A method for predicting compression failure region of composite stiffened panel in aircraft structure

Abstract

The application belongs to the technical field of aircraft structure design, and particularly relates to a method for predicting a compression damage region of a composite stiffened panel in an aircraft structure, comprising the following steps: step one, screening important related parameters related to the compression damage of the composite stiffened panel; step two, determining the value range of the important related parameters, sampling and constructing a sample set; step three, constructing a finite element model of the composite stiffened panel, and calculating the damage position of each sample in the sample set under compression load; step four, dividing the composite stiffened panel into multiple regions, and corresponding the damage position of each sample in the sample set under compression load to the corresponding region; step five, based on the corresponding region of each sample in the sample set, constructing a compression damage region prediction model of the composite stiffened panel; and step six, taking the important related parameters as input, and based on the compression damage region prediction model of the composite stiffened panel, calculating the damage region of the composite stiffened panel under compression load.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of aircraft structure design, and particularly relates to a method for predicting a compression failure region of a composite stiffened panel in an aircraft structure. BACKGROUND

[0002] The composite stiffened panel has become a core component of a modern aircraft load-bearing structure due to its excellent performance such as high specific strength and high specific stiffness, and mainly bears compression, bending and torsion loads, in which the compression load is the main factor of failure.

[0003] The typical failure modes of the composite stiffened panel are complex and diverse, including structural instability, interlaminar material failure and interface delamination. Taking a fuselage panel as an example, the structure belongs to a thin-walled stringer system, and when subjected to compression, the skin first appears buckling, and then the unstabilized stringer and the surrounding un-buckled skin jointly bear the load, finally leading to overall instability of the panel. Since the composite stiffened panel bears most of the load, once failure occurs, stress concentration will be caused, which will cause the load bearing of other components to rise sharply, resulting in serious consequences such as shortening of the service life of the aircraft or disintegration of the structure.

[0004] Efficient and accurate analysis of the compression failure behavior of the composite stiffened panel is of great significance to promote the application of the composite stiffened panel in aircraft structures. Currently, in aircraft structure design, engineering estimation method and finite element simulation method are usually used to predict the failure position of the composite stiffened panel under compression load.

[0005] The engineering estimation method is derived from early aircraft design, mainly through engineering handbooks and empirical formulas, based on historical data to induce key features, combined with test data to fit and construct an empirical model for estimation, to predict the failure position of the composite stiffened panel under compression load. The accuracy of the engineering estimation method highly depends on the historical test data which is continuously corrected, and the cycle is long, the cost is high, the cross-scene adaptability is insufficient, and the anisotropy and damage evolution are ignored to simplify the complex behavior of the composite material, which cannot show the damage evolution of the composite stiffened panel and is difficult to meet the demand of rapid iteration design of aircraft structure.

[0006] The finite element simulation method can accurately simulate the whole process of load response and failure of the composite stiffened panel through meshing and parameter setting, but the multi-scale characteristics and anisotropy of the composite material significantly increase the complexity of the finite element simulation modeling. For the prediction of the failure position of the large-scale composite stiffened panel under compression load, the fine finite element simulation model calculation needs to consume a large amount of computing resources, the calculation cost is high, the calculation cycle is long, and it is also difficult to meet the demand of rapid iteration design of aircraft structure.

[0007] The present application is proposed in view of the above technical defects. SUMMARY

[0008] The purpose of the present application is to provide a composite stiffened panel compression failure area prediction method in an aircraft structure, which can quickly and accurately predict the failure area of the composite stiffened panel under compression load, and effectively support the rapid iterative design of the aircraft structure.

[0009] The technical solution of the present application is:

[0010] A composite stiffened panel compression failure area prediction method in an aircraft structure, characterized in that it comprises:

[0011] Step one, screen out important related parameters related to the compression failure of the composite stiffened panel;

[0012] Step two, determine the value range of the important related parameters, sample and construct a sample set;

[0013] Step three, construct a finite element model of the composite stiffened panel, and calculate the failure position of each sample in the sample set under compression load;

[0014] Step four, divide the composite stiffened panel into multiple regions, and correspond the failure position of each sample in the sample set under compression load to the corresponding region;

[0015] Step five, based on the corresponding region of each sample in the sample set, construct a composite stiffened panel compression failure area prediction model;

[0016] Step six, taking the important related parameters as input, based on the composite stiffened panel compression failure area prediction model, calculate the failure area of the composite stiffened panel under compression load.

[0017] Optionally, in the above-mentioned composite stiffened panel compression failure area prediction method in an aircraft structure, in step one, the important related parameters related to the compression failure of the composite stiffened panel are screened out, including the tensile stiffness of the strip direction in the panel surface , the tensile stiffness perpendicular to the strip direction in the panel surface , the ratio of the tensile stiffness of the strip direction to the tensile stiffness perpendicular to the strip direction in the panel surface , the bending section modulus of the strip , the bending stiffness of the skin perpendicular to the strip direction , the ratio of the bending section modulus of the strip to the bending stiffness of the skin perpendicular to the strip direction , the equivalent in-plane tensile / compressive elastic modulus of the strip , the equivalent in-plane tensile / compressive elastic modulus of the skin , the equivalent in-plane tensile / compressive elastic modulus of the whole strip and skin , the width of the skin , the width of the lower edge strip , the width of the web Width of the upper edge strip Number of the rib strips Length of the rib strips .

[0018] Optionally, in the composite stiffened panel compression failure area prediction method of the aircraft structure, in step two, the parameter information of the composite stiffened panel in the aircraft structure is collected, the value range of the important related parameters is determined, sampling is performed, and a sample set is constructed.

[0019] Optionally, in the composite stiffened panel compression failure area prediction method of the aircraft structure, in step four, the composite stiffened panel is evenly divided into multiple areas in the transverse direction according to the number of rib strips, and evenly divided into five areas in the longitudinal direction.

[0020] Optionally, in the composite stiffened panel compression failure area prediction method of the aircraft structure, in step five, a convolutional neural network is used to construct a composite stiffened panel compression failure area prediction model.

[0021] Optionally, in the composite stiffened panel compression failure area prediction method of the aircraft structure, the grid search method or the neural architecture search method is used to iteratively optimize the composite stiffened panel compression failure area prediction model with accuracy as an index.

[0022] The present application has at least the following beneficial technical effects:

[0023] The present application provides a composite stiffened panel compression failure area prediction method for an aircraft structure. The method accurately predicts the failure position of the composite stiffened panel under compression load, converts it into a prediction of a larger failure area, constructs a composite stiffened panel compression failure area prediction model, and predicts the failure area of the composite stiffened panel under compression load. The method greatly improves the modeling and calculation efficiency under the condition of meeting the accuracy requirement, quickly and accurately predicts the failure area of the composite stiffened panel under compression load, and effectively supports the rapid iterative design of the aircraft structure. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is a schematic diagram of the composite stiffened panel compression failure area prediction method for an aircraft structure provided by the embodiments of the present application;

[0025] Figure 2 is a top view of the composite stiffened panel provided by the embodiments of the present application;

[0026] Figure 3 is a side view of the composite stiffened panel provided by the embodiments of the present application;

[0027] Figure 4This is a schematic diagram of dividing a composite stiffening plate into multiple regions, as provided in the embodiments of this application.

[0028] in:

[0029] 1-Skin; 2-Firming strip; 3-Upper edge strip; 4-Body plate; 5-Lower edge strip.

[0030] To better illustrate this embodiment, some content in the accompanying drawings may be omitted, enlarged, or reduced. They are for illustrative purposes only and should not be construed as limiting the scope of this application. Detailed Implementation

[0031] To make the technical solution and advantages of this application clearer, the technical solution of this application will be described in a clearer and more complete manner below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only some embodiments of this application, and are only used to explain this application, not to limit this application. It should be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, and other related parts can be referred to the general design.

[0032] Furthermore, unless otherwise defined, the technical or scientific terms used in this application description shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The word "comprising" as used in this application description indicates that the concept preceding the word encompasses the concepts listed following the word and their equivalents, without excluding other related concepts.

[0033] A method for predicting the compression failure zone of composite stiffened plates in aircraft structures, such as... Figure 1 As shown.

[0034] Step 1: Screen out the important relevant parameters associated with the compressive failure of composite stiffened plates.

[0035] Determine the configuration of the composite stiffened plate, such as Figures 1-2 As shown, it includes a skin 1 and a rib 2, wherein the rib 2 further includes an upper edge strip 3, a web 4, and a lower edge strip 5, and the material is CCF300.

[0036] Based on mechanical analysis and empirical formulas, correlation analysis using the mutual information method was employed to generate a correlation heatmap. Parameters with the highest correlation to compressive failure of composite stiffened plates were selected as key correlation parameters, including the tensile stiffness in the direction of the inner stiffeners. Tensile stiffness in the direction perpendicular to the stiffeners within the plate surface The ratio of the tensile stiffness of the inner stiffeners to the tensile stiffness perpendicular to the stiffeners. Section modulus of stiffeners Bending stiffness of the skin in the direction perpendicular to the stiffeners The ratio of the section modulus of the stiffener to the bending stiffness of the skin in the direction perpendicular to the stiffener. In-plane tensile / compressive modulus of elasticity of ribs In-plane tensile / compressive elastic modulus of skin The elastic modulus of tension / compression in the equivalent plane of the ribs and skin , width of the skin Width of the lower edge strip Width of the web Width of the upper edge strip Number of reinforcing bars and the length of the reinforcing bars .

[0037] Step 2: Determine the value range of important relevant parameters, conduct sampling, and construct a sample set.

[0038] Collect parameter information of composite stiffening plates in aircraft structures, determine the value range of important relevant parameters, conduct sampling, and construct a sample set.

[0039] Can Take a fixed value, Take 1~4, and... , , , , , , , , Take the numerical values ​​corresponding to the four layup sequences of the composite material. , , , Ten discrete values ​​were selected, and an orthogonal experimental design was used to sample them, resulting in 1904 samples, which were then used to construct a sample set.

[0040] Step 3: Construct a finite element model of the composite stiffened plate and calculate the failure location of each sample in the sample set under compressive load.

[0041] Referring to existing physical test specimens of composite stiffened plates, a finite element model is constructed to ensure that the details of the finite element model meet the analysis requirements, including structural shape, internal structure, and connection parts. Axial compression tests are conducted on the physical test specimens, and simulation calculations are performed based on the finite element model. The axial compression test results are compared with the simulation calculation results, the parameters of the finite element model are adjusted, and the finite element model is calibrated to ensure that the calculation accuracy of the finite element model meets the analysis requirements.

[0042] Based on the parameterized method, the finite element model corresponding to each sample in the sample set is constructed in batches, and the failure position of each sample under compression load is calculated. The failure position mentioned here is the initial failure position.

[0043] Step four, divide the composite stiffened plate into multiple regions, and correspond the failure position of each sample in the sample set under compression load to the corresponding region.

[0044] The composite stiffened plate can be divided into multiple regions in the transverse direction according to the number of ribs, and divided into five regions in the longitudinal direction. For a composite stiffened plate with two ribs, as shown in Figure 4 , a total of ten regions are divided, which can be numbered in sequence as 1~10.

[0045] Step five, based on the region corresponding to each sample in the sample set, a composite stiffened plate compression failure region prediction model is constructed.

[0046] The failure position of the composite stiffened plate under compression load is accurately predicted, the model used is complex, the training and prediction time is long, and the computing power requirement is high. However, in actual design, the prediction accuracy of the failure position of the composite stiffened plate under compression load does not need to be extremely accurate, only a certain accuracy is needed. Therefore, by dividing the composite stiffened plate into multiple regions, corresponding the failure position of each sample in the sample set under compression load to the corresponding region, and constructing the corresponding model, the failure region of the composite stiffened plate under compression load is predicted. The modeling and calculation efficiency can be greatly improved under the condition of meeting the accuracy requirement.

[0047] The prediction of the accurate failure position of the composite stiffened plate under compression load is converted into the prediction of a larger failure region, and the prediction task is converted into multi-classification prediction. However, the number of classification labels of each sample in the sample set is unevenly distributed, that is, the failure region is unevenly distributed.

[0048] Convolutional neural network is a deep learning model suitable for various complex tasks, with excellent feature extraction and classification information processing ability. Its architecture design is based on the principle of local receptive field of biological visual system, and through the hierarchical architecture of convolutional layer, pooling layer and full connection layer, an efficient feature learning framework is constructed. It has high processing efficiency for high-dimensional data and is good at processing complex correlation parameters.

[0049] The sample data can be preprocessed by Z-Score standardization, and the composite stiffened plate compression failure region prediction model can be constructed by convolutional neural network. The important related parameters of the composite stiffened plate can be used as input, and the failure region number can be used as output.

[0050] The strategy of combined sampling is used to train the composite stiffened plate compression failure area prediction model, that is, the majority class classification label samples are randomly undersampled, the minority class label samples are randomly oversampled, and the number of minority class label samples is not less than 1 / 5 of the number of majority class label samples, a training set is constructed, and the composite stiffened plate compression failure area prediction model is trained.

[0051] The confusion matrix and its related parameters can be selected, specifically including accuracy, weighted recall rate, weighted precision rate, and weighted score as evaluation parameters, to evaluate the composite stiffened plate compression failure area prediction model.

[0052] The grid search method or neural architecture search method can be used to iteratively optimize the composite stiffened plate compression failure area prediction model with accuracy as an index, to improve the prediction effect of the composite stiffened plate compression failure area prediction model. Specifically, the hidden layer structure, optimizer, learning rate, etc. of the composite stiffened plate compression failure area prediction model can be iteratively optimized, or the bottom structure of the composite stiffened plate compression failure area prediction model can be adjusted based on the convolutional neural network structure, such as adding a residual term, introducing Gaussian noise, etc., to improve the generalization ability of the composite stiffened plate compression failure area prediction model.

[0053] Step six, taking important related parameters as input, based on the composite stiffened plate compression failure area prediction model, calculating the failure area of the composite stiffened plate under compression load.

[0054] After completing the construction of the composite stiffened plate compression failure area prediction model, only the numerical values of the important related parameters of the composite stiffened plate need to be input into the composite stiffened plate compression failure area prediction model, and the failure area of the composite stiffened plate under compression load can be quickly calculated. The composite stiffened plate compression failure area prediction model can be iteratively optimized and reused as needed.

[0055] The above embodiment discloses a method for predicting the compression failure area of a composite stiffened plate in an aircraft structure. The design accurately predicts the failure position of the composite stiffened plate under compression load, converts it into a prediction of a larger failure area, constructs a composite stiffened plate compression failure area prediction model, and predicts the failure area of the composite stiffened plate under compression load. In the case of meeting the accuracy requirement, the modeling and calculation efficiency can be greatly improved, the failure area of the composite stiffened plate under compression load can be quickly and accurately predicted, and the rapid iterative design of the aircraft structure is effectively supported.

[0056] The technical scheme of the present application has been described in combination with the preferred embodiments shown in the drawings, and those skilled in the art should understand that the protection scope of the present application is obviously not limited to these specific embodiments, and those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical schemes after the changes or replacements will fall within the protection scope of the present application.

Claims

1. A method for predicting the compression failure zone of composite stiffened plates in aircraft structures, characterized in that, include: Step 1: Screen out the key relevant parameters associated with compressive failure of composite stiffened plates; Step 2: Determine the value range of important relevant parameters, conduct sampling, and construct a sample set; Step 3: Construct a finite element model of the composite stiffened plate and calculate the failure location of each sample in the sample set under compressive load; Step 4: Divide the composite stiffened plate into multiple regions and map the failure location of each sample in the sample set under compressive load to the corresponding region; Step 5: Based on the regions corresponding to each sample in the sample set, construct a prediction model for the compression failure region of the composite stiffened plate; Step 6: Using key relevant parameters as input, calculate the failure zone of the composite stiffened plate under compressive load based on the composite stiffened plate compression failure zone prediction model; In step five, a convolutional neural network is used to construct a prediction model for the compression failure zone of the composite stiffened plate.

2. The method for predicting the compression failure zone of composite stiffened plates in aircraft structures according to claim 1, characterized in that, In step one, key parameters related to compressive failure of composite stiffened plates are identified, including the tensile stiffness in the direction of the stiffeners within the plate surface. Tensile stiffness in the direction perpendicular to the stiffeners within the plate surface The ratio of the tensile stiffness of the inner stiffeners to the tensile stiffness perpendicular to the stiffeners. Section modulus of stiffeners Bending stiffness of the skin in the direction perpendicular to the stiffeners The ratio of the section modulus of the stiffener to the bending stiffness of the skin in the direction perpendicular to the stiffener. In-plane tensile / compressive modulus of elasticity of ribs In-plane tensile / compressive elastic modulus of skin The elastic modulus of tension / compression in the equivalent plane of the ribs and skin , width of the skin Width of the lower edge strip Width of the web Width of the upper edge strip Number of reinforcing bars and the length of the reinforcing bars .

3. The method for predicting the compression failure zone of composite stiffened plates in aircraft structures according to claim 1, characterized in that, In step two, parameter information of composite stiffening plates in aircraft structures is collected, the value range of important relevant parameters is determined, sampling is performed, and a sample set is constructed.

4. The method for predicting the compression failure zone of composite stiffened plates in aircraft structures according to claim 1, characterized in that, In step four, the composite stiffened plate is divided into multiple regions in the transverse direction according to the number of stiffeners, and into five regions in the longitudinal direction.

5. The method for predicting the compression failure zone of composite stiffened plates in aircraft structures according to claim 4, characterized in that, Using accuracy as an indicator, the prediction model for the compression failure zone of composite stiffened plates is iteratively optimized using either the grid search method or the neural architecture search method.

Images