CN113343375B - Prediction method for autoclave molding curing heat distribution - Google Patents
Prediction method for autoclave molding curing heat distribution Download PDFInfo
- Publication number
- CN113343375B CN113343375B CN202110480808.4A CN202110480808A CN113343375B CN 113343375 B CN113343375 B CN 113343375B CN 202110480808 A CN202110480808 A CN 202110480808A CN 113343375 B CN113343375 B CN 113343375B
- Authority
- CN
- China
- Prior art keywords
- temperature
- autoclave
- tool
- thermocouple
- curing
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- 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
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Heating, Cooling, Or Curing Plastics Or The Like In General (AREA)
- Curing Cements, Concrete, And Artificial Stone (AREA)
Abstract
The invention discloses a prediction method for autoclave molding curing heat distribution, belonging to the technical field of autoclave molding and comprising the following steps: a. preparing data; b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time length of the temperature which is set at the distance of each tank in the temperature uniformity data and the thermocouple temperature per minute; c. establishing and training a model; d. predicting data; e. and d, outputting a result, calculating the temperature difference of different areas of the parts molded and cured by the autoclave according to the temperature value of the thermocouple per minute in the molding and curing process of the autoclave, calculated in the step d, and judging the uniformity of the heat distribution on the surfaces of the parts which are combined into the autoclave. The invention can predict the molding curing thermal analysis of a plurality of parts and adjust the tank-entering combination of the parts with uneven thermal distribution, thereby effectively avoiding the problem of part quality caused by asynchronous part curing and improving the part quality.
Description
Technical Field
The invention relates to the technical field of autoclave molding, in particular to a method for predicting the curing heat distribution of autoclave molding.
Background
The autoclave curing process is a main method for producing composite material components, and the high temperature and the pressure generated by compressed gas in the autoclave are utilized to heat and pressurize a composite material blank to complete curing molding. The whole part curing and forming process involves complex heat exchange, thermochemical reaction and coupling effects of a tool in the autoclave, a flow field and a temperature field between the composite material component and the tank body. Due to the limited heating capacity of the autoclave, the complex flow field in the autoclave and the heat release of the curing reaction of the part, the temperature field distribution of the part is not uniform. When the temperature field distribution of the part is not uniform, the part is unsynchronized in solidification, and further, the residual stress and the residual deformation of the part are caused, so that the part is scrapped due to failure.
At present, a finite element simulation method is generally adopted to research the temperature and heat distribution of the autoclave, and a plurality of independent autoclave temperatures or single tool temperatures are simulated in an analog mode, but in order to improve production efficiency and reduce cost in actual production, almost a plurality of workpieces enter a tank together for curing and forming, simulation difficulty of a plurality of tool temperature fields is high, influence factors are more, and calculation time is long, so that prediction of the temperature and heat distribution of autoclave forming and curing by utilizing finite element simulation cannot meet actual production requirements.
Chinese patent document CN 105092636a, published 2015, 11/25 discloses a thermal distribution experimental method for a composite part tool, which is characterized by comprising the following steps:
step one, before a tool thermal distribution experiment, analyzing a curing curve according to different prepreg materials used in part forming and different part structural forms, and formulating corresponding curing parameter standards;
step two: laying heat release galvanic couple on the tool
When the thermocouple is laid, the whole tool is covered to the maximum extent, then the thermocouple is covered by two layers of air felts, and the air felts are also fixed on the tool by high-temperature adhesive tapes; finally, pasting a vacuum bag according to the part production;
step three: placing the tool in an autoclave:
the tools with the material, the structural form and the small volume difference are placed in the same autoclave, and the position of the tools in the autoclave is convenient for the airflow in the autoclave to flow to the maximum extent; when the number of the tools cannot enable the autoclave to be fully loaded, the tools are close to the opening of the autoclave;
step four: carrying out thermal distribution solidification on the tool, and simultaneously acquiring thermal distribution data in the solidification process by using a thermocouple:
step five: data acquisition and analysis:
after curing is finished, exporting the data of the heat distribution from curing equipment, wherein the exported curing data corresponds to the thermocouple position number, converting the acquired data into the heating rate during curing, and judging whether the heating rate of the data is qualified or not; and judging whether the constant temperature time of curing and the temperature of curing meet requirements.
The composite material part tool thermal distribution experimental method disclosed in the patent document cannot predict the molding curing thermal analysis of a plurality of parts, and cannot avoid the problem of part quality caused by asynchronous part curing.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a prediction method for the molding and curing heat distribution of an autoclave.
The invention is realized by the following technical scheme:
the method for predicting the curing heat distribution in autoclave molding is characterized by comprising the following steps of:
a. data preparation, namely analyzing, carding and identifying main influence factors of the surface heat distribution of a part in the autoclave curing molding process from structural parameters and temperature uniformity, and collecting data representing the main influence factors by reading temperature uniformity data and process parameter data which are collected in a database and reading structural parameters in a process digifax, wherein the temperature uniformity data comprises the temperature of the autoclave per minute, the running time of the temperature which is set from the beginning of the distance of each autoclave, the temperature of a thermocouple per minute, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool;
b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time length of the temperature which is set at the distance of each tank in the temperature uniformity data and the thermocouple temperature per minute;
c. model establishment and training: taking the normalized structure parameters, the temperature uniformity data and the degree of solidification as training data for establishing an SVM regression prediction model; constructing an SVM regression prediction model by taking the temperature of the autoclave per minute in the normalized structural parameters, the curing degree and the temperature uniformity data, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool as input parameters and taking the temperature of the thermocouple per minute in the temperature uniformity data as an output target parameter;
d. predicting data, acquiring parameters of the combined parts to be predicted, including the placement position of a tool in the autoclave, the position of a thermocouple arranged in the tool, structural parameters and process parameter data, and normalizing the structural parameters; calculating the autoclave temperature per minute and the operation duration according to the process parameter data, predicting a first minute thermocouple temperature value by using an SVM regression prediction model, calculating a second minute curing degree as an input attribute according to the first minute thermocouple temperature value, predicting a second minute thermocouple temperature value by using the SVM regression prediction model, iteratively calculating the thermocouple temperature per minute in the autoclave molding and curing process by using the SVM regression prediction model and the curing degree in sequence, and storing the thermocouple temperature per minute in a database;
e. and d, outputting a result, calculating the temperature difference of different areas of the parts molded and cured by the autoclave according to the temperature value of the thermocouple per minute in the molding and curing process of the autoclave, calculated in the step d, and judging the uniformity of the heat distribution on the surfaces of the parts which are combined into the autoclave.
In the step b, the degree of curing is calculated by formula 1 and formula 2;
in the formula, alpha is the curing degree, T is the thermocouple temperature value, and T is the operation time;
when t is 0, the degree of cure α is 0; r is 8.31434J/mol, A1=2.102×109min,A2=-2.102×109min,E1=8.07×104J/mol,E2=7.78×104J/mol,E3=5.66×104J/mol,A3=1.96×105min。
In the step d, the SVM regression prediction model is as follows:
wherein C is the inner product, K (x)i,xj) For the kernel function, a gaussian radial basis kernel function is adopted:
wherein x isjIs the kernel center and σ is the kernel width parameter.
In the step b, the structural parameter normalization is min-max normalization:
wherein x isiIs the original value of the sample and is,as a normalized value, xminIs the minimum value of the sample, xmaxIs the sample maximum.
In the temperature uniformity data, the placing position of the tool in the autoclave is represented by the distance from the tool to the door of the autoclave and the distance from the tool to the axial plane of the autoclave, and the thermocouple is arranged at the position of the tool and is represented by the distance from the thermocouple to the front section of the tool and the distance from the thermocouple to the symmetrical plane of the tool.
The process parameter data comprises a temperature rise rate VTemperature riseConstant temperature TconDuration of constant temperature tconAnd a cooling rate VTemperature reductionAnd a cooling-down end temperature TdownAnd the structural parameters comprise tool information and part information.
The tool information comprises tool length, tool width, tool height, tool quality, tool outer surface area, tool volume, tool heat exchange area and tool hot melting; the part information comprises part length, part width, part height, part quality, part surface area, part volume, part heat exchange area and part hot melting.
In the step d, calculating the autoclave temperature per minute and the operation duration according to the process parameter data comprises the following steps:
s1, calculating the autoclave temperature per minute in the heating stage according to the process parameter data;
Tt=T0+(t-1)×Vtemperature riseFormula 6
Wherein, TtAutoclave temperature, T, at the T minute0To start temperature, t is the temperature rise phase run minute, VTemperature riseThe rate of temperature rise; t ist<=Tcon,TconThe constant temperature is adopted;
s2, calculating the autoclave curing time in the constant temperature stage according to the process parameter data;
tconstant temperature=tWait for+tconFormula 7
Wherein, tConstant temperatureFor the duration of the actual autoclave constant temperature phase, tconThe constant temperature duration of the slowest thermocouple, tWait forThe temperature of the autoclave reaches a constant temperature TconThe temperature of the thermocouple reaches the constant temperature T from the slowestconThe time of (d);
s3, calculating the autoclave temperature per minute in the cooling stage according to the process parameter data;
Tt=Tcon-(t-1)×Vtemperature reductionFormula 8
Wherein, TtAutoclave temperature, T, at the T minutet<=Tcon,TconConstant temperature, t cooling phase run minutes, VTemperature reductionIs the rate of temperature decrease.
The SVM regression prediction model is a support vector machine regression prediction model.
The basic principle of the invention is as follows:
combing and identifying main influence factors of surface heat distribution in the part curing process according to an autoclave curing mechanism, collecting, cleaning and sorting the main influence factors, the real-time autoclave temperature and the thermocouple temperature, and training an SVM regression prediction model by taking the obtained data sample as an input variable of an SVM; the temperature value of the thermocouple per minute is iteratively predicted by utilizing an SVM regression prediction model and the degree of solidification, the temperature difference of different areas of each part in different tank feeding combinations in the hot pressing tank is analyzed, the prediction of the molding solidification thermal analysis of a plurality of parts is realized, the tank feeding combination of the parts with uneven thermal distribution is adjusted, and therefore the problem of part quality caused by asynchronous part solidification can be solved.
The beneficial effects of the invention are:
1. the method comprises the steps of a, preparing data, analyzing, carding and identifying main influence factors of the surface heat distribution of a part in the autoclave curing and forming process from structural parameters and temperature uniformity, and collecting data representing the main influence factors by reading temperature uniformity data and process parameter data which are collected in a database and reading structural parameters in a process digifax, wherein the temperature uniformity data comprise the temperature of an autoclave per minute, the running time of the temperature which is set at the beginning of the distance of each autoclave, the temperature of a thermocouple per minute, the placing position of a tool in the autoclave and the position of the thermocouple arranged in the tool; b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time length of the temperature which is set at the distance of each tank in the temperature uniformity data and the thermocouple temperature per minute; c. model establishment and training: taking the normalized structure parameters, the temperature uniformity data and the degree of solidification as training data for establishing an SVM regression prediction model; constructing an SVM regression prediction model by taking the autoclave temperature per minute in the normalized structure parameter, the curing degree and the temperature uniformity data, the placing position of the tool in the autoclave and the position of the thermocouple arranged on the tool as input parameters and taking the thermocouple temperature per minute in the temperature uniformity data as an output target parameter; d. data prediction, namely acquiring parameters of the combined part to be predicted, including the placement position of a tool in the autoclave, the position of a thermocouple arranged in the tool, structural parameters and process parameter data, and normalizing the structural parameters; calculating the autoclave temperature per minute and the operation time length according to the process parameter data, predicting a first minute thermocouple temperature value by using an SVM regression prediction model, calculating a second minute curing degree serving as an input attribute according to the first minute thermocouple temperature value, predicting a second minute thermocouple temperature value by using the SVM regression prediction model, iteratively calculating the thermocouple temperature per minute in the autoclave molding and curing process by using the SVM regression prediction model and the curing degree in sequence, and storing the thermocouple temperature per minute in a database; e. d, outputting a result, calculating the temperature difference of different areas of the autoclave molding and curing part according to the temperature value of the thermocouple per minute in the autoclave molding and curing process calculated in the step d, and judging the uniformity of the heat distribution on the surfaces of the parts which are combined into the autoclave; compared with the prior art, the temperature value of the autoclave molding curing thermocouple is predicted, and the temperature difference of different areas of each part in different autoclave inlet combinations is analyzed, so that the molding curing thermal analysis of a plurality of parts can be predicted, the parts with uneven thermal distribution are adjusted in the autoclave inlet combinations, the problem of part quality caused by asynchronous part curing is avoided, and the part quality is improved.
2. According to the method, the temperature per minute is obtained by utilizing an SVM algorithm and the iteration of the curing degree in the process of predicting the surface heat distribution of the part, the traditional finite element simulation technology is replaced by utilizing a big data analysis method to predict the heat distribution of the curing process of the multi-part tank-entering combined part, and compared with the finite element simulation method, the method is higher in calculation efficiency.
Drawings
The invention will be further described in detail with reference to the drawings and the following detailed description:
FIG. 1 is a block flow diagram of the present invention;
FIG. 2 is a block diagram of a data prediction process according to the present invention.
Detailed Description
Example 1
Referring to fig. 1 and 2, a method for predicting the distribution of curing heat in autoclave molding includes the following steps:
a. data preparation, namely analyzing, carding and identifying main influence factors of the surface heat distribution of a part in the autoclave curing molding process from structural parameters and temperature uniformity, and collecting data representing the main influence factors by reading temperature uniformity data and process parameter data which are collected in a database and reading structural parameters in a process digifax, wherein the temperature uniformity data comprises the temperature of the autoclave per minute, the running time of the temperature which is set from the beginning of the distance of each autoclave, the temperature of a thermocouple per minute, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool;
b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time length of the temperature which is set at the distance of each tank in the temperature uniformity data and the thermocouple temperature per minute;
c. model establishment and training: taking the normalized structure parameters, the temperature uniformity data and the degree of solidification as training data for establishing an SVM regression prediction model; constructing an SVM regression prediction model by taking the temperature of the autoclave per minute in the normalized structural parameters, the curing degree and the temperature uniformity data, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool as input parameters and taking the temperature of the thermocouple per minute in the temperature uniformity data as an output target parameter;
d. data prediction, namely acquiring parameters of the combined part to be predicted, including the placement position of a tool in the autoclave, the position of a thermocouple arranged in the tool, structural parameters and process parameter data, and normalizing the structural parameters; calculating the autoclave temperature per minute and the operation time length according to the process parameter data, predicting a first minute thermocouple temperature value by using an SVM regression prediction model, calculating a second minute curing degree serving as an input attribute according to the first minute thermocouple temperature value, predicting a second minute thermocouple temperature value by using the SVM regression prediction model, iteratively calculating the thermocouple temperature per minute in the autoclave molding and curing process by using the SVM regression prediction model and the curing degree in sequence, and storing the thermocouple temperature per minute in a database;
e. and d, outputting a result, calculating the temperature difference of different areas of the autoclave molding and curing part according to the temperature value of the thermocouple per minute in the autoclave molding and curing process calculated in the step d, and judging the uniformity of the heat distribution on the surfaces of the parts which are combined into the autoclave.
a. Data preparation, namely analyzing, combing and identifying main influence factors of the surface heat distribution of the part in the autoclave curing and forming process from the structural parameters and the temperature uniformity, and collecting data representing the main influence factors by reading the temperature uniformity data and the process parameter data which are collected in a database and reading the structural parameters in a process digifax, wherein the temperature uniformity data comprises the autoclave temperature per minute, the running time of the temperature which is set from the beginning of the distance of each autoclave, the thermocouple temperature per minute, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool; b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time length of the temperature which is set at the distance of each tank in the temperature uniformity data and the thermocouple temperature per minute; c. model establishment and training: taking the normalized structure parameters, the temperature uniformity data and the degree of solidification as training data for establishing an SVM regression prediction model; constructing an SVM regression prediction model by taking the temperature of the autoclave per minute in the normalized structural parameters, the curing degree and the temperature uniformity data, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool as input parameters and taking the temperature of the thermocouple per minute in the temperature uniformity data as an output target parameter; d. data prediction, namely acquiring parameters of the combined part to be predicted, including the placement position of a tool in the autoclave, the position of a thermocouple arranged in the tool, structural parameters and process parameter data, and normalizing the structural parameters; calculating the autoclave temperature per minute and the operation duration according to the process parameter data, predicting a first minute thermocouple temperature value by using an SVM regression prediction model, calculating a second minute curing degree as an input attribute according to the first minute thermocouple temperature value, predicting a second minute thermocouple temperature value by using the SVM regression prediction model, iteratively calculating the thermocouple temperature per minute in the autoclave molding and curing process by using the SVM regression prediction model and the curing degree in sequence, and storing the thermocouple temperature per minute in a database; e. d, outputting a result, calculating the temperature difference of different areas of the autoclave molding and curing part according to the temperature value of the thermocouple per minute in the autoclave molding and curing process calculated in the step d, and judging the heat distribution uniformity of the surfaces of the parts which are combined into the autoclave; compared with the prior art, the temperature value of the autoclave molding curing thermocouple is predicted, and the temperature difference of different areas of each part in different autoclave inlet combinations is analyzed, so that the molding curing thermal analysis of a plurality of parts can be predicted, the parts with uneven thermal distribution are adjusted in the autoclave inlet combinations, the problem of part quality caused by asynchronous part curing is avoided, and the part quality is improved.
Example 2
Referring to fig. 1 and 2, a method for predicting the distribution of curing heat in autoclave molding includes the following steps:
a. data preparation, namely analyzing, carding and identifying main influence factors of the surface heat distribution of a part in the autoclave curing molding process from structural parameters and temperature uniformity, and collecting data representing the main influence factors by reading temperature uniformity data and process parameter data which are collected in a database and reading structural parameters in a process digifax, wherein the temperature uniformity data comprises the temperature of the autoclave per minute, the running time of the temperature which is set from the beginning of the distance of each autoclave, the temperature of a thermocouple per minute, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool;
b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time length of the temperature which is set at the distance of each tank in the temperature uniformity data and the thermocouple temperature per minute;
c. model establishment and training: taking the normalized structure parameters, the temperature uniformity data and the degree of solidification as training data for establishing an SVM regression prediction model; constructing an SVM regression prediction model by taking the temperature of the autoclave per minute in the normalized structural parameters, the curing degree and the temperature uniformity data, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool as input parameters and taking the temperature of the thermocouple per minute in the temperature uniformity data as an output target parameter;
d. data prediction, namely acquiring parameters of the combined part to be predicted, including the placement position of a tool in the autoclave, the position of a thermocouple arranged in the tool, structural parameters and process parameter data, and normalizing the structural parameters; calculating the autoclave temperature per minute and the operation duration according to the process parameter data, predicting a first minute thermocouple temperature value by using an SVM regression prediction model, calculating a second minute curing degree as an input attribute according to the first minute thermocouple temperature value, predicting a second minute thermocouple temperature value by using the SVM regression prediction model, iteratively calculating the thermocouple temperature per minute in the autoclave molding and curing process by using the SVM regression prediction model and the curing degree in sequence, and storing the thermocouple temperature per minute in a database;
e. and d, outputting a result, calculating the temperature difference of different areas of the parts molded and cured by the autoclave according to the temperature value of the thermocouple per minute in the molding and curing process of the autoclave, calculated in the step d, and judging the uniformity of the heat distribution on the surfaces of the parts which are combined into the autoclave.
In the step b, the degree of curing is calculated by formula 1 and formula 2;
in the formula, alpha is the curing degree, T is the thermocouple temperature value, and T is the operation time;
when t is 0, the curing degree alpha is 0; r is 8.31434J/mol, A1=2.102×109min,A2=-2.102×109min,E1=8.07×104J/mol,E2=7.78×104J/mol,E3=5.66×104J/mol,A3=1.96×105min。
In the step d, the SVM regression prediction model is as follows:
wherein C is the inner product, K (x)i,xj) For the kernel function, a gaussian radial basis kernel function is used:
wherein x isjIs the kernel center and σ is the kernel width parameter.
In the step b, the structural parameter normalization is min-max normalization:
wherein x isiFor the original value of the sample, the value of,as a normalized value, xminIs the minimum value of the sample, xmaxIs the sample maximum.
In the temperature uniformity data, the placing position of the tool in the autoclave is represented by the distance from the tool to the autoclave door and the distance from the tool to the axial plane in the autoclave, and the thermocouple is arranged at the position of the tool and represented by the distance from the thermocouple to the front section of the tool and the distance from the thermocouple to the symmetrical plane of the tool.
The process parameter data comprises a temperature rise rate VTemperature riseConstant temperature TconConstant temperature duration tconAnd a cooling rate VTemperature reductionAnd a cooling-down finishing temperature TdownAnd the structural parameters comprise tool information and part information.
The tool information comprises tool length, tool width, tool height, tool quality, tool external surface area, tool volume, tool heat exchange area and tool hot melting; the part information comprises part length, part width, part height, part quality, part surface area, part volume, part heat exchange area and part hot melting.
Example 3
Referring to fig. 1 and 2, a method for predicting the curing heat distribution of autoclave molding includes the following steps:
a. data preparation, namely analyzing, carding and identifying main influence factors of the surface heat distribution of a part in the autoclave curing molding process from structural parameters and temperature uniformity, and collecting data representing the main influence factors by reading temperature uniformity data and process parameter data which are collected in a database and reading structural parameters in a process digifax, wherein the temperature uniformity data comprises the temperature of the autoclave per minute, the running time of the temperature which is set from the beginning of the distance of each autoclave, the temperature of a thermocouple per minute, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool;
b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time of the temperature which is set from the distance of each tank to the beginning in the temperature uniformity data and the thermocouple temperature per minute;
c. model establishment and training: taking the normalized structure parameters, the temperature uniformity data and the degree of solidification as training data for establishing an SVM regression prediction model; constructing an SVM regression prediction model by taking the temperature of the autoclave per minute in the normalized structural parameters, the curing degree and the temperature uniformity data, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool as input parameters and taking the temperature of the thermocouple per minute in the temperature uniformity data as an output target parameter;
d. predicting data, acquiring parameters of the combined parts to be predicted, including the placement position of a tool in the autoclave, the position of a thermocouple arranged in the tool, structural parameters and process parameter data, and normalizing the structural parameters; calculating the autoclave temperature per minute and the operation duration according to the process parameter data, predicting a first minute thermocouple temperature value by using an SVM regression prediction model, calculating a second minute curing degree as an input attribute according to the first minute thermocouple temperature value, predicting a second minute thermocouple temperature value by using the SVM regression prediction model, iteratively calculating the thermocouple temperature per minute in the autoclave molding and curing process by using the SVM regression prediction model and the curing degree in sequence, and storing the thermocouple temperature per minute in a database;
e. and d, outputting a result, calculating the temperature difference of different areas of the autoclave molding and curing part according to the temperature value of the thermocouple per minute in the autoclave molding and curing process calculated in the step d, and judging the uniformity of the heat distribution on the surfaces of the parts which are combined into the autoclave.
In the step b, the degree of curing is calculated by the formulas 1 and 2;
in the formula, alpha is the curing degree, T is the thermocouple temperature value, and T is the operation time;
when t is 0, the degree of cure α is 0; r is 8.31434J/mol, A1=2.102×109min,A2=-2.102×109min,E1=8.07×104J/mol,E2=7.78×104J/mol,E3=5.66×104J/mol,A3=1.96×105min。
In the step d, the SVM regression prediction model is as follows:
wherein C is the inner product, K (x)i,xj) For the kernel function, a gaussian radial basis kernel function is used:
wherein x isjIs the kernel center and σ is the kernel width parameter.
In the step b, the structural parameter normalization is min-max normalization:
wherein x isiFor the original value of the sample, the value of,as a normalized value, xminIs the minimum value of the sample, xmaxIs the sample maximum.
In the temperature uniformity data, the placing position of the tool in the autoclave is represented by the distance from the tool to the autoclave door and the distance from the tool to the axial plane in the autoclave, and the thermocouple is arranged at the position of the tool and represented by the distance from the thermocouple to the front section of the tool and the distance from the thermocouple to the symmetrical plane of the tool.
The process parameter data comprises a temperature rise rate VTemperature riseConstant temperature TconDuration of constant temperature tconAnd a cooling rate VTemperature reductionAnd a cooling-down end temperature TdownAnd the structural parameters comprise tool information and part information.
The tool information comprises tool length, tool width, tool height, tool quality, tool external surface area, tool volume, tool heat exchange area and tool hot melting; the part information comprises part length, part width, part height, part quality, part surface area, part volume, part heat exchange area and part hot melting.
In the step d, calculating the autoclave temperature and the operation time per minute according to the process parameter data comprises the following steps:
s1, calculating the autoclave temperature per minute in the temperature rise stage according to the process parameter data;
Tt=T0+(t-1)×Vtemperature riseFormula 6
Wherein, TtAutoclave temperature, T, at the T minute0To start temperature, t is the temperature rise phase run minute, VTemperature riseThe rate of temperature rise; t ist<=Tcon,TconIs at a constant temperature;
s2, calculating the autoclave curing time in the constant temperature stage according to the process parameter data;
tconstant temperature=tWait for+tconFormula 7
Wherein, tConstant temperatureIs the length of the constant temperature stage of the actual autoclave, tconThe constant temperature duration of the slowest thermocouple, tWait forThe temperature of the autoclave reaches a constant temperature TconThe temperature of the thermocouple reaches the constant temperature T from the slowestconThe time of (d);
s3, calculating the autoclave temperature per minute in the cooling stage according to the process parameter data;
Tt=Tcon-(t-1)×Vtemperature reductionFormula 8
Wherein, TtAutoclave temperature at T minute, Tt<=Tcon,TconConstant temperature, t cooling phase operating minutes, VTemperature reductionIs the rate of temperature decrease.
The temperature per minute is obtained by utilizing an SVM algorithm and the iteration of the degree of solidification in the process of predicting the surface heat distribution of the part, the heat distribution in the process of predicting the solidification of the multi-part tank-entering combined part is predicted by utilizing a big data analysis method to replace the traditional finite element simulation technology, and compared with the finite element simulation method, the method has higher calculation efficiency.
Claims (7)
1. The prediction method for the molding and curing heat distribution of the autoclave is characterized by comprising the following steps:
a. data preparation, namely analyzing, carding and identifying main influence factors of the surface heat distribution of a part in the autoclave curing molding process from structural parameters and temperature uniformity, and collecting data representing the main influence factors by reading temperature uniformity data and process parameter data which are collected in a database and reading structural parameters in a process digifax, wherein the temperature uniformity data comprises the temperature of the autoclave per minute, the running time of the temperature which is set from the beginning of the distance of each autoclave, the temperature of a thermocouple per minute, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool;
b. data preprocessing, namely normalizing the structural parameters, and calculating the curing degree according to the operating time of the temperature which is set from the distance of each tank to the beginning in the temperature uniformity data and the thermocouple temperature per minute;
c. model establishment and training: taking the normalized structure parameters, the temperature uniformity data and the degree of solidification as training data for establishing an SVM regression prediction model; constructing an SVM regression prediction model by taking the temperature of the autoclave per minute in the normalized structural parameters, the curing degree and the temperature uniformity data, the placing position of the tool in the autoclave and the position of the thermocouple arranged in the tool as input parameters and taking the temperature of the thermocouple per minute in the temperature uniformity data as an output target parameter;
d. predicting data, acquiring parameters of the combined parts to be predicted, including the placement position of a tool in the autoclave, the position of a thermocouple arranged in the tool, structural parameters and process parameter data, and normalizing the structural parameters; calculating the autoclave temperature per minute and the operation duration according to the process parameter data, predicting a first minute thermocouple temperature value by using an SVM regression prediction model, calculating a second minute curing degree as an input attribute according to the first minute thermocouple temperature value, predicting a second minute thermocouple temperature value by using the SVM regression prediction model, iteratively calculating the thermocouple temperature per minute in the autoclave molding and curing process by using the SVM regression prediction model and the curing degree in sequence, and storing the thermocouple temperature per minute in a database;
e. and d, outputting a result, calculating the temperature difference of different areas of the autoclave molding and curing part according to the temperature value of the thermocouple per minute in the autoclave molding and curing process calculated in the step d, and judging the uniformity of the heat distribution on the surfaces of the parts which are combined into the autoclave.
2. The autoclave molding curing heat distribution prediction method according to claim 1, characterized in that: in the step b, the degree of curing is calculated by formula 1 and formula 2;
in the formula, alpha is the curing degree, T is the thermocouple temperature value, and T is the operation time;
when t is 0, the degree of cure α is 0; r is 8.31434J/mol, A1=2.102×109min,A2=-2.102×109min,E1=8.07×104J/mol,E2=7.78×104J/mol,E3=5.66×104J/mol,A3=1.96×105min。
3. The autoclave molding curing heat distribution prediction method according to claim 1, characterized in that: in the step b, the structural parameter normalization is min-max normalization:
4. The autoclave molding curing heat distribution prediction method according to claim 1, characterized in that: in the temperature uniformity data, the placing position of the tool in the autoclave is represented by the distance from the tool to the door of the autoclave and the distance from the tool to the axial plane of the autoclave, and the thermocouple is arranged at the position of the tool and is represented by the distance from the thermocouple to the front section of the tool and the distance from the thermocouple to the symmetrical plane of the tool.
5. The autoclave molding curing heat distribution prediction method according to claim 1, characterized in that: the process parameter data comprises a temperature rise rate VTemperature riseConstant temperature TconConstant temperature duration tconAnd a cooling rate VTemperature reductionAnd a cooling-down end temperature TdownAnd the structural parameters comprise tool information and part information.
6. The autoclave molding curing heat distribution prediction method according to claim 5, characterized in that: the tool information comprises tool length, tool width, tool height, tool quality, tool external surface area, tool volume, tool heat exchange area and tool hot melting; the part information comprises part length, part width, part height, part quality, part surface area, part volume, part heat exchange area and part hot melting.
7. The autoclave molding curing heat distribution prediction method according to claim 1, characterized in that: in the step d, calculating the autoclave temperature per minute and the operation duration according to the process parameter data comprises the following steps:
s1, calculating the autoclave temperature per minute in the temperature rise stage according to the process parameter data;
Tt=T0+(t-1)×Vtemperature riseFormula 6
Wherein, TtAutoclave temperature, T, at the T minute0To start temperature, t is the temperature rise phase run minute, VTemperature riseThe rate of temperature rise; t is a unit oft<=Tcon,TconIs at a constant temperature;
s2, calculating the autoclave curing time in the constant temperature stage according to the process parameter data;
tconstant temperature=tWait for+tconFormula 7
Wherein, tConstant temperatureIs made ofLength of constant temperature phase of boundary autoclave, tconThe constant temperature duration of the slowest thermocouple, tWait forThe temperature of the autoclave reaches a constant temperature TconThe temperature of the thermocouple reaches the constant temperature T from the slowestconThe time of (d);
s3, calculating the autoclave temperature per minute in the cooling stage according to the process parameter data;
Tt=Tcon-(t-1)×Vtemperature reductionFormula 8
Wherein, TtAutoclave temperature at T minute, Tt<=Tcon,TconConstant temperature, t cooling phase run minutes, VTemperature reductionIs the rate of temperature decrease.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110480808.4A CN113343375B (en) | 2021-04-30 | 2021-04-30 | Prediction method for autoclave molding curing heat distribution |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110480808.4A CN113343375B (en) | 2021-04-30 | 2021-04-30 | Prediction method for autoclave molding curing heat distribution |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113343375A CN113343375A (en) | 2021-09-03 |
CN113343375B true CN113343375B (en) | 2022-07-15 |
Family
ID=77469213
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110480808.4A Active CN113343375B (en) | 2021-04-30 | 2021-04-30 | Prediction method for autoclave molding curing heat distribution |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113343375B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115169137A (en) * | 2022-07-20 | 2022-10-11 | 成都飞机工业(集团)有限责任公司 | Method, system, equipment and medium for predicting autoclave molding curing process |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111795721A (en) * | 2020-06-04 | 2020-10-20 | 成都飞机工业(集团)有限责任公司 | Real-time online detection method for autoclave curing data |
CN112182678A (en) * | 2020-09-24 | 2021-01-05 | 西北工业大学 | Autoclave forming method with co-designed curing quality and curing cost |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20210052218A1 (en) * | 2019-08-20 | 2021-02-25 | Patchd, Inc. | Systems and methods for sepsis detection and monitoring |
CN110991011B (en) * | 2019-11-14 | 2023-05-02 | 广东电网有限责任公司 | Transformer real-time hot spot temperature prediction method |
CN110889238B (en) * | 2019-12-06 | 2021-06-15 | 北京航空航天大学 | Autoclave management and control system driven by digital twin |
-
2021
- 2021-04-30 CN CN202110480808.4A patent/CN113343375B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111795721A (en) * | 2020-06-04 | 2020-10-20 | 成都飞机工业(集团)有限责任公司 | Real-time online detection method for autoclave curing data |
CN112182678A (en) * | 2020-09-24 | 2021-01-05 | 西北工业大学 | Autoclave forming method with co-designed curing quality and curing cost |
Also Published As
Publication number | Publication date |
---|---|
CN113343375A (en) | 2021-09-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113343375B (en) | Prediction method for autoclave molding curing heat distribution | |
US8343402B1 (en) | Consolidation of composite material | |
CN108062427A (en) | The method that gradient rate controlling based on numerical computations reduces turbine disk forging residual stress | |
CN104504185A (en) | Titanium alloy blooming and forging process method based on fracture criterion and finite element optimization | |
Huynh et al. | Minimizing Warpage for Macro-Size Fused Deposition Modeling Parts. | |
CN113919601A (en) | Resin process prediction method and device based on product performance and process data model | |
Yusoff et al. | Work measurement for process improvement in the car seat polyurethane injection manufacturing line | |
CN102682171A (en) | Method for compensating thermal deformation of mold for high-precision composite material antenna reflector unit | |
CN207498274U (en) | A kind of automation glass bending former | |
CN102452174B (en) | Air pressure shaping technology of carbon fiber product and glass fiber product and used die | |
CN110000246A (en) | A kind of large-scale complex thin-wall titanium alloy covering part stress relaxation straightening method in short-term | |
CN109676001A (en) | A kind of method that forming prepares aluminium alloy element product | |
CN108688114B (en) | Quick temperature-changing injection molding method and molding equipment thereof | |
CN116562184B (en) | Rapid prediction calculation method for mold temperature field in autoclave molding process | |
CN109732815A (en) | A kind of method that forming prepares fiber resin composite material component product | |
CN113409894B (en) | Prediction method for microstructure change of near-alpha type titanium alloy aviation die forging | |
CN106985326B (en) | A kind of multimode mold for moulded of plastics material | |
CN109177230A (en) | A method of improving composite product mass production efficiency | |
CN109562442A (en) | The life-span prediction method of mold | |
CN111730882A (en) | Analysis and tracing system and method for autoclave molding process data | |
Bromley | Investigation of integrally-heated tooling and thermal modeling methodologies for the rapid cure of aerospace composites | |
CN107915397A (en) | Bend glass production method for glass heat bender | |
JPH075715U (en) | Blade molding equipment | |
CN113283097A (en) | Die quenching and cooling multi-cycle analysis method based on CAE simulation technology | |
Jayaraman | DIMENSIONAL STABILITY OF LOW-COST THERMOPLASTIC COMPOSITE MOLDS |
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 |