A composite fan blade tenon defect identification and simulation method considering wrinkle angle distribution

Abstract

This invention provides a method for identifying and simulating defects in the tenon of composite fan blades, considering the distribution of wrinkle angles. It pertains to the field of evaluating the mechanical properties of tenon structures in resin-based composite fan blades. First, it acquires a ply cross-sectional image of the composite fan blade tenon, then performs image recognition to obtain the ply centerline coordinates. Subsequently, a curve segmentation method based on the second derivative is developed to separate the fiber wrinkled segments and normal bending segments in the ply, and wrinkle angles are measured. The mapping relationship between wrinkle angles and tenon spatial coordinates is characterized, and a finite element calculation model and calculation method considering the distribution of wrinkle defects are established. This invention achieves accurate identification of wrinkle defects generated during the manufacturing process of resin-based composite materials and accurate simulation of complex ply structures, improving the ability to evaluate and optimize the mechanical properties of fan blade tenon structures.

Description

Technical Field

[0001] This invention belongs to the field of mechanical performance evaluation of tenon structures in resin-based composite material fan blades, and specifically relates to a method for identifying and simulating defects in tenons of composite material fan blades that considers the distribution of wrinkle angles. Background Technology

[0002] Composite materials, due to their excellent properties such as lightweight, high strength, and corrosion resistance, have been widely used in aerospace, automotive manufacturing, and construction engineering. Particularly in the aerospace field, composite materials are used to manufacture critical components of aircraft, such as wings, fuselages, and engines. The performance of these components directly affects the safety and economic performance of the aircraft. However, during the manufacturing process of composite materials, the curing shrinkage characteristics of resin-based composites easily lead to wrinkling defects. These defects not only affect the geometric appearance quality of the material but, more importantly, reduce the mechanical properties of the structure, increase the risk of delamination and fiber breakage, thereby affecting the safety and reliability of the entire structure.

[0003] The tenon, as a crucial connecting structure for fixing composite fan blades to the rotor disk, bears centrifugal loads and aerodynamic bending moments during operation. It is an area prone to stress concentration and fatigue damage on the fan blades; therefore, the structural integrity of the tenon is fundamental to ensuring the safe operation of the engine. Compared to metal blade tenons, composite blade tenons have a larger number of ply layers. Thickness variations are achieved in the head section through numerous decreasing ply layers to meet shape design requirements. However, the complex ply design easily leads to defects in the tenon, among which wrinkling defects are one of the most common defects in laminated composite structures. These defects may be caused by factors such as pressure differences during molding, resin consolidation, local compression, and localized layer bending. The presence of wrinkles can lead to stress concentration in localized areas, increasing the risk of interlaminar crack initiation and propagation, thereby affecting the reliability and lifespan of the composite fan blades. Therefore, accurate identification and assessment of wrinkling defects are crucial for ensuring the safety of composite structures.

[0004] Traditional wrinkle detection methods mainly rely on manual visual inspection or simple image processing techniques, which suffer from low efficiency, high subjectivity, and difficulty in quantification. With the development of computer vision and image processing technologies, automated defect detection methods are gaining increasing attention. These methods acquire images of composite material tenon cross-sections, reconstruct ply information based on image recognition algorithms, and use numerical methods to identify and quantify wrinkles, thus improving the accuracy and efficiency of detection.

[0005] However, currently, there are many problems in identifying and simulating wrinkles and other defects that may arise during the manufacturing process of composite material fan tenons, including but not limited to:

[0006] 1) Accurate identification of wrinkle defects: Traditional wrinkle detection methods rely on manual visual inspection or simple image processing techniques. These methods are often inefficient and difficult to accurately identify and quantify the size, shape and distribution of wrinkle defects.

[0007] 2) Accurate Quantification of Wrinkle Angle: The wrinkle angle is the angle between the fiber wrinkle path and the theoretical direction, and it is also a key parameter for assessing the degree of fiber wrinkling. However, in the tenon area of ​​composite materials, in addition to the layup deviation caused by wrinkles, there will also be natural bending caused by changes in the tenon geometry. Bending is different from wrinkles, which are defects caused by manufacturing; it is a phenomenon expected in the design. Therefore, when performing wrinkle angle statistics, it is crucial to distinguish between wrinkled segments and normal bending segments.

[0008] 3) Spatial distribution characterization of wrinkle defects: Establish the mapping relationship between the size of the wrinkle angle and the spatial coordinates of the tenon in order to more accurately quantify the impact of wrinkle defects on the structural mechanical properties.

[0009] 4) Simulation calculation of composite tenon considering the distribution of wrinkle angle: A calculation model and method are needed that can take into account the non-uniformity of wrinkle defect distribution and evaluate the mechanical properties of composite material structures. Summary of the Invention

[0010] To address the aforementioned technical problems and overcome the limitations of existing methods in handling complex ply structures and wrinkle angle distributions, this invention provides a method for identifying and simulating tenon defects in composite material fan blades that considers wrinkle angle distribution. The method obtains high-resolution images of the tenon ply cross-section and uses image recognition methods to accurately reconstruct the ply centerline, acquiring its coordinate information. The second derivative of the ply centerline is used to assist in identifying wrinkled regions, and wrinkle angles are measured within these regions. A mapping relationship between wrinkle angles and tenon position coordinates is established, and corresponding mechanical models are applied to different tenon positions based on the size of the wrinkle angles, thus realizing a composite material tenon calculation method that considers wrinkle defects. This invention can improve the quality and efficiency of composite material structure fabrication and accelerate the application of resin-based composite materials in the field of aero-engines.

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] A method for identifying and simulating tenon defects in composite fan blades considering the distribution of wrinkle angles, comprising the following steps:

[0013] Step 1: Obtain a high-resolution image of the tenon ply cross section and use image recognition methods to accurately reconstruct the ply centerline and obtain its data coordinate information;

[0014] Step 2: Construct a piecewise method based on the second derivative of the curve to help identify the folded regions of the layup;

[0015] Step 3: Establish a method for calculating fold angles and determine the fold angles and their coordinates in the folded area of ​​the layer;

[0016] Step 4: Establish a proxy model of the tenon position coordinates and fold angle based on the data-driven method, and map the proxy model to the finite element simulation model of the tenon to provide fold angle information for subsequent simulation analysis;

[0017] Step 5: Establish a simulation model of the composite material tenon that reflects the distribution of the wrinkle angle, and adopt the corresponding mechanical model for each area of ​​the tenon based on the degree of wrinkling.

[0018] Furthermore, in step 1, the minimum resolution of the tenon structure sample image satisfies the following: clearly distinguishing different plies, the position of a single ply, and the thickness of the ply; the result of the reconstruction of the ply centerline accurately reflects the ply information of the original tenon structure; the centerline is the centerline of a single ply inside the tenon, which is determined by the upper and lower boundaries of the single ply in the laying direction.

[0019] Furthermore, in step 2, the segmentation method based on the second derivative of the curve includes: calculating the second derivative of the obtained ply centerline, and using the root mean square value of the second derivative of the curve as the dividing boundary to divide the ply centerline into non-folded segments and folded segments.

[0020] Furthermore, in step 3, the method for calculating the fold angle includes: firstly, solving for the tangent at each point on the fold segment of the ply centerline, and calculating the angle θ between the tangent and the straight line connecting the beginning and end coordinates of the fold segment, i.e., the fold angle.

[0021] Furthermore, in step 4, the surrogate model for establishing the tenon position coordinates and the fold angle based on the data-driven method is to establish the mapping relationship between the tenon position coordinates and the fold angle using a neural network method, and then reflect the established surrogate model on the tenon analysis model through the user material subroutine or in the form of a field.

[0022] Furthermore, in step 5, the actual distribution relationship of wrinkles and defects in the composite tenon is mapped to the finite element model based on the surrogate model; and based on the size of the wrinkle angle, a corresponding mechanical model is used for each integration point of the tenon to evaluate the impact of wrinkles and defects on the overall structural performance of the composite tenon.

[0023] The beneficial effects of this invention are as follows:

[0024] 1. This invention, based on a second-order derivative curve segmentation method, can effectively assist in identifying wrinkled regions in ply layups. This helps identify and resolve potential problems during the design phase, thereby improving the quality and reliability of composite tenons.

[0025] 2. Based on the definition of wrinkle angle, this invention proposes a method for calculating the wrinkle angle of a real tenon structure ply, which overcomes the traditional wrinkle quantification method that relies on manual vision or simple image processing technology, and greatly improves the efficiency and objectivity of wrinkle angle calculation for complex structures.

[0026] 3. This invention characterizes the mapping relationship between the tenon position coordinates and the fold angle based on a data-driven method, and takes into account the non-uniformity of the fold angle distribution. Compared with traditional calculation methods, it greatly improves the simulation accuracy. Attached Figure Description

[0027] Figure 1 This is a schematic diagram for identifying the center line of the composite material fan tenon ply of the present invention;

[0028] Figure 2 This is a schematic diagram of the centerline partitioning and wrinkle angle calculation of the composite material fan tenon ply of the present invention;

[0029] Figure 3 This is a schematic diagram showing the distribution of the wrinkle angle of the tenon in the composite material fan of the present invention;

[0030] Figure 4 A schematic diagram of the neural network model established for the wrinkle angle distribution of the composite material fan tenon of the present invention;

[0031] Figure 5 This is a schematic diagram of the wrinkle angle distribution mapped in the finite element model of the composite material fan tenon of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some examples of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] The present invention provides a method for identifying and simulating tenon defects in composite material fan blades that considers the distribution of wrinkle angles, comprising the following steps:

[0034] Step 1: Use a high-resolution imaging device to acquire images of the tenon area of ​​the composite material, ensuring sufficient detail for subsequent image processing and analysis, and obtain the coordinates of the ply centerline of the tenon based on image recognition methods.

[0035] Step 2: Calculate the second derivative of the ply centerline and use the root mean square value of the second derivative amplitude as the dividing line. Regions with second derivative values ​​less than this dividing line are considered non-wrinkled regions, thus aiding in the identification of wrinkled segments in the ply. By calculating the second derivative of the ply centerline in the image, the boundaries and severity of wrinkle defects can be determined.

[0036] Second-order derivative of the ply centerline The calculation method is as follows: ;

[0037] The method for calculating the dividing boundary of a curve is as follows: ;

[0038] Where x is the vertical coordinate value of the pixel point on the center line of the ply obtained by identification, y is the horizontal coordinate value of the pixel point on the center line of the ply obtained by identification, and i is the i-th pixel point. i Let be the vertical coordinate of the i-th pixel, and n be the total number of pixels along the center line of the layer. is the root mean square value of the second derivative of the pixel on the center line of the layer.

[0039] Step 3: For each point on the folded curve, find the tangent line and calculate the angle between the tangent line and the straight line connecting the beginning and end coordinates of the folded curve. That is, the fold angle.

[0040] Wrinkle angle The calculation formula is:

[0041] ;

[0042] Among them, (x0,y0), (x N ,y N These are the first and last coordinate points of the folded curve, respectively. Let be the first derivative value at the coordinates of the i-th point of the folded curve, and arccos is the inverse cosine function.

[0043] Step 4: Based on the BP (Back Propagation) neural network model, quantify the relationship between the distribution of tenon fold angles and spatial coordinate positions. The input of the training samples is the spatial coordinates of the tenon ply centerline, and the output is its fold angle. After training, a surrogate model that meets the accuracy requirements is obtained.

[0044] Step 5: Establish a finite element model of the composite tenon. Based on the wrinkle angle distribution relationship obtained in Step 4, divide the tenon model into regions and apply a corresponding mechanical model to each region. A neural network model of the relationship between the wrinkle angle and position coordinates is used to provide the wrinkle angle data of the corresponding region. Finally, different mechanical properties are assigned to regions with different wrinkle degrees to calculate their impact on the overall structural performance of the composite tenon.

[0045] Example:

[0046] An embodiment of the present invention provides a method for identifying and simulating tenon defects in composite material fan blades considering the distribution of wrinkle angles, comprising:

[0047] Step 1: Use a mobile phone or camera to acquire images of the composite material tenon. The minimum resolution of the tenon image must be sufficient to clearly distinguish different ply positions, individual ply locations, and ply thicknesses. The reconstructed ply centerline results must accurately reflect the ply information of the original tenon structure, such as... Figure 1 As shown.

[0048] Step 2: Calculate the second derivative of the obtained ply centerline, and use the root mean square value of the second derivative of the curve as the dividing boundary to split the ply centerline into non-folded and folded segments, such as... Figure 2 As shown.

[0049] Step 3: For each point on the aforementioned ply centerline fold segment, calculate the tangent line and the angle between the tangent line and the straight line connecting the beginning and end coordinates of the fold segment. For example, its fold angle, such as Figure 3 As shown.

[0050] Step 4: Establish a BP neural network model, which includes an input layer, hidden layers, and an output layer. The units in each layer are called neurons. The spatial coordinates of the tenon pavement centerline serve as the input, and the corresponding fold angle is the output. The input and output layers are connected by hidden layers; for example... Figure 4 As shown, neurons are connected to other neurons in adjacent layers through weights, biases, and activation functions, and finally, a neural network model with satisfactory accuracy can be obtained through training.

[0051] Step 5: Import the proxy model from Step 4 into the composite tenon finite element model. This allows mapping the distribution of the actual tenon's wrinkle angles into the finite element model. Based on the size of the wrinkle angles, appropriate mechanical models are applied to different locations of the tenon to further study the influence of wrinkle defects on the mechanical properties of the composite tenon. Figure 5 The distribution of the fold angles is shown in the figure.

[0052] The above embodiments are provided merely for the purpose of describing the present invention and are not intended to limit the scope of the invention. The scope of the invention is defined by the appended claims. Various equivalent substitutions and modifications made without departing from the spirit and principles of the invention should be covered within the scope of the invention.

Claims

1. A method for identifying and simulating tenon defects in composite material fan blades considering the distribution of wrinkle angles, characterized in that, Includes the following steps: Step 1: Obtain a high-resolution image of the tenon ply cross section and use image recognition methods to accurately reconstruct the ply centerline and obtain its data coordinate information; Step 2: Construct a piecewise method based on the second derivative of the curve to help identify the folded regions of the layup; The segmentation method based on the second derivative of the curve includes: calculating the second derivative of the obtained ply centerline, and using the root mean square value of the second derivative of the curve as the dividing boundary to divide the ply centerline into non-wrinkled segments and wrinkled segments, that is, the curve region with the second derivative value less than the dividing boundary is regarded as the non-wrinkled region, thereby helping to identify the wrinkled segment curve of the ply; by calculating the second derivative of the ply centerline in the image, the boundary and severity of the wrinkle defect are determined. Second-order derivative of the ply centerline The calculation method is as follows: ; The method for calculating the dividing boundary of a curve is as follows: ; Where x is the vertical coordinate value of the pixel point on the center line of the ply obtained by identification, y is the horizontal coordinate value of the pixel point on the center line of the ply obtained by identification, and i is the i-th pixel point. i Let be the vertical coordinate of the i-th pixel, and n be the total number of pixels along the center line of the layer. The root mean square value of the second derivative of the pixel on the center line of the layup; Step 3: Establish a method for calculating fold angles, and determine the fold angles and their coordinates in the folded region; the method for calculating fold angles includes: For each point on the folded curve, the tangent line is calculated, and the angle between the tangent line and the straight line connecting the beginning and end coordinates of the folded curve is determined. That is, the fold angle: ； Among them, (x0,y0), (x N ,y N These are the first and last coordinate points of the folded curve, respectively. is the first derivative value at the coordinates of the i-th point of the folded curve, and arccos is the inverse cosine function; Step 4: Establish a proxy model of the tenon position coordinates and fold angle based on the data-driven method, and map the proxy model to the finite element simulation model of the tenon to provide fold angle information for subsequent simulation analysis; Step 5: Establish a simulation model of the composite tenon that reflects the distribution of the fold angle, and adopt the corresponding mechanical model for each region of the tenon based on the degree of folding; map the actual fold defect distribution of the composite tenon to the finite element model based on the surrogate model; and adopt the corresponding mechanical model for each integration point of the tenon based on the size of the fold angle to evaluate the impact of fold defects on the overall structural performance of the composite tenon.

2. The method for identifying and simulating tenon defects in composite material fan blades considering wrinkle angle distribution as described in claim 1, characterized in that, In step 1, the minimum resolution of the tenon structure sample image satisfies the following: clearly distinguishing different plies, the position of a single ply, and the thickness of the ply; the result of the reconstruction of the ply centerline accurately reflects the ply information of the original tenon structure; the centerline is the centerline of a single ply inside the tenon, which is determined by the upper and lower boundaries of the single ply in the laying direction.

3. The method for identifying and simulating tenon defects in composite material fan blades considering wrinkle angle distribution as described in claim 1, characterized in that, In step 4, the proxy model for establishing the tenon position coordinates and the fold angle based on the data-driven method is to use a neural network method to establish the mapping relationship between the tenon position coordinates and the fold angle, and then reflect the established proxy model on the tenon analysis model through the user material subroutine or in the form of a field.

Images