A method for quantitatively evaluating performance degradation of a L-shaped laminated plate caused by a wrinkle defect

Abstract

This invention discloses a quantitative assessment method for the performance degradation caused by wrinkles in L-shaped laminates, relating to the field of composite material mechanics. The method involves preparing wrinkled and foldless specimens, performing four-point bending load tests on each specimen, and recording the results of the defective specimens at critical failure moments. Based on the actual morphology of the wrinkles, a parametric model of the wrinkles is constructed, establishing a numerical model corresponding to the wrinkle characteristics. Based on the results of the defective specimens, the numerical model is modified to obtain an interlaminar damage model. This interlaminar damage model is highly sensitive to the defect location. Based on the interlaminar damage model, load-displacement curves and failure modes of specimens with different defect locations are obtained, yielding the wrinkle location-performance degradation effect. This application uses the above method to provide crucial theoretical basis and data support for defect assessment, quality control, and location-based damage tolerance design of L-shaped composite components during the manufacturing process.

Description

Technical Field

[0001] This invention relates to the field of composite material mechanics, and in particular to a method for quantitatively evaluating the performance degradation caused by wrinkle defects in L-shaped laminates. Background Technology

[0002] Advanced fiber-reinforced resin matrix composites are increasingly used in high-end equipment manufacturing fields such as aerospace due to their significant advantages such as high specific strength and high specific modulus. Their application is shifting from secondary load-bearing structures to primary load-bearing structures. In new commercial aircraft (such as the Airbus A350 and Boeing 787), the amount of composite materials used has exceeded 50%. With the continuous growth in demand for lightweight structures, composite material structures are gradually developing towards integration and larger scale.

[0003] In aerospace structures, L-shaped components are typical elements connecting critical parts such as the fuselage frame, skin, and wing spars. They are extremely sensitive to out-of-plane loads, and their failure has attracted widespread attention. However, composite material parts are prone to manufacturing defects during the molding process, among which wrinkling defects (or fiber ripple defects) are common and highly detrimental. Especially in the manufacturing process of complex components with curvature (such as L-shaped laminates), the layup is subjected to complex stress states such as bending, shearing, and compression in the corner areas of the mold, which easily leads to the formation of out-of-plane wrinkles.

[0004] Wrinkle defects significantly alter the local stiffness and stress distribution of materials, leading to stress concentration and becoming the initiation point of damage. This severely weakens the load-bearing capacity of structures and even threatens their safety during service. Compared to metallic materials, composite materials are less likely to eliminate or mitigate the effects of such defects through "material yielding," exhibiting a higher sensitivity to defects. Therefore, a systematic study of the impact of wrinkle defects on the mechanical properties and failure behavior of L-shaped laminates is of crucial engineering significance for developing reasonable manufacturing defect acceptance criteria, conducting damage-tolerant design, and ensuring structural integrity.

[0005] The formation of wrinkles in composite materials is mainly due to the anisotropy of the materials and the complexity of the manufacturing process. The causes can be summarized as follows: (1) The thermal expansion coefficients of the components of the composite material and the mold are mismatched, which causes the fibers to shift as the resin softens during the curing process; (2) Different orientations of the lay-up produce inconsistent deformation and mutual constraints during the molding process, resulting in uneven distribution of interlayer pressure; (3) When manufacturing curved structures or parts, the fibers themselves need to be bent and laid, which can easily form wrinkles due to the stacking or slippage of the prepreg; (4) Human error in laying operation or improper control of process parameters can also introduce defects. According to the morphological characteristics of the wrinkles, they can be divided into two categories: the fibers bend in the plane of the single-layer board (in-plane wrinkles), and the fibers bend in the thickness direction (plane normal direction) of the laminate (out-of-plane wrinkles). Among them, out-of-plane wrinkles introduce significant out-of-plane stress and interlayer stress, which can more easily induce delamination and other damage, leading to premature failure of the structure, and thus become the focus and difficulty of research. Currently, significant progress has been made in the study of the impact of wrinkle defects on the properties of composite materials, but these studies mainly focus on flat specimens.

[0006] Studies by Adams et al. have shown that wrinkles significantly reduce the compressive properties of laminates. Hsiao and Daniel found that wrinkles lead to a decrease in the stiffness and strength of composite materials, with interlaminar shear being the dominant failure mode. Bloom et al. used DIC technology to observe a significant strain gradient in the wrinkled region, confirming that wrinkles cause early cracking, leading to a substantial decrease in tensile strength. El-Hajjar et al. pointed out that wrinkles also have a significant adverse effect on the tensile properties of multidirectional laminates.

[0007] Zhu Jun et al. established an analytical model for predicting the equivalent stiffness of laminated plates containing folds. Zeng Wenhao et al. used the Tsai-Wu criterion to predict the initial damage strength of laminated plates containing folds. Mukhopadhyay et al. and Hayman et al. analyzed the influence of folds on the compressive strength and failure mechanism of composite plates through finite element simulation combined with experiments.

[0008] However, research on fold defects in complex structures such as L-shaped curved beams is relatively limited. Hallander et al., Dodwell et al., and Lightfoot et al. explored the generation mechanism of folds during the forming process of bending members, pointing out that the corner area is a fold-prone area. Guillaume et al. found through CT scans that fold defects in the bending area of ​​cap profiles lead to premature structural failure. Feih et al.'s research showed that manufacturing defects can severely reduce the performance of L-shaped joints. In recent years, the team in this application has begun to systematically study the impact of folds on the bending performance of L-shaped laminates. Through experiments and numerical simulations, they found that folds significantly reduce the ultimate bearing capacity of structures and change the damage evolution path. However, current research conclusions are still relatively general, lacking a systematic quantitative analysis of the key geometric parameters of fold defects. This lack of systematicity is mainly reflected in the following aspects: most studies only focus on the weakening of macroscopic mechanical properties by the presence of folds, failing to deeply reveal the quantitative relationship between the geometric characteristics of folds (such as amplitude, wavelength, and distribution range) and the decrease in bearing capacity; and failing to reveal the different effects of folds on the failure mechanism due to their different positions in the R region. There is a lack of detailed numerical simulations and experimental verification regarding how wrinkles alter the complex stress state inside L-shaped laminates, particularly their specific impact on interlaminar stress concentration and damage initiation locations. Furthermore, the failure to establish a clear "defect parameter-performance reduction" relationship model makes it impossible to provide clear manufacturing defect tolerance standards for engineering applications, i.e., the critical size or acceptable range of wrinkle defects that are allowed.

[0009] Therefore, there is an urgent need for a quantitative assessment method for the performance degradation caused by wrinkle defects in L-shaped laminates to improve the above-mentioned problems. Summary of the Invention

[0010] To address the aforementioned issues, this application proposes a comprehensive and systematic quantitative assessment method for the performance degradation caused by wrinkles in L-shaped laminates. This method aims to overcome the limitations of existing assessment methods, which primarily focus on flat specimens and lack control over wrinkle defect parameters in L-shaped laminates. The specific details are as follows: A method for quantitatively evaluating the performance degradation caused by wrinkles in L-shaped laminates includes the following steps: S1. The target fold defect size and target fold defect location are preset. Based on the target fold defect size and target fold defect location, a folded sample is prepared, and a foldless sample is prepared at the same time. S2. Perform a four-point bending load test on the specimens with and without wrinkles, and record the results of the defective specimens at the critical failure moment. The results of the defective specimens include the crack initiation location and the displacement load curve. S3. Based on the actual morphology of the folds, perform parametric modeling of the folds of the sample and establish a numerical model of the sample corresponding to the fold characteristics of the sample. S4. Based on the results of the defective specimen, the numerical model of the specimen is modified to obtain the interlayer damage model of the specimen. The interlayer damage model of the specimen is highly sensitive to the location of the defect. S5. Based on the interlaminar damage model of the specimen, the load-displacement curves and failure modes of specimens at different defect locations are obtained, and the wrinkle location-performance reduction effect is obtained.

[0011] Preferably, the pre-prepared delamination defect in step S1, which involves using prepreg strips as padding during the preparation of a folded sample based on the target fold defect size and location, specifically includes: S101. First, cut the prepreg into the pre-designed size, and then lay it on the outward-convex Invar male mold to obtain the prepreg blank. The thermal expansion coefficients of the prepreg and the Invar male die are matched; During the installation process, each layer of prepreg is compacted using a roller, and every 6 layers are vacuum compacted for 15 minutes. S102. After the prepreg blank that has been laid on the Invar male mold is demolded from the male mold, it is transferred as a whole to the Invar female mold to complete the alignment of the prepreg blank with the Invar female mold. S102. On the prepreg blank inside the Invar mold, auxiliary materials are laid in sequence, sealed and then vacuum compacted. S103. Maintain negative pressure in the vacuum bag and send the Invar female mold into the autoclave for hot pressing. S103. Cut the cured motherboard into dimensions that meet the loading standard to obtain a sample containing wrinkles.

[0012] Preferably, during the installation process, a strip method is used to pre-create wrinkles in the laminate, and the specific steps include: S1011. Cut the prepreg into strips of different widths, with all strips having a 90° fiber orientation, and lay the strips at predetermined layup positions and angles; S1012. By adjusting the number of layers of horizontal strips and their position in the ply structure, the location of wrinkles can be controlled. S1013. By changing the width and number of horizontal bars, the size and severity of the wrinkles can be adjusted.

[0013] Preferably, in the four-point bending loading test in S2, key wrinkle parameters are defined, and influencing parameters are selected from the key wrinkle parameters based on the results of defective specimens. The key fold parameters include: Number of wrinkles on the main interface N This refers to the interface number with the highest fold height, and its value is consistent with the interface position where the insert is laid. fold height h, is defined as the vertical distance from the highest point of the fold to the original interface, and its value is positively correlated with the number of empty layers of the insert; pleat width ΔT The angle corresponding to the center of the R-zone is controlled by the width of the insert. Wrinkle angle θ This refers to the angle position corresponding to the highest point of the fold, which is determined by the placement of the insert. Number of inward radiating layers n i The number of interface layers affected by the wrinkles of the main interface and radiating to the inner side of the R region; Number of outward radiating layers n o The number of interface layers affected by the wrinkles of the main interface, extending to the inner and outer sides of the R region.

[0014] Preferably, the influencing parameters include the number of fold main interfaces, fold angle, fold height, and fold width. A fold configuration matrix is ​​constructed based on the influencing parameters to compare the effects of changes in fold size and position.

[0015] Preferably, the specific content of S3, which involves parametrically modeling the folds of the sample based on their actual morphology and establishing a numerical model of the sample corresponding to the fold characteristics, includes: The actual morphology of the folds was obtained based on high-precision microscopic observation; The actual morphology of the folds was parametrically characterized using Python scripts, and key geometric parameters were extracted, including fold height, fold width, and number of main interfaces. Numerical models of specimens are generated by mesh characterization based on key geometric parameters.

[0016] Preferably, in step S4, based on the defective sample results, the numerical model of the sample is modified to obtain the interlaminar damage model of the sample. This interlaminar damage model is highly sensitive to the defect location. Specific details include: S401. Defect Parameter Extraction and Equivalent Characterization: Based on the non-destructive testing results and cross-sectional observation data of the defective specimen, the geometric feature parameters of the fold defect are extracted, including the fold position, wavelength, amplitude and number of influence layers. The fold morphology is equivalently reconstructed using a parametric method so that the fold parameters in the finite element model are consistent with the actual specimen. S402. Finite element model construction for different defect locations: While maintaining consistent fold geometric parameters, fold defects were introduced in different key areas of the L-shaped laminate (including the inner side of the R-zone, the middle of the R-zone, the outer side of the R-zone, and the straight edge section) to establish four finite element models with different defect locations. Each model used the same material constitutive relation and interlayer interface properties to ensure that the single variable for comparative analysis was the defect location. S403. Introduction and parameter correction of interlaminar damage model: An interlaminar damage model based on interface elements is introduced into the finite element model. The initiation criteria and evolution law of interlaminar damage are defined. The key parameters are inverted and corrected in combination with the experimental results of the specimens, so that the initiation and propagation behavior of interlaminar damage in the simulation process matches the actual specimens. S404 Load-Displacement Response Comparison Analysis: Loading simulations are performed on each finite element model to obtain the corresponding load-displacement curves, which are then compared and analyzed with the experimental test curves of the defective specimens. The consistency between the numerical results and experimental results is evaluated by comparing the initial stiffness, peak load, and failure stage characteristics of the curves. S405. Model Validity Verification and Sensitivity Determination: When the load-displacement curve obtained from the finite element model has a high degree of agreement with the experimental curve, it indicates that the constructed interlaminar damage model can accurately predict the mechanical response of the structure by adjusting the geometric parameters of the folds and their spatial positions. On this basis, the response difference analysis between models at different defect locations verifies that the interlaminar damage model has a high sensitivity to changes in the defect position, thereby achieving an effective quantitative assessment of the performance degradation caused by fold defects.

[0017] Preferably, in S5, the load-displacement curves and failure modes of specimens at different defect locations are obtained based on the interlaminar damage model, and the specific content of the wrinkle location-performance reduction effect includes: 501. Model Construction and Working Condition Setting for Different Defect Locations: Based on the modified interlaminar damage model of the specimen, finite element models of folds located at different main interface locations were constructed while maintaining consistent fold geometric parameters. These included typical locations such as the inner layer (N=7), middle layer (N=11), and outer layer (N=16). At the same time, different fold deflection angle working conditions were set at the same main interface location to analyze the influence of fold angle on structural performance. 502. Load-displacement response acquisition: Apply the same loading boundary conditions as in the experiment to each defect model, conduct nonlinear progressive damage analysis, and obtain the load-displacement curves of each specimen; extract the key characteristic parameters of the curves, including initial stiffness, peak load, and failure stage response characteristics. 503. Quantitative Characterization of Performance Degradation: Using the defect-free specimen as a benchmark, a load-bearing capacity degradation factor (KDF) is defined. By comparing the peak load of each defective specimen with the peak load of the benchmark specimen, the degree of performance degradation under different fold positions and deflection angles is calculated. Among them, the KDF of the inner layer folded specimen is approximately 36.8%, the middle layer 0° folded specimen is approximately 19.8%, the middle layer 30° folded specimen is approximately 21.7%, and the outer layer folded specimen is approximately 9.4%. 504. Failure Mode Identification and Evolution Analysis: Based on simulation results, extract the initiation location and propagation path of interlaminar damage for each model, analyze the influence of different defect locations on the interlaminar damage evolution process, identify the dominant failure modes of the structure, and establish the correspondence between defect locations and failure behaviors. 505. Sensitivity analysis of fold location: By comparing the load-displacement curves and KDF results of specimens with different defect locations, the influence of fold location on the structural bearing capacity was analyzed. The results show that the impact of fold location on bearing capacity is most significant when the fold is located in the inner layer. As the defect location moves to the outer side, the degree of performance loss gradually decreases, indicating that the structure is significantly sensitive to the fold location. 506. Analysis of the influence of fold angle: Under the same main interface position, the mechanical response of folded specimens with different deflection angles was compared. The results showed that the load-displacement curves and peak loads under different deflection angles were relatively similar, and the corresponding KDF loss coefficients were close. This indicates that compared with the defect position, the fold angle has a relatively limited influence on the loss of bearing capacity. 507. Performance Reduction Effect Summary and Model Validation: Based on the above analysis results, a quantitative relationship between the location of folds and the reduction of structural performance is established. The interlaminar damage model constructed is verified to effectively characterize the performance degradation differences caused by different defect locations, providing a basis for the performance evaluation of fold defects in L-shaped laminates.

[0018] In summary, the present invention provides a quantitative evaluation method for the performance degradation caused by wrinkle defects in L-shaped laminates. Compared with traditional technologies, the present invention has the following advantages: 1. This application uses precisely controlled processes such as the "horizontal strip method" to artificially prefabricate wrinkles at different characteristic locations in the R region of an L-shaped laminate. Through systematic experiments, load-displacement curves and failure modes of samples at different defect locations are obtained, intuitively revealing the wrinkle location-performance reduction effect. 2. This application establishes a finite element model of an L-shaped laminate that accurately reflects the location and geometric characteristics of the folds. Using a progressive damage analysis method, it simulates the stress distribution and damage evolution under different fold parameters, explaining from a mechanical perspective why folds at different locations induce different failure behaviors. The focus is on the relative relationship between the fold location and the stress concentration zone. 3. Based on experimental and numerical results, this application extracts the laws governing the influence of wrinkle location on structural performance and attempts to construct a correlation framework of "wrinkle location-performance reduction". The research results will provide key theoretical basis and data support for defect assessment, quality control, and location-based damage tolerance design of L-shaped composite components during the manufacturing process.

[0019] The technical method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] Figure 1 Flow chart of specimen preparation for the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application; Figure 2 Schematic diagram of the dimensions for the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application; Figure 3 Schematic diagram of the fold interface for the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application; Figure 4 Photo of the actual morphology of the fold for the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application, Figure 4 in which (a) is the WN 7-0 fold specimen, Figure 4 in which (b) is the WN 11-0 fold specimen, Figure 4 in which (c) is the WN 16-0 fold specimen, Figure 4 in which (d) is the WN 11-30 fold specimen; Figure 5 For the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application, Figure 5 in which (a) is the WN 7-0 fold specimen, Figure 5 in which (b) is the WN 11-0 fold specimen, Figure 5 in which (c) is the WN 16-0 fold specimen, Figure 5 in which (d) is the WN 11-30 fold specimen; Figure 6 Load-displacement curve of the four-point bending experiment for the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application: (a) specimen without folds; (b) specimen with 0° folds in the inner layer; (c) specimen with 0° folds in the middle layer; (d) specimen with 0° folds in the outer layer; (e) specimen with 30° folds in the middle layer; (f) comparison of the load means of all types of specimens; Figure 7 Matrix of progressive crack evolution of the fold specimen for the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application; Figure 8 Load-displacement curve of the four-point bending simulation for the method of quantitatively evaluating the performance loss caused by fold defects in an L-shaped laminate of the present application: (a) specimen without folds; (b) specimen with 0° folds in the inner layer; (c) specimen with 0° folds in the middle layer; (d) specimen with 0° folds in the outer layer; (e) specimen with 30° folds in the middle layer; (f) comparison of the load means of all types of specimens; Figure 9Stress distributions of different wrinkled specimens in a quantitative evaluation method for performance loss caused by wrinkling defects in an L-shaped laminate of the present application under the same load Figure 9 In (a) is the WN 7-0 wrinkled specimen Figure 9 In (b) is the WN 11-0 wrinkled specimen Figure 9 In (c) is the WN 16-0 wrinkled specimen Figure 9 In (d) is the WN 11-30 wrinkled specimen Specific implementation manners

[0021] The technical method of the present invention will be further described below through the accompanying drawings and embodiments. It should be noted that: unless otherwise specifically stated, the relative arrangements, numerical expressions and values of the components and steps described in these embodiments do not limit the scope of the present application.

[0022] The following description of at least one exemplary embodiment is actually only illustrative and in no way restricts the present application or its application or use.

[0023] Technologies, systems and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, systems and devices should be regarded as part of the specification.

[0024] In all the examples shown and discussed here, any specific value should be construed as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0025] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the field to which the present invention belongs.

[0026] Embodiment 1 In this embodiment, the specimen preparation adopts the prepreg manual laying-autoclave vacuum bag molding process, and its process is as Figure 1 shown. The specific steps are as follows: First, cut the prepreg into a pre-designed size (180*200 mm), then lay and vacuum compact it on the outwardly convex invar male mold, and finally keep the negative pressure of the vacuum bag, and send the mold and the vacuum bag into the autoclave for hot pressing and forming. The size schematic diagram is as Figure 2 shown.

[0027] The prepreg is the T700 grade carbon fiber prepreg provided by Jiangsu Hengshen Co., Ltd. (model: EV201-35%-12KHF10-U-200gsm-1000). The mechanical property parameters of the single-layer prepreg are provided by the manufacturer and are shown in Table 1.

[0028] Table 1 Prepreg Material Parameters

[0029] The mold is made of Invar steel with an extremely low coefficient of thermal expansion, which is highly compatible with the coefficient of thermal expansion of the carbon fiber composite material. This allows the mold and the sample to expand and contract synchronously during the heating and cooling stages of the curing process, thereby ensuring extremely high dimensional accuracy of the sample while minimizing deformation caused by their interaction.

[0030] During the installation process, to avoid air pockets and voids, each layer of prepreg is compacted using a roller, and every 6 layers are vacuum compacted for 15 minutes.

[0031] To achieve precise control over the characteristics of wrinkle defects, a strip method is used to prefabricate wrinkle defects in the laminate, such as... Figure 2 As shown, the prepreg is first cut into strips of varying widths. To avoid cutting the fibers and maintain fiber continuity as much as possible, all strips are laid at a 90° fiber orientation. These prepreg strips are then laid at predetermined layup positions and angles. Because the strips form an additional thickness pad in local areas, when subsequent prepreg layers are laid, the upper layers cannot fully adhere to the substrate surface in these areas, resulting in localized warping and ultimately wrinkle defects. The location of wrinkles can be controlled by adjusting the number of strips laid and their position in the layup structure; the size and severity of wrinkles can be adjusted by changing the width and number of strips. Unlike methods in existing literature, this paper does not create voids in the layup. Instead, it reduces the impact of the strips on the overall thickness of the laminate through multiple vacuum compactions and the combined effect of resin flow and mold pressure during curing, thereby achieving controlled prefabrication of wrinkle defects while maintaining structural integrity.

[0032] After the sample is laid out, it is placed into an autoclave along with the mold, where it is cured at high temperature and pressure while maintaining negative pressure in the vacuum bag. The cured motherboard is then cut to dimensions that meet the loading standards to obtain the final sample.

[0033] This application defines six key fold parameters to quantify its geometric characteristics, such as Figure 3 As shown: (1) Number of main interfaces of folds ( N This refers to the interface number with the highest fold height, and its value corresponds to the interface position where the insert is laid (e.g., if laid between layers 9 and 10, then...). N =10); (2) Fold height ( h ), defined as the vertical distance from the highest point of the fold to the original interface, its value is positively correlated with the number of blank layers of the insert; (3) fold width ( ΔT ), with the included angle corresponding to the center of the R area, controlled by the width of the insert; (4) Wrinkle angle ( θ ), refers to the angular position corresponding to the highest point of the fold, which is determined by the embedding position of the insert; (5) number of inward radiating layers ( n i () and (6) outward radiation layers ( n o ), representing the number of interface layers affected by the main interface wrinkles and radiating to the inner and outer sides of the R region, respectively.

[0034] Of the six parameters mentioned above, the main ones that can be directly controlled through the process include the number of fold interfaces, fold angle, fold height, and fold width. To systematically study the influence of different fold morphologies on the mechanical behavior of laminates, this application constructs a fold configuration matrix based on these four parameters to compare the effects of changes in fold size and position.

[0035] It should be noted that due to the flow behavior of the resin during curing, coupled with the influence of voids in the layup, the wrinkle characteristics in the actual molded specimens may deviate from the theoretically preset values. Therefore, all analyses in this application are based on the wrinkle morphology actually measured after curing.

[0036] Table 2. Fold parameters for different fold configurations

[0037] The specimen naming convention adopted in this application uses the "WN layer position angle" format, the meaning of which can be clearly understood through specific examples. For example, the specimen "WN_11_0" indicates that the specimen contains a wrinkle defect (WN), the main interface of the wrinkle is located at the interface between the 11th and 12th layers (layer position 11), and the angle between the center line of the wrinkle and the reference axis of the laminate is 0° (angle 0), that is, the wrinkle direction is consistent with the reference direction. Another example is "WN_11_30", which indicates that the wrinkle is also located at the 11th / 12th interface, but with a 30° deflection angle. As a contrast, specimens without wrinkles are marked as "NW" (No Wrinkle). This naming system systematically identifies different wrinkle configurations through the two key parameters of layer position and angle, providing a clear classification basis for subsequent mechanical property comparison and analysis, as shown in Table 2.

[0038] According to ASTM D6415, a four-point bending fixture was used to load the specimens. The fixture consists of upper and lower loading components, with a cylindrical loading rod mounted on ball bearings. The test was conducted on an MTS universal testing machine, with the load applied via a hydraulic piston. Displacement control was used during loading at a rate of 0.5 mm / min, and the controller automatically acquired and recorded the displacement and load during the loading process. To capture crack initiation and propagation, a high-speed camera with an ultra-high sampling rate of 10,000 frames per second was used, providing high temporal resolution data for observing crack dynamics.

[0039] Figure 4 Microscopic images show the actual morphology of four typical wrinkle defects prefabricated in L-shaped laminates using the "insertion method". Each sub-figure (ad) presents the local morphology of the wrinkles at different circumferential angles (ΔT = 50°, 46°, 34°, 36°). The images show that the wrinkles exhibit a clear, continuous arc-shaped wavy structure with regular surface texture and distinct edge contours, indicating that the prefabrication process has good controllability and repeatability.

[0040] In all samples, the fold height was consistently controlled at the design value h = 0.21 mm, consistent with the preset height, indicating that the insert thickness and process pressure parameters were reasonably matched, effectively achieving precise control of the fold geometry. Simultaneously, the main interface position of the folds was accurately set through the layup sequence. The folds in the figure were all located at the preset interlayer interfaces, with no interface drift or interlayer penetration observed, verifying the reliability of the "insertion method" in locating the main interface.

[0041] Furthermore, the measured values ​​of the circumferential angles of the folds matched the design target well, and the angular deviations of each specimen were within acceptable limits. The fold morphology maintained a high degree of regularity at different angles, with no obvious local distortion or collapse, further demonstrating that the preparation method possesses good process stability in simultaneously controlling key geometric parameters such as fold height, main interface position, and circumferential angle. The fold prefabrication process adopted in this application can achieve controllable preparation of fold morphology, providing a reliable sample basis for subsequent research on the influence of fold defects with different geometric characteristics on mechanical properties.

[0042] Figure 5 A partial view of the parametric finite element model based on the actual morphology of the folds is shown. After obtaining the fold geometry through high-precision microscopic observation, a Python script was used to parametrically characterize the fold contour, extracting key geometric parameters including fold height, wavelength, and interface position. Based on this, a mesh model that closely matches the actual fold height was generated. As shown in the figure, the model successfully reproduces the arcuate undulation characteristics of the folds and the local deflection morphology of the fiber layup. The mesh is finely divided and closely matches the actual geometry, ensuring the accurate transfer of the stress field during the calculation. This modeling method achieves a high-fidelity conversion from the actual morphology to the simulation model, providing a reliable geometric foundation for subsequent microscopic stress analysis and damage evolution simulation under fold defects.

[0043] Figure 6 Load-displacement curves of L-shaped laminates with different fold configurations in four-point bending tests are shown. All specimens exhibit typical linear elastic rising segment and brittle fracture characteristics, with a cliff-like drop after reaching the peak load. Among them, the unfolded specimen (NW) has the highest peak load (approximately 114 N / mm).

[0044] After the introduction of wrinkles, the load-bearing capacity of the specimens was significantly reduced. Specimens with wrinkles located at different interface positions exhibited marked performance degradation: the wrinkled specimen WN located in the inner layer (N=7) showed the following degradation. 7-0 The load-bearing capacity was most severely reduced, with a KDF reduction factor of approximately 36.8%; the 0° folded specimen WN located in the middle layer (N=11) 11-0 The loss factor is approximately 19.8%; WN, a wrinkled specimen with the same main interface position but a 30° offset angle. 11-30 The loss factor is approximately 21.7%; while the folds in the sample WN are located in the outer layer (N=16). 16-0 The impact on load-bearing capacity is relatively small, with a KDF loss factor of approximately 9.4%. It is worth noting that when the folds are located on the same main interface, changes in the fold angle have a limited impact on load-bearing capacity loss—compared to WN. 11-0 With WN 11-30 It is evident that the peak loads of the two are not significantly different, and their KDF loss coefficients are similar.

[0045] The results reveal that the location of folds in the structure is a key factor affecting the reduction of load-bearing capacity. The closer the main interface is to the inner side of the L-shaped laminate corner, the greater the KDF reduction coefficient. While changes in the fold angle do not significantly affect the degree of reduction in ultimate load-bearing capacity, they do regulate the damage evolution path, thereby altering the failure mode of the structure. Overall, fold defects trigger interlayer delamination prematurely by inducing local stress concentration, and the severity of the damage is closely related to the interlayer location of the folds.

[0046] Figure 7 The system demonstrates the crack evolution process of L-shaped laminate specimens with different fold configurations under static load. The figure is arranged in a 3 (time series stages) × 4 (specimen type) matrix, clearly presenting the damage development process of four typical folded specimens from the initial state to final failure under four-point bending loads. Initial delamination refers to the crack initiation stage captured at an ultra-high sampling frequency of 10,000 frames / second, while stable delamination characterizes the final damage morphology recorded after delamination propagation reaches a steady state.

[0047] In the unloaded state (left column), the surfaces of each sample were smooth, the layer texture was clear, and no visible damage was observed. In the initial delamination state (middle column), microcracks in the matrix had initiated and aggregated. The initiation location of the cracks showed a direct spatial correlation with the geometry of the folds. Cracks preferentially originated at fold protrusions or abrupt changes in curvature, confirming the decisive role of fold defects as stress concentration sources in damage initiation. In the stable delamination state (right column), macroscopic cracks had fully propagated. It could be observed that the main crack path highly coincided with the orientation of the original folds. All cracks initiated from the folded region and mainly propagated to both sides along the extension direction of the folds, forming through-crack bands.

[0048] WN 7-0 (7 / 8 interface, 0° fold): After crack initiation, it propagates rapidly, with the largest propagation range. The crack path is relatively straight but the propagation distance is the longest. The crack front is relatively flat, indicating that the outer fold has the weakest constraint on crack propagation and the lowest damage tolerance. WN 11-0 (11 / 12 interface, 0° fold): The crack also originates in the folded region, and the propagation path shows slight fluctuations. The crack front exhibits localized tortuosity, indicating that the interface's ability to constrain crack propagation is weakened. WN 16-0 (16 / 17 interface, 0° fold): The crack initiates at the fold protrusion, with a relatively straight propagation path, steadily extending along the interlayer interface. The propagation range is moderate, and the crack front is relatively regular, indicating that the fold at this location has a strong constraint effect on crack propagation. WN 11-30 (11 / 12 interface, 30° deflection fold): The propagation path exhibits a clear asymmetric morphology, with the crack front deflected to one side and the propagation direction altered, demonstrating the significant regulatory effect of the fold angle on the crack propagation path. Although the folds in different groups differ in specific location, angle, and curvature, the crack propagation of all samples follows the same pattern: crack initiation and propagation are constrained within the pre-existing fold morphology and develop along its geometric path. This observation directly proves that the geometric characteristics of the pre-existing folds provide a predetermined weak path for damage evolution, dominate the crack initiation location and propagation direction, and are a key precursor factor determining the final failure mode of the composite laminate. In the stable delamination state (right column), the samples exhibit a typical interlaminar delamination failure morphology, with a clearly visible penetrating dark main crack propagating along the interlaminar interface.

[0049] This set of images visually reveals the core mechanism by which folded defects influence damage evolution: the local geometry of the folds is the dominant factor in crack initiation, and the protrusions or curvature changes are the preferred locations for stress concentration and damage initiation. The location of the folds (inner or outer layer) mainly controls the extent and degree of constraint of crack propagation. Folds located in the inner layer are more likely to induce large-scale, low-constraint delamination propagation, which is consistent with the WN (wound-nail) defect in mechanical testing. 7-0The result directly corresponds to the most severe loss of the specimen's load-bearing capacity (KDF=36.8%). Crack path control: The angle (offset angle) of the folds significantly affects the crack propagation path and direction, resulting in an asymmetric morphology at the crack front.

[0050] The visualization results directly confirm from the microscopic morphology that the geometric characteristics (location and angle) of the wrinkle defects, by controlling the initiation location and propagation behavior of cracks, ultimately determine the macroscopic failure mode and the degree of load-bearing capacity loss of the laminate.

[0051] Figure 8 The figure shows a comparison of normalized force-displacement curves obtained from experiments and simulations for L-shaped laminates with different fold configurations under four-point bending loads. In the figure, EXP represents the experimental curve, and FEA represents the finite element analysis curve. The black scatter points EXP in the figure are experimental data, and the red solid line FEA is the finite element simulation result. Subfigures (a) to (d) correspond to the unfolded specimen (NW) and specimens with four different fold positions and angles (WN7-0, WN11-0, WN11-30, and WN16-0), respectively.

[0052] Overall, the force-displacement curves of all specimens exhibit a typical linear elastic rising phase followed by a sharp drop in load after reaching the peak. The finite element model can reproduce the overall trend of the experimental curves well, especially showing a high degree of agreement in the elastic phase and near the peak load. In the post-peak phase, there is a certain deviation between the simulated curves and the peak values ​​of the experimental data for some specimens, mainly manifested in the simulated curve peak being lower than that of the experimental curve. This may be due to the variation in specimen thickness caused by the inserts during the preparation of the wrinkled specimens.

[0053] Overall, all curves exhibit typical linear elastic rising phases and a sharp drop in load after reaching the peak. The finite element model shows good consistency in reproducing the overall trend of the experimental curves, with a high degree of agreement between simulation and experiment. During the peak phase, the simulated curves for some wrinkled specimens show a slightly lower peak load than the experimental data. This difference may be related to the local thickness variation introduced by the "insertion method" during the preparation of the wrinkled specimens. This process variation may lead to non-uniformity in the thickness direction of the actual specimens, thus affecting their bending stiffness and load-bearing capacity. The simulation model assumes an ideal geometry and fails to fully reflect this thickness fluctuation.

[0054] Specifically, the simulation and experimental curves for the wrinkle-free specimen (NW) showed a high degree of agreement throughout, verifying the reliability of the model under defect-free conditions. In the wrinkled specimen, the WN located in the middle layer... 7-0 and WN 11-0 The simulation and experimental curves of the folded configuration show good agreement in terms of peak load and initial stiffness, indicating that the model can effectively reflect the influence of fold location on bearing capacity. Meanwhile, the outer WN layer... 16-0The simulated curve for the wrinkled specimen showed a slightly slower decline after the peak value, which may be related to the more significant interlayer delamination behavior during actual damage evolution. 11-30 The simulation and experimental curves of the sample (30° angular fold) both showed a relatively gentle downward trend after reaching the peak, and the model successfully captured the regulatory effect of the fold angle on the damage propagation path.

[0055] As shown in subfigure (f), the simulation model systematically reflects the reduction in load-bearing capacity caused by different fold configurations: the closer the fold is to the inner layer, the more significant the decrease in load-bearing capacity; the fold angle has a relatively small impact on the peak load. This comparison verifies the effectiveness of the established finite element model in predicting the bending mechanical behavior of L-shaped laminates with fold defects, providing a reliable tool for subsequent parametric analysis and damage tolerance assessment.

[0056] Figure 9 Simulation results of radial stress distribution along different analysis paths (path 1–path 3) of an L-shaped laminate with wrinkles under four-point bending loads are presented. In the figure, the black solid line represents the defect-free specimen (NW), and the red solid line represents specimens with different wrinkle configurations (WN). Overall, the presence of wrinkles significantly alters the internal stress distribution characteristics of the structure, especially causing obvious stress concentration in the area surrounding the wrinkles.

[0057] On path 1 (Figure a), the radial stress distribution of the defect-free specimen is relatively gentle, with a maximum stress of approximately 25 MPa. However, the stress in the wrinkled specimen WN... 7-0 (Main interface 7, fold angle 0°) Significant stress peaks appear in the normalized distance range of 0.4–0.6, with a maximum value of about 43 MPa, which is 72% higher than that of the defect-free sample, indicating that the folds form a significant stress concentration zone at this location.

[0058] On path 2 (Figure b), WN 11-0 The radial stress distribution of the specimen exhibits a "bimodal" characteristic under the influence of folds, with the two stress peaks occurring near normalized distances of 0.3 and 0.7, respectively, reaching a maximum of 38 MPa. Furthermore, the overall stress level is significantly higher than that of the defect-free specimen. This indicates that folds exert a dual interference on the interlayer stress transmission path, leading to a more complex stress distribution.

[0059] Another comparison along path 2 (Figure d) further verifies the influence of fold location on stress concentration: WN 16-0 The stress concentration effect was most significant in the sample (main interface 16, near the outer layer), with a peak stress close to 50 MPa, which is nearly 100% higher than that of the defect-free sample. The width of the stress concentration zone also increased, indicating that the closer the folds are to the surface, the stronger their disturbance to the local stress field.

[0060] On path 3 (Figure c), WN contains a 30° deflection angle fold. 11-30 The sample also exhibited a stress peak (approximately 35 MPa) near the normalized distance of 0.5, but the overall distribution was relatively gentle, with the peak stress decreasing by about 8% compared to WN11-0. This indicates that adjusting the fold angle can partially alleviate the stress concentration, but cannot completely eliminate its effect.

[0061] In summary, simulation results show that wrinkles significantly increase the interlayer radial stress level and induce stress concentration by locally creating fiber orientation anomalies and geometric discontinuities. The degree of stress concentration intensifies as the wrinkle location approaches the surface, while the deflection of the wrinkle angle can alleviate the stress peak to some extent. This result provides a mechanical explanation for the decrease in the load-bearing capacity of L-shaped laminates caused by wrinkles and offers a theoretical basis for suppressing the harmful effects of wrinkles through optimized layup design and process control.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical methods of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical methods of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical methods to deviate from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for quantitatively evaluating the performance degradation due to a wrinkle defect in an L-shaped laminated sheet, characterized by, Includes the following steps: S1. The target fold defect size and target fold defect location are preset. Based on the target fold defect size and target fold defect location, a folded sample is prepared, and a foldless sample is prepared at the same time. S2. Perform a four-point bending load test on the specimens with and without wrinkles, and record the results of the defective specimens at the critical failure moment. The results of the defective specimens include the crack initiation location and the displacement load curve. S3. Based on the actual morphology of the folds, perform parametric modeling of the folds of the sample and establish a numerical model of the sample corresponding to the fold characteristics of the sample. S4. Based on the results of the defective specimen, the numerical model of the specimen is modified to obtain the interlayer damage model of the specimen. The interlayer damage model of the specimen is highly sensitive to the location of the defect. S5. Based on the interlaminar damage model of the specimen, the load-displacement curves and failure modes of specimens at different defect locations are obtained, and the wrinkle location-performance reduction effect is obtained.

2. The method of claim 1, wherein the L-shaped laminate sheet is characterized in that, In S1, the process of preparing pleated specimens based on the target pleat defect size and location, using prepreg strips as padding to prefabricate delamination defects, specifically includes: S101. First, cut the prepreg into the pre-designed size, and then lay it on the outward-convex Invar male mold to obtain the prepreg blank. The thermal expansion coefficients of the prepreg and the Invar male die are matched; During the installation process, each layer of prepreg is compacted using a roller, and every 6 layers are vacuum compacted for 15 minutes. S102. After the prepreg blank that has been laid on the Invar male mold is demolded from the male mold, it is transferred as a whole to the Invar female mold to complete the alignment of the prepreg blank with the Invar female mold. S102. On the prepreg blank inside the Invar mold, auxiliary materials are laid in sequence, sealed and then vacuum compacted. S103. Maintain negative pressure in the vacuum bag and send the Invar female mold into the autoclave for hot pressing. S103. Cut the cured motherboard into dimensions that meet the loading standard to obtain a sample containing wrinkles.

3. The method of claim 2, wherein the L-shaped laminate sheet is characterized in that, During the installation process, the insert strip method is used to pre-create wrinkles in the laminate. The specific steps include: S1011. Cut the prepreg into strips of different widths, with all cross strips having a 90° fiber direction, and lay the cross strips at predetermined layup positions and angles; S1012. By adjusting the number of layers of horizontal strips and their position in the ply structure, the location of wrinkles can be controlled. S1013. By changing the width and number of horizontal bars, the size and severity of the wrinkles can be adjusted.

4. The method of claim 3, wherein the L-shaped laminate sheet is characterized in that, In the four-point bending loading test in S2, key wrinkle parameters are defined, and influencing parameters are obtained from the key wrinkle parameters based on the results of defective specimens. The key fold parameters include: fold main interface number N , the interface number with the largest fold height, the value of which is consistent with the interface position of the inserted strip crease height h defined as the vertical distance from the crease peak to the original interface, which positively correlates with the number of grafted layers Rug width ΔT The angle measurement of the R region center corresponds to the control of the strip width; folding angle θ , the angle position corresponding to the highest point of the fold, determined by the embedding position of the insert The number of layers of radiation inward n i The number of interface layers affected by the interface fold to the inside of the R region The number of layers radiated outward n o The number of layers affected by the interface fold to the inside and outside of the R region.

5. The method of claim 4, wherein the L-shaped laminate sheet is characterized in that, The influencing parameters include the number of fold main interfaces, fold angle, fold height, and fold width. A fold configuration matrix is ​​constructed based on the influencing parameters to compare the effects of changes in fold size and position.

6. The method of claim 5, wherein the L-shaped laminate sheet is characterized in that, The specific content of S3, which involves parametric modeling of the folds of the sample based on the actual morphology of the folds and establishing a numerical model of the sample corresponding to the fold characteristics, includes: The actual morphology of the folds was obtained based on high-precision microscopic observation; The actual morphology of the folds was parametrically characterized using Python scripts, and key geometric parameters were extracted, including fold height, fold width, and number of main interfaces. Numerical models of specimens are generated by mesh characterization based on key geometric parameters.

7. The method of claim 6, wherein the L-shaped laminate sheet is characterized in that, In S4, based on the results of defective specimens, the numerical model of the specimen is modified to obtain the interlaminar damage model of the specimen. The interlaminar damage model of the specimen is highly sensitive to the location of defects. Specific details include: S401. Defect Parameter Extraction and Equivalent Characterization: Based on the non-destructive testing results and cross-sectional observation data of the defective specimen, the geometric feature parameters of the fold defect are extracted, including the fold position, wavelength, amplitude and number of influence layers, and the fold morphology is equivalently reconstructed using a parametric method. S402. Construction of Finite Element Models for Different Defect Locations: While maintaining consistent fold geometry parameters, fold defects were introduced in different regions of the L-shaped laminate to establish four finite element models for different defect locations; each model used the same material constitutive relation and interlayer interface properties. S403. Introduction and parameter correction of interlaminar damage model: An interlaminar damage model based on interface elements is introduced into the finite element model. The initiation criteria and evolution law of interlaminar damage are defined. The key parameters are inverted and corrected in combination with the experimental results of the specimens, so that the initiation and propagation behavior of interlaminar damage in the simulation process matches the actual specimens. S404 Load-Displacement Response Comparison Analysis: Loading simulations are performed on each finite element model to obtain the corresponding load-displacement curves, which are then compared and analyzed with the experimental test curves of the defective specimens. The consistency between the numerical results and experimental results is evaluated by comparing the initial stiffness, peak load, and failure stage characteristics of the curves. S405. Model Validity Verification and Sensitivity Determination: When the load-displacement curve obtained from the finite element model has a high degree of agreement with the experimental curve, it indicates that the constructed interlaminar damage model can accurately predict the mechanical response of the structure by adjusting the geometric parameters of the folds and their spatial positions. On this basis, the response difference analysis between models at different defect locations verifies that the interlaminar damage model has a high sensitivity to changes in the defect position, thereby achieving an effective quantitative assessment of the performance loss due to fold defects.

8. The method of claim 7, wherein the L-shaped laminate sheet is characterized in that, S5 uses the interlaminar damage model to obtain load-displacement curves and failure modes for specimens at different defect locations, and obtains the specific content of the wrinkle location-performance reduction effect, including:

501. Model construction and working condition setting for different defect locations: Based on the modified interlaminar damage model of the specimen, finite element models of folds located at different main interface positions were constructed while keeping the geometric parameters of the folds consistent. At the same time, different fold deflection angle working conditions were set at the same main interface position to analyze the influence of the fold angle on the structural performance.

502. Load-displacement response acquisition: Apply the same loading boundary conditions as in the experiment to each defect model, conduct nonlinear progressive damage analysis, and obtain the load-displacement curves of each specimen; extract the key characteristic parameters of the curves, including initial stiffness, peak load, and failure stage response characteristics.

503. Quantitative characterization of performance loss: Using the defect-free specimen as a benchmark, define the load-bearing capacity loss coefficient. By comparing the peak load of each defective specimen with the peak load of the benchmark specimen, calculate the degree of performance loss under different fold positions and deflection angles.

504. Failure Mode Identification and Evolution Analysis: Based on simulation results, extract the initiation location and propagation path of interlaminar damage for each model, analyze the influence of different defect locations on the interlaminar damage evolution process, identify the dominant failure modes of the structure, and establish the correspondence between defect locations and failure behaviors.

505. Sensitivity analysis of fold location: By comparing the load-displacement curves and KDF results of specimens with different defect locations, the influence of fold location on the structural bearing capacity was analyzed. The results show that the impact of fold location on bearing capacity is most significant when the fold is located in the inner layer. As the defect location moves to the outer side, the degree of performance loss gradually decreases, indicating that the structure is significantly sensitive to the fold location.

506. Analysis of the influence of fold angle: Under the same main interface position, the mechanical response of folded specimens with different deflection angles was compared. The results showed that the load-displacement curves and peak loads under different deflection angles were relatively similar, and the corresponding KDF loss coefficients were close. This indicates that compared with the defect position, the fold angle has a relatively limited influence on the loss of bearing capacity.

507. Performance Reduction Effect Summary and Model Validation: Based on the above analysis results, a quantitative relationship between the location of folds and the reduction of structural performance is established. The interlaminar damage model constructed is verified to effectively characterize the performance degradation differences caused by different defect locations, providing a basis for the performance evaluation of fold defects in L-shaped laminates.

Images