Static pressure error calibration method and device based on density distribution clustering algorithm

Abstract

The invention provides a static pressure error calibration method and device based on a density distribution clustering algorithm. The method comprises the following steps: acquiring a static pressure error calibration coefficient vector CpS according to flight data; performing correlation analysis on each element in the CpS vector and corresponding flight data, determining factors influencing the CpS, and establishing a mapping relation matrix; performing normalization and density clustering analysis on the mapping relation matrix to find a plurality of clustering center data; scatter interpolation fitting is carried out on the clustering center data step by step according to the envelope range, a final static pressure error calibration coefficient mapping matrix is obtained, the static pressure error calibration coefficient mapping matrix is used for static pressure error calibration of airborne products, and the static pressure calibration efficiency and accuracy are improved.

Description

Technical Field

[0001] This invention belongs to the field of atmospheric static pressure error calibration technology, specifically relating to a static pressure error calibration method and apparatus based on a density distribution clustering algorithm. Background Technology

[0002] Air data systems play a crucial role in aircraft flight, essential for mission success and significantly impacting flight quality, navigation, and safety. Accurate measurement of airspeed and barometric altitude is paramount for safe and efficient aircraft operation. For instance, accurate airspeed measurement is necessary to prevent low-speed runaway (stall) and to avoid exceeding aerodynamic and structural limits at high speeds. Similarly, accurate barometric altitude measurement is essential to ensure obstacle avoidance and maintain the required minimum vertical separation along the flight path. During flight, the measured static pressure can be inaccurate due to variations in configuration, Mach number, angle of attack, and sideslip angle, resulting in discrepancies from the free-flow static pressure. Air data systems typically calculate barometric altitude using actual static pressure and indicate airspeed by calculating the difference between total pressure and static pressure. Inaccurate static pressure measurement fails to accurately reflect altitude and introduces errors into the indicated airspeed data, ultimately affecting flight quality and safety. Therefore, calibrating static pressure measurements is essential.

[0003] The core of static pressure calibration is establishing an error mapping relationship between measured static pressure and the true free-flow static pressure. This typically requires analyzing the difference between standard static pressure (such as the static pressure output from a standard pitot tube) and the static pressure measured on the fuselage, along with relevant data such as Mach and angle of attack, to generate static pressure calibration coefficients. The atmospheric data computer, based on current Mach and angle of attack data, calculates the static pressure calibration coefficients in real time through table lookup interpolation or curve fitting, thereby calibrating the measured static pressure and ultimately obtaining the true free-flow static pressure.

[0004] In practical applications, to obtain accurate static pressure calibration coefficients, professional engineers often need to screen and identify flight test data one by one, eliminating non-steady-state flights, such as those with large roll angle change rates, sudden jumps in abnormal static pressure values ​​of sensors, or environmental interference such as icing of static pressure orifices. Data representative of the current flight characteristics are then selected and substituted into the static pressure calibration coefficient calculation method to obtain the final effective static pressure calibration coefficient. After data screening, piecewise interpolation or curve fitting methods are used. This process often requires multiple screenings and verifications. Statistics show that this data processing process consumes a significant amount of time and effort, typically accounting for about 60% of the total time. Furthermore, because this process involves professional engineers manually labeling the data, different engineers may have different criteria for judging "valid data," leading to some deviation in the results. Summary of the Invention

[0005] This invention proposes a static pressure error calibration method and device based on a density distribution clustering algorithm, which improves the efficiency and accuracy of static pressure calibration.

[0006] The first aspect of this invention provides a hydrostatic error calibration method based on a density distribution clustering algorithm, comprising: Step 1: Obtain the static pressure error calibration coefficient vector Cp based on flight data. S ; Step 2: For Cp S Correlation analysis is performed on each element in the vector and its corresponding flight data to determine the factors affecting Cp. S Establish a mapping matrix based on the factors; Step 3: Standardize all dimensions of the data in the mapping matrix according to their weights or data units to obtain a normalized matrix; Step 4: Perform density clustering analysis on the normalized matrix to obtain high-density distributed data points; Step 5: Obtain cluster center data for each high-density distribution point; Step 6: Perform scatter interpolation fitting on the cluster center data step by step according to the envelope range to obtain the final static pressure error calibration coefficient mapping matrix. The static pressure error calibration coefficient mapping matrix is ​​used for static pressure error calibration of airborne products.

[0007] Optional, for Cp S Each element in the vector undergoes a correlation analysis with its corresponding flight data, including: For Cp S Correlation analysis is performed on each element in the vector with its corresponding flight data, including Mach number, angle of attack, and sideslip angle.

[0008] Optionally, the static pressure error calibration coefficient vector Cp can be obtained from the flight data. S ,include: Using formula Obtain the static pressure error calibration coefficient vector Cp S ; in, For free-flow static pressure, p To measure static pressure, q c It is dynamic pressure.

[0009] Optionally, density clustering analysis can be performed on the normalized matrix to obtain high-density distributed data points, including: Density clustering analysis was performed on the normalized matrix to remove outliers, resulting in high-density distributed data points.

[0010] A second aspect of the present invention provides a hydrostatic error calibration device based on a density distribution clustering algorithm, comprising: The static pressure error calibration coefficient vector acquisition module is used to obtain the static pressure error calibration coefficient vector Cp based on flight data. S ; The mapping matrix acquisition module is used to obtain Cp S Correlation analysis is performed on each element in the vector and its corresponding flight data to determine the factors affecting Cp. S Establish a mapping matrix based on the factors; The normalization matrix acquisition module is used to standardize all dimensions of data in the mapping relationship matrix according to weights or data units to obtain a normalized matrix. The clustering module is used to perform density clustering analysis on the normalized matrix to obtain high-density distributed data points. The cluster center acquisition module is used to acquire cluster center data for each high-density distribution point; The static pressure error calibration coefficient mapping matrix acquisition module is used to perform scatter interpolation fitting on the cluster center data step by step according to the envelope range to obtain the final static pressure error calibration coefficient mapping matrix. The static pressure error calibration coefficient mapping matrix is ​​used for static pressure error calibration of airborne products.

[0011] Optionally, the mapping matrix acquisition module is specifically used for: For Cp S Correlation analysis is performed on each element in the vector with its corresponding flight data, including Mach number, angle of attack, and sideslip angle.

[0012] Optionally, the static pressure error calibration coefficient vector acquisition module is specifically used for: Using formula Obtain the static pressure error calibration coefficient vector Cp S ; in, For free-flow static pressure, p To measure static pressure, q c It is dynamic pressure.

[0013] Optionally, the clustering module is specifically used for: Density clustering analysis was performed on the normalized matrix to remove outliers, resulting in high-density distributed data points.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a static pressure error calibration method and apparatus based on a density distribution clustering algorithm. First, cluster analysis is performed on relevant data in flight test data to obtain high-density distribution data as reference points for calibration, thereby achieving calibration of static pressure and other data. This method is a novel and efficient atmospheric data calibration method, applicable to aircraft with finite envelopes. Simulation and real flight verification using actual flight data demonstrate that the calibration coefficients generated using the proposed method can accurately calibrate static pressure and other data, exhibiting strong versatility and high data confidence. This invention studies a method for calibrating atmospheric static pressure errors in atmospheric data computing products. The aim is to obtain accurate static pressure calibration coefficients, efficiently eliminate outlier data points, select data representative of the current flight characteristics (i.e., data with high data distribution density), and ultimately form static pressure error coefficients. These coefficients are then used to calibrate static pressure, which is used to calculate important atmospheric data such as barometric altitude and calibrated airspeed. Using a density-based calibration method saves significant time spent manually screening and labeling data, effectively improves data accuracy, and greatly shortens the static pressure error calibration cycle. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the static pressure error calibration method based on density distribution clustering algorithm according to the present invention; Figure 2 This is a comparison chart of the actual static pressure and the calibrated static pressure of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be described in more detail below with reference to the accompanying drawings.

[0017] In the accompanying drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some, but not all, of the embodiments of the present invention.

[0018] The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0019] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0020] This invention provides a hydrostatic error calibration method based on a density distribution clustering algorithm, such as... Figure 1As shown, the core idea of ​​the density distribution-based clustering algorithm for calibrating hydrostatic pressure is that clustered data are usually located in high-density regions, while outliers are often located in low-density regions. Clustering is a commonly used unsupervised learning algorithm. Its advantage is that it requires minimal domain knowledge, can discover clusters of arbitrary shapes, and can effectively identify noisy data points, making it particularly suitable for irregularly shaped clusters and noisy datasets. This method includes: First, the calculation method for the static pressure error coefficient is determined. The core of static pressure calibration is to establish the error mapping relationship between the measured static pressure and the actual free-flow static pressure, and to eliminate the interference of the fuselage flow field through the calibration coefficient. Second, the error coefficients of all data are generated according to the static pressure error calibration coefficient calculation method, and the error coefficients are analyzed with relevant data such as Mach, angle of attack, and sideslip angle to determine the correlation. Next, cluster analysis is performed on the error coefficients and relevant data to obtain high-density distribution data. Finally, the high-density data is selected and statistically analyzed to obtain the final static pressure error coefficient. Based on the flight envelope, the coefficients are expanded to fill in the missing distribution points in the data, and applied to atmospheric data computer products for static pressure error calibration.

[0021] like Figure 1 As shown, a detailed description of the calibration process is provided: Step S1: The formula for the static pressure error coefficient is as follows: ; in, For free-flow static pressure, p To measure static pressure, q c It is dynamic pressure.

[0022] Step S2: Calculate the static pressure calibration coefficient for all data according to the formula in S1, and obtain Cp. S vector; Step S3: For Cp S Correlation analysis is performed on each element of the vector with data such as Mach number, angle of attack, and sideslip angle to identify factors affecting Cp. S Factors, establish a mapping matrix, such as Cp S Mach number relation or Cp S A matrix composed of Mach number and angle of attack; Step S4: Standardize all dimensions of the data in the matrix formed in S3 according to their weights or data units to obtain normalized data; Step S5: Perform density clustering analysis on the normalized data to remove outliers (such as intervals with large roll angle change rates, sudden jumps in abnormal static pressure values ​​of sensors, or environmental interference such as icing of static pressure holes), and obtain high-density distributed data points. For example, clustering uses the DBSCAN algorithm.

[0023] Step S6: Perform statistical analysis on the high-density distribution points in S5 to obtain the center data of different clusters and the scatter points corresponding to different static pressure error coefficient mapping relationships; Step S7: Perform scatter interpolation fitting on the scattered data step by step according to the envelope range to obtain the final error coefficient mapping matrix, which is then applied to airborne products for static pressure error calibration.

[0024] like Figure 2 As shown, Psc is the calibration static pressure, Ps is the actual static pressure, Err is the difference between the calibration static pressure and the actual static pressure (in kPa), and M and Aoa are the relevant variables affecting the calibration error coefficient. It can be seen that under different Mach numbers M and angles of attack Aoa, Err remains within ±0.02 kPa, which is significantly smaller than the error of -0.4 kPa to 0.2 kPa before calibration, indicating a good calibration effect.

[0025] The above description is merely an example of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A hydrostatic error calibration method based on density distribution clustering algorithm, characterized in that, include: Step 1: Obtain the static pressure error calibration coefficient vector Cp based on flight data. S ; Step 2: For Cp S Correlation analysis is performed on each element in the vector and its corresponding flight data to determine the factors affecting Cp. S Establish a mapping matrix based on the factors; Step 3: Standardize all dimensions of the data in the mapping matrix according to their weights or data units to obtain a normalized matrix; Step 4: Perform density clustering analysis on the normalized matrix to obtain high-density distributed data points; Step 5: Obtain cluster center data for each high-density distribution point; Step 6: Perform scatter interpolation fitting on the cluster center data step by step according to the envelope range to obtain the final static pressure error calibration coefficient mapping matrix. The static pressure error calibration coefficient mapping matrix is ​​used for static pressure error calibration of airborne products.

2. The hydrostatic error calibration method based on density distribution clustering algorithm according to claim 1, characterized in that, For Cp S Each element in the vector undergoes a correlation analysis with its corresponding flight data, including: For Cp S Correlation analysis is performed on each element in the vector with its corresponding flight data, including Mach number, angle of attack, and sideslip angle.

3. The hydrostatic error calibration method based on density distribution clustering algorithm according to claim 1, characterized in that, Obtain the static pressure error calibration coefficient vector Cp from the flight data. S ,include: Using formula Obtain the static pressure error calibration coefficient vector Cp S ; in, For free-flow static pressure, p To measure static pressure, q c It is dynamic pressure.

4. The hydrostatic error calibration method based on density distribution clustering algorithm according to claim 1, characterized in that, Density clustering analysis was performed on the normalized matrix to obtain high-density distributed data points, including: Density clustering analysis was performed on the normalized matrix to remove outliers, resulting in high-density distributed data points.

5. A hydrostatic error calibration device based on a density distribution clustering algorithm, characterized in that, include: The static pressure error calibration coefficient vector acquisition module is used to obtain the static pressure error calibration coefficient vector Cp based on flight data. S ; The mapping matrix acquisition module is used to obtain Cp S Correlation analysis is performed on each element in the vector and its corresponding flight data to determine the factors affecting Cp. S Establish a mapping matrix based on the factors; The normalization matrix acquisition module is used to standardize all dimensions of data in the mapping relationship matrix according to weights or data units to obtain a normalized matrix. The clustering module is used to perform density clustering analysis on the normalized matrix to obtain high-density distributed data points. The cluster center acquisition module is used to acquire cluster center data for each high-density distribution point; The static pressure error calibration coefficient mapping matrix acquisition module is used to perform scatter interpolation fitting on the cluster center data step by step according to the envelope range to obtain the final static pressure error calibration coefficient mapping matrix. The static pressure error calibration coefficient mapping matrix is ​​used for static pressure error calibration of airborne products.

6. The hydrostatic error calibration device based on density distribution clustering algorithm according to claim 5, characterized in that, The mapping matrix acquisition module is specifically used for: For Cp S Correlation analysis is performed on each element in the vector with its corresponding flight data, including Mach number, angle of attack, and sideslip angle.

7. The hydrostatic error calibration device based on density distribution clustering algorithm according to claim 5, characterized in that, The static pressure error calibration coefficient vector acquisition module is specifically used for: Using formula Obtain the static pressure error calibration coefficient vector Cp S ; in, For free-flow static pressure, p To measure static pressure, q c It is dynamic pressure.

8. The hydrostatic error calibration device based on density distribution clustering algorithm according to claim 5, characterized in that, The clustering module is specifically used for: Density clustering analysis was performed on the normalized matrix to remove outliers, resulting in high-density distributed data points.

Images