Deformation prediction method and system for plastic part

Abstract

The invention discloses a plastic part deformation prediction method and system, and relates to the field of manufacturing industry. A data preprocessing module; a feature extraction module; a deformation prediction model construction module; a prediction result output module; and a user interaction interface module. The deformation prediction model construction module constructs an accurate prediction model based on the key factors by using a machine learning algorithm, can effectively predict the deformation of the plastic part, further comprises a model verification sub-module and a model optimization sub-module, can strictly verify and optimize the prediction model, and can effectively predict the deformation of the plastic part. The prediction precision and generalization ability of the model are ensured, the advanced machine learning algorithms such as the support vector machine algorithm are adopted, and the comparison formula and the optimization formula are combined, so that the prediction model can be continuously optimized, and the prediction accuracy and reliability are improved.

Description

Technical Field

[0001] The present invention relates to the field of manufacturing industry, and in particular to a deformation prediction method and system for plastic parts. Background Art

[0002] In the manufacturing industry, the deformation problem of plastic parts has always been one of the key factors affecting product quality and production efficiency. During the processing, storage and use of plastic parts, deformation often occurs due to the influence of various factors such as material properties, processing conditions and external environment, resulting in a decrease in dimensional accuracy, which in turn affects the assembly performance and functional realization of the product.

[0003] Traditional deformation prediction methods for plastic parts mainly rely on empirical formulas and experimental data. These methods are not only time-consuming and labor-intensive, but also have limited prediction accuracy, and it is difficult to meet the modern manufacturing industry's demand for high precision and high efficiency. With the rapid development of machine learning technology, its application in various fields is becoming more and more extensive, providing new ideas and methods for deformation prediction of plastic parts.

[0004] However, there is currently no system on the market that can comprehensively consider the original size data, material property data, processing condition data, and external environment data of plastic parts, and use machine learning algorithms to perform high-precision deformation prediction. Therefore, how to develop an efficient and accurate plastic part deformation prediction method and system has become a technical problem that needs to be urgently solved in the current manufacturing industry.

[0005] Based on this, the present invention aims to provide a deformation prediction method and system for plastic parts, so as to achieve accurate prediction of the deformation of plastic parts and provide strong technical support for the manufacturing industry. Summary of the invention

[0006] Based on this, the purpose of the present invention is to provide a deformation prediction method and system for plastic parts to solve the technical problem that there is currently no system on the market that can comprehensively consider the original size data, material property data, processing condition data and external environment data of plastic parts and use machine learning algorithms to perform high-precision deformation prediction.

[0007] In view of the above problems, the present application provides a deformation prediction method and system for plastic parts.

[0008] In a first aspect of the present application, a deformation prediction system for plastic parts is provided, comprising: a data acquisition module for collecting original dimension data, material property data, processing condition data and external environment data of plastic parts; a data preprocessing module for cleaning, format conversion and standardization of the data collected by the data acquisition module; a feature extraction module for extracting key factors affecting the deformation of plastic parts from the preprocessed data, including but not limited to temperature, pressure, time, material elastic modulus and Poisson's ratio; a deformation prediction model construction module for constructing a prediction model for the deformation of plastic parts based on the key factors extracted by the feature extraction module using a machine learning algorithm; a prediction result output module for visually displaying or storing the prediction results output by the deformation prediction model construction module; and a user interaction interface module for receiving operation instructions input by a user and displaying system status, prediction results and error information.

[0009] The present invention is further configured such that the deformation prediction model construction module further includes: a model training submodule, which is used to train the initial prediction model using historical deformation data, wherein the historical deformation data includes the actual deformation of plastic parts under different conditions; a model verification submodule, which is used to verify the trained prediction model using a verification data set to evaluate the prediction accuracy and generalization ability of the model; and a model optimization submodule, which adjusts parameters or optimizes the structure of the prediction model according to the verification results to improve the prediction accuracy.

[0010] The present invention is further configured to also include: a deformation warning module, which is used to issue a deformation warning signal based on the prediction result output by the prediction result output module when the predicted deformation exceeds a preset threshold; an optimization suggestion generation module, which is used to generate optimization suggestions for processing conditions or material properties based on the deformation warning signal to reduce the deformation of plastic parts.

[0011] The second aspect of the present application provides a method for deformation prediction using a plastic component deformation prediction system, comprising the following steps: Step 1: using a data acquisition module to collect raw data of plastic components; Step 2: using a data preprocessing module to preprocess the raw data; Step 3: using a feature extraction module to extract key factors affecting the deformation of plastic components; Step 4: using a deformation prediction model construction module to construct a prediction model for the deformation of plastic components; Step 5: using a prediction result output module to output the prediction result; Step 6: based on the prediction result, using a deformation warning module and an optimization suggestion generation module to generate deformation warnings and optimization suggestions.

[0012] In summary, the present invention mainly has the following beneficial effects:

[0013] The present invention can comprehensively collect the original size data, material property data, processing condition data and external environment data of plastic parts through the data acquisition module, providing a rich information basis for subsequent deformation prediction. The data preprocessing module ensures the accuracy, consistency and availability of the data, providing reliable data guarantee for deformation prediction; the feature extraction module can extract the key factors affecting the deformation of plastic parts from the preprocessed data, such as temperature, pressure, time, material elastic modulus and Poisson's ratio, etc. These factors are crucial for deformation prediction. The deformation prediction model construction module uses a machine learning algorithm to construct an accurate prediction model based on these key factors, which can achieve effective prediction of the deformation of plastic parts. The present invention also includes a model verification submodule and a model optimization submodule, which can strictly verify and optimize the prediction model to ensure the prediction of the model. Accuracy and generalization ability. By adopting advanced machine learning algorithms such as support vector machine algorithms and combining comparison formulas and optimization formulas, the present invention can continuously optimize the prediction model and improve the accuracy and reliability of the prediction. The deformation warning module can issue a deformation warning signal in time according to the prediction results when the predicted deformation exceeds the preset threshold, reminding production personnel to take corresponding measures. The optimization suggestion generation module can generate optimization suggestions for processing conditions or material properties according to the deformation warning signal to reduce the deformation of plastic parts and improve product quality and production efficiency. The plastic parts deformation prediction system provided by the present invention has a friendly user interaction interface, can receive user input operation instructions, and display system status, prediction results and error information, so that users can easily operate the system. At the same time, the prediction method and process of the system are simple and clear, and easy to promote and apply. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a step diagram of the method for deformation prediction using the plastic component deformation prediction system of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0016] The following describes an embodiment of the present invention based on its overall structure.

[0017] A deformation prediction system for plastic parts, comprising:

[0018] The data acquisition module is a key component in the plastic parts deformation prediction system. The main function of this module is to collect various data of plastic parts, including but not limited to original size data, material property data, processing condition data and external environment data. These data are an important basis for subsequent data processing, feature extraction and deformation prediction.

[0019] Original dimension data refers to the initial dimension information of plastic parts before manufacturing or processing. These data are crucial for understanding the initial state of parts and predicting their deformation under different conditions; material property data involves various physical and chemical properties of the materials used in plastic parts, including but not limited to the elastic modulus, Poisson's ratio, thermal expansion coefficient, etc., which have an important influence on the deformation behavior of parts; processing condition data refers to various process parameters involved in the manufacturing or processing of plastic parts, including temperature, pressure, time, etc., which directly affect the processing quality and deformation of parts; external environment data refers to the environmental conditions of plastic parts during storage, transportation or use. These conditions may include temperature, humidity, light, etc., which also have a certain influence on the deformation of parts.

[0020] The data preprocessing module, whose main task is to perform a series of processing on the original data collected by the data acquisition module to ensure the accuracy and effectiveness of subsequent analysis. These processing steps include data cleaning, format conversion and standardization.

[0021] Data cleaning is the first step in data preprocessing, and its purpose is to identify and correct errors, anomalies, or missing values ​​in the data. In the deformation prediction of plastic parts, the original data may contain erroneous data due to equipment failure, sensor error, or human error. The data cleaning process will check this data and take appropriate measures to correct or delete it to ensure the accuracy of subsequent analysis. Format conversion refers to converting data from one format to another to meet the needs of subsequent analysis. In the deformation prediction system of plastic parts, the data acquisition module may collect data from different sources and different devices. These data may have different formats and units. The data preprocessing module will convert the format of these data and unify them into a standard format that the system can recognize, thereby facilitating subsequent data processing and analysis. Standardization processing refers to scaling or converting data according to certain rules to eliminate the dimensional differences between different data and make them have the same scale. In the deformation prediction of plastic parts, the original data may contain temperature, pressure, time, material elastic modulus and Poisson's ratio and other different types of data. These data may have great differences in numerical range and unit. If used directly for analysis, it may lead to inaccurate results. Therefore, the data preprocessing module will standardize these data and convert them into data with the same scale to ensure the accuracy and reliability of subsequent analysis.

[0022] The feature extraction module has the main task of extracting the key factors that have a significant impact on the deformation of plastic parts from the data that has been cleaned, format converted and standardized by the data preprocessing module. These factors, as the basis for the subsequent construction of the deformation prediction model, are directly related to the prediction accuracy and reliability of the model, including but not limited to temperature, pressure, time, material elastic modulus and Poisson's ratio.

[0023] Temperature is one of the important factors affecting the deformation of plastic parts. During the processing, temperature changes will cause the redistribution of internal stress in the material, thereby affecting the dimensional stability and shape accuracy of the parts. The feature extraction module can accurately capture and record the temperature data at different processing stages, providing key input for subsequent model construction. Pressure also has a significant impact on the deformation of plastic parts. In molding processes such as injection molding and pressing, the size and distribution of pressure directly determine the flow behavior and final molding state of the material. Therefore, the feature extraction module needs to accurately measure and record the pressure data during the processing. The time factor cannot be ignored in the deformation prediction of plastic parts. Long-term heating, cooling or pressure holding processes may cause changes in material properties, thereby affecting the deformation behavior of parts. The feature extraction module will record the key time points or time periods in the processing process so that the subsequent model can accurately reflect the effect of time on deformation. Material properties are intrinsic factors that affect the deformation of plastic parts. The elastic modulus reflects the ability of the material to resist deformation, while the Poisson's ratio describes the degree of lateral deformation of the material when subjected to force. These material property data are usually provided by material suppliers or obtained through experimental measurements. The feature extraction module will take these key material property data into consideration to build a more accurate prediction model.

[0024] The feature extraction module is usually implemented through technical means such as data analysis and machine learning algorithms. It can deeply mine and analyze the preprocessed data and identify those characteristic variables that are closely related to the deformation of plastic parts. In practical applications, the feature extraction module may combine multiple algorithms and technical means to improve the accuracy and efficiency of feature extraction.

[0025] The deformation prediction model building module is the core part of the entire system. It is responsible for building a prediction model for the deformation of plastic parts based on the key factors provided by the feature extraction module and using machine learning algorithms. This model can predict the deformation of plastic parts under different conditions and provide data support for subsequent deformation warnings and optimization suggestions.

[0026] When building the prediction model, the module uses a machine learning algorithm. Machine learning is a technology that can automatically learn from data and improve its performance. It is very suitable for dealing with complex and changeable problems such as deformation of plastic parts. By training a large amount of historical data, the machine learning algorithm can learn the complex relationship between the key factors affecting the deformation of plastic parts and the deformation amount, and build a prediction model based on this relationship. The feature extraction module has extracted the key factors affecting the deformation of plastic parts from the preprocessed data, including but not limited to temperature, pressure, time, material elastic modulus and Poisson's ratio. These factors are the basis for building a prediction model, and they can reflect the deformation law of plastic parts under different conditions.

[0027] The deformation prediction model building module further includes:

[0028] The model training submodule mainly trains the initial prediction model using historical deformation data, which contains the actual deformation of plastic parts under different conditions and is the basis for model learning. Historical deformation data: includes the actual deformation of plastic parts under different temperature, pressure, time, material properties and external environment conditions.

[0029] Obtain historical deformation data from the data acquisition module, pre-process the data to ensure data quality and consistency, use machine learning algorithms (such as support vector machines, neural networks, etc.) to build an initial prediction model, and input the historical deformation data as a training set into the initial prediction model for model training. The trained prediction model can predict the deformation of plastic parts based on the input feature vectors (such as temperature, pressure, time, etc.).

[0030] The prediction formula of the support vector machine algorithm used in the model training submodule is:

[0031]

[0032] Among them, x is the input feature vector, which is the input of the model and contains the key factors that affect the deformation of plastic parts, such as temperature, pressure, time, material elastic modulus and Poisson's ratio. These factors are extracted from the preprocessed data and are the basis for the model to make predictions;

[0033] sign is a sign function, which is a function used to determine the positive or negative of the expression in the brackets. In SVM (support vector machine), it is usually related to the decision boundary, which is used to determine which category the input feature vector belongs to (in this case, whether significant deformation will occur);

[0034] n is the number of support vectors, which are key data points used in the SVM algorithm to define the decision boundary. Its number represents the number of these key points, which are crucial to the construction and prediction ability of the model.

[0035] i is an index variable used to traverse all support vectors. During the training process of SVM, the model parameters need to be continuously adjusted to find the best decision boundary, and the index variable helps the algorithm complete this task;

[0036] y i is the label of the training sample, which is the actual deformation in the historical deformation data and is used to train the model. The positive or negative value of the label reflects the degree of deformation of the plastic parts under different conditions.

[0037] α iThe weight corresponding to the support vector is an important parameter in the SVM model. It represents the influence of each support vector on the decision boundary. By adjusting the weight, the algorithm can optimize the prediction ability of the model.

[0038] K(x i ,x) is the kernel function, which is a key concept in the SVM algorithm. It is used to map the input feature vector to a high-dimensional space in order to find the best decision boundary in the space. The choice of kernel function has an important impact on the performance and prediction ability of the model;

[0039] b is the bias term, which is another important parameter in the SVM model. It is used to adjust the position of the decision boundary. By adjusting the bias term, the algorithm can further optimize the prediction accuracy of the model.

[0040] In the deformation prediction system of plastic parts, this formula uses the support vector machine algorithm to train the initial prediction model using historical deformation data. By continuously adjusting the model parameters (such as weights, bias terms, etc.), the algorithm can find the optimal decision boundary, thereby achieving effective prediction of the deformation of plastic parts.

[0041] The model verification submodule uses a verification data set to verify the trained prediction model, which is used to evaluate the prediction accuracy and generalization ability of the trained prediction model. It uses a verification data set independent of the training set for testing to ensure that the model can not only perform well on known data, but also maintain stable prediction performance on unknown data. Verification data set: contains the actual deformation of plastic parts under different conditions, but does not overlap with the training set.

[0042] Obtain a validation data set from the data acquisition module or external data source, input the validation data set into the trained prediction model, make predictions, calculate the difference between the prediction results and the actual deformation, evaluate the prediction accuracy of the model, and use evaluation indicators (such as the determination coefficient R 2 , mean square error (MSE), etc.) to quantify the generalization ability of the model. Verification results: including prediction accuracy evaluation indicators, generalization ability evaluation indicators, and possible model performance analysis reports.

[0043] The comparison formula used by the model verification submodule is:

[0044]

[0045] Among them, R 2 is the coefficient of determination, (also called goodness of fit or coefficient of determination), which quantifies how well the model fits the data;

[0046] n is the number of samples, that is, the total number of sample points in the data set used for validation;

[0047] i is the index of the sample point, which is used to traverse all sample points;

[0048] y i is the actual deformation of the i-th sample point;

[0049] is the predicted deformation of the i-th sample point, that is, the deformation predicted by the model based on the input features; is the average value of the actual deformation, that is, the arithmetic mean of the actual deformation of all sample points.

[0050] Molecular part: It represents the sum of squares of prediction errors, that is, the sum of squares of the differences between the model prediction value and the actual value. The smaller this value is, the smaller the prediction error of the model is and the higher the prediction accuracy is. The denominator part: It represents the total sum of squares, that is, the sum of the squares of the differences between the actual deformation of all sample points and their average value. This value reflects the degree of fluctuation of the data itself. The coefficient of determination R is obtained by calculating the ratio of the sum of squares of the prediction error to the total sum of squares and subtracting this ratio from 1. 2 , R 2 The closer the value of is to 1, the higher the prediction accuracy of the model, that is, the model can better fit the actual data; 2 The smaller the value (but usually not less than 0), the lower the prediction accuracy of the model, that is, the greater the difference between the model and the actual data.

[0051] In the model validation submodule, this formula is used to evaluate the performance of the trained prediction model. By comparing the prediction results of the model on the validation data set with the actual results, the determination coefficient R can be calculated. 2 , so as to judge the prediction accuracy and generalization ability of the model. If R 2 If the value of is high (such as close to 1), it means that the model performs well on the validation data set and has high prediction accuracy and generalization ability; on the contrary, if R 2 If the value of is low, it means that the model may have problems such as overfitting or underfitting, and further adjustment and optimization are needed.

[0052] The model optimization submodule adjusts the parameters or optimizes the structure of the prediction model according to the verification results of the model verification submodule to improve the prediction accuracy. It may involve adjusting the model's hyperparameters, changing the model structure, or using more advanced machine learning algorithms. The verification results include prediction accuracy evaluation indicators, generalization ability evaluation indicators, etc.

[0053] Analyze the verification results, determine the deficiencies of the model, and adjust the model parameters (such as the C parameter of the support vector machine, kernel function parameters, etc.) or structure (such as adding hidden layers, changing the number of neurons, etc.) according to the analysis results, retrain the model, and verify it again to evaluate the optimization effect. Repeat the above steps until satisfactory prediction accuracy and generalization ability are achieved. Optimized prediction model: A model with higher prediction accuracy and better generalization ability after parameter adjustment or structural optimization.

[0054] The optimization formula used by the model optimization submodule is:

[0055]

[0056] Among them, J(θ) is the optimization objective function;

[0057] (h θ (x (i) ) is the output of the prediction model, that is, the deformation of the plastic component predicted by the model based on the input features;

[0058] y (i) is the actual deformation, which is the actual deformation of the plastic parts under different conditions, and is used to compare with the predicted value to evaluate the prediction accuracy of the model;

[0059] θ is the model parameter, including the weight and bias of the support vector machine, which are optimized during the model training process to minimize the prediction error;

[0060] γ is the regularization coefficient, which is used to control the complexity of the model and prevent overfitting. The regularization term is usually proportional to the sum of the squares of the model parameters. By adjusting the size of γ, a balance can be achieved between the prediction accuracy and generalization ability of the model.

[0061] m is the number of training samples, that is, the number of historical deformation data used to train the model;

[0062] n is the number of characteristics, that is, the number of key factors that affect the deformation of plastic parts, such as temperature, pressure, time, material elastic modulus and Poisson's ratio;

[0063] j is an index variable used to traverse all parameters in the parameter vector θ;

[0064] i is an index variable used to traverse all training samples.

[0065] In the model optimization submodule, the purpose of this formula is to minimize the difference between the output of the predicted model and the actual deformation by adjusting the model parameter θ, while considering the regularization term to control the complexity of the model. This usually involves an iterative process, continuously adjusting the value of θ through optimization algorithms such as gradient descent until a satisfactory prediction accuracy and generalization ability are achieved.

[0066] In practical applications, the optimization objective function may contain multiple parts, such as the sum of squares of prediction errors, regularization terms, etc. These parts are combined together through weighted summation and other methods to form a comprehensive optimization goal. The model optimization submodule will use appropriate optimization algorithms and strategies based on this optimization goal to adjust parameters or optimize the structure of the prediction model to improve its prediction accuracy and generalization ability.

[0067] Once the prediction model is built, verified and optimized, it can be used to predict the deformation of plastic parts under different conditions. These prediction results can be visualized or stored through the prediction result output module for user reference and decision-making.

[0068] Prediction result output module: The main function of this module is to visualize or store the prediction results output by the deformation prediction model construction module. This means that once the prediction model completes the prediction of the deformation of the plastic parts, the prediction result output module will receive the prediction data and present it in an intuitive way, or save it to the specified storage medium.

[0069] Visual display is usually in the form of charts, images or animations, making the prediction results more intuitive and easy to understand. Through charts, images and other forms, users can quickly understand the deformation of plastic parts under different conditions, which helps users discover deformation patterns and optimize processing conditions or material properties to reduce deformation. Storage is to save the predicted data in a database, file or other storage system for subsequent analysis, comparison or reference. Users can retrieve the stored prediction data at any time for further analysis or comparison, which helps users track the deformation of plastic parts, discover potential problems in a timely manner and take corresponding measures.

[0070] The user interaction interface module is used to receive the operation instructions input by the user and display the system status, prediction results and error information. The main function of this module is to receive the operation instructions input by the user. These instructions may include starting the system, selecting the prediction model, setting the prediction parameters, etc. The system will perform corresponding operations according to these instructions. At the same time, the user interaction interface module will also display the current status of the system in real time, such as data collection progress, model training status, prediction result output, etc., so that users can clearly understand the operation of the system. In addition, the module will also display relevant prediction results and error information during the prediction process or after the prediction results are output. The prediction results may include the predicted values ​​of the deformation of plastic parts under different conditions, and the error information may include data collection errors, model training failures, abnormal prediction results, etc. This information helps users to find and solve problems in a timely manner.

[0071] The system also includes:

[0072] The deformation warning module is used to send out a deformation warning signal when the predicted deformation exceeds a preset threshold value according to the prediction results output by the prediction result output module. The deformation warning module first receives the prediction results from the prediction result output module, which include the predicted deformation of the plastic parts under different conditions. Next, the deformation warning module compares the received predicted deformation with the preset deformation threshold value. This preset threshold value is set according to the use requirements, process specifications or industry experience of the plastic parts, and is used to determine whether the deformation is within the acceptable range. If the predicted deformation exceeds the preset threshold value, the deformation warning module will immediately send out a deformation warning signal, which can be a sound, light, screen display or other forms of notification to attract the attention of production personnel.

[0073] The optimization suggestion generation module is used to generate optimization suggestions for processing conditions or material properties according to the deformation warning signal to reduce the deformation of plastic parts. The optimization suggestion generation module first receives the warning signal from the deformation warning module, which indicates the possible deformation of plastic parts under specific conditions. Next, the optimization suggestion generation module will analyze the predicted deformation and try to find out the main factors causing deformation, such as processing temperature, pressure, time or material properties. Based on the analysis results, the optimization suggestion generation module will generate a series of optimization suggestions for processing conditions or material properties. These suggestions are intended to reduce the deformation of plastic parts by adjusting processing parameters or changing materials. Finally, the optimization suggestion generation module will output the generated optimization suggestions to users or related systems. These suggestions can be presented in text, charts or other forms to facilitate user understanding and implementation.

[0074] The optimization formula used by the optimization suggestion generation module is:

[0075]

[0076] Among them, T opt and P opt are the optimized temperature and pressure, which are the two key variables that the formula tries to solve, that is, it is hoped that the deformation of plastic parts can be reduced by adjusting these two parameters;

[0077] T base and P base are the base temperature and pressure, respectively, which are the initial conditions before any optimization suggestions are made and are usually determined based on historical data or experience;

[0078] ΔP and ΔT are the differences between the predicted deformation and the reference deformation, respectively. These differences reflect the gap between the deformation of the plastic component under the current conditions and the expected state;

[0079] Δt is the change in processing time, which is also an important factor affecting the deformation of plastic parts, but in this formula it is used as an adjustment factor rather than directly as an optimization variable;

[0080] E new and E base are the elastic modulus of the new material and the benchmark material, respectively. The elastic modulus is an important indicator of the material's ability to resist deformation, and is therefore a key factor to be considered during the optimization process;

[0081] V new and V base They are the Poisson’s ratio of the new material and the reference material, respectively. The Poisson’s ratio describes the degree of lateral deformation of the material when subjected to force, and also has an important influence on the deformation of plastic parts.

[0082] WB is the influencing factor of external environmental factors on deformation. It is a composite index that combines multiple external factors such as humidity, temperature, and light, and is used to reflect the influence of the external environment on the deformation of plastic parts.

[0083] β1, β2, β3, β4, β5 and μ1, μ2, μ3, μ4, μ5 are the adjustment coefficients of each factor respectively. These coefficients need to be determined through experiments or data analysis. They reflect the weight and sensitivity of different factors on the deformation of plastic parts. In practical applications, these coefficients can be adjusted and optimized according to specific circumstances to obtain better prediction effects and optimization suggestions.

[0084] This formula comprehensively considers multiple key factors that affect the deformation of plastic parts and gives a method for calculating the optimized temperature and pressure. In practical applications, the optimized temperature and pressure values ​​can be calculated based on the difference between the predicted deformation and the baseline deformation, as well as the adjustment coefficients of various factors, thereby providing production personnel with optimization suggestions for processing conditions or material properties to reduce the deformation of plastic parts and improve product quality and production efficiency.

[0085] The method for deformation prediction using a plastic component deformation prediction system comprises the following steps:

[0086] S1: Use the data acquisition module to collect the original data of plastic parts.

[0087] The data acquisition module is a key component of the plastic parts deformation prediction system. Its main function is to comprehensively collect various data of plastic parts, including but not limited to original dimension data, material property data, processing condition data and external environment data. These data are an important basis for subsequent data processing, feature extraction and deformation prediction. The original dimension data reflects the initial state of the parts, the material property data involves the physical and chemical properties of the materials used in the parts, the processing condition data includes process parameters such as temperature, pressure, time, etc., and the external environment data takes into account the environmental conditions of the parts during storage, transportation or use.

[0088] S2: Use the data preprocessing module to preprocess the original data.

[0089] The main task of the data preprocessing module is to perform a series of processing on the raw data collected by the data acquisition module to ensure the accuracy and effectiveness of subsequent analysis. These processing steps include data cleaning, format conversion and standardization. Data cleaning aims to identify and correct errors, anomalies or missing values ​​in the data to ensure the accuracy of the data. Format conversion is to convert data from one format to another to meet the needs of subsequent analysis. Standardization is to scale or convert data according to certain rules to eliminate the dimensional differences between different data and make them have the same scale, thereby ensuring the reliability of subsequent analysis.

[0090] S3: Use the feature extraction module to extract the key factors that affect the deformation of plastic parts.

[0091] The main task of the feature extraction module is to extract the key factors that have a significant impact on the deformation of plastic parts from the data processed by the data preprocessing module. These factors serve as the basis for the subsequent construction of the deformation prediction model and are directly related to the prediction accuracy and reliability of the model. In the deformation prediction of plastic parts, key factors may include temperature, pressure, time, material elastic modulus and Poisson's ratio, etc. The feature extraction module uses technical means such as data analysis and machine learning algorithms to conduct in-depth mining and analysis of the preprocessed data and identify characteristic variables that are closely related to the deformation of plastic parts.

[0092] S4: Use the deformation prediction model building module to build a prediction model for the deformation of plastic parts.

[0093] The deformation prediction model construction module is the core part of the entire system. It is responsible for building a prediction model for the deformation of plastic parts based on the key factors provided by the feature extraction module and using machine learning algorithms. This model can predict the deformation of plastic parts under different conditions and provide data support for subsequent deformation warnings and optimization suggestions. When building the prediction model, the module uses machine learning algorithms, such as support vector machines and neural networks. By training a large amount of historical data, the machine learning algorithm can learn the complex relationship between the key factors affecting the deformation of plastic parts and the deformation amount, and build a prediction model based on this relationship.

[0094] S5: Output the prediction results using the prediction result output module.

[0095] Once the prediction model is built, verified and optimized, it can be used to predict the deformation of plastic parts under different conditions. The prediction result output module is responsible for visualizing or storing the prediction results output by the deformation prediction model construction module, so that users can intuitively understand the deformation of plastic parts under different conditions, providing a reference for subsequent production and quality control.

[0096] S6: Based on the prediction results, deformation warning module and optimization suggestion generation module are used to generate deformation warning and optimization suggestions.

[0097] The deformation warning module and the optimization suggestion generation module are important components of the plastic parts deformation prediction system. The deformation warning module can issue a deformation warning signal in time according to the prediction results output by the prediction result output module when the predicted deformation exceeds the preset threshold, so that production personnel can take timely measures to avoid or reduce the deformation of plastic parts. The optimization suggestion generation module can generate optimization suggestions for processing conditions or material properties based on the deformation warning signal. These suggestions are aimed at reducing the deformation of plastic parts and improving product quality and production efficiency by adjusting processing conditions or selecting more appropriate material properties.

[0098] Those skilled in the art will understand that the various digital numbers such as the first and second involved in this application are only for the convenience of description, and are not used to limit the scope of this application, nor do they indicate a sequence of precedence. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated with each other are in an "or" relationship. "At least one" refers to one or more. At least two refers to two or more. "At least one", "any one" or similar expressions refer to any combination of these items, including any combination of single items (individuals) or plural items (individuals). For example, at least one item (individual, kind) of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0099] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in this application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0100] The steps of the method or algorithm described in this application can be directly embedded in hardware, software units executed by a processor, or a combination of the two. The software unit can be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, register, hard disk, removable disk, CD-ROM or any other form of storage medium in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. The processor and the storage medium can be arranged in an ASIC, and the ASIC can be arranged in a terminal. Optionally, the processor and the storage medium can also be arranged in different components in the terminal. These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0101] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A deformation prediction system for plastic parts, characterized in that: include: Data acquisition module, used to collect original size data, material property data, processing condition data and external environment data of plastic parts; A data preprocessing module is used to clean, convert and standardize the data collected by the data collection module; A feature extraction module is used to extract key factors affecting the deformation of plastic parts from the preprocessed data, including but not limited to temperature, pressure, time, material elastic modulus and Poisson's ratio; A deformation prediction model building module, using a machine learning algorithm to build a prediction model for deformation of plastic parts based on the key factors extracted by the feature extraction module; A prediction result output module, used for visually displaying or storing the prediction result output by the deformation prediction model building module; The user interaction interface module is used to receive operation instructions input by the user and display system status, prediction results and error information.

2. A deformation prediction system for plastic parts according to claim 1, characterized in that: The deformation prediction model building module further includes: A model training submodule, for training an initial prediction model using historical deformation data, wherein the historical deformation data includes actual deformation amounts of plastic parts under different conditions; The model validation submodule is used to validate the trained prediction model using the validation data set and evaluate the prediction accuracy and generalization ability of the model; The model optimization submodule adjusts the parameters or optimizes the structure of the prediction model according to the verification results to improve the prediction accuracy.

3. A deformation prediction system for plastic parts according to claim 2, characterized in that: The prediction formula of the support vector machine algorithm used by the model training submodule is: Where x is the input feature vector, sign is a sign function used to determine the positive or negative of the expression in the brackets, n is the number of support vectors, i is an index variable used to traverse all support vectors, and y is a i is the label of the training sample, α i is the weight corresponding to the support vector, K(x i ,x) is the kernel function and b is the bias term.

4. The deformation prediction system for plastic parts according to claim 2, characterized in that: The comparison formula used by the model verification submodule is: Among them, R 2 is the coefficient of determination, n is the number of samples, i is the index of the sample point, which is used to traverse all sample points, y i is the actual deformation, To predict the deformation, is the average value of the actual deformation.

5. The deformation prediction system for plastic parts according to claim 2, characterized in that: The optimization formula used by the model optimization submodule is: Among them, J(θ) is the optimization objective function, (h θ (x (i) ) is the output of the prediction model, y (i) is the actual deformation, θ is the model parameter, γ is the regularization coefficient, m is the number of training samples, n is the number of features, j is an index variable used to traverse all parameters in the parameter vector θ, and i is an index variable used to traverse all training samples.

6. The deformation prediction system for plastic parts according to claim 1, characterized in that: Also includes: A deformation warning module, configured to issue a deformation warning signal when the predicted deformation exceeds a preset threshold value according to the prediction result output by the prediction result output module; The optimization suggestion generation module is used to generate optimization suggestions for processing conditions or material properties according to the deformation warning signal to reduce the deformation of the plastic component.

7. A deformation prediction system for plastic parts according to claim 6, characterized in that: The optimization formula used by the optimization suggestion generation module is: Among them, T opt and P opt are the optimized temperature and pressure, T base and P base are the reference temperature and pressure, ΔP and ΔT are the differences between the predicted deformation and the reference deformation, Δt is the change in processing time, and E new and E base are the elastic modulus of the new material and the reference material, V new and V base are the Poisson's ratios of the new material and the benchmark material, respectively; WB is the influencing factor of the external environmental factors on the deformation, which can be a composite indicator integrating multiple external factors such as humidity, temperature, and light; β1, β2, β3, β4, β5 and μ1, μ2, μ3, μ4, μ5 are the adjustment coefficients of each factor, which need to be determined through experiments or data analysis.

8. A method for deformation prediction using the plastic component deformation prediction system according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: Use the data acquisition module to collect the original data of plastic parts; S2: preprocessing the raw data using a data preprocessing module; S3: Extract key factors affecting the deformation of plastic parts using feature extraction module; S4: construct a prediction model for deformation of plastic parts using a deformation prediction model building module; S5: Output the prediction result using the prediction result output module; S6: Based on the prediction results, deformation warning module and optimization suggestion generation module are used to generate deformation warning and optimization suggestions.

Images