Health monitoring method and system for angle of attack sensor blade sticking

Abstract

A health monitoring method and system for angle of attack sensor blade sticking, wherein the health monitoring method for angle of attack sensor blade sticking comprises: obtaining angle of attack detection data of historical flights, and obtaining an angle of attack threshold interval according to the angle of attack detection data in the cruise stage; obtaining angle of attack detection data of a first angle of attack sensor, a second angle of attack sensor and a third angle of attack sensor of a to-be-detected aircraft in the take-off stage, calculating the difference between the angle of attack detection data recorded by the three sensors at the same time in the take-off stage, and obtaining a 1-2 difference group, a 1-3 difference group and a 2-3 difference group; fitting the probability density functions of the above three difference groups respectively, and obtaining a 1-2 offset, a 1-3 offset and a 2-3 offset; calculating the difference between the above three offsets and the three offsets of the angle of attack sensor of a normal aircraft one by one, judging the health degree of the angle of attack sensor according to the difference, realizing long-term monitoring of the health condition of the angle of attack sensor blade, ensuring the flight safety of the aircraft and improving the aircraft maintenance efficiency.

Description

Technical Field

[0001] This invention relates to the field of aircraft anomaly detection, and in particular to the field of angle-of-attack sensor health monitoring methods and systems. Background Technology

[0002] Please see Figure 1 , Figure 1 This diagram illustrates the angle of attack (AOA) of an aircraft. For fixed-wing aircraft, the angle between the wing's direction relative to the airflow and the wing chord (different from the fuselage axis) is called the angle of attack (α angle), which is the reference for determining the wing's angle of attack in the airflow. Aircraft use angle-of-attack sensors to detect the ADIRU (Aircraft Design Utility Unit) and send the results to the flight control computer. Essentially, the angle of attack is used to calculate the aircraft's lift, thereby controlling the aircraft's safe flight. Since the angle of attack is fundamental to flight attitude, flight control, and lift, civil aircraft have instantaneous angle-of-attack calculations and monitoring for angle-of-attack sensors. However, there is a lack of long-term health monitoring for the angle-of-attack blades themselves. During flight, the aircraft adjusts its flight angle based on the angle of attack detected by the angle-of-attack sensor. When the angle-of-attack sensor malfunctions, it may cause unexpected aircraft grounding, lead to flight control difficulties, and even, in severe cases, cause aircraft to stop completely. Summary of the Invention

[0003] To address the aforementioned issues, this invention first utilizes big data to determine the normal angle-of-attack threshold range of the aircraft's angle-of-attack sensors. Then, it performs difference calculations on the detection data of the three angle-of-attack sensors of the aircraft under test during takeoff, obtaining three sets of difference values. These three sets are then fitted with probability density functions to obtain their distributions. Based on the distribution of differences in angle-of-attack detection data among the angle-of-attack sensors during takeoff, the operating status of the angle-of-attack sensors is determined. Ultimately, this enables long-term health monitoring of the angle-of-attack blades themselves, ensuring flight safety and reducing aircraft maintenance complexity.

[0004] This invention is achieved through the following technical solution:

[0005] On the one hand, this invention provides a health monitoring method for blade jamming in an angle-of-attack sensor, which includes:

[0006] Obtain historical angle of attack detection data and determine the angle of attack threshold range based on the historical angle of attack detection data during the cruise phase;

[0007] Acquire the angle of attack detection data of the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor of the aircraft under test during the takeoff phase, calculate the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase, and obtain the 1-2 difference group, the 1-3 difference group, and the 2-3 difference group.

[0008] The probability density function was fitted to the above three difference groups respectively to obtain the 1-2 offset, 1-3 offset, and 2-3 offset;

[0009] The three offsets mentioned above are compared one-to-one with the 1-2 offset, 1-3 offset, and 2-3 offset of the angle of attack sensor of a normal aircraft. The difference is used to determine the health status of the angle of attack sensor.

[0010] Furthermore, the health monitoring method for angle-of-attack sensor blade jamming provided by the present invention also includes:

[0011] The acquired angle-of-attack detection data is decoded and scaled in advance to unify the recording time interval of all angle-of-attack detection data.

[0012] Furthermore, after acquiring the angle of attack detection data of the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor during the takeoff phase of a single flight, the process also includes:

[0013] The number of all parameters in the angle of attack detection data that are not in the angle of attack threshold range is calculated. If the number of parameters that are not in the angle of attack threshold range exceeds the threshold, the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase is calculated.

[0014] Furthermore, the probability density function is specifically as follows:

[0015]

[0016] Where x is a single angle of attack detection value, f(x) is the total number of data with angle of attack detection value x, u is the mean, and σ is the variance.

[0017] On the other hand, the present invention also provides a health monitoring system for angle-of-attack sensor blade jamming, comprising:

[0018] Threshold range generation module: used to obtain angle of attack detection data of historical flights and obtain angle of attack threshold range based on the angle of attack detection data of historical flights during the cruise phase;

[0019] Angle of attack detection data difference calculation module: used to acquire the angle of attack detection data of the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor of the aircraft under test during the takeoff phase, calculate the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase, and obtain the 1-2 difference group, the 1-3 difference group, and the 2-3 difference group.

[0020] Angle of attack detection data offset calculation module: used to fit the probability density function to the above three difference groups respectively to obtain offset 1-2, offset 1-3, and offset 2-3;

[0021] Angle of attack sensor health assessment module: This module calculates the difference between the three offsets mentioned above and the 1-2, 1-3, and 2-3 offsets of the angle of attack sensor of a normal aircraft, and determines the health status of the angle of attack sensor based on the difference.

[0022] Furthermore, the health monitoring system for angle-of-attack sensor blade jamming also includes:

[0023] Data preprocessing module: used to decode and scale the acquired angle of attack detection data in advance, and to unify the recording time interval of all angle of attack detection data.

[0024] Furthermore, the angle of attack detection data difference calculation module also includes:

[0025] Fault prediction submodule: Used to calculate the number of all parameters in the angle of attack detection data that are not in the angle of attack threshold range. If the number of parameters that are not in the angle of attack threshold range exceeds the threshold, the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase is calculated.

[0026] On the other hand, the present invention also provides a computer device, the computer device including at least one memory and at least one processor;

[0027] The memory is used to store one or more programs; when the one or more programs are executed by the at least one processor, the at least one processor implements the steps of a health monitoring method for angle-of-attack sensor blade jamming as described in any of the preceding claims.

[0028] On the other hand, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a health monitoring method for angle-of-attack sensor blade jamming as described in any of the preceding claims.

[0029] In summary, the present invention provides a health monitoring method for angle-of-attack sensor blade jamming. Based on an angle-of-attack sensor health monitoring system, it first acquires angle-of-attack detection data for each historical flight of the target aircraft model. Based on the angle-of-attack detection data during the cruise phase, a normal angle-of-attack threshold range is determined. Then, angle-of-attack detection data from three angle-of-attack sensors detected during the takeoff phase of the aircraft under test are acquired. The number of data points exceeding the angle-of-attack threshold range is used to determine if an anomaly is possible. If an anomaly is suspected, the angle-of-attack detection data recorded by the three angle-of-attack sensors at the same time are subtracted pairwise to obtain three sets of difference values. The data from these three sets of difference values ​​are then subjected to three probability distribution fittings to find the distribution pattern of the difference values ​​and obtain three offsets. These three offsets are compared with pre-stored standard offsets. When the difference between two of the offsets and the standard offset exceeds a preset threshold, the angle-of-attack sensor involved in both offset calculations is identified, and a replacement of that sensor is indicated. This method effectively monitors the long-term health status of angle-of-attack sensor blades, ensuring aircraft flight safety and improving aircraft maintenance efficiency.

[0030] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description

[0031] Figure 1 This is a diagram illustrating the angle of attack of an aircraft.

[0032] Figure 2 This is a flowchart of a health monitoring method for an angle-of-attack sensor according to the present invention;

[0033] Figure 3 To execute Figure 2 A structural block diagram of a health monitoring system for angle-of-attack blade jamming;

[0034] Figure 4 The present invention describes the change of the angle of attack of an aircraft during takeoff over time. Detailed Implementation

[0035] Addressing the issue of angle-of-attack (AOA) sensor jamming in aircraft, the inventors conducted in-depth research into the causes of this jamming. First, friction tests were performed on the faulty AOA sensors, revealing that the friction force at all measuring points exceeded the standard. Further disassembly of the faulty AOA sensors revealed significant jamming during outer bearing rotation, and the inner bearing also exhibited jamming. Disassembly of the bearing dust cover revealed that the internal grease was dried and blackened. Cleaning revealed substantial deposits. Further disassembly and inspection of the disassembled components revealed a large amount of slowly accumulated deposits. These deposits affected bearing rotation, leading to AOA sensor jamming. When these deposits accumulated to the point of affecting normal aircraft flight, the AOA sensor needed replacement or repair. However, during aircraft maintenance, AOA sensors cannot be replaced every time or at fixed intervals; instead, they are only replaced after a malfunction occurs. Furthermore, when an AOA sensor malfunctions, it requires immediate grounding and emergency ordering of replacement parts, posing significant risks to flight safety and aircraft maintenance. To address the aforementioned issues, we urgently need a method for long-term monitoring of the health status of angle-of-attack sensors.

[0036] Based on this, the present invention provides a health monitoring method and system for an angle-of-attack sensor. Please refer to [link / reference]. Figure 2 and Figure 3 , Figure 2 This is a flowchart of a health monitoring method for an angle-of-attack sensor according to the present invention. Figure 3 To execute Figure 2 The structural block diagram of a health monitoring system for angle-of-attack blade jamming is shown. Specifically, the health monitoring method for an angle-of-attack sensor described in this invention includes:

[0037] S10: Obtain the angle of attack detection data of historical flights, and obtain the angle of attack threshold range based on the angle of attack detection data of historical flights during the cruise phase. Step S10 is executed by the threshold range generation module 10.

[0038] An aircraft's flight mission can generally be divided into three phases: takeoff, cruise, and landing. During the cruise phase, the aircraft moves steadily in the horizontal direction with minimal longitudinal displacement, thus it can be considered a stable operational process. Ideally, the angle of attack should be 0° during cruise. However, due to aerodynamic wave interference, boundary layer effects, and the influence of airflow from the fuselage or other external components, the angle of attack may fluctuate around 0°. Therefore, a normal angle of attack threshold range needs to be determined based on the angle of attack detection data during the cruise phase.

[0039] In another embodiment, the acquired angle-of-attack detection data is pre-decoded and scale-transformed to standardize the recording time interval of all angle-of-attack detection data. This step is performed by the data preprocessing module.

[0040] Existing aircraft crews collect data from various aircraft systems via FDIMU (Flight Data Interface Management Component) and transmit the data to DAR (Digital Integrated Data Recorder) and QAR (Flight Recorder). However, the data recorded by different aircraft crews is not uniform in format, differing in data type and byte length. For example, there are format differences of 1024 / 512 / 256 WPS for byte length. Finally, decoding software converts the data into a readable data type. Therefore, during data monitoring, it is necessary to first determine the aircraft model and the configuration of the aircraft's storage devices (FDIMU / DAR / QAR) and decode them according to different scenarios to reconstruct the recorded parameters for a flight segment. This data preprocessing affects whether subsequent steps are performed on a frame-per-second or multi-frame-per-second basis, but the inconsistent data format can easily affect the final result. Therefore, this invention pre-decodes and performs scale transformation on the data, unifying the format of all data and standardizing the recording time interval for angle-of-attack detection data to maintain data dimension consistency and ensure calculation accuracy. According to the data monitoring requirements, the acquisition time interval of the final angle-of-attack data must be converted to the second level.

[0041] S20: Acquire the angle of attack detection data from the first, second, and third angle of attack sensors of the aircraft under test during the takeoff phase. Calculate the pairwise differences between the angle of attack detection data recorded by the first, second, and third angle of attack sensors at the same time during the takeoff phase, obtaining difference groups 1-2, 1-3, and 2-3. Step S20 is executed by the angle of attack detection data difference calculation module.

[0042] During an aircraft's cruise, both takeoff and landing phases involve climb and descent. This means that takeoff and landing involve significant changes in angle of attack. The aircraft's climb and descent are adjusted based on these changes in angle of attack; therefore, the angle of attack detection data is dynamically changing during takeoff and landing. Please refer to [link / reference]. Figure 4 , Figure 4The change process of the angle of attack during the takeoff phase of an aircraft, as provided by this invention, shows that the takeoff process can be decomposed into a linear static part and a nonlinear part, corresponding to the process of the aircraft accelerating from taxiing to climbing, the nose accelerating from a stationary state, horizontally lifting, and the aircraft taking off. In this process, the initial angle of attack data will exceed the angle of attack threshold range. However, as the aircraft climbs, a normal angle of attack sensor will quickly reduce the output angle of attack detection value back to the angle of attack threshold range based on the aircraft's trajectory angle and velocity vector. However, due to the blade jamming of a faulty angle of attack sensor, the time for the output angle of attack detection value to return to the angle of attack threshold range is longer than that of a normal angle of attack sensor. However, due to different flight settings and routes, the flight altitude of the same aircraft may change, resulting in climb time. Simply looking at the time difference between the angle of attack detection value output by the angle of attack sensor and the angle of attack threshold range is not enough to accurately determine whether the angle of attack sensor is malfunctioning. Based on this, the inventors further analyzed that, considering that an aircraft is equipped with three angle of attack sensors, and that the angle of attack detection values ​​of each angle of attack sensor should be consistent or have very small differences under normal circumstances, the difference between the angle of attack detection data of the three angle of attack sensors can be used as the judgment parameter, thus ignoring the influence of different flight settings and routes on the judgment result.

[0043] In another embodiment, after acquiring the angle of attack detection data of the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor during the takeoff phase of a single flight, the method further includes:

[0044] The number of parameters outside the angle-of-attack threshold range in the angle-of-attack detection data is calculated. If the number of parameters outside the threshold range exceeds the threshold, the differences between each pair of angle-of-attack detection data recorded by the first, second, and third angle-of-attack sensors at the same time during takeoff are calculated, resulting in difference groups 1-2, 1-3, and 2-3. This judgment process is executed by the fault pre-judgment submodule.

[0045] Pre-screening flight data is performed because the angle of attack detection value output by a faulty angle of attack sensor takes longer to return to the angle of attack threshold range compared to a normal angle of attack sensor, resulting in a larger number of recorded parameters. Therefore, the number of all parameters outside the angle of attack threshold range can be counted to preliminarily determine whether the angle of attack sensor may be faulty. If the number of recorded parameters does not exceed the set threshold, it can be assumed that no jamming has occurred, thus saving computing resources.

[0046] S30: Perform probability density function fitting on the above three difference groups respectively to obtain the 1-2 offset, 1-3 offset, and 2-3 offset. Step S30 is executed by the angle of attack detection data offset calculation module 30.

[0047] The probability density function is:

[0048]

[0049] Where x is a single angle of attack detection value, f(x) is the total number of data points with angle of attack detection value x, u is the mean, and σ is the variance. Using a fitting tool, the mean value corresponding to each difference group is calculated as the offset. On the image, μ can show which regions the data in a set of data are concentrated. Therefore, the data distribution in the three difference groups is calculated to show the degree of deviation between the data detected by each angle of attack sensor and the data detected by other angle of attack sensors.

[0050] S40: Compare the above three offsets with the 1-2, 1-3, and 2-3 offsets of a normal aircraft's angle of attack sensor, and calculate the difference. Determine the health status of the angle of attack sensor based on the difference. Step S40 is executed by the angle of attack sensor health assessment module.

[0051] Using the 1-2, 1-3, and 2-3 offsets of a normal aircraft's angle-of-attack (AOA) sensor as a standard reference, different health ranges are preset. Based on the difference between the 1-2, 1-3, and 2-3 offsets of the aircraft under test and the standard 1-2, 1-3, and 2-3 offsets, the health level of each AOA sensor of the aircraft under test is determined. For example, we define the offset value in the range [-2.5, 2.5] as relatively good, belonging to the excellent zone (range A), where the AOA can quickly move to the correct position with the airflow; [-7.5, -2.5] and [2.5, 7.5] are the good zone (range B), where the AOA can quickly move to the correct position with the airflow; [-10, -7.5] and [7.5, 10] are the decay zone (range C), where there is already a certain degree of stagnation, and values ​​greater than 10 or less than -10 are replacement zones, requiring replacement. When the first angle-of-attack sensor malfunctions while the others are functioning normally, if the offset of 1-2 is greater than 10, then the offset of 1-3 should also be greater than 10, and the offset of 2-3 should be between [-2.5, 2.5]. This process ultimately monitors and determines the health status of each angle-of-attack sensor.

[0052] In summary, the present invention provides a health monitoring method for angle-of-attack sensor blade jamming. Based on an angle-of-attack sensor health monitoring system, it first acquires angle-of-attack detection data for each historical flight of the target aircraft model. Based on the angle-of-attack detection data during the cruise phase, a normal angle-of-attack threshold range is determined. Then, angle-of-attack detection data from three angle-of-attack sensors detected during the takeoff phase of the aircraft under test are acquired. The number of data points exceeding the angle-of-attack threshold range is used to determine if an anomaly is possible. If an anomaly is suspected, the angle-of-attack detection data recorded by the three angle-of-attack sensors at the same time are subtracted pairwise to obtain three sets of difference values. The data from these three sets of difference values ​​are then subjected to three probability distribution fittings to find the distribution pattern of the difference values ​​and obtain three offsets. These three offsets are compared with pre-stored standard offsets. When the difference between two of the offsets and the standard offset exceeds a preset threshold, the angle-of-attack sensor involved in both offset calculations is identified, and a replacement of that sensor is indicated. This method effectively monitors the health status of angle-of-attack sensor blades, improving aircraft maintenance efficiency.

[0053] Based on the same inventive concept described above, this invention also provides an electronic device, which can be a server, desktop computing device, or mobile computing device (e.g., laptop computing device, handheld computing device, tablet computer, netbook, etc.). This device includes one or more processors and a memory, wherein the processor executes a program to implement the aforementioned health monitoring method for angle-of-attack sensor blade jamming; the memory stores a computer program executable by the processor.

[0054] Based on the same inventive concept, the present invention also provides a computer-readable storage medium corresponding to the aforementioned embodiment of a health monitoring method for angle-of-attack sensor blade jamming. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps described in any of the above embodiments.

[0055] This invention can take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to: phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0056] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and the present invention also intends to include these modifications and variations.

Claims

1. A health monitoring method for blade jamming in an angle-of-attack sensor, comprising the following steps: Obtain historical angle of attack detection data and determine the angle of attack threshold range based on the historical angle of attack detection data during the cruise phase; Acquire the angle of attack detection data of the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor of the aircraft under test during the takeoff phase, calculate the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase, and obtain the 1-2 difference group, the 1-3 difference group, and the 2-3 difference group. The probability density function was fitted to the above three difference groups respectively to obtain the 1-2 offset, 1-3 offset, and 2-3 offset; The three offsets mentioned above are compared one-to-one with the 1-2 offset, 1-3 offset, and 2-3 offset of the angle of attack sensor of a normal aircraft. The difference is used to determine the health status of the angle of attack sensor.

2. The health monitoring method for blade jamming in an angle-of-attack sensor according to claim 1, characterized in that, Also includes: The acquired angle-of-attack detection data is decoded and scaled in advance to unify the recording time interval of all angle-of-attack detection data.

3. The health monitoring method for blade jamming in an angle-of-attack sensor according to claim 2, characterized in that, After acquiring the angle of attack detection data of the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor during the takeoff phase of a single flight, the following is also included: The number of all parameters in the angle of attack detection data that are not in the angle of attack threshold range is calculated. If the number of parameters that are not in the angle of attack threshold range exceeds the threshold, the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase is calculated.

4. The health monitoring method for blade jamming in an angle-of-attack sensor according to claim 3, characterized in that, The probability density function is as follows: Where x is a single angle of attack detection value, f(x) is the total number of data with angle of attack detection value x, u is the mean, and σ is the variance.

5. A health monitoring system for blade jamming in an angle-of-attack sensor, characterized in that, include: Threshold range generation module: used to obtain angle of attack detection data of historical flights and obtain angle of attack threshold range based on the angle of attack detection data of historical flights during the cruise phase; Angle of attack detection data difference calculation module: used to acquire the angle of attack detection data of the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor of the aircraft under test during the takeoff phase, calculate the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase, and obtain the 1-2 difference group, the 1-3 difference group, and the 2-3 difference group. Angle of attack detection data offset calculation module: used to fit the probability density function to the above three difference groups respectively to obtain offset 1-2, offset 1-3, and offset 2-3; Angle of attack sensor health assessment module: This module is used to calculate the difference between the above three offsets and the 1-2 offset, 1-3 offset, and 2-3 offset of the angle of attack sensor of a normal aircraft, and to determine the health status of the angle of attack sensor based on the difference value.

6. A health monitoring system for blade jamming of an angle-of-attack sensor according to claim 5, characterized in that, Also includes: Data preprocessing module: used to decode and scale the acquired angle of attack detection data in advance, and to unify the recording time interval of all angle of attack detection data.

7. A health monitoring system for blade jamming of an angle-of-attack sensor according to claim 6, characterized in that, The angle of attack detection data difference calculation module also includes: Fault prediction submodule: Used to calculate the number of all parameters in the angle of attack detection data that are not in the angle of attack threshold range. If the number of parameters that are not in the angle of attack threshold range exceeds the threshold, the difference between each pair of angle of attack detection data recorded by the first angle of attack sensor, the second angle of attack sensor, and the third angle of attack sensor at the same moment during the takeoff phase is calculated.

8. A computer device, characterized in that, include: At least one memory and at least one processor; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the at least one processor implements the steps of a health monitoring method for angle-of-attack sensor blade jamming as described in any one of claims 1 to 4.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of a health monitoring method for angle-of-attack sensor blade jamming as described in any one of claims 1 to 4.

Images