Composite material bonding residual stress prediction method considering manufacturing history and interface geometry

Abstract

The invention discloses a composite material bonding residual stress prediction method considering manufacturing history and interface geometry, and particularly relates to the field of composite material performance prediction. Comprising the following steps: constructing an initial cementing model of a composite material; obtaining a residual stress field of the panel in an upstream curing process, and mapping the residual stress field to the panel; fitting morphology deviation and local roughness of the surface of the honeycomb core, determining a low-frequency error field and a high-frequency local error field, and superposing the low-frequency error field and the high-frequency local error field to obtain a total error field of the surface of the honeycomb core; mapping the total error field into a glue layer thickness field with spatial variation, and reconstructing the initial glue joint model to obtain a glue joint model; a thermal-chemical-mechanical equation is constructed, simulation of the curing process is carried out based on the cementing model, the thermal-chemical-mechanical equation is solved in a coupling mode, and a total residual stress field and a deformation field are determined. Based on the method, the prediction accuracy can be remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of composite material performance prediction, and particularly relates to a composite material bonding residual stress prediction method considering manufacturing history and interface geometry. BACKGROUND

[0002] The composite panel-honeycomb core sandwich structure is widely used in aerospace, precision instruments and other fields with extremely high requirements for structural lightweight and topography precision due to its excellent specific strength, specific stiffness and designability. In the manufacturing process of the structure, bonding is a key assembly link. During the curing process of the adhesive, significant residual stress will be generated due to the chemical shrinkage of the adhesive, the mismatch of the thermal expansion coefficients between the material components, and the evolution of the modulus of the adhesive layer with the degree of curing. The residual stress causes the product to warp and deform, and stress concentration occurs at the interface, increasing the risk of debonding failure, and seriously affecting the final performance and service reliability of the product.

[0003] At present, the residual stress analysis of the bonding process mainly relies on computer simulation based on the finite element method. However, the existing mainstream analysis model has two inherent limitations: The historical mechanical effects of the manufacturing chain are ignored: traditional models usually assume that the panel is in an ideal state without initial stress at the beginning of bonding. In fact, after the composite panel undergoes the previous manufacturing processes such as prepreg laying, curing, and demolding, there is an inherent residual stress field inside the panel determined by the process history, which is spatially distributed and non-uniform. Ignoring this pre-stress field is equivalent to breaking the mechanical continuity of the manufacturing chain, resulting in a fundamental difference between the initial conditions of the analysis and the actual situation, and the prediction results are biased from the beginning.

[0004] The real bonding interface geometry is simplified: the traditional model considers the bonding interface after honeycomb machining as an ideal geometric plane and assumes that the adhesive layer thickness is uniform. However, the inevitable honeycomb machining errors in actual manufacturing will cause significant spatial non-uniformity in the actual adhesive layer thickness. This geometric non-uniformity will strongly modulate the spatial distribution of chemical shrinkage and thermal stress, and the traditional model cannot reflect this key physical mechanism, leading to distorted predictions of stress transmission paths and final surface accuracy.

[0005] In summary, the existing prediction methods mainly focus on the rapid generation of geometric structures, and the underlying model is still based on the ideal material constitutive relation and the assumption of no initial stress, resulting in low accuracy of residual stress and deformation prediction. SUMMARY

[0006] The main purpose of the present application is to provide a composite material bonding residual stress prediction method considering manufacturing history and interface geometry, aiming to solve the problem of low accuracy of residual stress and deformation prediction in the existing method.

[0007] To achieve the above objectives, this application provides a method for predicting residual stress in composite material bonding that considers manufacturing history and interface geometry. The method includes: constructing an initial bonding model of the composite material, comprising a panel, an adhesive layer, and a honeycomb core; obtaining the residual stress field of the panel during upstream curing and mapping the residual stress field onto the panel; fitting the surface morphology deviation of the honeycomb core and determining the low-frequency error field; obtaining the characteristic dimensions of the honeycomb core, using these dimensions as the upper limit of the frequency band to simulate the local roughness of the honeycomb core surface and determining the high-frequency local error field; superimposing the low-frequency and high-frequency error fields to obtain the total error field of the honeycomb core surface and mapping the total error field to a spatially varying adhesive layer thickness field; reconstructing the initial bonding model based on the adhesive layer thickness field to obtain the bonding model; constructing thermo-chemical-mechanical equations; simulating the curing process based on the bonding model; coupling and solving the thermo-chemical-mechanical equations; and determining the total residual stress field and deformation field.

[0008] Optionally, fitting the morphological deviations of the honeycomb core surface and determining the low-frequency error field includes: using a Zernike polynomial to fit the low-frequency, slowly changing morphological deviations and determining the low-frequency error field; wherein the morphological deviations include warping and bending.

[0009] Alternatively, in the polar coordinate system, the low-frequency error field can be determined as follows:

[0010] In the formula, For the first The coefficients of the Zernike polynomial, For the corresponding Zernike basis functions, These are normalized polar coordinates.

[0011] Optionally, the characteristic dimensions of the honeycomb core are obtained, and the characteristic dimensions of the honeycomb core are used as the upper limit of the frequency band to simulate the local roughness of the surface of the honeycomb core and determine the high-frequency local error field. This includes: obtaining the characteristic dimensions of each cell of the honeycomb core, using the characteristic dimensions as the upper limit of the frequency band, simulating the local roughness related to the processing technology using a bandpass coherent texture model, and determining the high-frequency local error field.

[0012] Optionally, the high-frequency local error field can be determined as follows:

[0013] In the formula, For the first The amplitude of each sinusoidal component, For the first The spatial frequency of a sinusoidal component For the first wave direction angle of the sinusoidal component, is the first random phase angle of the sinusoidal component; The upper limit of the frequency band is related to the characteristic size of the honeycomb lattice.

[0014] Optionally, the thermo-chemical-mechanical equation of the bonding model comprises: a curing kinetics equation for characterizing the evolution law of the adhesive curing degree with curing time and temperature; a thermo-chemical strain equation for characterizing the total strain increment of the adhesive layer; and an adhesive layer constitutive relation equation for characterizing the change of the elastic modulus of the adhesive layer with the curing degree.

[0015] Optionally, the deformation field comprises a surface shape RMS value and a surface shape PV value.

[0016] To achieve the above-mentioned object, the application further provides a composite material bonding residual stress prediction device considering manufacturing history and interface geometry, comprising: a model construction module for constructing an initial bonding model of a composite material, the initial bonding model comprising a panel, an adhesive layer and a honeycomb core; a stress introduction module for obtaining a residual stress field of the panel in an upstream curing process and mapping the residual stress field to the panel; an error simulation module for fitting the surface topography deviation of the honeycomb core and determining a low-frequency error field; obtaining the characteristic size of the honeycomb core, taking the characteristic size of the honeycomb core as the upper limit of the frequency band, simulating the local roughness of the surface of the honeycomb core and determining a high-frequency local error field; superimposing the low-frequency error field and the high-frequency error field to obtain a total error field of the surface of the honeycomb core, and mapping the total error field to a spatially varying adhesive layer thickness field; a model reconstruction module for reconstructing the initial bonding model based on the adhesive layer thickness field to obtain a bonding model; and a simulation module for constructing a thermo-chemical-mechanical equation, simulating the curing process based on the bonding model, coupling the thermo-chemical-mechanical equation and determining a total residual stress field and a deformation field.

[0017] Compared with the prior art, the application has the following beneficial effects: The composite material bonding residual stress prediction method considering manufacturing history and interface geometry of the application can construct a high-fidelity physical prediction framework by introducing the inherent residual stress field of the panel and accurately characterizing the real machining surface topography of the honeycomb core, so that the residual stress and deformation prediction results are closer to the physical actuality, and the prediction accuracy can be significantly improved; by obtaining the residual stress field of the panel in the upstream curing process, fitting the surface topography deviation and local roughness of the honeycomb core, and through simulation calculation, the influence of the manufacturing history and interface geometry on the composite material bonding residual stress and deformation can be quantified, and the understanding of the complex bonding mechanism can be deepened. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 Fig. 1 is a flowchart of a composite material bonding residual stress prediction method considering manufacturing history and interface geometry according to the application; Figure 2 A frequency-domain-modal decomposition diagram of machining surface error of a honeycomb core; Figure 3 A residual stress state diagram before and after the introduction of step S20 in Example 1; Figure 4 A comparison diagram of final deformations predicted for a comparative example and examples; Figure 5 A comparison diagram of predicted deformations and experimentally measured deformations for examples.

[0019] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0021] A first embodiment of the present application provides a composite adhesive residual stress prediction method considering manufacturing history and interface geometry, as shown in Figure 1 and specifically comprising the following steps: Step S1, constructing an initial adhesive model of the composite material, the initial adhesive model comprising a panel, an adhesive layer and a honeycomb core; Since the panel of the composite material has an inherent residual stress field determined by the process history and spatially distributed unevenly after experiencing the previous prepreg laying, curing, demolding and other manufacturing links, the present embodiment maps the inherent residual stress field as the initial stress on the initial adhesive model, in the following specific manner.

[0022] Step S2, obtaining the residual stress field of the panel in the upstream curing process, and mapping the residual stress field to the panel; In the present embodiment, the inherent residual stress field of the panel generated in the upstream curing process (such as prepreg curing and demolding) is defined as the initial stress state of the present adhesive analysis.

[0023] Specifically, in the finite element software, the stress field is mapped to the corresponding grid nodes of the panel in the adhesive model through the initial condition introduction function.

[0024] In the present embodiment, by considering the inherent residual stress field generated by the manufacturing history, the mechanical continuity of the manufacturing chain can be ensured, and the deviation of the prediction result caused by the difference between the initial conditions of the analysis and the actual situation can be avoided.

[0025] Since the honeycomb core machining error in actual manufacturing will cause significant spatial non-uniformity of the actual glue layer thickness, that is, the surface error of the honeycomb core. The embodiment is measured by a three-dimensional topography measuring instrument in a Cartesian coordinate system The surface error data of the honeycomb core actually measured below Frequency-domain-modal decomposition and characterization (decomposition schematic as Figure 2 shown) are carried out to determine the surface error data of the honeycomb core, including low-frequency overall topography error and high-frequency local roughness error, in the following specific manner.

[0026] Step S3, fitting the topography deviation of the surface of the honeycomb core and determining the low-frequency error field; obtaining the characteristic size of the honeycomb core, taking the characteristic size of the honeycomb core as the upper limit of the frequency band, simulating the local roughness of the surface of the honeycomb core, and determining the high-frequency local error field; Specifically, step S31, using Zernike polynomials to fit the low-frequency and slowly changing topography deviation and determine the low-frequency error field; wherein the topography deviation includes warping and bending. Zernike polynomials are obtained by coordinate transformation, mapping the measurement area into a unit circle, and using the first term (usually ) Zernike polynomial to fit, to obtain the error field in polar coordinates , and through coordinate inverse transformation, a continuous low-frequency error field in the Cartesian coordinate system is generated. The expression of the error field in polar coordinates is:

[0027] In the formula, a is the coefficient of the first term Zernike polynomial, is the corresponding Zernike basis function, is the normalized polar coordinate, (θ is the actual distance from the point to the center of the circle, is the radius of the workpiece), is the polar angle.

[0028] Step S32, obtaining the characteristic size of each honeycomb cell of the honeycomb core, taking the characteristic size as the upper limit of the frequency band, and using a bandpass coherent texture model to simulate the local roughness related to the machining process in the Cartesian coordinate system , and determining the high-frequency local error field. The determination method of the high-frequency local error field is as follows: ​

[0029] wherein, is the amplitude (unit: μm), is the spatial frequency (unit: mm -1 ), is the wave direction angle (unit: radian), is the random phase angle (unit: radian), is not greater than the upper limit of the frequency band , the upper limit of the frequency band , is the characteristic size of the honeycomb lattice to ensure that the local texture wavelength to be characterized is not less than the size of the honeycomb lattice unit.

[0030] In this embodiment, the topography deviation and local roughness error of the surface of the honeycomb core generated in the machining process are considered and fitted into a low-frequency error field and a high-frequency local error field, which can reflect the modulation of the spatial distribution of geometric unevenness on chemical shrinkage and thermal stress, avoid causing distortion in the prediction of stress transmission path and surface accuracy, and thus improve the accuracy of residual stress prediction.

[0031] Step S4, superimposing the low-frequency error field and the high-frequency error field to obtain the total error field of the surface of the honeycomb core , and mapping the total error field into a spatially varying adhesive layer thickness field;

[0032] Map this total error field into a spatially varying adhesive layer thickness field :

[0033] wherein, is the nominal thickness of the adhesive layer set by the process, i.e. the thickness of the uniform adhesive layer expected to be formed under ideal conditions.

[0034] Step S5, reconstructing the initial bonding model based on the adhesive layer thickness field to obtain a bonding model; Step S6, constructing a thermal-chemical-mechanical equation, simulating the curing process based on the bonding model, coupling the thermal-chemical-mechanical equation, and determining the total residual stress field and the deformation field.

[0035] wherein, the deformation field includes the surface RMS value and the surface PV value. The curing kinetics parameters in the thermal-chemical-mechanical equation are obtained in advance through DSC testing. The thermal-chemical-mechanical equation of the bonding model includes: a curing kinetics equation for characterizing the curing degree of the adhesive with the evolution law of curing time and temperature , and the expression is:

[0036] wherein, and is the temperature dependent reaction rate constant following the Arrhenius law; is the pre-exponential factor, is the activation energy, is the universal gas constant, is the reaction order.

[0037] Thermo-chemical strain equation for characterizing the total strain increment of the adhesive layer , which is expressed as:

[0038] wherein, is the thermal strain, is the chemical shrinkage strain, , and are the thermal expansion coefficients of the adhesive and the substrate, respectively, is the temperature change amount; , is the chemical shrinkage coefficient of the adhesive, is the degree of cure change amount.

[0039] Adhesive layer constitutive relation equation for characterizing the elastic modulus of the adhesive layer with the change of the degree of cure, which is expressed as:

[0040] wherein, is the initial modulus of the uncured adhesive, is the final modulus of the fully cured adhesive, is the exponential parameter controlling the modulus evolution pattern.

[0041] The total residual stress field calculated by the finite element software should include three parts, which are specifically as follows:

[0042] In the formula, is the residual stress field of the panel, is the stress newly generated by the curing simulation process on the substrate of , is the nonlinear coupling stress caused by the slight change in the geometry of the bonding interface due to the initial deformation of the panel caused by the initial stress of the panel.

[0043] A second embodiment of the present invention provides a composite material adhesive bonding residual stress prediction device considering manufacturing history and interface geometry, comprising: a model building module for building an initial adhesive bonding model of the composite material, the initial adhesive bonding model including a panel, an adhesive layer, and a honeycomb core; a stress introduction module for acquiring the residual stress field of the panel during the upstream curing process and mapping the residual stress field onto the panel; an error simulation module for fitting the surface morphology deviation of the honeycomb core and determining the low-frequency error field; acquiring the characteristic dimensions of the honeycomb core, using the characteristic dimensions of the honeycomb core as the upper limit of the frequency band, simulating the local roughness of the surface of the honeycomb core, and determining the high-frequency local error field; superimposing the low-frequency error field and the high-frequency error field to obtain the total error field of the surface of the honeycomb core, and mapping the total error field to a spatially varying adhesive layer thickness field; a model reconstruction module for reconstructing the initial adhesive bonding model based on the adhesive layer thickness field to obtain the adhesive bonding model; and a simulation module for constructing thermo-chemical-mechanical equations, simulating the curing process based on the adhesive bonding model, coupling and solving the thermo-chemical-mechanical equations, and determining the total residual stress field and deformation field.

[0044] Example Step S10: A three-dimensional initial adhesive bonding model of the composite material is established in the general-purpose finite element software Abaqus / Standard. This model includes the panel, the honeycomb core (using an equivalent solid model and assigning orthogonal anisotropic properties), and the adhesive layer.

[0045] Step S20: Use the residual stress field obtained from the simulation of the upstream panel curing manufacturing process as the initial stress field. In Abaqus, using *INITIAL CONDITIONS, TYPE=STRESS, the mapping is precisely applied to the mesh nodes of the current glued model. (See...) Figure 3 ,from Figure 3 (a) shows the inherent residual stress field cloud map after the panel is cured and demolded, and its distribution has significant non-uniformity; Figure 3 Figure (b) shows the stress contour plot of the rear panel portion after importing this stress field as an initial condition into the adhesive bonding model. (Comparison) Figure 3 As can be seen in (a) and (b), the inherent residual stress field of the panel is completely and accurately inherited into the adhesive bonding analysis model, and its spatial distribution pattern is consistent. This operation uses the stress state determined by the manufacturing history as the starting point for the adhesive bonding analysis, rather than an ideal zero-stress state, thus fundamentally ensuring the physical authenticity of the initial conditions of the simulation.

[0046] Step S30, low-frequency overall topography characterization: in the normalized polar coordinate system, the first 10 Zernike polynomials are used to fit the overall warping, bending and other low-frequency errors, and the key Zernike coefficients are as follows: the defocus term coefficient is +8.3 μm, the 0° astigmatism term coefficient is -4.0 μm, and the 45° astigmatism term coefficient is +3.6 μm.

[0047] High-frequency local error characterization: in the Cartesian coordinate system, the bandpass coherence texture model is used to simulate the local roughness related to the machining tool marks, and the center spatial frequency is set to 0.1 mm -1 , and the frequency band width is 0.05 mm -1 .

[0048] The above low and high frequency error fields are superimposed to obtain the total error field of the honeycomb core surface , which is mapped into the glue layer thickness field, wherein is 120 μm.

[0049] Step S40, in the finite element software, the glue layer geometry is reconstructed according to the thickness field, and the glue joint model is obtained; Step S50, on the basis of the glue joint model, the heat-chemical-mechanical control equation is coupled and executed to perform complete solidification process simulation, wherein the solidification schedule is: heating from 25°C to 130°C at 0.5°C / min, holding for 240 minutes, and then cooling to 60°C at 0.5°C / min, and finally the high-precision residual stress field and structural deformation field are calculated, and the results are shown in Table 1 and Figures 4-5 .

[0050] Comparative Example The stress prediction results of the traditional model are shown in Table 1 and Figure 4 . The prediction method of the traditional model does not include steps 20 and 30.

[0051]

[0052] As shown in Table 1, the surface shape RMS value (24.4 μm) predicted by the method of the embodiment is highly consistent with the experimental measurement value (26.5 μm), and the relative error is only 7.9%. The prediction result (8.2 μm) of the traditional model of the comparative example deviates greatly from the measured value, and the actual deformation is seriously underestimated. Comparing Figure 4 (a) and (b), it can be seen from the deformation cloud map that the method of the embodiment successfully predicts the main deformation modes such as low-order astigmatism existing in the actual structure, and the prediction result of the comparative example is completely distorted. Comparing Figure 5 (a) and (b), the deformation field predicted by the embodiment has high consistency with the experimental measurement result in spatial distribution.

[0053] By the above manner, the prediction method can realize accurate prediction of the composite material bonding residual stress and deformation by systematically introducing the manufacturing history (panel initial stress) and the real interface geometry (spatial variable thickness).

[0054] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the specification and drawings of the present application, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for predicting adhesive residual stress in composite materials considering manufacturing history and interface geometry, characterized in that, include: An initial bonding model for the composite material is constructed, the initial bonding model including a panel, an adhesive layer, and a honeycomb core; Obtain the residual stress field of the panel during the upstream curing process, and map the residual stress field onto the panel; The morphological deviations of the honeycomb core surface were fitted, and the low-frequency error field was determined. The characteristic dimensions of the honeycomb core are obtained, and the characteristic dimensions of the honeycomb core are used as the upper limit of the frequency band to simulate the local roughness of the surface of the honeycomb core and determine the high-frequency local error field. The low-frequency error field and the high-frequency error field are superimposed to obtain the total error field of the surface of the honeycomb core, and the total error field is mapped to a spatially varying adhesive layer thickness field. The initial bonding model is reconstructed based on the adhesive layer thickness field to obtain the bonding model. Thermo-chemical-mechanical equations are constructed, and the curing process is simulated based on the adhesive bonding model. The thermo-chemical-mechanical equations are solved in a coupled manner, and the total residual stress field and deformation field are determined.

2. The method for predicting residual stress in composite material bonding considering manufacturing history and interface geometry according to claim 1, characterized in that, The fitting of the surface topography deviation of the honeycomb core and the determination of the low-frequency error field include: The Zernike polynomial is used to fit the low-frequency, slowly changing morphological deviations and determine the low-frequency error field; wherein the morphological deviations include warping and bending.

3. The method for predicting residual stress in composite material bonding considering manufacturing history and interface geometry according to claim 2, characterized in that, In polar coordinates, the low-frequency error field is determined as follows: In the formula, For the first The coefficients of the Zernike polynomial, For the corresponding Zernike basis functions, These are normalized polar coordinates.

4. The method for predicting residual stress in composite material bonding considering manufacturing history and interface geometry according to claim 1, characterized in that, The process of obtaining the characteristic dimensions of the honeycomb core, using these characteristic dimensions as the upper limit of the frequency band, simulating the local roughness of the honeycomb core surface, and determining the high-frequency local error field includes: The characteristic size of each cell in the honeycomb core is obtained, and this characteristic size is used as the upper limit of the frequency band. A bandpass coherent texture model is used to simulate the local roughness related to the processing technology and to determine the high-frequency local error field.

5. The method for predicting residual stress in composite material bonding considering manufacturing history and interface geometry according to claim 4, characterized in that, The high-frequency local error field is determined as follows: In the formula, For the first The amplitude of each sinusoidal component, For the first The spatial frequency of a sinusoidal component For the first The wave direction angle of a sinusoidal component, For the first Random phase angle of a sinusoidal component; The upper limit of the frequency band is related to the feature size of the cellular grid.

6. The method for predicting residual stress in composite material bonding considering manufacturing history and interface geometry according to claim 1, characterized in that, The thermo-chemical-mechanical equations of the adhesive bonding model include: Curing kinetics equations are used to characterize the evolution of the degree of curing of adhesives with curing time and temperature; Thermochemical strain equations are used to characterize the total strain increment of the adhesive layer; The constitutive equation of the adhesive layer is used to characterize the change of the elastic modulus of the adhesive layer with the degree of curing.

7. The method for predicting adhesive residual stress in composite materials considering manufacturing history and interface geometry according to claim 1, characterized in that, The deformation field includes the surface RMS value and the surface PV value.

8. A device for predicting residual stress in composite material bonding that considers manufacturing history and interface geometry, characterized in that, include: The model building module is used to build an initial bonding model of the composite material, which includes a panel, an adhesive layer, and a honeycomb core. The stress introduction module is used to acquire the residual stress field of the panel during the upstream curing process and map the residual stress field onto the panel. An error simulation module is used to fit the morphological deviation of the honeycomb core surface and determine the low-frequency error field; obtain the characteristic size of the honeycomb core, use the characteristic size of the honeycomb core as the upper limit of the frequency band, simulate the local roughness of the honeycomb core surface, and determine the high-frequency local error field; superimpose the low-frequency error field and the high-frequency error field to obtain the total error field of the honeycomb core surface, and map the total error field into a spatially varying adhesive layer thickness field; The model reconstruction module is used to reconstruct the initial bonding model based on the adhesive layer thickness field to obtain the bonding model. The simulation module is used to construct thermo-chemical-mechanical equations, simulate the curing process based on the adhesive model, solve the thermo-chemical-mechanical equations in a coupled manner, and determine the total residual stress field and deformation field.

Images