Fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics

Abstract

This invention relates to an early fault warning method for fixed-wing aircraft based on the physical coupling and correlation characteristics of multiple parameters. The method includes: dividing flight parameter data into coupled parameter groups with strong causal or adjoint relationships according to the physical working principles of each subsystem of the fixed-wing aircraft; filtering fault-free flight sorties from a historical database and extracting flight parameter data from steady-state level flight cruise segments; constructing a health baseline and correlation envelope model based on the coupled parameter groups in the flight parameter data; acquiring real-time flight parameter data when the aircraft is in steady-state level flight cruise condition, and calculating the correlation deviation based on the health baseline and correlation envelope model; determining whether the physical correlation relationship of the coupled parameter groups has broken based on the correlation deviation, and triggering an early warning when a break occurs. This invention enables accurate early fault warning.

Description

Technical Field

[0001] This invention relates to the field of fault prediction and health management (PHM) technology, and in particular to an early fault warning method for fixed-wing aircraft based on the multi-parameter physical coupling correlation characteristics. Background Technology

[0002] During long-term operation, the core components of fixed-wing aircraft, such as the power system and flight control system, inevitably experience performance degradation. Current flight parameter monitoring primarily relies on single-parameter threshold criteria, triggering an alarm when a single parameter, such as engine speed, temperature, or vibration, exceeds a preset alarm threshold. However, this traditional monitoring method has the following significant drawbacks:

[0003] Unable to provide early warning of hidden faults: In the early stages of faults (such as minor engine air leakage, decreased combustion efficiency, or gradual sensor drift), the changes in individual parameters are still within the normal threshold range, and the system cannot provide timely warnings until the fault worsens to the point of exceeding the limit, thus losing the window of opportunity for early intervention.

[0004] Isolating the physical logic between parameters: Aircraft subsystems have strong physical coupling characteristics. For example, an increase in throttle command will inevitably lead to an increase in engine speed, which in turn leads to an increase in exhaust temperature. Monitoring a single parameter in isolation completely ignores the inherent causal relationships between parameters and loses key coupling information reflecting the health status of the system.

[0005] Existing algorithms lack interpretability: Although some deep learning-based multivariate anomaly detection methods can process multiple parameters simultaneously, their black-box nature means they can only output anomaly scores and cannot explain to maintenance personnel which set of physical relationships broke and the direction of the break, making it difficult to directly guide troubleshooting; at the same time, such methods consume a lot of computational resources and are not easy to deploy on airborne embedded devices. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide an early fault warning method for fixed-wing aircraft based on the multi-parameter physical coupling correlation characteristics, which can achieve accurate early fault warning.

[0007] The technical solution adopted by this invention to solve its technical problem is: to provide an early fault warning method for fixed-wing aircraft based on multi-parameter physical coupling correlation characteristics, comprising the following steps:

[0008] Based on the physical working principles of each subsystem of a fixed-wing aircraft, flight parameter data are divided into coupled parameter groups with strong causal or adjoint relationships.

[0009] Fault-free flight sorties are selected from historical databases, and flight parameter data of steady-state level flight cruise segments are extracted. Based on the coupled parameter groups in the flight parameter data, a model based on a healthy baseline and associated envelope is constructed.

[0010] When the aircraft is in steady-state level flight cruise, real-time flight parameter data is acquired, and the correlation deviation is calculated based on the health baseline and correlation envelope model.

[0011] The physical relationship of the coupling parameter group is determined based on the correlation deviation, and an early warning is triggered when a break occurs.

[0012] The criteria for intercepting flight parameter data during steady-state cruising are: the absolute value of the aircraft's altitude change rate is less than a preset altitude change rate threshold, and the absolute value of the pitch angle is less than a preset angle threshold.

[0013] The construction of a health baseline and correlation envelope model based on the coupled parameter set in the flight parameter data specifically involves: using the independent variables in the coupled parameter set as the x-axis and the dependent variables as the y-axis, fitting a central correlation line under healthy conditions; and calculating the longitudinal residuals and residual standard deviations from each healthy data point to the central correlation line. Using the central correlation line as a reference, shift upwards and downwards. This constitutes a health-related envelope, in which, This is the preset scaling factor.

[0014] The calculation of the correlation deviation based on the health baseline and correlation envelope model specifically includes:

[0015] Extract the real-time coupling parameter set from the real-time flight parameter data;

[0016] The independent variables in the real-time coupling parameter set are substituted into the health baseline and correlation envelope model to obtain the calculation results, and the difference between the dependent variable in the real-time coupling parameter set and the calculation results is calculated to obtain the real-time residual.

[0017] Determine whether the absolute value of the real-time residual is greater than the shift of the health-related envelope. ;

[0018] If the absolute value of the real-time residual is greater than the shift of the health-related envelope... Based on the absolute value of the real-time residual and the shift of the health-related envelope band. Calculate the correlation deviation and record the sign of the real-time residual.

[0019] The correlation deviation is calculated as follows: ,in, For correlation deviation, For real-time residuals, The amount of translation of the health-related envelope.

[0020] The step of determining whether the physical association relationship of the coupling parameter group has broken based on the association deviation specifically includes:

[0021] Set the number of continuous cruise condition sampling points included in the sliding time window and the threshold for the proportion of abnormal points;

[0022] Within the current time window, count the number of data points with a correlation deviation greater than zero;

[0023] Determine whether the ratio of the number of data points with a correlation deviation greater than zero to the number of continuous cruise condition sampling points included in the sliding time window exceeds the outlier ratio threshold.

[0024] If the ratio of the number of data points with a correlation deviation greater than zero to the number of continuous cruise condition sampling points included in the sliding time window exceeds the outlier ratio threshold, it is determined that the physical correlation of the coupled parameter group has broken.

[0025] After triggering the warning upon the occurrence of a rupture, the system also includes:

[0026] The sign of the real-time residuals of data points with a correlation deviation greater than zero within the current time window is statistically analyzed, and the physical attribution of the fault is determined based on the proportion of the signs of the real-time residuals.

[0027] The technical solution adopted by the present invention to solve its technical problem is: to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the steps of the above-mentioned method for early fault warning of fixed-wing aircraft based on multi-parameter physical coupling correlation characteristics.

[0028] The technical solution adopted by the present invention to solve its technical problem is: to provide a computer-readable storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the above-mentioned method for early fault warning of fixed-wing aircraft based on multi-parameter physical coupling correlation characteristics are implemented.

[0029] Beneficial effects

[0030] By employing the above-mentioned technical solutions, this invention has the following advantages and positive effects compared with existing technologies: This invention breaks through the limitations of traditional single-parameter threshold monitoring. Even when the absolute values ​​of each parameter have not reached the alarm threshold, it identifies the breakdown of the physical correlation between coupled parameter groups, thus capturing signs of system performance degradation in advance and providing sufficient time for condition-based maintenance. This invention divides parameter groups based on the physical logic of aircraft subsystems. The early warning results correspond to clear correlation deviation directions, directly locating the fault mechanism and component range, avoiding the shortcomings of black-box algorithms in guiding troubleshooting. The core computation of this invention only includes data fitting and residual threshold determination, consuming extremely low computational resources and allowing for convenient deployment on airborne embedded devices and ground monitoring terminals. This invention relies on steady-state condition screening and correlation envelope determination mechanisms to effectively filter out transient condition interference, significantly reducing the false alarm rate. Attached Figure Description

[0031] Figure 1 This is a flowchart of the first embodiment of the fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics.

[0032] Figure 2 This is a schematic diagram of the health baseline and associated envelope model in the first embodiment of the present invention;

[0033] Figure 3 This is a schematic diagram of real-time correlation deviation calculation and early fault capture in the first embodiment of the present invention. Detailed Implementation

[0034] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0035] The first embodiment of the present invention relates to an early fault warning method for fixed-wing aircraft based on multi-parameter physical coupling correlation characteristics. This embodiment takes a piston engine mounted on a certain type of fixed-wing aircraft as an example. Figure 1 As shown, the specific steps include:

[0036] Step 1: Based on the physical operating principles of each subsystem of the fixed-wing aircraft, the flight parameter data is divided into coupled parameter groups with strong causal or adjoint relationships. In this step, the engine speed is confirmed. With exhaust temperature It is a set of coupled parameters with a strong physical causal relationship. The physical logic between the two is: under steady-state cruise conditions, the higher the speed, the higher the exhaust temperature will inevitably be, showing a positive correlation.

[0037] Step 2: Filter out fault-free flight sorties from the historical database and extract flight parameter data for the steady-state level flight cruise segment. Construct a health baseline and associated envelope model based on the coupled parameter group in the flight parameter data.

[0038] In this step, data from the first 50 fault-free sorties of the aircraft are selected. Based on cruise operating condition criteria (absolute altitude change rate < 0.5 m / s, absolute pitch angle < 5°), the level flight segment is extracted, and 5000 valid data points are obtained. .

[0039] like Figure 2 As shown, the 5000 extracted valid data pairs By performing least squares fitting, the correlation line between health centers was obtained: .

[0040] Calculate and statistically analyze the longitudinal residuals from each data point to the central correlation line to obtain the standard deviation of the residuals. Take the preset ratio coefficient. Then the translation amount Therefore, the upper envelope is determined to be: The lower envelope is: .

[0041] Step 3: When the aircraft is in steady-state level flight cruise, acquire real-time flight parameter data and calculate the correlation deviation based on the health baseline and correlation envelope model.

[0042] During a real-time mission, the aircraft entered the cruise phase. When the engine speed... At that time, the expected exhaust temperature is calculated based on the healthy baseline and the associated envelope model: Its envelope range is .exist At what time, the measured exhaust temperature The difference between the temperature and the desired exhaust temperature is calculated to obtain the real-time residual. At -10℃, the absolute value of this real-time residual is less than the translation amount. This indicates that it is within the envelope range, therefore the correlation deviation is... .exist At what time, the measured exhaust temperature The difference between the temperature and the desired exhaust temperature is calculated to obtain the real-time residual. At -120℃, the absolute value of this real-time residual is greater than the translation amount. This indicates that it is outside the envelope zone, and at this time according to Calculate the correlation deviation to obtain =300%, and record the real-time residual. The symbol, that is This indicates that the exhaust temperature is too low.

[0043] Step 4: Determine whether the physical relationship of the coupling parameter group has been broken based on the correlation deviation, and trigger an early warning when a break occurs.

[0044] This step first sets the number of continuous cruise condition sampling points included in the sliding time window and the outlier ratio threshold; then, within the current time window, it counts the number of data points with a correlation deviation greater than zero; finally, it determines whether the ratio of the number of data points with a correlation deviation greater than zero to the number of continuous cruise condition sampling points included in the sliding time window exceeds the outlier ratio threshold.

[0045] For example, the number of continuous cruise condition sampling points included in the sliding time window is set to... That is, the length of the sliding window is M=30 points, and an abnormal proportion threshold is set. When in Of the 30 consecutive sampling points after time step 1, 20 sampling points satisfy the condition. Furthermore, the signs of the real-time residuals at all 20 sampling points are -1. At this point, the ratio of the abnormal sampling points to the length of the sliding window is calculated as 20 / 30 = 66.7%, which is greater than the abnormality ratio threshold. The system determines that the physical correlation between engine speed and exhaust temperature has broken down, triggering an early warning.

[0046] After triggering the warning, this implementation method can also determine the physical cause of the fault based on the proportion of the signs of the real-time residuals. When the signs of the vast majority of abnormal points are -1, it indicates that the exhaust temperature is systematically low, which is attributed to decreased combustion efficiency, exhaust manifold leakage, or poor exhaust valve sealing. When the signs of the vast majority of abnormal points are +1, it indicates that the exhaust temperature is systematically high, which is attributed to early knocking, rich combustion chamber, or abnormal ignition advance angle. In this implementation method, since the signs of the real-time residuals of the 20 sampling points are all -1, the fault is attributed to decreased combustion efficiency, exhaust manifold leakage, or poor exhaust valve sealing. Subsequent inspection confirmed a minor crack in the exhaust manifold, consistent with the warning. Throughout the entire process, the engine speed and absolute temperature did not reach the red line, and the warning was triggered several flights in advance.

[0047] It is worth mentioning that for coupled parameter sets exhibiting nonlinear characteristics, a quadratic polynomial can be used to fit the central correlation line, and the envelope definition and deviation calculation logic are completely consistent with the linear implementation.

[0048] It is easy to see that this invention breaks through the limitations of traditional single-parameter threshold monitoring. Even when the absolute values ​​of each parameter have not reached the alarm threshold, it identifies the breakdown of the physical correlation between coupled parameter groups, thus capturing signs of system performance degradation in advance and providing sufficient time for condition-based maintenance. This invention divides parameter groups based on the physical logic of aircraft subsystems, and the early warning results correspond to clear correlation deviation directions, directly locating the fault mechanism and component range, avoiding the shortcomings of black-box algorithms in guiding troubleshooting. The core calculations of this invention only involve data fitting and residual threshold determination, consuming extremely low computational resources, and can be easily deployed on airborne embedded devices and ground monitoring terminals. Based on steady-state condition screening and correlation envelope determination mechanisms, this invention effectively filters out transient condition interference, significantly reducing the false alarm rate.

[0049] The second embodiment of the present invention relates to an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics of the first embodiment.

[0050] The third embodiment of the present invention relates to a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics of the first embodiment.

[0051] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0052] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0053] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction methods implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0054] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0055] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics, characterized in that, The method comprises the following steps: According to the physical working principle of each subsystem of the fixed-wing aircraft, the flight parameter data is divided into a coupled parameter group with strong causal or accompanying relationship; From the historical database, the flight parameter data of the stable steady flight cruise section is intercepted, and a health baseline and correlation envelope model is constructed based on the coupled parameter group in the flight parameter data; When the aircraft is in a stable steady flight cruise condition, real-time flight parameter data is obtained, and a correlation deviation is calculated according to the health baseline and correlation envelope model; According to the correlation deviation, it is judged whether the physical correlation relationship of the coupled parameter group is broken, and a warning is triggered when the breakage occurs.

2. The fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics according to claim 1, characterized in that, When the flight parameter data of the stable steady flight cruise section is intercepted, the interception criterion is that the absolute value of the height change rate of the aircraft is less than a preset height change rate threshold, and the absolute value of the pitch angle is less than a preset angle threshold.

3. The fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics according to claim 1, characterized in that, The health baseline and correlation envelope model is constructed based on the coupled parameter group in the flight parameter data, specifically: the independent variable in the coupled parameter group is taken as the horizontal coordinate, and the dependent variable is taken as the vertical coordinate, and a central correlation line in the healthy state is fitted by using data; calculating longitudinal residuals of each health data point to the center line of association and a standard deviation of residuals ; upwardly and downwardly from the center correlation line constitute a healthy correlation envelope band, wherein is a preset proportionality coefficient.

4. The fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics according to claim 1, characterized in that, The correlation deviation is calculated according to the health baseline and correlation envelope model, specifically including: The real-time coupled parameter group is extracted from the real-time flight parameter data; The independent variable in the real-time coupled parameter group is brought into the health baseline and correlation envelope model to obtain a calculation result, and the difference between the dependent variable in the real-time coupled parameter group and the calculation result is calculated to obtain a real-time residual error; determining whether the absolute value of the real-time residual is greater than a shift amount of a healthy correlation envelope band ; if the absolute value of the real-time residue is greater than a shift of the healthy correlation envelope band calculating a correlation deviation from the absolute value of the real-time residue and a shift of the healthy correlation envelope band and recording the sign of the real-time residue.

5. The fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics according to claim 4, characterized in that, The correlation deviation is calculated in the following manner: wherein, is the correlation deviation, is the real-time residual, is the amount of translation of the healthy correlation envelope band.

6. The fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics according to claim 1, characterized in that, According to the correlation deviation, it is judged whether the physical correlation relationship of the coupled parameter group is broken, specifically including: Set the number of continuous cruise condition sampling points contained in the sliding time window and the abnormal point proportion threshold; In the current time window, the number of data points with correlation deviation greater than zero is counted; It is judged whether the ratio of the number of data points with correlation deviation greater than zero to the number of continuous cruise condition sampling points contained in the sliding time window exceeds the abnormal point proportion threshold; If the ratio of the number of data points with correlation deviation greater than zero to the number of continuous cruise condition sampling points contained in the sliding time window exceeds the abnormal point proportion threshold, it is determined that the physical correlation relationship of the coupled parameter group is broken.

7. The fixed-wing aircraft early fault warning method based on multi-parameter physical coupling correlation characteristics according to claim 1, characterized in that, After triggering the warning when the breakage occurs, it further includes: The signs of the real-time residual errors of the data points with correlation deviation greater than zero in the current time window are counted, and the physical attribution of the fault is determined according to the proportion of the signs of the real-time residual errors.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the early fault warning method of the fixed-wing aircraft based on the physical coupling correlation characteristics of multiple parameters according to any one of claims 1-7.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the early fault warning method of the fixed-wing aircraft based on the physical coupling correlation characteristics of multiple parameters according to any one of claims 1-7.

Images