A process for enhancing the dielectric strength of non-conductive composite materials

Abstract

The present invention relates to the technical field of dielectric strength, and discloses a process method for enhancing the dielectric strength of non-conductive composite materials, comprising the following steps: evaluating the maximum pressure that the existing non-conductive composite materials can withstand, and selecting a pressure not greater than the maximum pressure to set the autoclave; sampling in the test piece where the non-conductive composite materials are applied, and performing ply parameter analysis according to the original structure of the sampled part; and predicting the lowest void ratio of different process parameter combinations by learning from historical production data. The present invention predicts a better parameter combination using an algorithm without adjusting any structural design, and then adjusts the pressurization process requirements according to the parameter combination, increases the structural density, and thus reduces the internal void ratio, thereby achieving an improvement in the dielectric strength of the non-conductive composite material itself, which is beneficial to the applicability of non-conductive materials in lightning protection.

Description

Technical Field

[0001] The present invention relates to the technical field of dielectric strength, and more particularly to a process for enhancing the dielectric strength of a non-conductive composite material. Background Art

[0002] Non-conductive composite materials are prone to structural perforation in lightning electric field environments. With the increasing application of new materials in aviation, aerospace, ships, ground radar and other fields, the requirements for electrical properties of non-conductive composite materials are getting higher and higher. For example, the frequency of use of aircraft radomes and antenna covers is increasing. In order to reduce the impact of the structure on radio waves, the thickness of the cover is getting thinner and thinner, and its own lightning resistance is becoming more and more limited. At the same time, compared with the traditional structure of radar antenna covers, the degree of integration is more obvious, and the distance between the internal antenna and the cover is almost zero, which makes it easier for radar and antenna covers to be perforated under the action of high lightning electric fields.

[0003] The application of this method can effectively improve the dielectric strength of the non-conductive composite material structure itself, that is, with the same structural material organization form and thickness, the dielectric strength of the structure produced by this process method is higher than that of the ordinary form. Summary of the invention

[0004] The invention provides a process for enhancing the dielectric strength of a non-conductive composite material, and solves the technical problems in the related art.

[0005] The present invention provides a process for enhancing the dielectric strength of a non-conductive composite material, comprising the following steps:

[0006] Step S100: Evaluating the maximum pressure that the existing non-conductive composite material can withstand , and choose a pressure not greater than the maximum The autoclave is set at a pressure of

[0007] Step S200: sampling a test piece in which the non-conductive composite material is applied, and performing ply parameter analysis according to the original structure of the sampled portion;

[0008] Step S300: predicting the lowest void ratio of different process parameter combinations by learning historical production data;

[0009] The process parameters include hot pressing parameters, layering parameters and post-processing parameters;

[0010] Hot pressing parameters include pressure, temperature and holding time;

[0011] Laying parameters include the number of layers, laying direction and laying method;

[0012] Post-processing parameters include annealing time, annealing temperature and annealing humidity;

[0013] Step S400: according to the predicted process parameter combination, the part void ratio is controlled within 1.5%, and the process parameters are set to produce a test piece;

[0014] Step S500: preparing a comparative test block while preparing the test piece. After the test piece and the comparative test block are solidified, non-destructive testing is required. The test piece and the comparative test block are transported by a transport device and transported to a testing platform for non-destructive testing.

[0015] Step S600: Collect data from the preparation process, use data analysis algorithms to generate a detailed evaluation report, including process performance indicators and potential improvement points, and write a process evaluation report.

[0016] Furthermore, by studying historical production data, the steps to predict the lowest void ratio for different process parameter combinations are as follows:

[0017] Step S310: collecting various process parameters and their corresponding void fraction data recorded in the previous production process;

[0018] Step S320: Process the above process parameters and void ratios, remove invalid or erroneous data points, and convert or create new feature variables as needed;

[0019] Step S330: using a machine learning model to fit the relationship between process parameters and void fraction, and using historical data to train the selected machine learning model;

[0020] Given a set of input and output , the goal of model training is to find a set of parameters , so that the model best fit to the data;

[0021] The linear regression model is expressed as follows:

[0022]

[0023] in, Indicates different process parameters, is the weight corresponding to each parameter, is the error term;

[0024] Step S340: Define the objective function, based on the void ratio Minimize as the goal, use gradient descent to minimize , thus obtaining the optimal parameters ;

[0025] The basic form of gradient descent is ;

[0026] in is the learning rate, is the objective function with respect to the parameter The partial derivative of .

[0027] Furthermore, the predicted process parameter combinations are as follows:

[0028] S410: Setting target void ratio ;

[0029] S420: Based on the prediction result in step S300, a set of process parameter combinations is selected. , the combination includes:

[0030] Hot pressing parameters: ;

[0031] Laying parameters: ;

[0032] Post-processing parameters: ;

[0033] in: Indicates stress; Indicates temperature; Indicates the holding time; Indicates the number of plies; Indicates the direction of ply laying; Indicates the laying method; Indicates annealing treatment time; represents the annealing temperature; Indicates annealing humidity;

[0034] S430: According to the selected hot pressing parameter combination , set the working conditions of the autoclave;

[0035] S440: According to the determined laying parameters , prepare the composite material layup so that the number, direction and pattern of layup meet the expected design requirements;

[0036] S450: According to the selected post-processing parameters , perform necessary annealing treatment on the specimen;

[0037] S460: After all the process parameters in the above S410-S450 are set, prepare the test piece.

[0038] Furthermore, the porosity of the specimen prepared in S460 is within the target range, as represented by the optimization model:

[0039]

[0040] in, is the porosity obtained according to the prediction model in step S300, and is the target void ratio;

[0041] Through this optimization model, we can find a set of optimal process parameters , making the predicted void fraction close to the target value.

[0042] Furthermore, the specific steps of step S500 are as follows:

[0043] S510: Prepare a series of comparative test blocks by using the same materials and process parameters as the test pieces and changing a single variable;

[0044] S520: placing the test piece and the comparison test block under the same conditions for curing, unifying the hot pressing process;

[0045] S530: After the specimen and the comparison test block are cured, they are ready for non-destructive testing;

[0046] S540: Use handling equipment to safely transport the test pieces and reference test blocks to the nondestructive testing platform; the test pieces and reference test blocks must not be damaged or deformed during the handling process;

[0047] S550: Perform nondestructive testing on the test pieces and comparison test blocks, and record the test results.

[0048] Further, in step S550, nondestructive testing includes ultrasonic testing, radiographic testing, and magnetic particle testing methods, and internal defects and voids are checked by nondestructive testing.

[0049] Furthermore, the specific content of step S600 is as follows:

[0050] Data preprocessing: clean the data to ensure that there are no missing values ​​or outliers that affect the analysis results;

[0051] Feature engineering: extracting features for modeling;

[0052] Model building: Use algorithms to predict void fraction;

[0053] Model training and validation: Train the model and evaluate the model performance using cross-validation techniques;

[0054] Parameter adjustment: adjust model parameters to improve prediction accuracy;

[0055] Results interpretation and report generation: Interpret the model results and generate a process evaluation report.

[0056] Furthermore, the model is established as follows:

[0057] Linear Regression Model:

[0058]

[0059] The loss function is minimized:

[0060]

[0061] in, is the coefficient, is the eigenvector, is the error term.

[0062] A system for enhancing the dielectric strength of a non-conductive composite material, used for performing one or more steps of the aforementioned process for enhancing the dielectric strength of a non-conductive composite material, comprising:

[0063] Material evaluation and parameter setting module: evaluates the physical properties of non-conductive composite materials, including the maximum pressure they can withstand, and sets the working pressure of the autoclave to not exceed the maximum pressure the material can withstand;

[0064] Sampling and ply analysis module: take samples from the test piece, analyze the ply parameters of the sampled part, and record and analyze the number, direction and method of ply layup;

[0065] Data analysis and void ratio prediction module: Utilizes historical production data and uses machine learning to predict void ratios under different process parameters, and outputs the process parameter combination with the lowest void ratio;

[0066] Process parameter optimization and test piece manufacturing module: optimize the process parameters according to the prediction results, the void ratio of the product does not exceed 1.5%, and manufacture the test pieces according to the optimized parameters to manufacture the test pieces that meet the requirements;

[0067] Nondestructive testing module: Perform nondestructive testing on the prepared test pieces and comparison test blocks at the same time and after curing, and output the test results of the test pieces and comparison test blocks.

[0068] Evaluation report generation module: collects data from the preparation process and uses data analysis algorithms to generate an evaluation report that includes process performance indicators and potential improvement points.

[0069] A storage medium stores non-transitory computer-readable instructions for executing one or more steps of the aforementioned process for enhancing the dielectric strength of a non-conductive composite material.

[0070] The beneficial effects of the present invention are:

[0071] The present invention predicts a better parameter combination by using an algorithm without adjusting any structural design, and then adjusts the pressurization process requirements according to the parameter combination to increase the density of the structure, thereby reducing the internal void ratio, thereby achieving an improvement in the dielectric strength of the non-conductive composite material itself, which is beneficial to the applicability of the non-conductive material in lightning protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a flow chart of a process for improving the dielectric strength of a non-conductive composite material proposed by the present invention;

[0073] Figure 2 It is a structural block diagram of a system for enhancing the dielectric strength of a non-conductive composite material proposed by the present invention;

[0074] Figure 3 It is a data table of the curing process pressure and the dielectric strength after molding of the present invention;

[0075] Figure 4 It is a line graph showing the relationship between the molding pressure and the dielectric strength of the present invention.

[0076] In the figure: 101, material evaluation and parameter setting module; 102, sampling and layer analysis module; 103, data analysis and void ratio prediction module; 104, process parameter optimization and specimen manufacturing module; 105, non-destructive testing module; 106, evaluation report generation module. DETAILED DESCRIPTION

[0077] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that the discussion of these implementations is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the present specification. Various examples may omit, replace, or add various processes or components as needed. In addition, the features described in some examples may also be combined in other examples.

[0078] At least one embodiment of the present invention discloses a process for enhancing the dielectric strength of a non-conductive composite material, such as Figure 1-Figure 3 As shown, the following steps are included:

[0079] Step S100: Evaluating the maximum pressure that the existing non-conductive composite material can withstand , and choose a pressure not greater than the maximum The autoclave is set at a pressure of

[0080] 1. Collect relevant information about non-conductive composite materials

[0081] Material Type ; Material thickness is (Unit: mm); Material width is (Unit: mm).

[0082] 2. Test the compressive strength of non-conductive composite materials

[0083] The number of test samples is ;

[0084] Test sample size: According to standard size, such as ASTM D695 or ISO 14125, etc.

[0085] Test conditions: including loading rate, etc.

[0086] 3. Calculate the average compressive strength

[0087] Compressive strength of a single sample: (Unit: MPa)

[0088] Average compressive strength:

[0089] 4. Determine the maximum pressure the material can withstand

[0090] Safety factor: (Generally the value range is 1.2-2.0)

[0091] Maximum pressure that the material can withstand:

[0092] 5. Set the autoclave pressure

[0093] Actual set pressure of autoclave: ,Require

[0094] In one embodiment of the present invention, the results of the parameter evaluation are as follows:

[0095] Average compressive strength:

[0096] Safety factor:

[0097] The calculation results are:

[0098] If the pressure of the autoclave can be set, e.g. , then the condition is met .

[0099] It should be noted that in determining When preparing composite materials, the mechanical limitations of the autoclave and other physical properties of the composite material (such as temperature sensitivity, etc.) must also be considered.

[0100] In actual operation, it may be necessary to adjust based on experience to achieve the desired effect.

[0101] Through the above steps, the pressure of the autoclave can be reasonably set to ensure that the composite material will not be damaged due to excessive pressure during the hot pressing process, and the void ratio can be effectively reduced to improve product quality.

[0102] Step S200: sampling a test piece in which the non-conductive composite material is applied, and performing ply parameter analysis according to the original structure of the sampled portion;

[0103] In one embodiment of the present invention, the specific steps are as follows:

[0104] 1. Select a representative sample

[0105] Sample location: Select samples from different areas of the test piece (such as center, edge) to ensure that the samples are representative.

[0106] Sample size: (The quantity is determined based on statistical requirements and the size of the test piece).

[0107] 2. Sample Preparation

[0108] Sample size: As required by standard test methods, such as ASTM or ISO standards.

[0109] Sample preparation: Ensure that the sample surface is clean and free of contamination or damage.

[0110] 3. Analysis of ply parameters

[0111] Number of layers: (Thickness of each layer and total number of layers).

[0112] Laying direction: (The direction of each layer of fibers is usually expressed as an angle, such as 0°, 90°, +45°, -45°).

[0113] Layup method: describes how each layer is laid, such as unidirectional fabric layers, bidirectional fabric layers, or 3D weave.

[0114] 4. Recording and Measurement

[0115] Interlayer bond quality: can be assessed by ultrasonic testing or other non-destructive testing methods.

[0116] Layer thickness measurement: Use a microscope or other precision measuring tools to measure the thickness of each layer .

[0117] 5. Data collation

[0118] Average layer thickness:

[0119]

[0120] Inter-layer orientation change: records the angle change of each layer relative to the previous layer.

[0121] 6. Analysis and Evaluation

[0122] Inter-layer consistency: Check whether all layers are evenly distributed and whether there are any defects or gaps.

[0123] Ply Angle Deviation: Calculates the deviation between the actual ply orientation and the designed orientation.

[0124] In one embodiment of the present invention, the ply parameters are analyzed as follows:

[0125] The measured data are as follows: Number of plies The thickness of each layer is ; And then calculate: ;

[0126] If you need to calculate the deviation of a layer relative to the design angle, for example, the design angle is 45 degrees and the actual angle is 46 degrees, then the deviation is 1 degree.

[0127] Step S300: predicting the lowest void ratio of different process parameter combinations by learning historical production data;

[0128] The process parameters include hot pressing parameters, layering parameters and post-processing parameters;

[0129] Hot pressing parameters include pressure, temperature and holding time;

[0130] Laying parameters include the number of layers, laying direction and laying method;

[0131] Post-processing parameters include annealing time, annealing temperature and annealing humidity;

[0132] In one embodiment of the present invention, the specific steps are as follows:

[0133] 1. Collect historical data

[0134] Data source: Collect various process parameters and their corresponding void ratio data recorded in the previous production process.

[0135] Parameter range: Hot pressing parameters (pressure ,temperature , holding time ), ply parameters (number of plies , ply direction , Lamination method ), post-processing parameters (annealing time , annealing temperature , annealing humidity ).

[0136] 2. Data Preprocessing

[0137] Data cleaning: Removing invalid or erroneous data points.

[0138] Feature Engineering: Transform or create new feature variables as needed.

[0139] 3. Build the model

[0140] Select model: Select an appropriate machine learning model (such as regression analysis, decision tree, neural network, etc.) to fit the relationship between process parameters and void fraction.

[0141] For the problem of predicting void fraction, regression models such as linear regression, polynomial regression or more complex machine learning models such as support vector machine (SVM), random forest or deep learning models can be used.

[0142] Train Model: Use historical data to train the selected model.

[0143] Given a set of input and output (here is the porosity), the goal of model training is to find a set of parameters , so that the model best fit to the data.

[0144] Linear Regression Model

[0145]

[0146] in, Indicates different process parameters (e.g. , etc), is the weight corresponding to each parameter, is the error term.

[0147] Polynomial regression model

[0148] If the linear model is not accurate enough, you can try a higher order polynomial model:

[0149]

[0150] Validate the model: Use techniques such as cross-validation to assess the accuracy of the model.

[0151] Use techniques such as cross-validation to assess the generalization ability of the model and ensure that the model is not overfitting or underfitting.

[0152] 4. Parameter Optimization

[0153] Define the objective function: void ratio Minimization is the goal.

[0154] Refine your search

[0155] Use optimization algorithms (such as gradient descent, quasi-Newton method, etc.) to minimize , thus obtaining the optimal parameters .

[0156] Gradient Descent Algorithm

[0157] The basic form of gradient descent is in is the learning rate, is the objective function with respect to the parameter The partial derivative of .

[0158] In one embodiment of the present invention, as follows:

[0159] Find the minimum The parameter combination ( .

[0160] Define an objective function , usually with the goal of minimizing the prediction error of the void fraction, such as the mean squared error (MSE):

[0161]

[0162] in, is the sample size, and They are Features and labels of samples.

[0163] 5. Model Evaluation

[0164] Performance indicators: Use indicators such as mean square error (MSE), coefficient of determination R2, etc. to measure the predictive ability of the model.

[0165] Sensitivity analysis: Analyze the influence of different parameters on the void ratio.

[0166] Evaluate the extent to which each process parameter affects the void fraction. This can be done by calculating the partial derivatives of each parameter or using other statistical methods such as principal component analysis (PCA).

[0167] 6. Application of results

[0168] Recommended parameter combination: Based on the results predicted by the model, one or more sets of process parameters that can achieve the lowest void ratio are recommended.

[0169] Implementation suggestions: Provide specific process parameter setting guidelines for the next step of actual production testing.

[0170] In one embodiment of the present invention, a simple linear regression model is used to predict the void fraction , then the model is in the following form:

[0171]

[0172] in, is the weight coefficient, is the bias term.

[0173] The goal of the model is to minimize the gap between the predicted value and the actual value, that is, to minimize the loss function.

[0174] In practical applications, more complex models such as polynomial regression or nonlinear models may also be considered, depending on the data characteristics and the requirements for prediction accuracy.

[0175] Step S400: according to the predicted process parameter combination, the part void ratio is controlled within 1.5%, and the process parameters are set to produce a test piece;

[0176] 1. Determine the target void ratio

[0177] Setting target void ratio .

[0178] 2. Process parameter combination

[0179] Based on the prediction result in step S300, a set of process parameter combinations is selected. , the combination includes:

[0180] Hot pressing parameters:

[0181] Laying parameters:

[0182] Post-processing parameters:

[0183] in: Indicates stress; Indicates temperature; Indicates the holding time; Indicates the number of plies; Indicates the direction of ply laying; Indicates the laying method; Indicates annealing treatment time; represents the annealing temperature; Indicates annealing humidity.

[0184] 3. Set the autoclave parameters

[0185] According to the selected hot pressing parameters combination , set the working conditions of the autoclave to ensure that the process requirements are met.

[0186] 4. Layer design

[0187] According to the determined laying parameters , prepare the layup of composite materials, ensuring that the number, direction and method of layup meet the expected design requirements.

[0188] 5. Post-processing

[0189] According to the selected post-processing parameters , carry out necessary annealing treatment on the specimen to ensure its physical and mechanical properties.

[0190] 6. Preparation of test pieces

[0191] After all the above process parameters are set, start preparing the test piece to ensure that the entire process follows the set parameters to achieve the target void ratio requirements.

[0192] In one embodiment of the present invention, in order to ensure that the void ratio of the specimen is within the target range, this process is represented by an optimization model:

[0193]

[0194] in, is the porosity obtained according to the prediction model in step S300, and is the target void ratio.

[0195] Through the above model, find a set of optimal process parameters , so that the predicted void fraction is as close to the target value as possible.

[0196] Step S500: preparing a comparative test block while preparing the test piece. After the test piece and the comparative test block are solidified, non-destructive testing is required. The test piece and the comparative test block are transported by a transport device and transported to a testing platform for non-destructive testing.

[0197] 1. Preparation of comparison test blocks

[0198] A series of comparative test pieces are prepared using the same materials and process parameters as the test pieces, but keeping certain parameters constant, such as changing only a single variable (such as temperature, pressure, etc.). This helps to evaluate the impact of different process parameters on the quality of the final product.

[0199] 2. Curing

[0200] The specimens and control blocks were cured under the same conditions to ensure that they underwent the same hot pressing process.

[0201] 3. Nondestructive testing preparation

[0202] After the specimens and reference blocks are cured, they are ready for non-destructive testing. Non-destructive testing usually includes but is not limited to ultrasonic testing (UT), radiographic testing (RT), magnetic particle testing (MT) and other methods to check for internal defects and voids.

[0203] 4. Handling

[0204] Use special handling equipment to safely transport the test pieces and reference test blocks to the non-destructive testing platform. During the handling process, ensure that the test pieces and test blocks will not be damaged or deformed.

[0205] 5. Implement non-destructive testing

[0206] Perform non-destructive testing on the test pieces and comparison test blocks, and record the test results.

[0207] It should be noted that the specimen is represented by , the comparison block is expressed as , the curing parameters (which may include temperature, time and pressure) are expressed as , the nondestructive testing method is expressed as , the nondestructive testing results are expressed as :

[0208] For the results of nondestructive testing, the results are expressed as , Indicates the test piece And comparison block Results obtained after non-destructive testing.

[0209] Test results Including information such as void ratio, crack length, location, etc., which will be used for subsequent analysis and report generation.

[0210] In this way, the NDT results of each test piece and its corresponding reference block can be systematically recorded and tracked, leading to a better understanding of the impact of different process parameters on the final product quality.

[0211] Step S600: Collect data from the preparation process, use data analysis algorithms to generate a detailed evaluation report, including process performance indicators and potential improvement points, and write a process evaluation report.

[0212] 1. Data Collection

[0213] Collect all data from step S100 to S500, including but not limited to hot pressing parameters (pressure ,temperature , holding time ), ply parameters (number of plies , ply direction , Lamination method ), post-processing parameters (annealing time , annealing temperature , annealing humidity ) and non-destructive testing results (such as porosity , crack situation wait).

[0214] 2. Data Analysis

[0215] The collected data are processed using data analysis algorithms to identify the key factors affecting the void ratio.

[0216] Involving statistical analysis, regression analysis, or machine learning models;

[0217] Analyze the differences between the test pieces and the reference blocks to determine which process parameter adjustments are most effective.

[0218] Among them, the specific content of the data analysis algorithm is:

[0219] (1) Data preprocessing: Clean the data to ensure that there are no missing values ​​or outliers that affect the analysis results.

[0220] For missing values, use the mean value to fill For outliers, you can use (interquartile range) method for identification.

[0221] Outlier identification formula: and ,in and are the first and third quartiles, .

[0222] (2) Feature engineering: extracting meaningful features for modeling.

[0223] Calculate the correlation matrix , find features that are highly correlated with the target variable (such as void fraction).

[0224] Correlation coefficient:

[0225]

[0226] (3) Model building: Use appropriate algorithms to predict void fraction.

[0227] Assume that a linear regression model is used:

[0228]

[0229] in is the coefficient, is the eigenvector, is the error term.

[0230] The loss function is minimized:

[0231]

[0232] (4) Model training and validation: Train the model and evaluate model performance using cross-validation techniques.

[0233] Using mean square error or coefficient of determination as evaluation criteria.

[0234] (5) Parameter adjustment: Adjust model parameters to improve prediction accuracy.

[0235] Use grid search or random search techniques to find the best hyperparameter configuration.

[0236] For example, in Search for the optimal value within the range .

[0237] (6) Result interpretation and report generation: Interpret the model results and generate a process evaluation report.

[0238] The feature importance score F is used to explain which process parameters have the greatest impact on the void fraction.

[0239] The feature importance score can be the absolute value of the regression coefficient Or other methods for evaluating feature contribution.

[0240] Through the above steps, a detailed evaluation report can be obtained, which includes: key indicators of process performance (such as void fraction), potential improvement points (which process parameters need to be adjusted) and recommended parameter settings to further optimize performance.

[0241] Improve the quality of your non-conductive composites by adjusting process parameters according to the recommendations in the report.

[0242] 3. Evaluation report generation

[0243] Generate a detailed assessment report based on the analysis results. The report should include the following:

[0244] Process performance indicators: comparison between actual values ​​and target values ​​of key performance indicators such as porosity, strength, and hardness.

[0245] Potential improvement points: Based on the problem areas found by data analysis, make improvement suggestions.

[0246] Data visualization: Display data analysis results through charts (such as bar charts, scatter plots, trend charts, etc.) to make the report easier to understand.

[0247] Conclusions and recommendations: Summarize the main findings of the entire research process and put forward suggestions for further research or optimization.

[0248] In one embodiment of the present invention, examples of void ratios are as follows:

[0249] Void ratio The calculation formula is:

[0250]

[0251] in, represents the void volume, Indicates the total material volume;

[0252] The deviation analysis of the performance indicators is expressed as:

[0253]

[0254] in, It is The actual value of the performance indicator, is the target value.

[0255] If regression analysis is used to predict void fraction, the model is established:

[0256]

[0257] in, It is a function that represents the relationship between void fraction and all process parameters.

[0258] Through the above steps and mathematical expressions, data can be systematically collected and analyzed, and a detailed evaluation report can be generated to help understand the performance of the existing process and point out possible improvements.

[0259] In one embodiment of the present invention, according to steps S100-S600, a 350mm×350mm composite board made of FQW199 / 5224A epoxy resin is taken as an example to compare the DC dielectric strength under different pressures under the same design and thickness conditions.

[0260] It can be seen that after the FQW199 / 5224A epoxy resin system structure panels were molded using four pressures of 0.1MPa, 0.2MPa, 0.4MPa, and 0.6MPa, the dielectric strength of the composite panels at 0.2MPa, 0.4MPa, and 0.6MPa increased by 31.52%, 60.64%, and 102.13% respectively compared with the composite panels cured at normal pressure. The experiment proves that this process can enhance the dielectric strength of non-conductive composite materials.

[0261] like Figure 2 A system for enhancing the dielectric strength of a non-conductive composite material is shown, comprising:

[0262] Material evaluation and parameter setting module: evaluates the physical properties of non-conductive composite materials, especially the maximum pressure they can withstand, and sets the working pressure of the autoclave not to exceed the maximum pressure of the material;

[0263] Sampling and ply analysis module: take samples from the test piece, analyze the ply parameters of the sampled part, and record and analyze the number, direction and method of ply layup;

[0264] Data analysis and void ratio prediction module: Use historical production data to predict void ratios under different process parameters through machine learning or other prediction algorithms, and predict the minimum void ratio under different combinations of process parameters;

[0265] Process parameter optimization and test piece manufacturing module: optimize the process parameters according to the prediction results to ensure that the void ratio of the final product does not exceed 1.5%, and manufacture the test pieces according to the optimized parameters to manufacture the test pieces that meet the requirements;

[0266] Nondestructive testing module: Perform nondestructive testing on the prepared test pieces and comparison test blocks at the same time and after curing, and output the test results of the test pieces and comparison test blocks.

[0267] Evaluation report generation module: collects data from the entire preparation process and uses data analysis algorithms to generate an evaluation report that includes process performance indicators and potential improvement points.

[0268] At least one embodiment of the present disclosure provides a storage medium storing non-transitory computer-readable instructions for executing one or more steps in the aforementioned process for enhancing the dielectric strength of a non-conductive composite material.

[0269] The computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

[0270] The above describes an embodiment of the present invention, but this embodiment is not limited to the above-mentioned specific implementation mode. The above-mentioned specific implementation mode is merely illustrative and not restrictive. Under the guidance of this embodiment, ordinary technicians in this field can also make more forms of equivalent embodiments, all of which are within the protection of this embodiment.

Claims

1. A process for enhancing the dielectric strength of a non-conductive composite material, characterized in that: The following steps are involved: Step S100: Evaluating the maximum pressure that the existing non-conductive composite material can withstand , and choose a pressure not greater than the maximum The autoclave is set at a pressure of Step S200: sampling a test piece in which the non-conductive composite material is applied, and performing ply parameter analysis according to the original structure of the sampled portion; Step S300: predicting the lowest void ratio of different process parameter combinations by learning historical production data; The process parameters include hot pressing parameters, layering parameters and post-processing parameters; Hot pressing parameters include pressure, temperature and holding time; Laying parameters include the number of layers, laying direction and laying method; Post-processing parameters include annealing time, annealing temperature and annealing humidity; Step S400: according to the predicted process parameter combination, the part void ratio is controlled within 1.5%, and the process parameters are set to produce a test piece; Step S500: preparing a comparative test block while preparing the test piece. After the test piece and the comparative test block are solidified, non-destructive testing is required. The test piece and the comparative test block are transported by a transport device and transported to a testing platform for non-destructive testing. Step S600: Collect data from the preparation process, use data analysis algorithms to generate a detailed evaluation report, including process performance indicators and potential improvement points, and write a process evaluation report.

2. A process for enhancing the dielectric strength of a non-conductive composite material according to claim 1, characterized in that: By studying historical production data, the steps to predict the lowest void ratio for different process parameter combinations are as follows: Step S310: collecting various process parameters and their corresponding void fraction data recorded in the previous production process; Step S320: Process the above process parameters and void ratios, remove invalid or erroneous data points, and convert or create new feature variables as needed; Step S330: using a machine learning model to fit the relationship between process parameters and void fraction, and using historical data to train the selected machine learning model; Given a set of input and output , the goal of model training is to find a set of parameters , so that the model best fit to the data; The linear regression model is expressed as follows: ； in, Indicates different process parameters, is the weight corresponding to each parameter, is the error term; Step S340: Define the objective function, based on the void ratio Minimize as the goal, use gradient descent to minimize , thus obtaining the optimal parameters ; The basic form of gradient descent is ; in is the learning rate, is the objective function with respect to the parameter The partial derivative of .

3. A process for enhancing the dielectric strength of a non-conductive composite material according to claim 2, characterized in that: The predicted process parameter combinations are as follows: S410: Setting target void ratio ; S420: Based on the prediction result in step S300, a set of process parameter combinations is selected. , the combination includes: Hot pressing parameters: ; Laying parameters: ; Post-processing parameters: ; in: Indicates stress; Indicates temperature; Indicates the holding time; Indicates the number of plies; Indicates the direction of ply laying; Indicates the laying method; Indicates annealing treatment time; represents the annealing temperature; Indicates annealing humidity; S430: According to the selected hot pressing parameter combination , set the working conditions of the autoclave; S440: According to the determined laying parameters , prepare the composite material layup so that the number, direction and pattern of layup meet the expected design requirements; S450: According to the selected post-processing parameters , perform necessary annealing treatment on the specimen; S460: After all the process parameters in the above S410-S450 are set, prepare the test piece.

4. A process for enhancing the dielectric strength of a non-conductive composite material according to claim 3, characterized in that: The porosity of the specimens prepared in S460 is within the target range, as represented by the optimization model: ； in, is the porosity obtained according to the prediction model in step S300, and is the target void ratio; Through this optimization model, we can find a set of optimal process parameters , making the predicted void fraction close to the target value.

5. A process for enhancing the dielectric strength of a non-conductive composite material according to claim 4, characterized in that: The specific steps of step S500 are as follows: S510: Use the same material and process parameters as the test piece and change a single variable to prepare a series of comparison test pieces; S520: placing the test piece and the comparison test block under the same conditions for curing, unifying the hot pressing process; S530: After the specimen and the comparison test block are cured, they are ready for non-destructive testing; S540: Use handling equipment to safely transport the test pieces and reference test blocks to the nondestructive testing platform; the test pieces and reference test blocks must not be damaged or deformed during the handling process; S550: Perform nondestructive testing on the test pieces and comparison test blocks, and record the test results.

6. A process for enhancing the dielectric strength of a non-conductive composite material according to claim 5, characterized in that: In step S550, nondestructive testing includes ultrasonic testing, radiographic testing, and magnetic particle testing methods, and internal defects and voids are checked through nondestructive testing.

7. A process for enhancing the dielectric strength of a non-conductive composite material according to claim 6, characterized in that: The specific content of step S600 is as follows: Data preprocessing: clean the data to ensure that there are no missing values ​​or outliers that affect the analysis results; Feature engineering: extracting features for modeling; Model building: Use algorithms to predict void fraction; Model training and validation: Train the model and evaluate the model performance using cross-validation techniques; Parameter adjustment: adjust model parameters to improve prediction accuracy; Results interpretation and report generation: Interpret the model results and generate a process evaluation report.

8. A process for enhancing the dielectric strength of a non-conductive composite material according to claim 7, characterized in that: The contents of the model are as follows: Linear Regression Model: ； The loss function is minimized: ； in, is the coefficient, is the eigenvector, is the error term.

9. A system for enhancing the dielectric strength of a non-conductive composite material, characterized in that: One or more steps of the process for enhancing the dielectric strength of a non-conductive composite material according to any one of claims 1 to 8 include: Material evaluation and parameter setting module: evaluates the physical properties of non-conductive composite materials, including the maximum pressure they can withstand, and sets the working pressure of the autoclave to not exceed the maximum pressure the material can withstand; Sampling and ply analysis module: take samples from the test piece, analyze the ply parameters of the sampled part, and record and analyze the number, direction and method of ply layup; Data analysis and void ratio prediction module: Utilizes historical production data and uses machine learning to predict void ratios under different process parameters, and outputs the process parameter combination with the lowest void ratio; Process parameter optimization and test piece manufacturing module: optimize the process parameters according to the prediction results, the void ratio of the product does not exceed 1.5%, and manufacture the test pieces according to the optimized parameters to manufacture the test pieces that meet the requirements; Nondestructive testing module: Perform nondestructive testing on the prepared test pieces and comparison test blocks at the same time and after curing, and output the test results of the test pieces and comparison test blocks; Evaluation report generation module: collects data from the preparation process and uses data analysis algorithms to generate an evaluation report that includes process performance indicators and potential improvement points.

10. A storage medium storing non-transitory computer-readable instructions for executing one or more steps of the process for enhancing the dielectric strength of a non-conductive composite material as claimed in any one of claims 1 to 8.

Images