Target parameter acquisition method and device for aircraft equivalent stress prediction
By acquiring and classifying parameters by the aircraft's digital main line, combined with dimensionality reduction processing, the problems of data processing complexity and parameter redundancy in the aircraft's structural stress prediction are solved, achieving more efficient prediction accuracy.
Patent Information
- Application Number
- CN202411850041.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-16
- Publication Date
- 2025-04-11
AI Technical Summary
During the aircraft design stage, finite element analysis is extremely labor-intensive and time-consuming due to the complexity of the model and the large amount of computing. When establishing machine learning models, it is difficult to avoid the introduction of redundant parameters or the missed effective parameters, resulting in overfitting or underfitting problems.
The structural stress-related parameters of the target aircraft model are obtained through the aircraft digital main line, and the classification parameters are the first type and the second type parameters, the original parameter set is constructed, and the redundant and invalid parameters are removed through two dimensionality reduction processing to obtain the target parameter set.
It effectively reduces the complexity and time-consuming data processing in aircraft structural stress prediction, removes redundant parameters, and improves the prediction accuracy and efficiency of the model.
Smart Images
Figure CN119808472A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of feature engineering technology, and in particular to a method and device for acquiring target parameters for aircraft equivalent stress prediction. Background Art
[0002] Aircraft are used in high-speed and high-load environments, and have high requirements for structural strength. At the same time, due to the constraints of engine power and load-bearing capacity requirements, their own weight cannot be too large. Therefore, there are a large number of thin-walled structures in aircraft. Thin-walled structures are composed of thin plates, thin shells and slender rods, which can withstand large loads with less weight and less material. In order to ensure that the aircraft achieves the target performance and efficiency, and at the same time control the overall structural stress margin to remain within a safe range, it is necessary to study and analyze the strength and reliability of the aircraft structure during the design phase. At present, this process is highly dependent on finite element analysis. However, due to factors such as the complex geometric structure, large volume, and large number of constructions of the aircraft, the workload of finite element meshing of the model is complex, and the solution and post-processing process has a huge amount of calculations and is time-consuming.
[0003] In the related art, in order to improve the efficiency of finite element simulation in the auxiliary design stage, a data-driven method is usually adopted to organize existing parameters (such as speed, acceleration, temperature, etc.) as input, actual stress or simulated stress as output, and use machine learning models to predict stress changes in aircraft structures under extreme conditions (such as critical design speed, critical temperature, etc.). However, due to the complexity of aircraft structures and the large number of units involved in the development of aircraft models, the number of simulation data in the design stage and the actual measured data in the test stage involved are large and scattered, and the parameters that affect the structural strength are inconsistent in different models of aircraft. Therefore, when selecting input parameters for establishing a machine learning model, redundant parameters may be introduced or valid parameters may be omitted, resulting in overfitting or underfitting problems. Therefore, how to avoid introducing redundant parameters or omitting valid parameters when selecting input parameters for establishing a machine learning model is a problem that needs to be solved urgently. Summary of the invention
[0004] In order to solve the above technical problem or at least partially solve the above technical problem, the present application provides a method and device for obtaining target parameters for aircraft equivalent stress prediction.
[0005] In a first aspect, the present application provides a method for obtaining target parameters for aircraft equivalent stress prediction, the method comprising:
[0006] Based on the aircraft digital main line, obtaining aircraft structural stress related parameters of the target aircraft model; the aircraft digital main line is used to provide full information of multiple aircraft models;
[0007] Based on the correlation degree between each relevant parameter and the aircraft structural stress, the aircraft structural stress related parameters are classified to obtain first type parameters and second type parameters; the correlation degree between the first type parameters and the aircraft structural stress is higher than the correlation degree between the second type parameters and the aircraft structural stress;
[0008] Constructing an original parameter set based on the first type parameters and the second type parameters;
[0009] Performing a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set; the first dimensionality reduction process is used to reduce the feature dimension of each second type parameter;
[0010] The intermediate parameter set is subjected to a second dimensionality reduction process to obtain a target parameter set; the second dimensionality reduction process is used to remove parameters in the intermediate parameter set that are invalid for equivalent stress prediction. As an optional implementation of an embodiment of the present invention, the original parameter set is constructed based on the first type parameter and the second type parameter, including:
[0011] Based on the first type of parameters, a first parameter set is obtained; the first parameter set includes precise parameter values obtained by performing mean calculation on the time series data of each first type of parameter;
[0012] Based on the second type parameters, obtaining a second parameter set; the second parameter set includes fuzzy parameter values obtained by fuzzifying each second type parameter;
[0013] An original parameter set is constructed based on the first parameter set and the second parameter set.
[0014] As an optional implementation manner of the embodiment of the present invention, the acquiring a first parameter set based on the first type parameter includes:
[0015] According to the test data time series file of the first type parameters, average calculation is performed on the time series data of each first type parameter to obtain the precise value of each first type parameter;
[0016] A first parameter set is constructed based on the precise values of the first type parameters.
[0017] As an optional implementation manner of the embodiment of the present invention, the second type of parameters includes quantifiable parameters and non-quantifiable parameters;
[0018] The acquiring a second parameter set based on the second type parameter includes:
[0019] Performing fuzzy coding classification on the non-quantifiable parameters to obtain a non-quantifiable parameter set; the non-quantifiable parameter set includes fuzzy values of each non-quantifiable parameter;
[0020] According to the test data time series file of the quantifiable parameters, characteristic value calculation is performed on the time series data of the quantifiable parameters in the time domain, the frequency domain, and the energy domain respectively to obtain the characteristic vector of each quantifiable parameter;
[0021] Inputting the feature vector of each quantifiable parameter into a preset support vector machine model to obtain a fuzzy value of each quantifiable parameter;
[0022] A second parameter set is constructed based on the fuzzy values of the non-quantifiable parameters and the fuzzy values of the quantifiable parameters.
[0023] As an optional implementation manner of the embodiment of the present invention, performing a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set includes:
[0024] Calculating feature importance of feature vectors of the quantifiable parameters, removing low information density features in the quantifiable parameters, and obtaining a first feature set;
[0025] Based on the regression feature elimination method and the principal component analysis method, high correlation features in the first feature set are removed to obtain a quantifiable parameter set; the quantifiable parameter set includes key feature vectors of each quantifiable parameter;
[0026] An intermediate parameter set is acquired based on the first parameter set, the quantizable parameter set, and the non-quantizable parameter set.
[0027] As an optional implementation manner of the embodiment of the present invention, performing a second dimensionality reduction process on the intermediate parameter set to obtain a target parameter set includes:
[0028] Performing a second dimensionality reduction process on the intermediate parameter set based on a regression feature elimination algorithm to obtain a dimensionality reduction parameter set;
[0029] Dividing the dimension reduction parameter set into a training set and a test set;
[0030] Inputting the training set into a preset stress prediction model to obtain predicted equivalent stress;
[0031] The preset stress prediction model is optimized based on a preset algorithm until the difference between the predicted equivalent stress output by the preset stress prediction model and the actual equivalent stress included in the test set is less than a preset value, and a target parameter set and a target stress prediction model are determined.
[0032] As an optional implementation of the embodiment of the present invention, the acquisition of aircraft structural stress related parameters of the target aircraft model based on the aircraft digital main line includes:
[0033] Based on the model-based system engineering method and asset management shell technology, an aircraft meta-model containing all the information of the aircraft throughout its life cycle is established;
[0034] Building an aircraft digital thread containing full information of multiple aircraft models based on a mapping relationship between model attributes stored in the aircraft meta-model and model attributes stored in multiple data resource management systems;
[0035] Based on the aircraft digital main line, obtain full information of the target aircraft model;
[0036] Data cleaning is performed on the full amount of information of the target aircraft model, attribute information irrelevant to the aircraft structural stress in the full amount of information is deleted, and aircraft structural stress parameters and structural stress-related parameters of the target aircraft model are obtained.
[0037] In a second aspect, the present application provides a target parameter acquisition device for aircraft equivalent stress prediction, comprising:
[0038] An acquisition module, used to acquire aircraft structural stress related parameters of a target aircraft model based on an aircraft digital main line; the aircraft digital main line is used to provide full information of multiple aircraft models;
[0039] a classification module, configured to classify the aircraft structural stress related parameters based on the correlation degree between each related parameter and the aircraft structural stress, to obtain first type parameters and second type parameters; the correlation degree between the first type parameters and the aircraft structural stress is higher than the correlation degree between the second type parameters and the aircraft structural stress;
[0040] A construction module, configured to construct an original parameter set based on the first type parameters and the second type parameters;
[0041] A dimensionality reduction module, configured to perform a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set; the first dimensionality reduction process is configured to reduce the feature dimension of each second type parameter;
[0042] A determination module is used to perform a second dimensionality reduction process on the intermediate parameter set to obtain a target parameter set; the second dimensionality reduction process is used to remove parameters in the intermediate parameter set that are invalid for equivalent stress prediction.
[0043] As an optional implementation of the embodiment of the present invention, the building module includes:
[0044] A first acquisition unit, configured to acquire a first parameter set based on the first type of parameters; the first parameter set includes precise parameter values obtained by performing mean calculation on time series data of each first type of parameter;
[0045] A second acquisition unit, configured to acquire a second parameter set based on the second type parameters; the second parameter set includes fuzzy parameter values obtained by fuzzifying each second type parameter;
[0046] A construction unit is used to construct an original parameter set based on the first parameter set and the second parameter set.
[0047] As an optional implementation manner of the embodiment of the present invention, the first acquisition unit is specifically configured to:
[0048] According to the test data time series file of the first type parameters, average calculation is performed on the time series data of each first type parameter to obtain the precise value of each first type parameter;
[0049] A first parameter set is constructed based on the precise values of the first type parameters.
[0050] As an optional implementation manner of the embodiment of the present invention, the second type parameter includes a quantifiable parameter and a non-quantifiable parameter; and the second acquisition unit is specifically configured to:
[0051] The acquiring a second parameter set based on the second type parameter includes:
[0052] Performing fuzzy coding classification on the non-quantifiable parameters to obtain a non-quantifiable parameter set; the non-quantifiable parameter set includes fuzzy values of each non-quantifiable parameter;
[0053] According to the test data time series file of the quantifiable parameters, characteristic value calculation is performed on the time series data of the quantifiable parameters in the time domain, the frequency domain, and the energy domain respectively to obtain the characteristic vector of each quantifiable parameter;
[0054] Inputting the feature vector of each quantifiable parameter into a preset support vector machine model to obtain a fuzzy value of each quantifiable parameter;
[0055] A second parameter set is constructed based on the fuzzy values of the non-quantifiable parameters and the fuzzy values of the quantifiable parameters.
[0056] As an optional implementation manner of the embodiment of the present invention, the dimension reduction module is specifically used to:
[0057] Calculating feature importance of feature vectors of the quantifiable parameters, removing low information density features in the quantifiable parameters, and obtaining a first feature set;
[0058] Based on the regression feature elimination method and the principal component analysis method, high correlation features in the first feature set are removed to obtain a quantifiable parameter set; the quantifiable parameter set includes key feature vectors of each quantifiable parameter;
[0059] An intermediate parameter set is acquired based on the first parameter set, the quantizable parameter set, and the non-quantizable parameter set.
[0060] As an optional implementation manner of the embodiment of the present invention, the determining module is specifically used to:
[0061] Performing a second dimensionality reduction process on the intermediate parameter set based on a regression feature elimination algorithm to obtain a dimensionality reduction parameter set;
[0062] Dividing the dimension reduction parameter set into a training set and a test set;
[0063] Inputting the training set into a preset stress prediction model to obtain predicted equivalent stress;
[0064] The preset stress prediction model is optimized based on a preset algorithm until the difference between the predicted equivalent stress output by the preset stress prediction model and the actual equivalent stress included in the test set is less than a preset value, and a target parameter set and a target stress prediction model are determined.
[0065] As an optional implementation of the embodiment of the present invention, the acquisition module is specifically used to:
[0066] Based on the model-based system engineering method and asset management shell technology, an aircraft meta-model containing all the information of the aircraft throughout its life cycle is established;
[0067] Building an aircraft digital thread containing full information of multiple aircraft models based on a mapping relationship between model attributes stored in the aircraft meta-model and model attributes stored in multiple data resource management systems;
[0068] Based on the aircraft digital main line, obtain full information of the target aircraft model;
[0069] Data cleaning is performed on the full amount of information of the target aircraft model, attribute information irrelevant to the aircraft structural stress in the full amount of information is deleted, and aircraft structural stress parameters and structural stress-related parameters of the target aircraft model are obtained.
[0070] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method provided in the first aspect of the present application when executing the computer program.
[0071] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method provided in the first aspect of the present application are implemented.
[0072] Compared with the prior art, the technical solution provided by this application has the following advantages:
[0073] The present application provides a method for obtaining target parameters for aircraft equivalent stress prediction. Since the aircraft digital main line is used to provide full information of multiple aircraft models, the aircraft structural stress related parameters of the target aircraft model can be obtained based on the aircraft digital main line. Through the digital main line's associated traceability query of test data, it is ensured that the relevant parameters and data of each version of the design and simulation stages of the specific model can be obtained, which solves the problem of various information not being interoperable throughout the product life cycle. Based on the degree of correlation between each relevant parameter and the aircraft structural stress, the aircraft structural stress related parameters are classified to obtain the first type of parameters and the second type of parameters. Since the correlation between the first type of parameters and the aircraft structural stress is higher than the correlation between the second type of parameters and the aircraft structural stress, the characteristic dimension of the second type of parameters is reduced, and the characteristic vectors with low correlation with the aircraft structural stress in each second type of parameters are removed to obtain the intermediate parameter set, and the intermediate parameter set is further reduced in dimension to remove the parameters in the intermediate parameter set that are invalid for equivalent stress prediction to obtain the target parameter set. Through two dimensionality reduction processes, redundant parameters in the full set of parameters of a specific model of aircraft can be removed, solving the problems faced in the process of using data-driven machine learning models to predict the structural strength of aircraft, such as multiple data sources, complex distribution, and parameter processing methods that rely heavily on experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0074] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0075] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0076] Figure 1 It is a flowchart of a method for obtaining target parameters for aircraft equivalent stress prediction in one embodiment;
[0077] Figure 2 It is a structural schematic diagram of a target parameter acquisition device for aircraft equivalent stress prediction in one embodiment;
[0078] Figure 3 FIG. 1 is a schematic diagram of the structure of a computer device in an embodiment. DETAILED DESCRIPTION
[0079] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the scheme of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.
[0080] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only part of the embodiments of the present application, rather than all of the embodiments.
[0081] Terminology explanation:
[0082] The equivalent stress of aircraft structure refers to the simplification of complex load conditions into equivalent single force in structural design and analysis for the convenience of calculation and analysis. In aviation structural engineering, the concept of equivalent stress is very important and is used to predict the behavior of structures under various individual conditions.
[0083] Feature engineering is an important step in data science and machine learning, which involves extracting, building, selecting, and transforming features (also called attributes or variables) from raw data so that machine learning algorithms can learn more effectively.
[0084] In one embodiment, Figure 1 As shown, a method for obtaining target parameters for aircraft equivalent stress prediction is provided. In an embodiment of the present disclosure, the method includes the following steps:
[0085] S11. Based on the aircraft digital main line, obtain aircraft structural stress related parameters of the target aircraft model.
[0086] The aircraft digital main line is used to provide full information of multiple aircraft models.
[0087] Specifically, after determining the model of the research object, the digital main line is used to achieve the end-to-end engineering horizontal integration and related traceability of the model, and collect and extract all the design, manufacturing, and testing phase data of the target model product, including design specifications, model parameters and simulation results of each version, processes and materials, test data files, etc.
[0088] In some embodiments, the above step S11 (obtaining aircraft structural stress-related parameters of the target aircraft model based on the aircraft digital main line) can be implemented as follows:
[0089] A. Based on the model-based system engineering method and asset management shell technology, an aircraft meta-model containing all the information of the aircraft throughout its life cycle is established.
[0090] Specifically, we first use the model-based systems engineering (MBSE) approach to uniformly model all the information involved in the entire life cycle of historical aircraft (including the entire process from demand, design, simulation, production and manufacturing, testing, and operation and maintenance; which can be further refined to the performance indicators of demand, two-dimensional drawings and three-dimensional models of design, multidisciplinary collaborative simulation parameter settings and simulation results, the manufacturer's machine tool information and material information involved in the parts production and assembly process, the residual stress of the aircraft structure before the test, the physical information signals of all sensors used in the test, the charging conditions on the day of the test, and the weather environment including wind force, humidity, temperature, etc.).
[0091] In the life cycle of different aircraft models, if there is no unified planning in advance, the information such as the tables recorded by the personnel involved in the planning, design, simulation, and testing processes, the software used, the location where the information is stored, and the industrial software that the information reading depends on will be scattered to various departments of various units. This will cause the test personnel to inquire and connect with all the personnel involved in the test phase at that time when they want to obtain specific test data, and the efficiency of collecting data will be extremely slow. Unified modeling is to record all test data with a set of meta models. The meta model here can be understood as the information model of the record model. For example, a time series signal test data file with the suffix sie is a file placed on a hard disk if it is not recorded. However, after unified modeling, with the database as the basic support, a meta model including the test name to which the file belongs, the personnel responsible for laying sensors, the sensor type, the hardware brand of the record, the physical and network addresses of the file storage, and other attributes are established, as well as similar attributes for any other entities including parts drawings, three-dimensional models, etc. With the support of the graph database, it is possible to quickly retrieve and find the full amount of information related to the test or any process in the entire life cycle in a similar way.
[0092] B. Based on the mapping relationship between the model attributes stored in the aircraft meta-model and the model attributes stored in multiple data resource management systems, construct an aircraft digital main line containing full information of multiple aircraft models.
[0093] Among them, the multiple data resource management systems include: a product data management system, a test data management system and an enterprise resource information management system; the full amount of information includes data of the entire life cycle of the aircraft.
[0094] Specifically, a mapping relationship is established between the attributes of the aircraft metamodel and the model attributes stored in the current system, and information stored in existing systems such as PDM (Product Data Management), TDM (Test Data Management), and ERP (Enterprise Resource Planning) is extracted (these information management systems have different attributes due to different users, but in essence they manage data from all aspects of the entire life cycle, such as design, testing, and enterprise asset management. Therefore, there are differences in attribute naming and the number of attributes included. The significance of the digital main line is to unify the format of the data model attributes of the above systems, break through the boundaries, and realize the correlation and traceability of the entire process data), and map them into a unified authoritative data source based on the metamodel.
[0095] Develop a unified data sharing and application service engine. Support the connection of various model data throughout the entire life cycle of aircraft equipment, quickly build graph relationships through graph databases to achieve model information association and traceability, conduct data governance, and provide data analysis services.
[0096] C. Based on the aircraft digital main line, obtain the full amount of information of the target aircraft model.
[0097] D. Perform data cleaning on the full amount of information of the target aircraft model, delete the attribute information irrelevant to the aircraft structural stress in the full amount of information, and obtain the aircraft structural stress related parameters of the target aircraft model.
[0098] Specifically, data cleaning is performed on the entire information of the target aircraft model, and attributes such as personnel and date that are obviously irrelevant to the structural stress are excluded to obtain aircraft structural stress-related parameters of the target aircraft model.
[0099] The disclosed embodiment ensures that relevant parameters and data of all versions of the design and simulation stages of a specific model can be obtained through the associative traceability query of the test data through the digital main line, thereby solving the problem of non-interoperability of PDM, TDM and other information throughout the product life cycle.
[0100] S12. Classify the aircraft structural stress related parameters based on the degree of correlation between each related parameter and the aircraft structural stress to obtain first type parameters and second type parameters.
[0101] The correlation degree between the first type parameter and the aircraft structure stress is higher than the correlation degree between the second type parameter and the aircraft structure stress.
[0102] Specifically, based on historical experience, the aircraft structural stress-related parameters can be divided into important parameters and unimportant parameters, namely, first type parameters and second type parameters, according to the degree of correlation with the aircraft structural stress.
[0103] Exemplarily, since there are limited parameters that have a significant impact on stress, such as speed, acceleration, and the outer shell temperature of the aircraft, the number of first-type parameters is relatively small. During the historical test process, the first-type parameters are recorded by the preset hardware unit to form a test data file, and based on historical experience, these parameters are determined to have a significant impact on stress, so their precise values are calculated based on the test data file.
[0104] The number of second type parameters is very large, but their influence is limited.
[0105] S13. Construct an original parameter set based on the first type parameters and the second type parameters.
[0106] In some embodiments, the above step S13 (building an original parameter set based on the first type parameters and the second type parameters) can be implemented as follows:
[0107] (1) Based on the first type of parameters, obtain a first parameter set.
[0108] The first parameter set includes precise parameter values obtained by performing mean calculation on the time series data of each first type parameter.
[0109] In some embodiments, the above step (1) can be implemented as follows:
[0110] According to the test data time series file of the first type parameters, average calculation is performed on the time series data of each first type parameter to obtain the precise value of each first type parameter;
[0111] A first parameter set is constructed based on the precise values of the first type parameters.
[0112] Among them, the test data time series file refers to the file that records the data at each time point during the test process. This file is usually used to store and analyze test data in order to study and evaluate the test results.
[0113] Specifically, according to the test data time series file of the first type parameters, the time series data of each first type parameter is averaged to obtain the precise value of each first type parameter. Based on the precise value of each first type parameter, a first parameter set is constructed.
[0114] For example, for sensors and simulation parameters that are closely related to the equivalent stress of a specific model, such as time series signals such as velocity and acceleration, angular velocity and angular acceleration, and air pressure, the precise values of the first type parameters are directly obtained by integrating and averaging or calculating the median value over time. Then, the first parameter set is constructed based on the precise values of each first type parameter.
[0115] (2) Based on the second type parameters, obtain a second parameter set.
[0116] The second parameter set includes fuzzy parameter values obtained by fuzzifying each second type parameter. The second type parameters include quantifiable parameters and non-quantifiable parameters. For example, quantifiable parameters are atmospheric temperature, oil level height, etc.
[0117] In some embodiments, the above step (2) can be implemented as follows:
[0118] a. Perform fuzzy coding classification on the non-quantifiable parameters to obtain a non-quantifiable parameter set; the non-quantifiable parameter set includes the fuzzy value of each non-quantifiable parameter.
[0119] Exemplarily, non-quantifiable parameters are, for example, the weather conditions of the day (sunny, cloudy, rainy, etc.). For this type of data, a non-quantifiable parameter set is obtained through fuzzy coding classification. Among them, fuzzy coding classification can be a classification based on fuzzy rules, that is, each feature is assigned a membership value, and the extraction and design of classification rules must meet mutually exclusive rules and exhaustive rules. Further, for example, the atmospheric temperature can be fuzzy processed into high temperature, low temperature, etc., and high temperature and low temperature can be represented by 0 and 1 respectively. The oil level can be fuzzy processed into full oil, high oil, low oil, empty oil, etc.
[0120] b. According to the test data time series file of the quantifiable parameters, characteristic value calculation is performed on the time series data of the quantifiable parameters in the time domain, frequency domain and energy domain respectively to obtain the characteristic vector of each quantifiable parameter.
[0121] For example, the quantifiable parameters can be calculated in the time domain to obtain the maximum value, minimum value, extreme value, skewness, kurtosis, etc.; the quantifiable parameters can be calculated in the frequency domain to obtain the center frequency of different frequency bands, etc.; the quantifiable parameters can be calculated in the energy domain to obtain the energy ratio of each frequency band relative to the whole frequency band, etc. The eigenvector composed of the eigenvalues calculated from the time series file (such as the eigenvector composed of the maximum value, minimum value, etc., if there are n eigenvalues, the dimension of the eigenvector is n).
[0122] c. Input the feature vector of each quantifiable parameter into a preset support vector machine model to obtain a fuzzy value of each quantifiable parameter.
[0123] Exemplarily, after the feature vector obtained in the above steps is input into a preset support vector machine model, the output classification result is obtained, such as acceleration, idling, deceleration, etc., which can be represented by 0, 1, and 2 respectively.
[0124] d. Constructing a second parameter set based on the fuzzy values of the non-quantifiable parameters and the fuzzy values of the quantifiable parameters.
[0125] Specifically, the second parameter set includes: a fuzzy value of a non-quantifiable parameter and a fuzzy value of each quantifiable parameter.
[0126] (3) Construct an original parameter set based on the first parameter set and the second parameter set.
[0127] Specifically, based on the precise values of important parameters and the fuzzy values of unimportant parameters, an original parameter set is constructed, and the original parameter set is used as the input of the preset stress prediction model.
[0128] S14. Perform a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set.
[0129] The first dimensionality reduction process is used to reduce the feature dimension of each second type parameter.
[0130] Exemplarily, for a certain second-type parameter, its eigenvalues calculated include: maximum value, minimum value, extreme value, skewness, center frequency of different frequency bands, the energy ratio of each frequency band relative to the entire frequency band, etc., and the eigenvector composed of the above multiple eigenvalues of the parameter is reduced in dimension to solve the problem that there are too many quantifiable parameters in the second-type parameters and the manual classification speed is too slow.
[0131] It can be understood that the purpose of the first dimensionality reduction process is to obtain a feature vector that can accurately convert the quantifiable parameters in the second type parameters into non-quantifiable labels. This step of dimensionality reduction process is to reduce the feature dimensions in each second type parameter without deleting any parameters. That is, the number of parameters in the intermediate parameter set is not reduced compared to the original parameter set, but the feature dimensions in each second type parameter are reduced.
[0132] In some embodiments, the above step S14 (performing a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set) can be implemented as follows:
[0133] The feature importance is calculated for the feature vectors of the quantifiable parameters, and low information density features in the quantifiable parameters are removed to obtain a first feature set.
[0134] Specifically, for time series signals such as temperature sensors, optical sensors, and internal component impact sensors, which describe a large number of physical quantities that are not closely related to the overall structural strength of the aircraft, directly using them as feature vector dimensions is likely to cause the convergence speed and effect of the machine learning model to decline, resulting in an overfitting model. Therefore, it is necessary to identify these parameters and classify them into specific working conditions.
[0135] For each determined working condition, the time domain, frequency domain, and energy domain eigenvalues of the signal in the time period under the working condition are calculated and combined into a high-dimensional working condition identification feature vector. The feature importance is calculated based on the random forest algorithm. A threshold is set to filter out features with feature importance lower than the threshold to obtain the first feature set.
[0136] Based on the regression feature elimination method and the principal component analysis method, the high correlation features in the first feature set are removed to obtain the target feature set of the quantifiable parameters.
[0137] The target feature set of the quantifiable parameters includes key feature vectors of the quantifiable parameters.
[0138] Specifically, feature importance only involves a single feature. However, due to the correlation in some feature calculation formulas, there may be two or more highly correlated feature combinations, which may cause overfitting of the working condition identification model. Also, due to the existence of highly correlated feature combinations, the deleted feature with the lowest feature importance may actually have the greatest impact on the model. Therefore, the regression feature elimination method is used to input the feature set into the Gaussian kernel support vector machine model to exclude highly correlated features.
[0139] Exemplarily, after selecting the optimal variable set, PCA is used to further eliminate the correlation of eigenvectors in the feature space, obtain the optimal working condition identification feature subset, and input it into the parameter-adjusted support vector machine to achieve rapid and large-scale identification of non-important parameters.
[0140] An intermediate parameter set is acquired based on the first parameter set, the quantizable parameter set, and the non-quantizable parameter set.
[0141] Specifically, the intermediate parameter set is composed of the precise values of the important parameters, the parameter set after the dimension reduction processing of the quantifiable parameters among the unimportant parameters, and the unquantifiable parameter set.
[0142] S15. Perform a second dimensionality reduction process on the intermediate parameter set to obtain a target parameter set.
[0143] The second dimensionality reduction process is used to remove parameters in the intermediate parameter set that are invalid for equivalent stress prediction.
[0144] For example, 100 parameters are associated from the digital main line, of which 30 are first type parameters, 40 are quantifiable parameters of the second type parameters, and 30 are non-quantifiable parameters of the second type parameters. The first dimensionality reduction process is essentially to reduce the dimensionality of the feature vectors of the 40 signal-containing quantifiable parameters respectively, so as to improve the classification accuracy of the accurate signal into a fuzzy label. However, after the first dimensionality reduction process, there are still 40 quantifiable parameters of the second type parameters, but the dimension of the feature vector of each parameter of the 40 parameters is reduced. That is, the number of parameters of the intermediate parameter set obtained after the first dimensionality reduction process has not been reduced, but the characteristic dimension of each second type parameter has been reduced. The second dimensionality reduction process is to reduce the dimensionality of the 100 parameters after the above-mentioned first dimensionality reduction process. After the second dimensionality reduction process, the parameters that are invalid for predicting aircraft stress or the parameters that are easy to cause overfitting are eliminated, and the number of parameters will be reduced, and the number of parameters may become 20 or 10, etc.
[0145] In some embodiments, the above step S15 (performing a second dimensionality reduction process on the intermediate parameter set to obtain a target parameter set) can be implemented as follows:
[0146] A second dimensionality reduction process is performed on the intermediate parameter set based on a regression feature elimination algorithm to obtain a dimensionality reduction parameter set.
[0147] Among them, the regression feature elimination algorithm includes recursive feature elimination and its variants. For example, select a suitable learning algorithm as the basic model and set the target number of features; use the basic model to train the original features and calculate the importance score of the features; sort the features according to the scores and select several features with the lowest scores as the features to be eliminated; eliminate the features to be eliminated from the feature set to obtain a new feature subset; if the number of features reaches the preset target or all features are eliminated, end the algorithm; otherwise, return to the feature importance evaluation step; select the remaining features as the final feature subset. In addition, cross-validation can be used to find the optimal number of features. If reducing features will cause performance loss, no features will be removed.
[0148] Exemplarily, the intermediate parameter set after the dimensionality reduction processing is further subjected to a second dimensionality reduction processing to eliminate data that is invalid for accurately predicting aircraft stress or that easily causes overfitting, thereby obtaining a dimensionality reduction training set.
[0149] The dimension reduction parameter set is divided into a training set and a test set.
[0150] Exemplarily, the dimension reduction parameter set is divided into a training set and a test set.
[0151] The training set is input into a preset stress prediction model to obtain the predicted equivalent stress.
[0152] Specifically, the preset stress prediction model is used to obtain the predicted equivalent stress.
[0153] The preset stress prediction model is optimized based on a preset algorithm until the difference between the predicted equivalent stress output by the preset stress prediction model and the actual equivalent stress included in the test set is less than a preset value, and a target parameter set and a target stress prediction model are determined.
[0154] The preset algorithms include but are not limited to genetic algorithms, entropy optimization algorithms, etc.
[0155] Specifically, the equivalent stress obtained in the test process is used as the output, and the genetic algorithm is used to optimize the weights and thresholds of the preset stress prediction model. Through multiple training of the regression feature elimination method, the parameters that are invalid for predicting the aircraft stress or the parameters that easily cause overfitting are eliminated, and finally the most effective parameters for predicting the aircraft stress are obtained. Before performing the finite element simulation design, the changed parameters and physical quantities can be input to obtain the stress predicted by the model, which can be used as reference information for the finite element simulation design, reducing the workload of multiple simulations.
[0156] The present application provides a method for obtaining target parameters for aircraft equivalent stress prediction. The present application provides a method for obtaining target parameters for aircraft equivalent stress prediction. Since the aircraft digital main line is used to provide full information of multiple aircraft models, the aircraft structural stress related parameters of the target aircraft model can be obtained based on the aircraft digital main line. Through the digital main line, the correlation and tracing query of the test data ensures that the relevant parameters and data of each version of the design and simulation stages of the full amount of a specific model can be obtained, which solves the problem that various information in the entire life cycle of the product is not interoperable. Based on the degree of correlation between each relevant parameter and the aircraft structural stress, the aircraft structural stress related parameters are classified to obtain the first type parameters and the second type parameters. Since the correlation between the first type parameters and the aircraft structural stress is higher than the correlation between the second type parameters and the aircraft structural stress, the characteristic dimension of the second type parameters is reduced, and the characteristic vectors with low correlation with the aircraft structural stress in each second type parameter are removed to obtain the intermediate parameter set, and the intermediate parameter set is further reduced in dimension to remove the parameters that are invalid for equivalent stress prediction in the intermediate parameter set to obtain the target parameter set. Through two dimensionality reduction processes, redundant parameters in the full set of parameters of a specific model of aircraft can be removed, solving the problems faced in the process of using data-driven machine learning models to predict the structural strength of aircraft, such as multiple data sources, complex distribution, and parameter processing methods that rely heavily on experience.
[0157] In one embodiment, Figure 2 As shown, a target parameter acquisition device 200 for aircraft equivalent stress prediction is provided, comprising:
[0158] An acquisition module, used to acquire aircraft structural stress related parameters of a target aircraft model based on an aircraft digital main line; the aircraft digital main line is used to provide full information of multiple aircraft models;
[0159] a classification module, configured to classify the aircraft structural stress related parameters based on the correlation degree between each related parameter and the aircraft structural stress, to obtain first type parameters and second type parameters; the correlation degree between the first type parameters and the aircraft structural stress is higher than the correlation degree between the second type parameters and the aircraft structural stress;
[0160] A construction module, configured to construct an original parameter set based on the first type parameters and the second type parameters;
[0161] A dimensionality reduction module, configured to perform a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set; the first dimensionality reduction process is configured to reduce the feature dimension of each second type parameter;
[0162] A determination module is used to perform a second dimensionality reduction process on the intermediate parameter set to obtain a target parameter set; the second dimensionality reduction process is used to remove parameters in the intermediate parameter set that are invalid for equivalent stress prediction.
[0163] As an optional implementation of the embodiment of the present invention, the building module includes:
[0164] A first acquisition unit, configured to acquire a first parameter set based on the first type of parameters; the first parameter set includes precise parameter values obtained by performing mean calculation on time series data of each first type of parameter;
[0165] A second acquisition unit, configured to acquire a second parameter set based on the second type parameters; the second parameter set includes fuzzy parameter values obtained by fuzzifying each second type parameter;
[0166] A construction unit is used to construct an original parameter set based on the first parameter set and the second parameter set.
[0167] As an optional implementation manner of the embodiment of the present invention, the first acquisition unit is specifically configured to:
[0168] According to the test data time series file of the first type parameters, average calculation is performed on the time series data of each first type parameter to obtain the precise value of each first type parameter;
[0169] A first parameter set is constructed based on the precise values of the first type parameters.
[0170] As an optional implementation manner of the embodiment of the present invention, the second type parameter includes a quantifiable parameter and a non-quantifiable parameter; and the second acquisition unit is specifically configured to:
[0171] The acquiring a second parameter set based on the second type parameter includes:
[0172] Performing fuzzy coding classification on the non-quantifiable parameters to obtain a non-quantifiable parameter set; the non-quantifiable parameter set includes fuzzy values of each non-quantifiable parameter;
[0173] According to the test data time series file of the quantifiable parameters, characteristic value calculation is performed on the time series data of the quantifiable parameters in the time domain, the frequency domain, and the energy domain respectively to obtain the characteristic vector of each quantifiable parameter;
[0174] Inputting the feature vector of each quantifiable parameter into a preset support vector machine model to obtain a fuzzy value of each quantifiable parameter;
[0175] A second parameter set is constructed based on the fuzzy values of the non-quantifiable parameters and the fuzzy values of the quantifiable parameters.
[0176] As an optional implementation manner of the embodiment of the present invention, the dimension reduction module is specifically used to:
[0177] Calculating feature importance of feature vectors of the quantifiable parameters, removing low information density features in the quantifiable parameters, and obtaining a first feature set;
[0178] Based on the regression feature elimination method and the principal component analysis method, high correlation features in the first feature set are removed to obtain a quantifiable parameter set; the quantifiable parameter set includes key feature vectors of each quantifiable parameter;
[0179] An intermediate parameter set is acquired based on the first parameter set, the quantizable parameter set, and the non-quantizable parameter set.
[0180] As an optional implementation manner of the embodiment of the present invention, the determining module is specifically used to:
[0181] Performing a second dimensionality reduction process on the intermediate parameter set based on a regression feature elimination algorithm to obtain a dimensionality reduction parameter set;
[0182] Dividing the dimension reduction parameter set into a training set and a test set;
[0183] Inputting the training set into a preset stress prediction model to obtain predicted equivalent stress;
[0184] The preset stress prediction model is optimized based on a preset algorithm until the difference between the predicted equivalent stress output by the preset stress prediction model and the actual equivalent stress included in the test set is less than a preset value, and a target parameter set and a target stress prediction model are determined.
[0185] As an optional implementation of the embodiment of the present invention, the acquisition module is specifically used to:
[0186] Based on the model-based system engineering method and asset management shell technology, an aircraft meta-model containing all the information of the aircraft throughout its life cycle is established;
[0187] Building an aircraft digital thread containing full information of multiple aircraft models based on a mapping relationship between model attributes stored in the aircraft meta-model and model attributes stored in multiple data resource management systems;
[0188] Based on the aircraft digital main line, obtain full information of the target aircraft model;
[0189] Data cleaning is performed on the full amount of information of the target aircraft model, attribute information irrelevant to the aircraft structural stress in the full amount of information is deleted, and aircraft structural stress parameters and structural stress-related parameters of the target aircraft model are obtained.
[0190] The present application provides a target parameter acquisition device for aircraft equivalent stress prediction. Since the aircraft digital main line is used to provide full information of multiple aircraft models, the aircraft structural stress related parameters of the target aircraft model can be obtained based on the aircraft digital main line. Through the digital main line's associated traceability query of test data, it is ensured that the relevant parameters and data of each version of the design and simulation stages of the specific model can be obtained, which solves the problem of various information not being interoperable throughout the product life cycle. Based on the degree of correlation between each relevant parameter and the aircraft structural stress, the aircraft structural stress related parameters are classified to obtain the first type of parameters and the second type of parameters. Since the correlation between the first type of parameters and the aircraft structural stress is higher than the correlation between the second type of parameters and the aircraft structural stress, the characteristic dimension of the second type of parameters is reduced, and the characteristic vectors with low correlation with the aircraft structural stress in each second type of parameters are removed to obtain the intermediate parameter set, and the intermediate parameter set is further reduced in dimension to remove the parameters in the intermediate parameter set that are invalid for equivalent stress prediction to obtain the target parameter set. Through two dimensionality reduction processes, redundant parameters in the full set of parameters of a specific model of aircraft can be removed, solving the problems faced in the process of using data-driven machine learning models to predict the structural strength of aircraft, such as multiple data sources, complex distribution, and parameter processing methods that rely heavily on experience.
[0191] For the specific limitations on the target parameter acquisition device for aircraft equivalent stress prediction, please refer to the limitations on the target parameter acquisition method for aircraft equivalent stress prediction applied to computer equipment, which will not be repeated here. Each module in the above-mentioned target parameter acquisition device for aircraft equivalent stress prediction can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.
[0192] In one embodiment, a computer device is provided, whose internal structure diagram can be as follows: Figure 3 As shown. The computer device includes a processor 31 and a memory 32 connected via a system bus. The processor 31 of the computer device is used to provide computing and control capabilities. The memory 32 of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. When the computer program is executed by the processor, a method for obtaining target parameters for aircraft equivalent stress prediction is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.
[0193] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0194] In one embodiment, the target parameter acquisition method for aircraft equivalent stress prediction provided by the present application can be implemented in the form of a computer program. The computer program can be used in Figure 3 The computer device is run on the computer device shown. The memory of the computer device can store various program modules constituting the mobile terminal of the computer device, and the computer program composed of the various program modules enables the processor to execute the steps of the target parameter acquisition method for aircraft equivalent stress prediction of the computer device of each embodiment of the present application described in this specification.
[0195] It should be noted that, in this article, the terms "include", "comprises" or any other variations 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. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0196] Those of ordinary skill in the art can understand that the implementation of all or part of the process in the example method described above can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the process of the embodiments of the above methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include ROM (Read-Only Memory), tape, floppy disk, flash memory or optical storage, etc. Volatile memory may include RAM (Random Access Memory) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as SRAM (Static Random Access Memory) and DRAM (Dynamic Random Access Memory).
[0197] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0198] The above is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features applied for herein.
Claims
1. A method for obtaining target parameters for aircraft equivalent stress prediction, characterized in that: The method comprises: Based on the aircraft digital main line, obtain aircraft structural stress related parameters of the target aircraft model; the aircraft digital main line is used to provide full information of multiple aircraft models; Based on the correlation degree between each relevant parameter and the aircraft structural stress, the aircraft structural stress related parameters are classified to obtain first type parameters and second type parameters; the correlation degree between the first type parameters and the aircraft structural stress is higher than the correlation degree between the second type parameters and the aircraft structural stress; Constructing an original parameter set based on the first type parameters and the second type parameters; Performing a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set; the first dimensionality reduction process is used to reduce the feature dimension of each second type parameter; A second dimensionality reduction process is performed on the intermediate parameter set to obtain a target parameter set; the second dimensionality reduction process is used to remove parameters in the intermediate parameter set that are invalid for equivalent stress prediction.
2. The method according to claim 1, characterized in that The constructing an original parameter set based on the first type parameters and the second type parameters includes: Based on the first type of parameters, a first parameter set is obtained; the first parameter set includes precise parameter values obtained by performing mean calculation on the time series data of each first type of parameter; Based on the second type parameters, obtaining a second parameter set; the second parameter set includes fuzzy parameter values obtained by fuzzifying each second type parameter; An original parameter set is constructed based on the first parameter set and the second parameter set.
3. The method according to claim 2, characterized in that The acquiring a first parameter set based on the first type parameter includes: According to the test data time series file of the first type parameters, average calculation is performed on the time series data of each first type parameter to obtain the precise value of each first type parameter; A first parameter set is constructed based on the precise values of the first type parameters.
4. The method according to claim 2, characterized in that: The second type of parameters includes quantifiable parameters and non-quantifiable parameters; The acquiring a second parameter set based on the second type parameter includes: Performing fuzzy coding classification on the non-quantifiable parameters to obtain a non-quantifiable parameter set; the non-quantifiable parameter set includes fuzzy values of each non-quantifiable parameter; According to the test data time series file of the quantifiable parameters, characteristic value calculation is performed on the time series data of the quantifiable parameters in the time domain, the frequency domain, and the energy domain respectively to obtain the characteristic vector of each quantifiable parameter; Inputting the feature vector of each quantifiable parameter into a preset support vector machine model to obtain a fuzzy value of each quantifiable parameter; A second parameter set is constructed based on the fuzzy values of the non-quantifiable parameters and the fuzzy values of the quantifiable parameters.
5. The method according to claim 4, characterized in that The performing a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set includes: Calculating feature importance of feature vectors of the quantifiable parameters, removing low information density features in the quantifiable parameters, and obtaining a first feature set; Based on the regression feature elimination method and the principal component analysis method, high correlation features in the first feature set are removed to obtain a quantifiable parameter set; the quantifiable parameter set includes key feature vectors of each quantifiable parameter; An intermediate parameter set is acquired based on the first parameter set, the quantizable parameter set, and the non-quantizable parameter set.
6. The method according to claim 1, characterized in that The performing a second dimensionality reduction process on the intermediate parameter set to obtain a target parameter set includes: Performing a second dimensionality reduction process on the intermediate parameter set based on a regression feature elimination algorithm to obtain a dimensionality reduction parameter set; Dividing the dimension reduction parameter set into a training set and a test set; Inputting the training set into a preset stress prediction model to obtain predicted equivalent stress; The preset stress prediction model is optimized based on a preset algorithm until the difference between the predicted equivalent stress output by the preset stress prediction model and the actual equivalent stress included in the test set is less than a preset value, and a target parameter set and a target stress prediction model are determined.
7. The method according to claim 1, characterized in that The step of obtaining aircraft structural stress-related parameters of a target aircraft model based on the aircraft digital main line includes: Based on the model-based system engineering method and asset management shell technology, an aircraft meta-model containing all the information of the aircraft throughout its life cycle is established; Building an aircraft digital thread containing full information of multiple aircraft models based on a mapping relationship between model attributes stored in the aircraft meta-model and model attributes stored in multiple data resource management systems; Based on the aircraft digital main line, obtain full information of the target aircraft model; Data cleaning is performed on the full amount of information of the target aircraft model, attribute information irrelevant to the aircraft structural stress in the full amount of information is deleted, and aircraft structural stress parameters and structural stress-related parameters of the target aircraft model are obtained.
8. A target parameter acquisition device for aircraft equivalent stress prediction, characterized in that: include: An acquisition module, used for acquiring aircraft structural stress related parameters of a target aircraft model based on an aircraft digital main line; The aircraft digital main line is used to provide full information of multiple aircraft models; a classification module, configured to classify the aircraft structural stress related parameters based on the correlation degree between each related parameter and the aircraft structural stress, to obtain first type parameters and second type parameters; the correlation degree between the first type parameters and the aircraft structural stress is higher than the correlation degree between the second type parameters and the aircraft structural stress; A construction module, configured to construct an original parameter set based on the first type parameters and the second type parameters; A dimensionality reduction module, configured to perform a first dimensionality reduction process on the second type parameters included in the original parameter set to obtain an intermediate parameter set; the first dimensionality reduction process is configured to reduce the feature dimension in each second type parameter; A determination module is used to perform a second dimensionality reduction process on the intermediate parameter set to obtain a target parameter set; the second dimensionality reduction process is used to remove parameters in the intermediate parameter set that are invalid for equivalent stress prediction.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the target parameter acquisition method for aircraft equivalent stress prediction described in any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the target parameter acquisition method for aircraft equivalent stress prediction described in any one of claims 1 to 7 is implemented.