Digital model identification method based on avionics system

Abstract

The invention discloses a digital model identification method based on an avionics system, and belongs to the technical field of avionics systems, and the method comprises the following steps: S1, constructing an avionics system digital model, and determining avionics system influence factors for digital model identification; s2, performing simulation analysis on the avionics system digital model, and determining a simulation analysis result of the avionics system digital model; s3, performing digital model identification on the avionics system digital model simulation analysis result, and determining an avionics system digital model identification result; and S4, generating an avionics system digital model identification optimization report, and timely solving the abnormal problem of the avionics system digital model. According to the method, the problem that the use effect is poor due to the fact that the digital model of the avionics system cannot be effectively identified and abnormal problems cannot be found and solved in time in the prior art is solved. According to the method, the digital model of the avionics system can be effectively identified, abnormal problems can be found and solved in time, and the use effect can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of avionics systems, and in particular to a digital model identification method based on avionics systems. Background Art

[0002] Traditional aircraft design and manufacturing methods can no longer meet the requirements of modern aircraft for flight performance and reliability; therefore, the use of digital design has become the development trend of the current aviation industry; in aircraft systems, avionics systems are a crucial component, and their performance directly affects flight safety. Therefore, it is of great significance to identify the digital model of the avionics system and ensure its reliability and stability under various working conditions to improve flight safety and reduce operating costs.

[0003] The Chinese patent with publication number CN115185493A discloses a model-based avionics system architecture design method, which includes the following steps: determining the design requirements of the avionics system architecture; establishing the meta-models required in the avionics system field in the functional layer, logical layer, physical layer, and message layer respectively; constructing a mapping matrix between the design requirements and the functions, designing the models of each layer, establishing the vertical mapping relationship between the models of each layer, and forming a complete avionics system architecture model; carrying out avionics system architecture design modeling according to the design requirements and the meta-models of each layer, comprehensively realizing and covering the requirements, and configuring resources on the multi-layer model to ensure that the current system architecture can meet the requirements of security, reliability, real-time performance, etc.; However, the patent has the following defects:

[0004] The existing digital models of avionics systems cannot be effectively authenticated, and abnormal problems cannot be discovered and solved in a timely manner, resulting in poor use results. Summary of the invention

[0005] The purpose of the present invention is to provide a digital model identification method based on an avionics system, which can effectively identify the digital model of the avionics system, can promptly discover and solve abnormal problems, can improve the use effect, and solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The digital model identification method based on the avionics system includes the following steps:

[0008] S1: Use modeling software and mathematical tools to build a digital model of the avionics system, define identification rules, and determine the influencing factors of the avionics system for digital model identification;

[0009] S2: Conduct simulation analysis on the avionics system digital model based on the influencing factors of the avionics system identified by different digital models, and determine the simulation analysis results of the avionics system digital model;

[0010] S3: Based on the avionics system digital model simulation analysis standard, perform digital model identification on the avionics system digital model simulation analysis results to determine the avionics system digital model identification results;

[0011] S4: Based on the identification results of the avionics system digital model, an avionics system digital model identification optimization report is generated to intelligently guide management personnel to adjust and optimize the avionics system digital model and promptly resolve abnormal problems with the avionics system digital model.

[0012] Preferably, in S1, the digital model of the avionics system is constructed by performing the following operations:

[0013] Collect the hardware specifications and software configuration parameters of the avionics system and determine the design parameters of the avionics system;

[0014] Process the avionics system design parameters to determine the avionics system characteristic data;

[0015] Based on the characteristic data of the avionics system, and using modeling software and mathematical tools, the avionics system is three-dimensionally modeled to determine the digital model of the avionics system.

[0016] Preferably, in S1, the avionics system design parameters are processed to perform the following operations:

[0017] Obtain avionics system design parameters;

[0018] Clean the avionics system design parameters, including:

[0019] Conduct consistency checks on avionics system design parameters;

[0020] According to the data consistency requirements, the inconsistent data in the avionics system design parameters that are useless for the digital model identification based on the avionics system are removed;

[0021] Process invalid and missing values ​​of avionics system design parameters;

[0022] According to the requirements of data validity and integrity, invalid values ​​and missing values ​​in the avionics system design parameters that are useless for the digital model identification based on the avionics system are removed, and the avionics system design parameters that are useful for the digital model identification based on the avionics system are determined;

[0023] Obtain avionics system design parameters useful for digital model identification based on avionics system;

[0024] Extract the features of the avionics system design parameters that are useful for identification based on the digital model of the avionics system;

[0025] Extract features that can reflect the digital model identification based on avionics systems;

[0026] Determine the avionics system characteristic data.

[0027] Preferably, the removal of invalid values ​​and missing values ​​in the avionics system design parameters that are useless for the digital model identification based on the avionics system comprises the following operations:

[0028] After removing useless invalid values ​​and missing values, the number of data of avionics system design parameters useful for digital model identification based on avionics system is extracted;

[0029] Comparing the data quantity with a preset data quantity threshold;

[0030] When the number of data is lower than a preset data number threshold, the weight values ​​of the invalid values ​​and missing values ​​that are removed are extracted;

[0031] Extract the weight value of the valid value that is closest to the data generation time of the invalid value and missing value to be removed;

[0032] Obtain a weight value threshold by using the weight values ​​of the removed invalid values ​​and missing values ​​combined with the weight value of the valid value whose data generation time is closest to the removed invalid values ​​and missing values;

[0033] The weight value threshold is obtained by the following formula:

[0034]

[0035] Among them, S represents the weight value threshold; n represents the number of invalid values ​​removed; m represents the number of missing values ​​removed; w 01i represents the weight value of the i-th invalid value removed; w 01xi represents the weight value of the valid value closest to the data generation time of the i-th invalid value to be removed; w 02i represents the weight value of the i-th missing value to be removed; w 02xi The weight value of the valid value closest to the data generation time of the i-th missing value to be removed; E 01 and E 02 represent the first adjustment coefficient and the second adjustment coefficient respectively, and the first adjustment coefficient is obtained by the following formula:

[0036]

[0037] Among them, E 01 represents the first adjustment coefficient; w 01z represents the median weight value of n invalid values ​​that have been removed; w 01xz Indicates the middle value of the weight of the valid values ​​with the closest data generation time corresponding to the n invalid values ​​removed;

[0038] At the same time, the second adjustment coefficient is obtained by the following formula:

[0039]

[0040] Among them, E 02 represents the second adjustment coefficient; w 02z represents the median weight value of the m missing values ​​removed; w 02xz The middle value of the weight of the valid values ​​of the data generation time closest to the m missing values ​​to be removed;

[0041] The weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with the preset weight value thresholds, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are used to fill the corresponding invalid values ​​and missing values.

[0042] Preferably, the weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with a preset weight value threshold, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the corresponding invalid values ​​and missing values ​​are filled with the filling values ​​corresponding to the invalid values ​​and missing values, and the following operations are performed:

[0043] Compare the weight value of the invalid value with the weight value threshold, and extract the invalid value corresponding to the weight value exceeding the weight value threshold as the target invalid value;

[0044] Extracting a valid value having the same weight value as each target invalid value from the valid values ​​as a first reference valid value;

[0045] extracting a data generation time interval between the first reference valid value and each target invalid value;

[0046] Obtaining a fill value corresponding to each target invalid value by using the data value of the first reference valid value in combination with the data generation time interval between the first reference valid value and each target invalid value;

[0047] The fill value corresponding to each target invalid value is obtained by the following formula:

[0048]

[0049] Among them, X w represents the fill value corresponding to each target invalid value; a represents the number of data of the first reference valid value corresponding to each target invalid value; X 01i represents the data value of the i-th first reference valid value; X 01i+1 represents the data value of the (i+1)th first reference valid value; X 01brepresents the standard deviation of the data values ​​of a first reference valid values; T g01i represents the data generation time interval between the i-th first reference valid value and each target invalid value;

[0050] The target invalid values ​​are filled using the filling values ​​corresponding to each target invalid value.

[0051] Preferably, the weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with a preset weight value threshold, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the corresponding invalid values ​​and missing values ​​are filled with the filling values ​​corresponding to the invalid values ​​and missing values, and the following operations are also performed:

[0052] Compare the weight value of the missing value with the weight value threshold, and extract the missing value corresponding to the weight value exceeding the weight value threshold as the target missing value;

[0053] Extracting a valid value having the same weight value as each target missing value from the valid values ​​as a second reference valid value;

[0054] Extracting a target invalid value that has been filled and has the same weight value as the target missing value as a third reference valid value;

[0055] Extracting the data generation time interval between the second reference valid value and the third reference valid value and each target missing value;

[0056] Obtain a filling value corresponding to each target missing value by using the second reference valid value and the third reference valid value in combination with the data generation time interval between the second reference valid value and the third reference valid value and each target missing value;

[0057] The filling value corresponding to each target missing value is obtained by the following formula:

[0058]

[0059] Among them, X q represents the filling value corresponding to each target missing value; b represents the number of data of the second reference valid value; c represents the number of data of the third reference valid value; X 02i represents the data value of the i-th second reference valid value; X wi The data value representing the i-th third reference valid value; T g02i represents the data generation time interval between the i-th second reference valid value and each target missing value; T g03i represents the data generation time interval between the ith third reference valid value and each target missing value; X 02zrepresents the middle value of the data value of b second reference valid values; X wz represents the middle value of the data value of c third reference valid values; X wb represents the standard deviation of the data values ​​of c third reference valid values; X wp represents the average value of the data values ​​of c third reference valid values; X 02b represents the standard deviation of the data value of b second reference valid values; X 02p represents the average value of the data values ​​of b second reference valid values;

[0060] The target missing values ​​are filled using the filling values ​​corresponding to each target missing value.

[0061] Preferably, in S1, the influencing factors of the avionics system for digital model identification are determined by performing the following operations:

[0062] According to the structure and function of the digital model of the avionics system and the actual flight situation of the aircraft, the identification rules are defined to check whether the digital model of the avionics system can correctly realize all the expected functions, comprehensively consider which factors will affect the behavior of the avionics system, and determine the influencing factors of the avionics system for digital model identification;

[0063] Among them, the factors affecting the avionics system in digital model identification include: speed, altitude, load, temperature, pressure and electromagnetic interference.

[0064] Preferably, in S2, the simulation analysis result of the digital model of the avionics system is determined, and the following operations are performed:

[0065] Obtain the influencing factors of avionics system for digital model verification;

[0066] Input the influencing factors of the avionics system identified by different digital models into the digital model of the avionics system;

[0067] The avionics system digital model is simulated and analyzed based on the influencing factors of the avionics system identified by different digital models. By comparing the performance of the avionics system under different influencing factors of the avionics system, the simulation analysis results of the avionics system digital model are determined.

[0068] Preferably, in S3, the avionics system digital model identification result is determined by performing the following operations:

[0069] According to the digital model identification requirements based on the avionics system, the avionics system digital model simulation analysis standards are pre-set, and the pre-set avionics system digital model simulation analysis standards are stored;

[0070] According to the digital model identification requirements of the avionics system, the stored digital model simulation analysis standards of the avionics system are indexed, and the indexed digital model simulation analysis standards of the avionics system are retrieved;

[0071] Based on the avionics system digital model simulation analysis standard, the avionics system digital model simulation analysis results are subjected to digital model identification to determine the avionics system digital model identification results.

[0072] Preferably, in S3, based on the avionics system digital model simulation analysis standard, performing digital model authentication on the avionics system digital model simulation analysis result includes:

[0073] When the simulation analysis result of the avionics system digital model is within the simulation analysis standard range of the avionics system digital model, the avionics system digital model identification result is that the avionics system digital model is normal and can correctly realize all expected functions;

[0074] When the avionics system digital model simulation analysis results are not within the avionics system digital model simulation analysis standard range, the avionics system digital model identification result is that the avionics system digital model is abnormal and cannot correctly implement all expected functions.

[0075] Preferably, in S4, the management personnel are intelligently guided to adjust and optimize the digital model of the avionics system by performing the following operations:

[0076] Obtain the results of the digital model identification of the avionics system;

[0077] Based on the identification results of the avionics system digital model, the avionics system digital model is mined and analyzed to generate an avionics system digital model identification optimization report;

[0078] The avionics system digital model identification and optimization report is transmitted to management personnel in a visual form, and management personnel are intelligently guided to adjust and optimize the avionics system digital model and solve abnormal problems of the avionics system digital model in a timely manner.

[0079] Compared with the prior art, the present invention has the following beneficial effects:

[0080] 1. The present invention adopts modeling software and mathematical tools to construct a digital model of an avionics system. According to the structure and function of the digital model of the avionics system and in combination with the actual flight conditions of the aircraft, identification rules are defined to determine the influencing factors of the avionics system identified by the digital model. Based on the influencing factors of the avionics system identified by different digital models, the digital model of the avionics system is simulated and analyzed to determine the simulation analysis results of the digital model of the avionics system.

[0081] 2. The present invention is based on the avionics system digital model simulation analysis standard, performs digital model identification on the avionics system digital model simulation analysis results, determines the avionics system digital model identification results, and generates an avionics system digital model identification optimization report, which intelligently guides management personnel to adjust and optimize the avionics system digital model, and promptly resolves abnormal problems of the avionics system digital model. It can effectively identify the digital model of the avionics system, promptly discover and resolve abnormal problems, and improve the use effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is an operation flow chart of the digital model identification method based on the avionics system of the present invention;

[0083] Figure 2 The present invention is a flowchart of an algorithm for performing digital model identification on the results of the digital model simulation analysis of the avionics system based on the digital model simulation analysis standard of the avionics system. DETAILED DESCRIPTION

[0084] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] In order to solve the existing problem that the digital model of the avionics system cannot be effectively identified, abnormal problems cannot be discovered and solved in time, resulting in poor use effect, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0086] The digital model identification method based on the avionics system includes the following steps:

[0087] S1: Use modeling software and mathematical tools to build a digital model of the avionics system, define identification rules, and determine the influencing factors of the avionics system for digital model identification;

[0088] In this embodiment, the digital model of the avionics system is constructed by performing the following operations:

[0089] Collect the hardware specifications and software configuration parameters of the avionics system and determine the design parameters of the avionics system;

[0090] Process the avionics system design parameters to determine the avionics system characteristic data;

[0091] Based on the characteristic data of the avionics system, and using modeling software and mathematical tools, the avionics system is three-dimensionally modeled to determine the digital model of the avionics system.

[0092] It should be noted that modern aircraft use a unified processor to uniformly process information from various avionics equipment on the aircraft, combine equipment with the same or similar functions into one component, and display relevant parameters on the display. Relevant information is transmitted between various avionics equipment through the airborne data bus, so that the performance of all avionics equipment on the entire aircraft reaches a higher level. Such a system is called an avionics system. The development of avionics systems has gone through several important stages, from the initial discrete structure, to the centralized structure, to the integrated structure, and finally to the current highly integrated structure. This development has made avionics systems more intelligent and efficient, greatly improving the performance and safety of aircraft.

[0093] Specifically, the avionics system mainly includes the following parts: 1) Communication system: responsible for voice and data transmission between the aircraft and the outside world, ensuring stable communication between the aircraft and the ground; 2) Navigation system: real-time collection and measurement of aircraft motion information through a variety of navigation sensors to ensure safe and accurate flight; 3) Display system: provides the flight crew with comprehensive and clear flight information display to help the flight crew accurately and timely grasp the flight dynamics and aircraft operating conditions; 4) Flight control system: responsible for the stability and maneuverability of the aircraft, ensuring the stability and safety of the aircraft under various flight conditions; 5) Weather radar: used to detect weather conditions during flight to help pilots make appropriate flight decisions; 6) Aircraft management system: comprehensive management of various functions of the aircraft to ensure the overall performance and efficiency of the aircraft.

[0094] Among them, the hardware specifications of the avionics system mainly include the following aspects: 1) Display components: The multi-function display in the avionics system supporting equipment is available in a variety of sizes, including 10.4 inches, 8 inches and 6 inches. The display screen is a TFT panel with anti-reflection function and can be clearly read even under direct sunlight; its resolution is 1024×768, 800×640 and 640×480 respectively; 2) Communication components: Atmospheric data computer: It is used to measure the static pressure, total pressure and external atmospheric temperature of the aircraft during flight, calculate and output indications Parameters such as airspeed, true airspeed, relative pressure altitude, absolute pressure altitude, vertical ascent and descent speed; Attitude and heading reference system: used to measure the pitch, roll and heading of the aircraft with an accuracy of 2°, 2° and 2° respectively; Ultra-short wave radio: the working range is 117.975-137.0MHz; 3) Storage and interface: Storage interface: The multi-function display has 1 SD card interface for expanded storage; Communication interface: including 2 Ethernet interfaces and 14 serial interfaces; Discrete IO interface: with 8 discrete IO interfaces.

[0095] In this embodiment, the avionics system design parameters are processed and the following operations are performed:

[0096] Obtain avionics system design parameters;

[0097] Clean the avionics system design parameters, including:

[0098] Conduct consistency checks on avionics system design parameters;

[0099] According to the data consistency requirements, the inconsistent data in the avionics system design parameters that are useless for the digital model identification based on the avionics system are removed;

[0100] Process invalid and missing values ​​of avionics system design parameters;

[0101] According to the requirements of data validity and integrity, invalid values ​​and missing values ​​in the avionics system design parameters that are useless for the digital model identification based on the avionics system are removed, and the avionics system design parameters that are useful for the digital model identification based on the avionics system are determined;

[0102] It should be noted that by cleaning the avionics system design parameters, inconsistent data, invalid values ​​and missing values ​​in the avionics system design parameters that are useless for the identification of the digital model based on the avionics system can be removed.

[0103] Among them, data cleaning refers to the process of preprocessing data in data governance work. The purpose is to correct or delete erroneous, duplicated and inconsistent parts in the data to ensure the accuracy and completeness of the data. Through data cleaning, the data can be converted into high-quality data to facilitate subsequent data analysis and use.

[0104] Obtain avionics system design parameters useful for digital model identification based on avionics system;

[0105] Extract the features of the avionics system design parameters that are useful for identification based on the digital model of the avionics system;

[0106] Extract features that can reflect the digital model identification based on avionics systems;

[0107] Determine the avionics system characteristic data.

[0108] It should be noted that by extracting the features of the avionics system design parameters useful for the digital model identification based on the avionics system, the characteristic data of the avionics system can be determined, which is convenient for the subsequent determination of the avionics system influencing factors for the digital model identification.

[0109] Specifically, the removal of invalid values ​​and missing values ​​in the avionics system design parameters that are useless for the digital model identification based on the avionics system is performed by performing the following operations:

[0110] After removing useless invalid values ​​and missing values, the number of data of avionics system design parameters useful for digital model identification based on avionics system is extracted;

[0111] Comparing the data quantity with a preset data quantity threshold;

[0112] When the number of data is lower than a preset data number threshold, the weight values ​​of the invalid values ​​and missing values ​​that are removed are extracted;

[0113] Extract the weight value of the valid value that is closest to the data generation time of the invalid value and missing value to be removed;

[0114] Obtain a weight value threshold by using the weight values ​​of the removed invalid values ​​and missing values ​​combined with the weight value of the valid value whose data generation time is closest to the removed invalid values ​​and missing values;

[0115] The weight value threshold is obtained by the following formula:

[0116]

[0117] Among them, S represents the weight value threshold; n represents the number of invalid values ​​removed; m represents the number of missing values ​​removed; w 01i represents the weight value of the i-th invalid value removed; w 01xi represents the weight value of the valid value closest to the data generation time of the i-th invalid value to be removed; w 02i represents the weight value of the i-th missing value to be removed; w 02xi The weight value of the valid value closest to the data generation time of the i-th missing value to be removed; E 01 and E 02 represent the first adjustment coefficient and the second adjustment coefficient respectively, and the first adjustment coefficient is obtained by the following formula:

[0118]

[0119] Among them, E 01 represents the first adjustment coefficient; w 01z represents the median weight value of n invalid values ​​that have been removed; w 01xz Indicates the middle value of the weight of the valid values ​​with the closest data generation time corresponding to the n invalid values ​​removed;

[0120] At the same time, the second adjustment coefficient is obtained by the following formula:

[0121]

[0122] Among them, E 02 represents the second adjustment coefficient; w02z represents the median weight value of the m missing values ​​removed; w 02xz The middle value of the weight of the valid values ​​of the data generation time closest to the m missing values ​​to be removed;

[0123] The weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with the preset weight value thresholds, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are used to fill the corresponding invalid values ​​and missing values.

[0124] The technical effect of the above technical solution is: by removing useless invalid values ​​and missing values, the accuracy and reliability of the avionics system design parameter data used for digital model identification based on the avionics system are ensured. This helps to improve the accuracy and predictive ability of model identification. Weight analysis and filling processing of missing and invalid data further improves the integrity of the data and reduces the model deviation or error caused by incomplete data. By comparing the number of data with the preset threshold and further analyzing based on the weight value, the technical solution can more effectively utilize the available data and avoid the problem of insufficient model training or inaccurate results caused by insufficient data. The setting of the weight value threshold and the calculation of the adjustment coefficient enable the importance and time proximity of different data points to be considered in the data filling process, thereby improving the overall utilization efficiency of the data. Through the weight analysis and filling processing of invalid values ​​and missing values, the technical solution can enhance the robustness of the model so that it can still maintain good performance when facing incomplete or missing data. The setting of the weight value threshold and the introduction of the adjustment coefficient enable the model to more accurately reflect the actual distribution and characteristics of the data when filling the data, thereby improving the generalization ability and adaptability of the model. This technical solution avoids the tedious manual data cleaning and filling process through clear weight value calculation and filling value acquisition steps, and improves the computational efficiency of data processing and model training. At the same time, through automated weight analysis and filling processing, the possibility of human intervention and errors is reduced, further improving the overall computational efficiency. This technical solution makes the data filling and model training process more transparent and explainable through clear weight value calculation process and filling value acquisition method. This helps users understand the behavior and performance of the model, as well as the impact of data on the model, so as to make more informed decisions and adjustments.

[0125] In summary, this technical solution can achieve multiple technical effects in terms of performance indicators, including improved data integrity and accuracy, improved data utilization efficiency, enhanced model robustness, improved computing efficiency, and enhanced interpretability.

[0126] Specifically, the weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with the preset weight value threshold, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the corresponding invalid values ​​and missing values ​​are filled with the filling values ​​corresponding to the invalid values ​​and missing values, and the following operations are performed:

[0127] Compare the weight value of the invalid value with the weight value threshold, and extract the invalid value corresponding to the weight value exceeding the weight value threshold as the target invalid value;

[0128] Extracting a valid value having the same weight value as each target invalid value from the valid values ​​as a first reference valid value;

[0129] extracting a data generation time interval between the first reference valid value and each target invalid value;

[0130] Obtaining a fill value corresponding to each target invalid value by using the data value of the first reference valid value in combination with the data generation time interval between the first reference valid value and each target invalid value;

[0131] The fill value corresponding to each target invalid value is obtained by the following formula:

[0132]

[0133] Among them, X w represents the fill value corresponding to each target invalid value; a represents the number of data of the first reference valid value corresponding to each target invalid value; X 01i represents the data value of the i-th first reference valid value; X 01i+1 represents the data value of the (i+1)th first reference valid value; X 01b represents the standard deviation of the data values ​​of a first reference valid values; T g01i represents the data generation time interval between the i-th first reference valid value and each target invalid value;

[0134] The target invalid values ​​are filled using the filling values ​​corresponding to each target invalid value.

[0135] The technical effect of the above technical solution is: by comparing the weight value of the invalid value with the preset weight value threshold, and extracting the target invalid value exceeding the threshold for special processing, the technical solution can selectively fill in the invalid values ​​that are more important (i.e., higher weight) to the model or analysis. This helps to improve the quality and reliability of the overall data set. The method of obtaining the filling value by using the first reference valid value with the same weight value as the target invalid value and combining the data generation time interval takes into account the time series characteristics of the data and the relationship between adjacent data points. By combining multiple factors (such as data value, number of data, standard deviation and time interval), the calculation of the filling value is more accurate and can better reflect the possible true value of the target invalid value. Accurate filling of invalid values ​​helps to reduce noise and outliers in the data set, thereby improving the performance of the model trained based on these data. More accurate filling values ​​mean that the model can learn more realistic data distribution and characteristics, thereby improving the accuracy of prediction or classification. The technical solution avoids the cumbersome manual data cleaning and filling process through clear weight value comparison and filling value calculation steps. Automated filling processing reduces the possibility of human intervention and errors, while improving the overall data processing efficiency. This technical solution can handle invalid values ​​with different weights and different time intervals, showing good adaptability and flexibility. This means that it can be widely used in different data sets and scenarios without a lot of adjustments or modifications to the algorithm. The calculation process of the fill value is based on clear formulas and factors (such as data value, number of data, standard deviation, and time interval), which makes the filling process more transparent and explainable. Users or developers can more easily understand the source and calculation process of the fill value, and thus have a deeper understanding of the data set and model.

[0136] In summary, this technical solution can achieve multiple technical effects in terms of performance indicators, including improved data quality, improved fill value accuracy, optimized model performance, improved computing efficiency, enhanced adaptability and flexibility, and improved interpretability. These effects work together in the data analysis and model training process, helping to improve overall data processing efficiency and model performance.

[0137] Specifically, the weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with the preset weight value threshold, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the corresponding invalid values ​​and missing values ​​are filled with the filling values ​​corresponding to the invalid values ​​and missing values, and the following operations are performed:

[0138] Compare the weight value of the missing value with the weight value threshold, and extract the missing value corresponding to the weight value exceeding the weight value threshold as the target missing value;

[0139] Extracting a valid value having the same weight value as each target missing value from the valid values ​​as a second reference valid value;

[0140] Extracting a target invalid value that has been filled and has the same weight value as the target missing value as a third reference valid value;

[0141] Extracting the data generation time interval between the second reference valid value and the third reference valid value and each target missing value;

[0142] Obtain a filling value corresponding to each target missing value by using the second reference valid value and the third reference valid value in combination with the data generation time interval between the second reference valid value and the third reference valid value and each target missing value;

[0143] The filling value corresponding to each target missing value is obtained by the following formula:

[0144]

[0145] Among them, X q represents the filling value corresponding to each target missing value; b represents the number of data of the second reference valid value; c represents the number of data of the third reference valid value; X 02i represents the data value of the i-th second reference valid value; X wi The data value representing the i-th third reference valid value; T g02i represents the data generation time interval between the i-th second reference valid value and each target missing value; T g03i represents the data generation time interval between the ith third reference valid value and each target missing value; X 02z represents the middle value of the data value of b second reference valid values; X wz represents the middle value of the data value of c third reference valid values; X wb represents the standard deviation of the data values ​​of c third reference valid values; X wp represents the average value of the data values ​​of c third reference valid values; X 02b represents the standard deviation of the data value of b second reference valid values; X 02p represents the average value of the data values ​​of b second reference valid values;

[0146] The target missing values ​​are filled using the filling values ​​corresponding to each target missing value.

[0147] The technical effect of the above technical solution is: by comparing the weight value of the missing value with the preset weight value threshold, and extracting the target missing value exceeding the threshold for filling processing, the technical solution can ensure the integrity of important data. This helps to reduce the analysis deviation or model performance degradation caused by missing data. The second reference valid value with the same weight value as the target missing value and the target invalid value that has been filled (as the third reference valid value) are used, and the filling value is calculated in combination with the data generation time interval. This method comprehensively considers multiple factors, including data value, number of data, standard deviation, average value and time interval. This comprehensive consideration makes the calculation of the filling value more accurate and can more accurately reflect the possible true value of the target missing value. Accurate filling of missing values ​​helps to reduce noise and outliers in the data set, thereby improving the robustness of the model trained based on these data. More accurate filling values ​​mean that the model can better handle incomplete data and show more stable performance in practical applications. The technical solution realizes automated data filling processing through clear weight value comparison and filling value calculation steps. This reduces the need for manual intervention and improves the overall efficiency of data processing. At the same time, the solution can handle data sets of different sizes and complexities, showing good scalability. This technical solution can handle missing values ​​with different weights and different time intervals, showing good adaptability and flexibility. This means that it can be widely used in different data sets and scenarios without a lot of adjustments or modifications to the algorithm. By calculating the fill-in value and applying it, the technical solution can also be used as part of data quality monitoring and improvement. By evaluating the effect of the fill-in value, we can further understand the problems in the data set and take corresponding measures to improve it. The calculation process of the fill-in value is based on clear formulas and multiple factors (such as data value, number of data, standard deviation, mean value, and time interval), which makes the filling process more transparent and explainable. Users or developers can more easily understand the source and calculation process of the fill-in value, so as to have a deeper understanding of the data set and model.

[0148] In summary, this technical solution can achieve multiple technical effects in terms of performance indicators, including enhanced data integrity, improved fill value accuracy, improved model robustness, improved computing efficiency and scalability, enhanced adaptability and flexibility, and promotion of data quality monitoring and improvement. These effects work together in the data analysis and model training process, helping to improve overall data processing efficiency and model performance.

[0149] In this embodiment, the influencing factors of the avionics system for digital model identification are determined by performing the following operations:

[0150] According to the structure and function of the digital model of the avionics system and the actual flight situation of the aircraft, the identification rules are defined to check whether the digital model of the avionics system can correctly realize all the expected functions, comprehensively consider which factors will affect the behavior of the avionics system, and determine the influencing factors of the avionics system for digital model identification;

[0151] Among them, the factors affecting the avionics system in digital model identification include: speed, altitude, load, temperature, pressure and electromagnetic interference.

[0152] S2: Conduct simulation analysis on the avionics system digital model based on the influencing factors of the avionics system identified by different digital models, and determine the simulation analysis results of the avionics system digital model;

[0153] In this embodiment, to determine the simulation analysis results of the avionics system digital model, the following operations are performed:

[0154] Obtain the influencing factors of avionics system for digital model verification;

[0155] Input the influencing factors of the avionics system identified by different digital models into the digital model of the avionics system;

[0156] The avionics system digital model is simulated and analyzed based on the influencing factors of the avionics system identified by different digital models. By comparing the performance of the avionics system under different influencing factors of the avionics system, the simulation analysis results of the avionics system digital model are determined.

[0157] S3: Based on the avionics system digital model simulation analysis standard, perform digital model identification on the avionics system digital model simulation analysis results to determine the avionics system digital model identification results;

[0158] In this embodiment, the avionics system digital model identification result is determined by performing the following operations:

[0159] According to the digital model identification requirements based on the avionics system, the avionics system digital model simulation analysis standards are pre-set, and the pre-set avionics system digital model simulation analysis standards are stored;

[0160] According to the digital model identification requirements of the avionics system, the stored digital model simulation analysis standards of the avionics system are indexed, and the indexed digital model simulation analysis standards of the avionics system are retrieved;

[0161] Based on the avionics system digital model simulation analysis standard, the avionics system digital model simulation analysis results are subjected to digital model identification to determine the avionics system digital model identification results.

[0162] In this embodiment, based on the avionics system digital model simulation analysis standard, the avionics system digital model simulation analysis result is subjected to digital model authentication, including:

[0163] When the simulation analysis result of the avionics system digital model is within the simulation analysis standard range of the avionics system digital model, the avionics system digital model identification result is that the avionics system digital model is normal and can correctly realize all expected functions;

[0164] When the avionics system digital model simulation analysis results are not within the avionics system digital model simulation analysis standard range, the avionics system digital model identification result is that the avionics system digital model is abnormal and cannot correctly implement all expected functions.

[0165] It should be noted that, based on the avionics system digital model simulation analysis standard, digital model appraisal is performed on the avionics system digital model simulation analysis results, and the avionics system digital model appraisal results can be determined. Based on the avionics system digital model appraisal results, an avionics system digital model appraisal optimization report can be generated to intelligently guide management personnel to adjust and optimize the avionics system digital model and promptly resolve abnormal problems of the avionics system digital model. The digital model of the avionics system can be effectively appraised, abnormal problems can be discovered and resolved in a timely manner, and the use effect can be improved.

[0166] S4: Based on the identification results of the avionics system digital model, an avionics system digital model identification optimization report is generated to intelligently guide management personnel to adjust and optimize the avionics system digital model and promptly resolve abnormal problems with the avionics system digital model.

[0167] In this embodiment, the intelligent guidance management personnel adjust and optimize the digital model of the avionics system and perform the following operations:

[0168] Obtain the results of the digital model identification of the avionics system;

[0169] Based on the identification results of the avionics system digital model, the avionics system digital model is mined and analyzed to generate an avionics system digital model identification optimization report;

[0170] The avionics system digital model identification and optimization report is transmitted to management personnel in a visual form, and management personnel are intelligently guided to adjust and optimize the avionics system digital model and solve abnormal problems of the avionics system digital model in a timely manner.

[0171] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0172] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A digital model identification method based on an avionics system, characterized in that: The steps include: S1: Use modeling software and mathematical tools to build a digital model of the avionics system, define identification rules, and determine the influencing factors of the avionics system for digital model identification; The avionics system design parameters are collected, the avionics system design parameters are processed, and the weight value threshold is obtained by combining the weight value of the removed invalid value and missing value with the weight value of the valid value closest to the data generation time of the removed invalid value and missing value; Compare the weight values ​​corresponding to the invalid values ​​and missing values ​​with the preset weight value threshold, obtain the filling values ​​corresponding to the invalid values ​​and missing values ​​through the comparison result, and use the filling values ​​corresponding to the invalid values ​​and missing values ​​to fill the corresponding invalid values ​​and missing values; S2: Conduct simulation analysis on the avionics system digital model based on the influencing factors of the avionics system identified by different digital models, and determine the simulation analysis results of the avionics system digital model; S3: Based on the avionics system digital model simulation analysis standard, perform digital model identification on the avionics system digital model simulation analysis results to determine the avionics system digital model identification results; S4: Based on the identification results of the avionics system digital model, an avionics system digital model identification optimization report is generated to intelligently guide management personnel to adjust and optimize the avionics system digital model and promptly resolve abnormal problems with the avionics system digital model.

2. The digital model identification method based on the avionics system according to claim 1 is characterized in that: In S1, a digital model of the avionics system is constructed, and the following operations are performed: Collect the hardware specifications and software configuration parameters of the avionics system and determine the design parameters of the avionics system; Process the avionics system design parameters to determine the avionics system characteristic data; Based on the characteristic data of the avionics system, and using modeling software and mathematical tools, the avionics system is three-dimensionally modeled to determine the digital model of the avionics system.

3. The digital model identification method based on avionics system according to claim 2 is characterized in that: In S1, the avionics system design parameters are processed and the following operations are performed: Obtain avionics system design parameters; Clean the avionics system design parameters, including: Conduct consistency checks on avionics system design parameters; According to the data consistency requirements, the inconsistent data in the avionics system design parameters that are useless for the digital model identification based on the avionics system are removed; Process invalid and missing values ​​of avionics system design parameters; According to the requirements of data validity and integrity, invalid values ​​and missing values ​​in the avionics system design parameters that are useless for the digital model identification based on the avionics system are removed, and the avionics system design parameters that are useful for the digital model identification based on the avionics system are determined; Obtain avionics system design parameters useful for digital model identification based on avionics system; Extract the features of the avionics system design parameters that are useful for identification based on the digital model of the avionics system; Extract features that can reflect the digital model identification based on avionics systems; Determine the avionics system characteristic data.

4. The digital model identification method based on avionics system according to claim 3 is characterized in that: The removal of invalid values ​​and missing values ​​in the avionics system design parameters that are useless for the digital model identification based on the avionics system comprises the following operations: After removing useless invalid values ​​and missing values, the number of data of avionics system design parameters useful for digital model identification based on avionics system is extracted; Comparing the data quantity with a preset data quantity threshold; When the number of data is lower than a preset data number threshold, the weight values ​​of the invalid values ​​and missing values ​​that are removed are extracted; Extract the weight value of the valid value that is closest to the data generation time of the invalid value and missing value to be removed; Obtain a weight value threshold by using the weight values ​​of the removed invalid values ​​and missing values ​​combined with the weight value of the valid value whose data generation time is closest to the removed invalid values ​​and missing values; The weight value threshold is obtained by the following formula: ； Among them, S represents the weight value threshold; n represents the number of invalid values ​​removed; m represents the number of missing values ​​removed; w 01i represents the weight value of the i-th invalid value removed; w 01xi represents the weight value of the valid value closest to the data generation time of the i-th invalid value to be removed; w 02i represents the weight value of the i-th missing value to be removed; w 02xi The weight value of the valid value closest to the data generation time of the i-th missing value to be removed; E 01 and E 02 represent the first adjustment coefficient and the second adjustment coefficient respectively, and the first adjustment coefficient is obtained by the following formula: ； Among them, E 01 represents the first adjustment coefficient; w 01z represents the median weight value of n invalid values ​​that have been removed; w 01xz Indicates the middle value of the weight of the valid values ​​with the closest data generation time corresponding to the n invalid values ​​removed; At the same time, the second adjustment coefficient is obtained by the following formula: ； Among them, E 02 represents the second adjustment coefficient; w 02z represents the median weight value of the m missing values ​​removed; w 02xz The middle value of the weight of the valid values ​​of the most recent data generation time corresponding to the m missing values ​​to be removed; The weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with the preset weight value thresholds, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are used to fill the corresponding invalid values ​​and missing values.

5. The digital model identification method based on avionics system according to claim 4 is characterized in that: The weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with the preset weight value thresholds, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the corresponding invalid values ​​and missing values ​​are filled with the filling values ​​corresponding to the invalid values ​​and missing values, and the following operations are performed: Compare the weight value of the invalid value with the weight value threshold, and extract the invalid value corresponding to the weight value exceeding the weight value threshold as the target invalid value; Extracting a valid value having the same weight value as each target invalid value from the valid values ​​as a first reference valid value; extracting a data generation time interval between the first reference valid value and each target invalid value; Obtaining a fill value corresponding to each target invalid value by using the data value of the first reference valid value in combination with the data generation time interval between the first reference valid value and each target invalid value; The fill value corresponding to each target invalid value is obtained by the following formula: ； Among them, X w represents the fill value corresponding to each target invalid value; a represents the number of data of the first reference valid value corresponding to each target invalid value; X 01i represents the data value of the i-th first reference valid value; X 01i+1 represents the data value of the (i+1)th first reference valid value; X 01b represents the standard deviation of the data values ​​of a first reference valid values; T g01i represents the data generation time interval between the i-th first reference valid value and each target invalid value; The target invalid values ​​are filled using the filling values ​​corresponding to each target invalid value.

6. The digital model identification method based on avionics system according to claim 5 is characterized in that: The weight values ​​corresponding to the invalid values ​​and missing values ​​are compared with the preset weight value threshold, and the filling values ​​corresponding to the invalid values ​​and missing values ​​are obtained through the comparison results, and the corresponding invalid values ​​and missing values ​​are filled with the filling values ​​corresponding to the invalid values ​​and missing values, and the following operations are performed: Compare the weight value of the missing value with the weight value threshold, and extract the missing value corresponding to the weight value exceeding the weight value threshold as the target missing value; Extracting a valid value having the same weight value as each target missing value from the valid values ​​as a second reference valid value; Extracting a target invalid value that has been filled and has the same weight value as the target missing value as a third reference valid value; Extracting the data generation time interval between the second reference valid value and the third reference valid value and each target missing value; Obtain a filling value corresponding to each target missing value by using the second reference valid value and the third reference valid value in combination with the data generation time interval between the second reference valid value and the third reference valid value and each target missing value; The filling value corresponding to each target missing value is obtained by the following formula: ； Among them, X q represents the filling value corresponding to each target missing value; b represents the number of data of the second reference valid value; c represents the number of data of the third reference valid value; X 02i represents the data value of the i-th second reference valid value; X wi The data value representing the i-th third reference valid value; T g02i represents the data generation time interval between the i-th second reference valid value and each target missing value; T g03i represents the data generation time interval between the ith third reference valid value and each target missing value; X 02z represents the middle value of the data value of b second reference valid values; X wz represents the middle value of the data value of c third reference valid values; X wb represents the standard deviation of the data values ​​of c third reference valid values; X wp represents the average value of the data values ​​of c third reference valid values; X 02b represents the standard deviation of the data value of b second reference valid values; X 02p represents the average value of the data values ​​of b second reference valid values; The target missing values ​​are filled using the filling values ​​corresponding to each target missing value.

7. The digital model identification method based on avionics system according to claim 6 is characterized in that: In S1, the influencing factors of the avionics system for digital model identification are determined, and the following operations are performed: According to the structure and function of the digital model of the avionics system and the actual flight situation of the aircraft, the identification rules are defined to check whether the digital model of the avionics system can correctly realize all the expected functions, comprehensively consider which factors will affect the behavior of the avionics system, and determine the influencing factors of the avionics system for digital model identification; Among them, the factors affecting the avionics system in digital model identification include: speed, altitude, load, temperature, pressure and electromagnetic interference.

8. The digital model identification method based on avionics system according to claim 7 is characterized in that: In S2, the simulation analysis result of the digital model of the avionics system is determined, and the following operations are performed: Obtain the influencing factors of avionics system for digital model verification; Input the influencing factors of the avionics system identified by different digital models into the digital model of the avionics system; The avionics system digital model is simulated and analyzed based on the influencing factors of the avionics system identified by different digital models. By comparing the performance of the avionics system under different influencing factors of the avionics system, the simulation analysis results of the avionics system digital model are determined.

9. The digital model identification method based on avionics system according to claim 8 is characterized in that: In S3, the avionics system digital model identification result is determined, and the following operations are performed: According to the digital model identification requirements based on the avionics system, the avionics system digital model simulation analysis standards are pre-set, and the pre-set avionics system digital model simulation analysis standards are stored; According to the digital model identification requirements of the avionics system, the stored digital model simulation analysis standards of the avionics system are indexed, and the indexed digital model simulation analysis standards of the avionics system are retrieved; Based on the avionics system digital model simulation analysis standard, the avionics system digital model simulation analysis results are digitally modeled and identified to determine the avionics system digital model identification results; When the simulation analysis result of the avionics system digital model is within the simulation analysis standard range of the avionics system digital model, the avionics system digital model identification result is that the avionics system digital model is normal and can correctly realize all expected functions; When the avionics system digital model simulation analysis results are not within the avionics system digital model simulation analysis standard range, the avionics system digital model identification result is that the avionics system digital model is abnormal and cannot correctly implement all expected functions.

10. The digital model identification method based on avionics system according to claim 9 is characterized in that: In S4, the management personnel are intelligently guided to adjust and optimize the digital model of the avionics system and perform the following operations: Obtain the results of the digital model identification of the avionics system; Based on the identification results of the avionics system digital model, the avionics system digital model is mined and analyzed to generate an avionics system digital model identification optimization report; The avionics system digital model identification and optimization report is transmitted to management personnel in a visual form, and management personnel are intelligently guided to adjust and optimize the avionics system digital model and solve abnormal problems of the avionics system digital model in a timely manner.

Images