CN112632813B - Optimization method of curing system of large-thickness resin-based composite material - Google Patents
Optimization method of curing system of large-thickness resin-based composite material Download PDFInfo
- Publication number
- CN112632813B CN112632813B CN202011410760.1A CN202011410760A CN112632813B CN 112632813 B CN112632813 B CN 112632813B CN 202011410760 A CN202011410760 A CN 202011410760A CN 112632813 B CN112632813 B CN 112632813B
- Authority
- CN
- China
- Prior art keywords
- temperature
- curing
- composite material
- prepreg
- heat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 239000000463 material Substances 0.000 title claims abstract description 38
- 238000005457 optimization Methods 0.000 title claims abstract description 35
- 239000000805 composite resin Substances 0.000 title claims abstract description 21
- 238000004321 preservation Methods 0.000 claims abstract description 78
- 239000002131 composite material Substances 0.000 claims abstract description 51
- 238000004088 simulation Methods 0.000 claims abstract description 29
- 230000008569 process Effects 0.000 claims abstract description 18
- 238000009826 distribution Methods 0.000 claims abstract description 15
- 239000011347 resin Substances 0.000 claims abstract description 15
- 229920005989 resin Polymers 0.000 claims abstract description 15
- 238000013528 artificial neural network Methods 0.000 claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 11
- 238000001382 dynamic differential scanning calorimetry Methods 0.000 claims abstract description 11
- 238000004519 manufacturing process Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000001723 curing Methods 0.000 claims description 136
- 239000003795 chemical substances by application Substances 0.000 claims description 18
- 238000006243 chemical reaction Methods 0.000 claims description 14
- 238000010438 heat treatment Methods 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 13
- 238000007711 solidification Methods 0.000 claims description 11
- 230000008023 solidification Effects 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 9
- 238000009413 insulation Methods 0.000 claims description 8
- 238000006116 polymerization reaction Methods 0.000 claims description 8
- 238000005844 autocatalytic reaction Methods 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 5
- 230000003197 catalytic effect Effects 0.000 claims description 4
- 238000002474 experimental method Methods 0.000 claims description 4
- 239000000835 fiber Substances 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 238000012546 transfer Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000004132 cross linking Methods 0.000 abstract description 10
- 238000000465 moulding Methods 0.000 abstract description 5
- 238000012706 support-vector machine Methods 0.000 abstract 2
- 230000006872 improvement Effects 0.000 description 9
- 238000012795 verification Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000001938 differential scanning calorimetry curve Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000007789 gas Substances 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013035 low temperature curing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 230000036632 reaction speed Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000000518 rheometry Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/26—Composites
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Reinforced Plastic Materials (AREA)
- Moulding By Coating Moulds (AREA)
Abstract
The invention discloses an optimization method of a curing system of a large-thickness resin-based composite material. According to the method, the curing kinetic equation of the prepreg is obtained by using the dynamic differential scanning calorimetry experimental data of the prepreg; then, establishing a temperature distribution model of the composite material in the curing process by using numerical analysis software and finite element simulation software, and realizing numerical simulation calculation of the temperature distribution and the curing degree in the manufacturing process of the composite material; based on a Support Vector Machine (SVM) neural network algorithm, a proxy model is established to solve an optimization problem, and further optimization of process parameters is achieved. The curing system optimized by the method can ensure that the central curing temperature rise of the composite material is respectively reduced by 54 percent and 71 percent at the peak values of the two heat-preservation steps, the curing degree of the composite material at the central low temperature is obviously reduced, the large-range long chain crosslinking curing of resin molecules is ensured, the mechanical property of the composite material molding is further ensured, and the method can be widely applied to the manufacturing of large-thickness resin-based composite material parts such as wind power, ships, fan blades and the like.
Description
Technical Field
The invention belongs to the field of composite material molding and manufacturing, and relates to an optimization method of a curing system of a large-thickness resin-based composite material.
Background
The composite material structure with variable cross section and super large thickness is widely used in the fields of composite material fan blades of large bypass ratio aviation turbofan engines, wind power blades, ships and the like. Because the thermal conductivity of the composite material is poor in the curing process and the heat release phenomenon is caused along with the curing, the curing system different from that of the thin-wall composite material is adopted for the large-thickness composite material in order to ensure the curing uniformity and control the residual stress.
The large-thickness curing system recommended by prepreg suppliers is only suitable for a certain thickness range, and whether the large-thickness curing system is suitable for parts with the thickness of more than 70mm, such as tenons of fan blades of dual-channel engines, is a problem which needs to be researched urgently. The solidification reaction speed of the inside, reduction center of high temperature heat preservation step messenger's solidification temperature more evenly transmits the combined material in the accessible increases, makes the inside solidification exothermic can distribute out more in time, avoids too high solidification temperature rise.
However, studies have shown that excessive increase of the curing degree of the composite material at a temperature lower than the ideal curing temperature causes the molecular crosslinking structure of the resin matrix to be a dispersed local short-chain crosslinking structure rather than a large-scale long-chain crosslinking structure, thereby seriously affecting the mechanical properties of the molding. Therefore, the selection of the curing parameters of the ultra-thick composite material needs to find a balance between the control of the curing temperature rise of the center and the guarantee of the large-range long-chain crosslinking curing of resin molecules so as to guarantee the mechanical property of the composite material molding.
In order to solve the problems, the invention provides an optimization method of a curing system of a large-thickness resin-based composite material, which is used for establishing a temperature distribution model of the curing process of the composite material by combining numerical analysis and finite element simulation from the curing dynamics and rheological properties of the prepreg, accurately and efficiently optimizing the curing system, and is particularly suitable for prepregs without a curing dynamics model and a curing system of the large-thickness composite material.
Disclosure of Invention
In order to solve the technical problem, the invention provides an optimization method capable of effectively reducing the central curing temperature rise of the large-thickness composite material.
The technical scheme adopted by the invention is as follows:
a method for optimizing a curing system of a resin-based composite material with large thickness is used for optimizing parameters of low-temperature heat-preservation steps in a two-step curing system of prepreg, and comprises the following steps:
s1: performing data fitting on an autocatalytic reaction kinetic model by using dynamic differential scanning calorimetry experimental data of the target prepreg to obtain a curing kinetic model of the prepreg;
s2: obtaining a viscosity temperature curve of the prepreg, and determining a temperature range of which the viscosity is lower than a threshold value according to the viscosity temperature curve;
s3: based on a curing kinetic model and a viscosity-temperature curve, adopting numerical analysis software to analyze the curing behavior of the prepreg under different isothermal conditions in the temperature range, taking the influence of a local short-chain cross-linked structure on the mechanical performance as a judgment standard, and selecting the temperature of a low-temperature heat-preservation step in the temperature range;
s4: under the condition of keeping the temperature of a high-temperature heat-preservation step and the time length of the high-temperature heat-preservation step in an original prepreg curing system unchanged, establishing a temperature distribution model of a composite material curing process made of prepreg by using Abaqus finite element simulation software, and realizing numerical simulation calculation of temperature distribution and curing degree in the composite material manufacturing process so as to obtain a plurality of groups of simulation data; wherein each group of analog data comprises the temperature rise rate v and the low-temperature heat preservation step time length t1The peak temperature T of the low-temperature heat-preservation step at the central point in the composite material is obtained by lower simulationmax1High temperature insulation step peak temperature Tmax2And a degree of cure;
s5: training the SVM neural network by using the simulation data as training data under the condition of keeping the temperature of the high-temperature heat-preservation step and the time length of the high-temperature heat-preservation step in the original prepreg curing system unchanged to obtain an agent model, wherein the agent model is obtained by using the simulation data as training dataAt the temperature rising rate v and the low-temperature heat preservation step time length t1Outputting the peak temperature T of the low-temperature heat-preservation step of the composite material for inputmax1High temperature insulation step peak temperature Tmax2And the degree of solidification at the end of the low temperature insulating step;
s6: solving an optimization problem aiming at the agent model through a multi-objective optimization algorithm to obtain an optimal solution of two input values in the agent model which enables an objective function to be minimum, wherein the objective function is a linear weighted sum of three output values of the agent model; using the optimal solution of the temperature and the heating rate v of the low-temperature heat-preservation step and the time length t of the low-temperature heat-preservation step1The optimal solution of (2) is used as the low-temperature heat-preservation step temperature, the heating rate and the low-temperature heat-preservation step duration in the parameters of the optimal curing system of the target prepreg, so that the optimization of the curing system of the large-thickness resin-based composite material is realized.
As a further improvement of the method of the present invention, the optimization method is applicable to prepregs with a high-thickness composite cure system or prepregs without a cure kinetic model and without a high-thickness composite cure system.
As a further improvement of the method of the invention, the kinetic model of the autocatalytic reaction of the prepreg in S1 is shown in formula 1:
In the formula: alpha is the degree of cure of the material, d alpha/dt is the cure reaction rate, k1And k2Respectively a non-catalytic polymerization reaction rate constant and an autocatalytic polymerization reaction rate constant, m and n are reaction series, A is a pre-factor, R is a universal gas constant, T is an absolute temperature, and E is reaction activation energy of a curing reaction.
As a further improvement of the method, in the dynamic differential scanning calorimetry experiment in S1, the selected temperature rise rate is any temperature between 1-20 ℃/min, preferably 3 ℃/min, 5 ℃/min, 10 ℃/min, 15 ℃/min or 20 ℃/min; the dynamic differential scanning calorimetry experimental data should include temperature, heat flow rate, and rate of temperature rise.
As a further improvement of the method of the present invention, the threshold value in S2 is 10 to 50pa · S.
As a further improvement of the method of the present invention, in the judgment criteria described in S3, the highest temperature at which the increase in the degree of solidification within 60 minutes of isothermal heating does not exceed 5% is preferably used as the low temperature holding step temperature.
As a further improvement of the method of the present invention, the temperature distribution model of the composite material curing process described in S4 employs the following heat transfer model:
in the formula: ρ and CpRespectively, material density and material specific heat capacity, kxx、kyyAnd kzzRespectively the thermal conductivity of the material in three directions,
the heat release rate of the curing is calculated by the following formula:
in the formula vfIs the fiber volume fraction, HrIs the total heat release per mass of resin cured.
As a further improvement of the method, in S6, the time cost and the influence of the central curing temperature rise on the performance are comprehensively considered, and according to the difference of the resin system and the thickness, the curing temperature rise rate below 15mm is preferably 1.5-3 ℃, and the curing temperature rise rate above 15mm is preferably 0.5-2 ℃.
As a further improvement of the method, in S5, the simulation data are sampled by Latin hypercube sampling method (LHS), and the sampling points are 100 groups, wherein 90 groups are used for establishing the proxy model, and 10 groups are used for verifying the proxy model.
As a further improvement of the method of the present invention, the SVM neural network described in S5 can be solved by the following formula:
in the formula, K (x, x)i) Is a function, or positive definite kernel, meaning that there is a mapping, x, from the input space to the feature spaceiIs the i-th feature vector, yiIs a class mark, αiAs lagrange multiplier, b*For the optimal solution, N is the total number of categories.
By means of the scheme, the invention at least has the following advantages: on one hand, the steps of the method are not only suitable for the prepreg of the existing curing kinetic model, but also suitable for the prepreg without the curing kinetic model and the curing system of the large-thickness composite material, and the application range is wide; on the other hand, the method achieves a balance between the control of the local short chain cross-linking structure and the central curing temperature rise, provides an optimization method aiming at the curing system of the resin-based composite material with large thickness through finite element simulation and SVM neural network algorithm, and is accurate and efficient.
The above description is only an outline of the technical solution of the present invention, and in order to clearly understand the technical means of the present invention and to implement the method according to the content of the description, the following detailed description will be given to the method for optimizing the curing system of the resin-based composite material with large thickness, which is provided by the present invention, with reference to the accompanying drawings and the embodiments.
Drawings
FIG. 1 is a dynamic DSC curve of a prepreg provided by the present invention at different temperature rise rates;
FIG. 2 is a viscosity-temperature curve for a prepreg provided by a supplier;
FIG. 3 is a verification diagram of an actual curing process of a prepreg curing kinetic model provided by the present invention;
FIG. 4 is a graph of isothermal curing of prepregs based on numerical analysis, as provided by the present invention;
FIG. 5 is a finite element analysis component model provided by the present invention;
FIG. 6 is a graph of temperature at the center of a large thickness composite provided by the present invention that was not optimized and optimized by the method of the present invention versus degree of cure.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
In the present invention, the autocatalytic reaction kinetic model, the Kissinger method, the immobilizer method, the SVM neural network algorithm, and the like are proper terms known to those skilled in the art, and are not limited to any specific terms.
Generally, a supplier of prepregs will provide a corresponding cure regime (MRCC) for the prepreg, which includes five parameters: temperature T of low-temperature heat-preservation stepmax1High temperature insulating step temperature Tmax2Time t of low-temperature heat-preservation step1Time t of high-temperature heat-preservation step2And a temperature rise rate v. However, as mentioned above, the large thickness curing system recommended by the prepreg supplier is only applicable to a certain thickness range, and the recommended curing system parameters are not necessarily applicable when the thickness exceeds a certain range, and it is necessary to optimize the thickness. Among the five parameters, the high-temperature heat-preservation step parameter is sensitive to the influence of the final curing effect, and the recommended parameter of a supplier needs to be adopted preferentially, but the low-temperature heat-preservation step parameter can be optimized to a certain extent, so that the use under a large-thickness scene is met.
The invention provides an optimization method of a curing system of a resin-based composite material with large thickness, which is used for optimizing low-temperature heat-preservation step parameters in a two-step curing system of prepreg and comprises the following steps:
s1: and (3) performing data fitting on the autocatalytic reaction kinetic model by using the dynamic differential scanning calorimetry experimental data of the target prepreg to obtain a curing kinetic model of the prepreg.
In this step, the dynamic model of the autocatalytic reaction of the prepreg is shown in the following two formulas:
In the formula: alpha is the degree of cure of the material, d alpha/dt is the cure reaction rate, k1And k2Respectively a non-catalytic polymerization reaction rate constant and an autocatalytic polymerization reaction rate constant, m and n are reaction series, A is a pre-factor, R is a universal gas constant, T is an absolute temperature, and E is reaction activation energy of a curing reaction.
In the two formulas, which formula model is specifically selected needs to be determined by combining the type of the prepreg.
The specific method of the dynamic differential scanning calorimetry experiment is the prior art, and is not repeated, and the temperature rise rate selected in the dynamic differential scanning calorimetry experiment is any temperature between 1 and 20 ℃/min, preferably 3 ℃/min, 5 ℃/min, 10 ℃/min, 15 ℃/min or 20 ℃/min. In addition, dynamic differential scanning calorimetry experimental data should include temperature, heat flow rate, and ramp rate.
After the curing kinetic model is obtained, the curing kinetic model of the prepreg can be subjected to actual process verification to ensure that the model is feasible, and if the accuracy of the model is poor, correction is required. The process verification can be carried out by a solidification stopping method, a plurality of sampling points are selected at the heat preservation step for verification and comparison of the measured value and the model predicted value, and preferably, one sampling point is arranged at intervals of 20 or 30 min.
S2: and acquiring a viscosity temperature curve of the prepreg by using a rotational rheology method or a supplier data manual, and determining a temperature range with the viscosity lower than a threshold value according to the viscosity temperature curve. The threshold value here may be determined to be an optimum value according to the type of prepreg, and is generally 10 to 50pa · s.
S3: and (3) analyzing the curing behavior of the prepreg under different isothermal conditions in the temperature range by adopting numerical analysis software based on a curing kinetic model and a viscosity-temperature curve, taking the influence of a local short-chain cross-linking structure on the mechanical performance as a judgment standard, and selecting the temperature of the low-temperature heat-preservation step in the temperature range.
In the step, the influence of resin fluidity and low-temperature curing on performance needs to be comprehensively considered, a temperature range with the viscosity lower than a threshold value (such as the range lower than 10-50 pa · s) in a viscosity-temperature curve is selected, curing degree-time curves at different isothermal temperatures (with 10 ℃ as an interval) are fitted according to a curing kinetic equation, and the time is selected to be 300min, so that the curing behavior of the prepreg is analyzed. The numerical analysis software should have the capability of solving nonlinear differential equations, and matlab commercial mathematical software is preferred. And reasonably setting the number of the heat preservation steps and the heat preservation time based on the result obtained by the analysis, and preferably selecting 1 or 2 heat preservation steps. In order to reduce the influence of the local short-chain cross-linked structure on the mechanical performance, the above judgment standard is preferably set as the highest temperature with the solidification degree not exceeding 5% within isothermal 60min is selected as the temperature of the low-temperature heat-preservation step; the heat preservation time is preferably 30-180 min according to different thicknesses, 30-120 min below 15mm is recommended, 60-180 min above 15mm is recommended, and the specific time is determined by optimizing the subsequent steps.
S4: under the condition of keeping the high-temperature heat preservation step temperature and the high-temperature heat preservation step duration in the original prepreg curing system unchanged (namely, the high-temperature heat preservation step temperature and the high-temperature heat preservation step duration in the subsequent simulation process are both fixed to the original recommended values in the curing system provided by a prepreg supplier), an Abaqus finite element simulation software is utilized to establish a temperature distribution model of the prepreg-made composite material curing process, the numerical simulation calculation of the temperature distribution and the curing degree in the composite material manufacturing process is realized, and thus a plurality of groups of simulation data are obtained. Wherein each group of analog data comprises the temperature rise rate v and the low-temperature heat preservation step time length t1The peak temperature T of the low-temperature heat-preservation step at the central point in the composite material is obtained by lower simulationmax1High temperature insulation step peak temperature Tmax2And a degree of cure. Heating rate v and low-temperature heat-preservation step duration t in each group of simulation data1All are different.
In this step, the following heat transfer model is used as the temperature distribution model of the composite material curing process:
in the formula: ρ and CpRespectively, material density and material specific heat capacity, kxx、kyyAnd kzzRespectively the thermal conductivity of the material in three directions,
the heat release rate of the curing is calculated by the following formula:
in the formula vfIs the fiber volume fraction, HrIs the total heat release per mass of resin cured.
The simulation data may be sampled by a latin hypercube sampling method (LHS) based on a numerical simulation of the temperature distribution model. For example, in one embodiment, 100 sets of sampling points are used, and a total of 100 sets of simulation data are obtained, wherein 90 sets can be used for building the proxy model and 10 sets are used for verifying the proxy model.
S5: similarly, under the condition of keeping the temperature of the high-temperature heat preservation step and the time length of the high-temperature heat preservation step in the original curing system of the prepreg unchanged, training the SVM neural network by taking the obtained simulation data as training data to obtain an agent model, wherein the agent model is used for obtaining the agent model according to the temperature rise rate v and the time length t of the low-temperature heat preservation step1Outputting the peak temperature T of the low-temperature heat-preservation step of the composite material for inputmax1High temperature insulation step peak temperature Tmax2And the degree of cure at the end of the low temperature insulating step.
The specific structure of the SVM neural network belongs to the prior art, and the SVM algorithm can be solved by the following formula:
in the formula, K (x, x)i) Is thatA function, or positive definite kernel, means that there is a mapping, x, from the input space to the feature spaceiIs the i-th feature vector, yiIs a class mark, αiAs lagrange multiplier, b*For the optimal solution, N is the total number of categories.
S6: solving an optimization problem aiming at the agent model through a multi-objective optimization algorithm to obtain an optimal solution of two input values in the agent model which enables an objective function to be minimum, wherein the objective function is a linear weighted sum of three output values of the agent model; using the optimal solution of the temperature and the heating rate v of the low-temperature heat-preservation step and the time length t of the low-temperature heat-preservation step1The optimal solution of (2) is used as the low-temperature heat-preservation step temperature, the heating rate and the low-temperature heat-preservation step duration in the parameters of the optimal curing system of the target prepreg, so that the optimization of the curing system of the large-thickness resin-based composite material is realized.
That is, after the SVM neural network is introduced into the multi-objective optimization algorithm, the optimization objectives are the solidification degree at the end of the low-temperature heat preservation step and the peak temperature T of the two heat preservation stepsmax1、Tmax2The optimization variable is the low-temperature heat-preservation step duration t1And a temperature rise rate v. Because a plurality of optimization targets are provided, the multi-target optimization algorithm is a uniform target method, a linear weighting method is adopted to convert the multi-target optimization problem into a target function, and the solidification degree and the peak temperatures of two heat preservation steps can be normalized and then subjected to linear weighting to serve as the target function.
In the step, in the optimization process of the curing system, the time cost and the influence of the central curing temperature rise on the performance need to be comprehensively considered, and the optimal curing temperature rise rate is determined according to the difference between the resin system and the thickness, wherein the curing temperature rise rate below 15mm is preferably 1.5-3 ℃, and the curing temperature rise rate above 15mm is 0.5-2 ℃.
The optimization method is not only suitable for the prepreg with a large-thickness composite material curing system, but also suitable for the prepreg without a curing kinetic model and the large-thickness composite material curing system.
In order to further enable those skilled in the art to better understand the specific implementation process and technical effects of the optimization method, the method is applied to a specific embodiment.
Example (b):
a method for optimizing a curing system of a large-thickness composite material comprises the following steps:
(1) 10-15 mg of prepreg sample is placed in an aluminum crucible, the nitrogen flow is 50mL/min, five heating rates of 3, 5, 10, 15 and 20 ℃/min are selected in the example, and dynamic DSC curves under different heating rates are obtained, as shown in figure 1.
(2) Determination of exothermic peak temperatures T of dynamic DSC curves at different ramp rates by means of FIG. 1pThe curing reaction activation energy E was determined by the Kissinger method.
(3) Using the autocatalytic cure reaction kinetic model, a cure kinetic model of the prepreg was constructed by nonlinear fitting software (the nonlinear fitting software of this example is preferably an origin function mapping tool) in the form shown below:
wherein alpha is the degree of curing of the material, d alpha/dt is the rate of curing reaction, k1And k2Fitting a obtained curing kinetic model, which is a non-catalytic polymerization reaction rate constant and an autocatalytic polymerization reaction rate constant, m and n are reaction series, A is a pre-index factor, R is a universal gas constant, T is an absolute temperature, and E is reaction activation energy of a curing reaction, and is shown as the following formula:
(4) the viscosity-temperature curve provided by the supplier is shown in FIG. 2, and the viscosity of the prepreg is determined to be in a lower range of 100-140 ℃.
(5) And (3) carrying out actual process verification on the curing kinetic model of the prepreg by using a curing stopping method. Five sampling points are selected on the 180 ℃ heat preservation step, one sampling point is arranged at intervals of 30min, the curing degree test value of the sample is compared with the predicted value of the curing kinetic model, and the verification result is shown in figure 3, which shows that the curing kinetic model is feasible.
(6) Based on a curing kinetic model and a viscosity temperature curve, the curing behavior of the prepreg under different isothermal conditions of 100 ℃, 110 ℃, 120 ℃, 130 ℃ and 140 ℃ in a range with lower viscosity is analyzed by adopting an ode function in numerical analysis software matlab, and the result is shown in fig. 4, in order to reduce the influence of a local short-chain cross-linking structure on the mechanical property, a first heat preservation step is selected to be arranged at 110 ℃ according to the judgment standard, namely the temperature of the low-temperature heat preservation step is set to be 110 ℃.
(7) Selecting a composite material component made of the prepreg shown in fig. 5 as a finite element analysis model, wherein the effective size is 150 multiplied by 100 multiplied by 36.8mm, the layering is unidirectional 0-degree layering along the length direction of the component, and the temperature and the curing degree of the internal central point A of the component are calculated in total of 200 layers. Based on an LHS sampling method, 100 groups of sampling points are selected, a subroutine is written by using finite element simulation software Abaqus to establish a temperature distribution model of the composite material curing process, and numerical simulation calculation of temperature distribution and curing degree in the manufacturing process of 100 groups of sampling points is realized. The temperature profile of the composite material during curing is obtained by a heat transfer model shown below:
where ρ and CpRespectively, material density and material specific heat capacity, kxx、kyyAnd kzzRespectively the thermal conductivity of the material in three directions,
for curing heat release rate, obtained by the following formula:
in the formula vfIs the fiber volume fraction, HrIs the total heat release per mass of resin cured.
Thus, 100 sets of sample points obtain 100 sets of simulation data, of which 90 sets can be used for building the proxy model and 10 sets are used for verifying the proxy model. Each group of simulation data comprises the temperature rise rate v and the low-temperature heat preservation step time length t1The peak temperature Tmax of the low-temperature heat preservation step at the central point in the composite material is obtained by lower simulation1And the peak temperature Tmax of the high-temperature heat preservation step2And a degree of cure. Heating rate v and low-temperature heat-preservation step duration t in each group of simulation data1All are different.
(8) In order to balance the curing temperature rise of a control center and guarantee the large-range long-chain crosslinking curing of resin molecules so as to guarantee the mechanical property of composite material forming, a proxy model is established by utilizing an SVM neural network algorithm based on 100 sets of simulation data in the step (7), and the proxy model is used for establishing the proxy model according to the temperature rise rate v and the low-temperature heat preservation step duration t1Outputting the peak temperature T of the low-temperature heat-preservation step of the composite material for inputmax1High temperature insulation step peak temperature Tmax2And the degree of cure at the end of the low temperature insulating step. After training with the SVM neural network for 90 sets of simulated data, the validation was passed in the remaining 10 sets of simulated data. It should be noted that when the SVM neural network is trained, the temperature of the high-temperature heat preservation step and the duration of the high-temperature heat preservation step are kept consistent with the two parameter values in the original prepreg curing system.
And then, importing the trained agent model into an optimization algorithm based on a unified objective method, and solving an optimization problem to obtain an optimal solution of two input values in the agent model with the minimum objective function. The target function is the linear weighted sum of three output values of the proxy model, and the three output values are weighted after being normalized to be used as the target function. In the objective function of this embodiment, the peak temperature T of the low temperature holding stepmax1High temperature insulation step peak temperature Tmax2And the weight of the degree of curing at the end of the low-temperature heat-preservation step is 0.3, 0.5 and 0.2 respectively.
Thereby, the optimal solution of the temperature of the low-temperature heat-preservation step obtained in the step (6), the temperature-rise rate v obtained in the step (8) and the low-temperature heat-preservation step time length t can be obtained1Replacement of prepregAnd keeping the temperature, the heating rate and the time length of the low-temperature heat preservation step in the original curing system parameters to obtain optimized optimal curing system parameters, thereby realizing the optimization of the curing system of the large-thickness resin matrix composite material.
In this example, the round post-cure system was optimized as: firstly, the temperature is increased from room temperature to 110 ℃ at the speed of 1.3 ℃/min, the temperature is kept for 35min at 110 ℃, then the temperature is increased to 180 ℃ at the speed of 1.3 ℃/min, the temperature is kept for 120min, and finally the temperature is reduced to the room temperature.
(9) In order to verify the effect of the invention, the temperature and curing degree curve of the point A after Optimization in the step (8) is extracted, and compared with the temperature and curing degree curve of the point A under a curing system (MRCC) provided by a supplier, the temperature rise of the central curing of the composite material is respectively reduced by 54% and 71% at the peak value of two heat preservation steps, the curing degree under the central low temperature is obviously reduced, and the large-range long-chain crosslinking curing of resin molecules is ensured so as to ensure the mechanical property of the composite material molding.
The above-described embodiments are merely preferred embodiments of the present invention, which should not be construed as limiting the invention. Various changes and modifications may be made by one of ordinary skill in the pertinent art without departing from the spirit and scope of the present invention. Therefore, the technical scheme obtained by adopting the mode of equivalent replacement or equivalent transformation is within the protection scope of the invention.
Claims (9)
1. An optimization method of a curing system of a resin-based composite material with large thickness is used for optimizing parameters of low-temperature heat-preservation steps in a two-step curing system of prepreg, and is characterized in that: the method comprises the following steps:
s1: performing data fitting on an autocatalytic reaction kinetic model by using dynamic differential scanning calorimetry experimental data of the target prepreg to obtain a curing kinetic model of the prepreg;
s2: acquiring a viscosity temperature curve of the prepreg, and determining a temperature range with the viscosity lower than a threshold value according to the viscosity temperature curve;
s3: based on a curing kinetic model and a viscosity-temperature curve, adopting numerical analysis software to analyze the curing behavior of the prepreg under different isothermal conditions in the temperature range, taking the influence of a local short-chain cross-linked structure on the mechanical performance as a judgment standard, and selecting the temperature of a low-temperature heat-preservation step in the temperature range; in the judgment standard, selecting the highest temperature with the solidification degree not exceeding 5% within isothermal 60min as the temperature of the low-temperature heat-preservation step;
s4: under the condition of keeping the temperature of a high-temperature heat-preservation step and the time length of the high-temperature heat-preservation step in an original prepreg curing system unchanged, establishing a temperature distribution model of a composite material curing process made of prepreg by using Abaqus finite element simulation software, and realizing numerical simulation calculation of temperature distribution and curing degree in the composite material manufacturing process so as to obtain a plurality of groups of simulation data; wherein each group of analog data comprises the temperature rise rate v and the low-temperature heat preservation step time length t1The peak temperature T of the low-temperature heat-preservation step at the central point in the composite material is obtained by lower simulationmax1High temperature insulation step peak temperature Tmax2And a degree of cure;
s5: training the SVM neural network by using the simulation data as training data under the condition of keeping the temperature of the high-temperature heat-preservation step and the time length of the high-temperature heat-preservation step in the original prepreg curing system unchanged to obtain an agent model, wherein the agent model is used for training the SVM neural network at the temperature-rise rate v and the low-temperature heat-preservation step time length t1Outputting the peak temperature T of the low-temperature heat-preservation step of the composite material for inputmax1High temperature insulation step peak temperature Tmax2And the degree of cure at the end of the low temperature insulating step;
s6: solving an optimization problem aiming at the agent model through a multi-objective optimization algorithm to obtain an optimal solution of two input values in the agent model which enables an objective function to be minimum, wherein the objective function is a linear weighted sum of three output values of the agent model; using the optimal solution of the temperature and the heating rate v of the low-temperature heat-preservation step and the time length t of the low-temperature heat-preservation step1As the optimal solution of the low-temperature heat-preservation step temperature, the heating rate and the low temperature in the optimal curing system parameters of the target prepregThe temperature is kept for a long time, and the optimization of the curing system of the resin matrix composite material with large thickness is realized.
2. The method of optimizing a cure system for a high build resin-based composite material of claim 1, wherein the method is applied to prepregs with a high build composite cure system or prepregs without a cure kinetic model and without a high build composite cure system.
3. The method for optimizing the curing system of the large-thickness resin-based composite material as claimed in claim 1, wherein the dynamic model of the autocatalytic reaction of the prepreg in S1 is represented by the following formula:
In the formula: alpha is the degree of cure of the material, d alpha/dt is the cure reaction rate, k1And k2Respectively a non-catalytic polymerization reaction rate constant and an autocatalytic polymerization reaction rate constant, m and n are reaction series, A is a pre-factor, R is a universal gas constant, T is an absolute temperature, and E is reaction activation energy of a curing reaction.
4. The optimization method of the curing system of the large-thickness resin-based composite material according to claim 1, wherein the dynamic differential scanning calorimetry experiment selects any temperature with a heating rate of 1-20 ℃/min, preferably 3 ℃/min, 5 ℃/min, 10 ℃/min, 15 ℃/min or 20 ℃/min; the dynamic differential scanning calorimetry experimental data should include temperature, heat flow rate, and rate of temperature rise.
5. The method for optimizing the curing system of the large-thickness resin-based composite material as claimed in claim 1, wherein the threshold value in S2 is 10-50 pa-S.
6. The method for optimizing the curing system of the large-thickness resin-based composite material as claimed in claim 1, wherein the temperature distribution model of the curing process of the composite material as described in S4 adopts the following heat transfer model:
in the formula: ρ and CpRespectively, material density and material specific heat capacity, kxx、kyyAnd kzzRespectively the thermal conductivity of the material in three directions,
the heat release rate of the curing is calculated by the following formula:
in the formula vfIs the fiber volume fraction, HrIs the total heat release per mass of resin cured.
7. The method for optimizing the curing system of the large-thickness resin-based composite material according to claim 1, wherein the influence of time cost and central curing temperature rise on the performance is comprehensively considered in S6, and according to the difference between the resin system and the thickness, the curing temperature rise rate below 15mm is preferably 1.5-3 ℃, and the curing temperature rise rate above 15mm is preferably 0.5-2 ℃.
8. The method for optimizing the curing system of the large-thickness resin-based composite material as claimed in claim 1, wherein the simulation data in S5 is sampled by Latin hypercube sampling method, the sampling points are 100 groups, wherein 90 groups are used for establishing the proxy model, and 10 groups are used for verifying the proxy model.
9. The method for optimizing the curing system of the resin-based composite material with large thickness as claimed in claim 1, wherein the SVM neural network described in S5 is solved by the following formula:
in the formula, K (x, x)i) Is a function, or positive definite kernel, meaning that there is a mapping, x, from the input space to the feature spaceiIs the ith feature vector, yiIs a mark of a class, and is,
as lagrange multiplier, b*For the optimal solution, N is the total number of categories.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011410760.1A CN112632813B (en) | 2020-12-03 | 2020-12-03 | Optimization method of curing system of large-thickness resin-based composite material |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011410760.1A CN112632813B (en) | 2020-12-03 | 2020-12-03 | Optimization method of curing system of large-thickness resin-based composite material |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112632813A CN112632813A (en) | 2021-04-09 |
| CN112632813B true CN112632813B (en) | 2022-05-31 |
Family
ID=75308250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011410760.1A Active CN112632813B (en) | 2020-12-03 | 2020-12-03 | Optimization method of curing system of large-thickness resin-based composite material |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112632813B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113378427B (en) * | 2021-05-11 | 2023-03-10 | 三峡大学 | Calculation method for evaluating wind load fracture resistance of branches and trunks of arbor |
| CN113607772B (en) * | 2021-08-04 | 2022-09-09 | 西北工业大学 | Method and system for determining damage of toughened composite material during curing molding |
| CN113807028B (en) * | 2021-10-15 | 2024-09-27 | 华东理工大学 | Optimization method and optimization system for epoxy resin curing process |
| CN114202262B (en) * | 2022-02-21 | 2022-07-01 | 德州联合拓普复合材料科技有限公司 | Prepreg process improvement method and system based on neural network and storage medium |
| CN116631549B (en) * | 2023-07-25 | 2023-09-22 | 北京理工大学 | Optimization method, device, equipment and medium for composite material curing system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101523394A (en) * | 2006-10-06 | 2009-09-02 | 日立化成工业株式会社 | Method for analyzing fluidity of resin material including particles and fluidity analysis system |
| CN102490370A (en) * | 2011-11-15 | 2012-06-13 | 中国人民解放军国防科学技术大学 | Liquid model molding technology for preparing polymer matrix composite material |
| WO2015072040A1 (en) * | 2013-11-18 | 2015-05-21 | 株式会社日立製作所 | Resin flow behavior calculation method, and resin flow behavior calculation program |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150111141A1 (en) * | 2013-10-22 | 2015-04-23 | Xerox Corporation | Bio-Based Toner Resin with Increased Fusing Performance |
| CN110197009B (en) * | 2019-05-09 | 2021-02-23 | 西北工业大学 | Prediction method for curing reaction of resin-based composite material |
-
2020
- 2020-12-03 CN CN202011410760.1A patent/CN112632813B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101523394A (en) * | 2006-10-06 | 2009-09-02 | 日立化成工业株式会社 | Method for analyzing fluidity of resin material including particles and fluidity analysis system |
| CN102490370A (en) * | 2011-11-15 | 2012-06-13 | 中国人民解放军国防科学技术大学 | Liquid model molding technology for preparing polymer matrix composite material |
| WO2015072040A1 (en) * | 2013-11-18 | 2015-05-21 | 株式会社日立製作所 | Resin flow behavior calculation method, and resin flow behavior calculation program |
Non-Patent Citations (2)
| Title |
|---|
| Improvement of electrical characteristics of fluorinated perylene diimide thin-film transistors by gate dielectric surface treatment;Li-Gong Yang 等;《2007 Asia Optical Fiber Communication and Optoelectronics Conference》;20071126;第248-250页 * |
| 基于DSC法的RTM工艺用6421双马树脂固化反应分析;郭启微 等;《宇航材料工艺》;20120815;第100-104页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112632813A (en) | 2021-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112632813B (en) | Optimization method of curing system of large-thickness resin-based composite material | |
| CN107742049B (en) | Design method of radiating system of armored vehicle under altitude-variable working condition | |
| CN107239634B (en) | A kind of aero-engine transient process modeling method | |
| CN104734147A (en) | Probability energy flow analysis method for integrated energy system (IES) | |
| CN105302987B (en) | A kind of method of equivalent prediction Thermomechanical Fatigue Life | |
| CN104698323B (en) | A kind of dry-type distribution transformer accelerated aging test method | |
| CN110161329B (en) | Quench resistance simulation prediction method and system for second-generation high-temperature superconducting tape | |
| CN105158085A (en) | Compound polyimide retainer storage life prediction method | |
| CN105550390B (en) | A kind of across scale heat analysis equivalent method of the fiber reinforced composite material of multicriterion | |
| Sun et al. | Microstructure evolution modeling and simulation for dynamic recrystallization of Cr12MoV die steel during hot compression based on real metallographic image | |
| Mahdi et al. | Numerical investigations of the thermal behavior of a HAWT nacelle using ANSYS FLUENT | |
| Deshmukh et al. | Transient thermodynamic modeling of air cooler in supercritical CO2 Brayton cycle for solar molten salt application | |
| CN116959598A (en) | Alloy grain growth prediction method controlled by saturated grain size | |
| Chen et al. | Simulation of flow of aluminum alloy 3003 under hot compressive deformation | |
| Wang et al. | Data mining optimization of laidback fan-shaped hole to improve film cooling performance | |
| CN113109190B (en) | Short crack-based life prediction method under multi-axis thermomechanical load | |
| CN111031613A (en) | Active thermal control method for power device in complex convection environment | |
| CN113690891A (en) | Analytic method-based power-heat interconnection comprehensive energy system probability power flow determination method | |
| Zhang et al. | Temperature dependent tensile strength modeling and analysis of shape memory polymers with physics‐based energy equivalence principle | |
| Zhu et al. | Study on MRLS Direction Pipe's Multi-objective Optimization and Multi-attribute Decision Making | |
| Zhang et al. | Residual Stress Evolution of 7050 Aluminum Alloy during Thermal Processing and Its Effects on Processing Deformation and Mechanical Properties | |
| CN117744397B (en) | Optimal energy flow computing method, device and storage medium for electric-thermal interconnection system | |
| CN103744739B (en) | A kind of method improving hard disk reliability based on multiobjectives decision | |
| Lee et al. | Effect of flow imbalance on the operational performance of the KSTAR PF superconducting magnets system | |
| CN113866008B (en) | Creep life prediction method based on threshold stress and tensile strength |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |