during flight can be captured with greater precision, providing high-quality data support for anomaly detection. Furthermore, real-time data is aligned and stored by timestamp, ensuring data integrity and traceability, providing a solid foundation for subsequent analysis.

[0088] Step S3 includes the following sub-steps:

[0089] Step S301 : Calculate dynamic statistical data according to the first analysis result. The dynamic statistical data includes a sliding window mean and a standard deviation.

[0090] The mathematical expression of the sliding window mean is:

[0091] ;

[0092] in, is the sliding window mean, is the sliding window length, For the Real-time flight parameters at all times, is the time point index within the window, The current time point.

[0093] The mathematical expression of standard deviation is:

[0094] ;

[0095] in, is the standard deviation.

[0096] Step S301 calculates dynamic statistics based on the first analysis results. This step quantifies the average level and dispersion of flight parameters within a sliding window. The sliding window mean reflects the overall trend of the flight parameter, while the standard deviation reveals the fluctuation range of the parameter. These data provide a key basis for establishing an adaptive baseline.

[0097] Step S302: establishing an adaptive baseline based on dynamic statistical data.

[0098] Step S302 establishes an adaptive baseline based on dynamic statistical data. This step combines the real-time status of flight parameters with historical data to create a reference standard that can adapt to changing flight conditions. The establishment of an adaptive baseline makes anomaly detection more flexible and accurate, enabling more effective identification of flight anomalies and providing a strong guarantee for flight safety.

[0099] Step S3 uses the first analysis results to construct an adaptive baseline to more accurately monitor the real-time status of flight parameters. By calculating dynamic statistics, we can capture trends and fluctuations in flight parameters over time, thereby establishing an adaptive baseline that reflects the current flight status. This step provides a more realistic reference standard for subsequent anomaly detection, improving detection accuracy and reliability.

[0100] Step S4 includes the following sub-steps:

[0101] Step S401: construct a first anomaly detection model based on the adaptive baseline, and use a local outlier factor algorithm to calculate the ratio of the local density to the neighboring density of the historical parameters. The mathematical expression of the local outlier factor algorithm is:

[0102] ;

[0103] in, for point The local outlier factor, is any data point in the real-time flight parameters, for point The adjacent collection, for point Neighborhood points, is a natural number greater than 0, for point The local reachable density of for point The local reachable density of .

[0104] Step S401 constructs a first anomaly detection model based on the adaptive baseline and uses a local outlier factor algorithm to calculate the ratio of the local density of historical parameters to the neighboring density. This step quantifies the dispersion of flight parameters in historical data, providing an effective method for identifying outliers. The application of the local outlier factor algorithm enables the model to more sensitively capture local variations in parameters, improving the sensitivity of anomaly detection.

[0105] Step S402: Classify the models according to the flight phases, and output a cruise phase model, a landing phase model, and a takeoff phase model.

[0106] Step S402 categorizes the models according to flight phase, outputting cruise, landing, and takeoff models. This step takes into account the differences in flight parameters across different flight phases, providing a more realistic classification standard for the models. By constructing models tailored to each flight phase, the parameter characteristics of each phase can be more accurately reflected, improving the accuracy and specificity of anomaly detection.

[0107] Step S403 , calculate the Mahalanobis distance between the current height value and the time series baseline, analyze the Mahalanobis distance based on the abnormal distance prediction classification logic, and mark it as an abnormal prediction result if the distance exceeds a preset Mahalanobis distance threshold.

[0108] The abnormal result distance prediction classification logic is:

[0109] If the Mahalanobis distance satisfies the Mahalanobis distance threshold that is greater than or equal to twice, it is marked as confirmed abnormal.

[0110] If the Mahalanobis distance satisfies the condition that the Mahalanobis distance is greater than or equal to the Mahalanobis distance threshold and less than twice the Mahalanobis distance threshold, it is marked as a suspected anomaly.

[0111] If the Mahalanobis distance is less than the Mahalanobis distance threshold, it is marked as normal.

[0112] Confirmed anomalies, suspected anomalies, and normal conditions are combined and output as anomaly prediction results.

[0113] Step S403 calculates the Mahalanobis distance between the current height value and the time series baseline and analyzes the Mahalanobis distance based on the anomaly distance prediction classification logic. This step accurately identifies anomalies by quantifying the degree of difference between the current parameter value and the baseline. The application of anomaly distance prediction classification logic enables the model to classify and label anomalies based on their severity, providing clear guidance for subsequent alarm responses. By combining confirmed anomalies, suspected anomalies, and normal conditions into an anomaly prediction result, the entire anomaly detection process becomes clearer and more organized.

[0114] Step S4 constructs an anomaly detection model based on an adaptive baseline, enabling precise monitoring of flight parameters across different flight phases. By employing a local outlier factor algorithm and Mahalanobis distance analysis, it accurately identifies anomalies during flight and classifies and labels them based on their severity. This step not only improves the accuracy and efficiency of anomaly detection but also provides strong support for subsequent alert responses and flight safety assessments.

[0115] Step S5 includes the following sub-steps:

[0116] Step S501: Select an alarm level based on the abnormality prediction result:

[0117] If the abnormal prediction result is a confirmed abnormality, an alarm signal with an alarm level of level 1 will be triggered.

[0118] If the abnormal prediction result is a suspected abnormality, the triggering alarm level is an alarm signal that triggers a level 2 alarm.

[0119] If the abnormal prediction result is normal, no alarm is triggered.

[0120] Step S501 selects an alert level based on the anomaly prediction results. This step intelligently triggers the corresponding alert level by accurately determining the anomaly prediction results. A confirmed anomaly triggers a Level 1 alert, indicating an urgent situation requiring immediate action. A suspected anomaly triggers a Level 2 alert, prompting the pilot to pay attention and prepare countermeasures. If the prediction results are normal, no alert is triggered, avoiding unnecessary interference. This step ensures the accuracy and pertinence of alert signals, improving the efficiency of flight safety monitoring.

[0121] Step S502: Combine the warning level and the corresponding real-time flight parameters and transmit them to the flight control terminal.

[0122] Step S502 combines the alert level and corresponding real-time flight parameters and transmits them to the flight control terminal. This step combines alert information with real-time flight parameters to provide pilots with comprehensive flight status information. By transmitting this information to the flight control terminal, pilots can promptly understand any abnormalities during flight and make accurate judgments and decisions based on the alert level and real-time parameters. This step ensures the timely transmission and effective use of information, providing a strong guarantee for flight safety.

[0123] Step S5 intelligently selects the warning level based on the abnormality prediction results, and transmits the warning level and real-time flight parameters to the flight control terminal, thus achieving timely response and effective management of abnormal situations during flight. This step not only improves the accuracy and efficiency of flight safety monitoring,

[0124] The method further comprises:

[0125] Step S601: Verify whether the altitude change rate exceeds the performance envelope in combination with the aircraft dynamics model. The constraints are:

[0126] ;

[0127] in, is the height change rate, is the maximum airspeed, is the maximum angle of attack.

[0128] Step S601 verifies whether the rate of altitude change exceeds the performance envelope, using the aircraft dynamics model. This step utilizes the aircraft dynamics model to precisely calculate and analyze flight parameters, accurately determining whether the rate of altitude change is within a safe range. This verification process allows for the timely detection and correction of potential flight parameter anomalies, ensuring flight safety and stability.

[0129] Step S602: If the verification in step S601 fails, the current detection result is overwritten and an alarm signal with an alarm level of triggering a level 1 alarm is triggered.

[0130] The mathematical expression of the Mahalanobis distance threshold is:

[0131] ;

[0132] in, is the Mahalanobis distance threshold, is the critical value of the chi-square distribution.

[0133] If verification fails in step S601, the current test result is overwritten and a Level 1 alert is triggered. This step immediately initiates the highest-priority alert, alerting the pilot to take urgent action. By overwriting the current test result and triggering a Level 1 alert, pilots are provided with critical information immediately, effectively addressing potential flight safety risks.

[0134] This method establishes an adaptive baseline by calculating dynamic statistical data. This baseline reflects changes in the flight environment and altimeter performance in real time, thereby improving the accuracy and adaptability of anomaly detection. Compared with traditional fixed threshold methods, dynamic baselines better adapt to data changes in different flight phases and environmental conditions, reducing false positives and missed negatives.

[0135] This method combines multiple flight parameters, including historical altitude values, altitude change rate, signal strength, and ambient noise, to extract and analyze spatiotemporal features, achieving multimodal data fusion detection. This multimodal fusion approach provides a more comprehensive reflection of the altimeter's performance, improving the accuracy and reliability of anomaly detection. Furthermore, by constructing a first anomaly detection model and employing advanced technologies such as the local outlier factor algorithm, the intelligent level of anomaly detection is further enhanced.

[0136] This method also incorporates physical validation using an aircraft dynamics model, further confirming the accuracy of anomaly detection results by verifying whether the rate of altitude change exceeds the performance envelope. This physical validation mechanism not only enhances the reliability of anomaly detection but also enables timely identification and correction of potential risks caused by equipment failures or data errors, providing a strong guarantee for flight safety.

[0137] This method selects different levels of warning signals based on the anomaly prediction results and transmits them to the flight control terminal in real time, allowing pilots to quickly understand the altimeter's performance status and take appropriate countermeasures. This real-time response mechanism helps reduce the risk of flight accidents and improve flight safety and reliability.

[0138] Example 2, reference Figure 2 , provides a radio altimeter test data anomaly detection system, including a data analysis module, a real-time acquisition module, a baseline establishment module, a model construction module and an anomaly alarm module.

[0139] The data analysis module is used to obtain historical test data and perform spatiotemporal feature extraction and analysis, and output a first analysis result.

[0140] The data analysis module extracts and analyzes historical test data through in-depth spatiotemporal feature extraction, accurately capturing the changing patterns of flight parameters at different stages. The output of this initial analysis provides critical data support and decision-making basis for subsequent modules, helping to improve the accuracy and reliability of the entire anomaly detection system.

[0141] The real-time acquisition module is used to acquire real-time flight parameters according to the first analysis result and store the real-time flight parameters in a test database.

[0142] Based on the first analysis results output by the data analysis module, the real-time acquisition module intelligently collects real-time flight parameters and stores them in the test database. This step ensures the timeliness and integrity of the data, providing a rich and accurate data source for subsequent anomaly detection.

[0143] The baseline establishment module is used to calculate dynamic statistical data according to the first analysis result, and establish an adaptive baseline based on the dynamic statistical data.

[0144] The baseline establishment module uses the first analysis results to calculate dynamic statistics and establishes an adaptive baseline based on this data. This adaptive baseline can adjust as flight conditions change, more accurately reflecting the current flight status and providing a realistic reference standard for subsequent anomaly detection.

[0145] The model building module is used to build the first anomaly detection model, obtain the real-time flight stage, match the normal and output the anomaly prediction results.

[0146] The model building module constructs the first anomaly detection model and is able to obtain real-time flight phase information, match real-time flight parameters with the model, and output anomaly prediction results. By using advanced algorithms and models, this module can achieve real-time monitoring of flight parameters and anomaly identification, providing a strong guarantee for flight safety.

[0147] The abnormal warning module is used to select the warning level according to the abnormal prediction results and input it to the flight control end.

[0148] The Abnormal Alert Module intelligently selects an alert level based on the abnormality prediction results output by the Model Building Module and transmits the alert information to the flight control terminal. Through timely and accurate alerts, the module alerts pilots to abnormal conditions during flight and enables them to take appropriate measures to ensure flight safety. This module also provides strong support for subsequent flight safety assessments and data analysis.

[0149] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0150] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for detecting anomalies in radio altimeter test data, characterized in that: The steps include: Step S1, obtaining historical test data and performing spatiotemporal feature extraction and analysis, and outputting a first analysis result, including: Step S101, retrieving historical test data from a test database, wherein the historical test data includes flight parameters during takeoff, cruising, and landing phases, and the flight parameters include historical altitude values, historical altitude change rates, historical signal strengths, and historical ambient noise; Step S102, performing quantized wavelet decomposition on the historical test data and outputting a time-frequency energy matrix; Step S103: extracting spatiotemporal features from the time-frequency energy matrix and calculating feature correlation coefficients, outputting a first analysis result, wherein the first analysis result includes a takeoff phase feature correlation coefficient, a cruise phase feature correlation coefficient, and a landing phase feature correlation coefficient; Step S2: collecting real-time flight parameters according to the first analysis result and storing the real-time flight parameters in a test database; Step S3, calculating dynamic statistical data according to the first analysis result, and establishing an adaptive baseline based on the dynamic statistical data; Step S4, constructing a first anomaly detection model, obtaining the real-time flight phase and outputting anomaly prediction results; Step S5: Select an alarm level based on the abnormality prediction result and input it to the flight control terminal.

2. The radio altimeter test data anomaly detection method according to claim 1, wherein: The step S2 includes the following sub-steps: Step S201, parsing the first analysis result: If the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient, or the landing phase characteristic correlation coefficient is less than a preset baseline correlation coefficient, it is determined to be an environmentally sensitive phase, and real-time flight parameters corresponding to the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient, or the landing phase characteristic correlation coefficient are collected and a high-frequency sampling mode is set. The real-time flight parameters include real-time altitude value, altitude change rate, signal strength, and environmental noise; If the takeoff phase characteristic correlation coefficient, the cruise phase characteristic correlation coefficient, or the landing phase characteristic correlation coefficient is greater than or equal to the preset benchmark correlation coefficient, it is determined to be a stable phase and the preset standard sampling frequency is used; Step S202: align the real-time flight parameters by timestamp and store them in the test database.

3. The radio altimeter test data anomaly detection method according to claim 2, wherein: The step S3 includes the following sub-steps: Step S301, calculating dynamic statistical data according to the first analysis result, wherein the dynamic statistical data includes a sliding window mean and a standard deviation; The mathematical expression of the sliding window mean is: ； in, is the sliding window mean, is the sliding window length, For the Real-time flight parameters at all times, is the time point index within the window, is the current time point; The mathematical expression of the standard deviation is: ； in, is the standard deviation; Step S302: establishing an adaptive baseline based on dynamic statistical data.

4. The radio altimeter test data anomaly detection method according to claim 3, wherein: The step S4 includes the following sub-steps: Step S401: construct a first anomaly detection model based on the adaptive baseline, and use a local outlier factor algorithm to calculate the ratio of the local density to the neighboring density of the historical parameters. The mathematical expression of the local outlier factor algorithm is: ； in, for point The local outlier factor, is any data point in the real-time flight parameters, for point The adjacent collection, for point Neighborhood points, is a natural number greater than 0, for point The local reachable density of for point The local reachable density of Step S402: classify the models according to the flight phases, and output a cruise phase model, a landing phase model, and a takeoff phase model; Step S403 , calculate the Mahalanobis distance between the current height value and the time series baseline, analyze the Mahalanobis distance based on the abnormal distance prediction classification logic, and mark it as an abnormal prediction result if the distance exceeds a preset Mahalanobis distance threshold.

5. The radio altimeter test data anomaly detection method according to claim 4, wherein: The abnormal distance prediction classification logic is: If the Mahalanobis distance satisfies the Mahalanobis distance threshold that is greater than or equal to twice, it is marked as confirmed abnormal; If the Mahalanobis distance satisfies the condition that the Mahalanobis distance is greater than or equal to the Mahalanobis distance threshold and less than twice the Mahalanobis distance threshold, it is marked as a suspected anomaly; If the Mahalanobis distance is less than the Mahalanobis distance threshold, it is marked as normal; Confirmed anomalies, suspected anomalies, and normal conditions are combined and output as anomaly prediction results.

6. The radio altimeter test data anomaly detection method according to claim 5, wherein: The step S5 includes the following sub-steps: Step S501: Select an alarm level based on the abnormality prediction result: If the abnormal prediction result is a confirmed abnormality, an alarm signal with a level 1 alarm is triggered; If the abnormal prediction result is a suspected abnormality, the triggering alarm level is a level 2 alarm signal; If the abnormal prediction result is normal, no alarm is triggered; Step S502: Combine the warning level and the corresponding real-time flight parameters and transmit them to the flight control terminal.

7. The radio altimeter test data anomaly detection method according to claim 6, wherein: Also includes: Step S601: Verify whether the altitude change rate exceeds the performance envelope in combination with the aircraft dynamics model. The constraints are: ； in, is the height change rate, is the maximum airspeed, is the maximum angle of attack; Step S602: If the verification in step S601 fails, the current detection result is overwritten and an alarm signal with an alarm level of triggering a level 1 alarm is triggered.

8. The method for detecting abnormality in radio altimeter test data according to claim 7, wherein: The mathematical expression of the Mahalanobis distance threshold is: ； in, is the Mahalanobis distance threshold, is the critical value of the chi-square distribution.

9. A radio altimeter test data anomaly detection system, applied to the radio altimeter test data anomaly detection method according to any one of claims 1 to 8, characterized in that: It includes data analysis module, real-time acquisition module, baseline establishment module, model building module and abnormal alarm module; The data analysis module is used to obtain historical test data and perform spatiotemporal feature extraction and analysis to output a first analysis result, including: retrieving historical test data from a test database, the historical test data including flight parameters for the takeoff phase, the cruise phase, and the landing phase, the flight parameters including historical altitude values, historical altitude change rates, historical signal strengths, and historical environmental noise; performing quantized wavelet decomposition on the historical test data to output a time-frequency energy matrix; performing spatiotemporal feature extraction on the time-frequency energy matrix and calculating feature correlation coefficients to output a first analysis result, the first analysis result including a takeoff phase feature correlation coefficient, a cruise phase feature correlation coefficient, and a landing phase feature correlation coefficient; The real-time acquisition module is used to collect real-time flight parameters according to the first analysis result and store the real-time flight parameters in a test database; The baseline establishment module is used to calculate dynamic statistical data according to the first analysis result, and establish an adaptive baseline based on the dynamic statistical data; The model building module is used to build a first anomaly detection model, obtain the real-time flight stage, match the normal and output the anomaly prediction result; The abnormality warning module is used to select the warning level according to the abnormality prediction result and input it to the flight control end.

Images