A software control system and method for resin material production and processing

Abstract

The present invention discloses a software control system and method for the production and processing of new resin materials, specifically in the field of resin production control algorithms, for the production of ultra-low dielectric resins that may involve complex chemical reactions and physical processes. Neural networks are particularly suitable for capturing these nonlinear and complex relationships due to their high fidelity and accuracy, ensuring that the dielectric constant is within the required range; in the production process of dielectric resins, slight changes in raw materials or fluctuations in production parameters may affect the properties of the final product. Neural networks can predict the impact of these changes on the dielectric constant in real time, allowing the production team to make immediate adjustments; and for the production of ultra-low dielectric resins, neural networks provide a way to make decisions based on actual production data rather than just experience.

Description

Technical Field

[0001] The invention relates to the field of resin production control algorithms, in particular to a software control system and method for resin material production processing. Background Art

[0002] A dielectric resin is a material with a specific dielectric constant. The dielectric constant is a measure of the response of a substance to an electric field or its capacitance properties. The dielectric constant affects the performance of circuit components, such as capacitors, in a circuit. Ultra-low dielectric resins are very valuable in certain high-tech applications, especially in microelectronics and radio frequency (RF) applications, where strict requirements are placed on the dielectric properties of the material. The production process of ultra-low dielectric resins usually involves selecting suitable monomers for polymerization, specific curing and cross-linking, and then a series of post-processing steps, such as heat treatment or chemical modification, to ensure that the resin material has the desired electrical and physical properties. For the existing ultra-low dielectric resin production process, traditional control methods often rely on the experience of engineers or operators. Such decisions may not make full use of all available data and may lead to suboptimal results or unstable production. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide a software control system and method for resin material production and processing, so as to achieve the effects of orderly production and intelligent planning.

[0004] Among them, a software control system for resin material production and processing includes:

[0005] A data acquisition module, used to collect raw material composition data, production parameter data and target output data, wherein the target output data is dielectric constant data;

[0006] The neural network module is used to predict new data points based on the MLP architecture and obtain the predicted dielectric constant by giving a new raw material ratio and production parameters;

[0007] The result prediction module is used to make decisions on production parameters or raw material ratios based on the predicted results;

[0008] Wherein, the neural network module specifically includes:

[0009] A data preprocessing unit is used to normalize raw material composition data, production parameter data and target output data, and scale the range of all numerical features to [0, 1] or convert their distribution using standardization;

[0010] A neural network building unit is used to build an MLP neural network, including: an input layer, a hidden layer and an output layer, wherein the input layer includes four features and four neurons, and the four features include the proportion of raw material A , Raw material B ratio ,temperature and pressure ; The output layer includes a neuron for outputting the regression task of predicting the dielectric constant;

[0011] The model training unit is used to train the neural network using the data in the dataset and adopt the mean square error as the loss function.

[0012] Furthermore, the neural network establishment unit also includes:

[0013] The input layer-hidden layer calculation subunit is used to calculate the output of the hidden layer. The specific calculation method is as follows:

[0014] ;

[0015] Among them, the represents the output of the hidden layer, represents the activation function, represents a column vector, represents the weight matrix from the input layer to the hidden layer, represents the bias vector; the column vector Including the values ​​of raw material A percentage, raw material B percentage, temperature and pressure, that is, ;

[0016] The hidden layer-output layer calculation subunit is used to calculate the output of the output layer. The specific calculation method is as follows:

[0017] ;

[0018] Among them, the represents the dielectric constant predicted by the model, represents the weight matrix from the hidden layer to the output layer, Represents the bias vector.

[0019] Furthermore, the model training unit also includes:

[0020] The loss value calculation subunit averages the errors of all data points according to the mean square error loss function to obtain a single loss value. The specific calculation method is as follows:

[0021] ;

[0022] Among them, the Indicates loss, represents the number of data points, Indicates The true value of the data point, Indicates The model prediction value of data points, Represents the square of the difference between the true value and the predicted value;

[0023] The neural network training subunit trains the neural network according to the back propagation algorithm.

[0024] A software control method for resin material production and processing, the method is based on any one of the software control systems for resin material production and processing, and comprises the following steps:

[0025] S1. Collecting raw material composition data, production parameter data and target output data, wherein the target output data is dielectric constant data;

[0026] S2. According to the MLP architecture, new data points are predicted, and the predicted dielectric constant is obtained by giving a new raw material ratio and production parameters;

[0027] S3. Make decisions on production parameters or raw material ratios based on the predicted results;

[0028] Wherein, the step S2 specifically includes the following sub-steps:

[0029] S201. Normalize the raw material composition data, production parameter data, and target output data, scaling the range of all numerical features to [0, 1] or converting their distribution using standardization;

[0030] S202. Establish an MLP neural network, including: an input layer, a hidden layer and an output layer, wherein the input layer includes four features and four neurons, and the four features include the proportion of raw material A , Raw material B ratio ,temperature and pressure ; The output layer includes a neuron for outputting the regression task of predicting the dielectric constant;

[0031] S203. Train the neural network using the data in the dataset and use mean square error as the loss function.

[0032] Furthermore, the step S202 specifically includes the following sub-steps:

[0033] S2021. Calculate the output of the hidden layer. The specific calculation method is as follows:

[0034] ;

[0035] Among them, the represents the output of the hidden layer, represents the activation function, represents a column vector, represents the weight matrix from the input layer to the hidden layer, represents the bias vector; the column vector Including the values ​​of raw material A percentage, raw material B percentage, temperature and pressure, that is, ;

[0036] S2022. Calculate the output of the output layer. The specific calculation method is as follows:

[0037] ;

[0038] Among them, the represents the dielectric constant predicted by the model, represents the weight matrix from the hidden layer to the output layer, Represents the bias vector.

[0039] Furthermore, the step S203 specifically includes the following sub-steps:

[0040] S2031. Average the errors of all data points according to the mean square error loss function to obtain a single loss value. The specific calculation method is as follows:

[0041] ;

[0042] Among them, the Indicates loss, represents the number of data points, Indicates The true value of the data point, Indicates The model prediction value of data points, Represents the square of the difference between the true value and the predicted value;

[0043] S2032. Training a neural network based on a back-propagation algorithm.

[0044] The beneficial effects of the present invention are:

[0045] The present invention proposes a software control system and method for resin material production and processing, which is aimed at the complex chemical reactions and physical processes involved in the production of ultra-low dielectric resins. Neural networks are particularly suitable for capturing these nonlinear and complex relationships due to their high fidelity and accuracy, ensuring that the dielectric constant is within the required range; in the production process of dielectric resins, slight changes in raw materials or fluctuations in production parameters may affect the properties of the final product. Neural networks can predict the impact of these changes on the dielectric constant in real time, allowing the production team to make adjustments immediately; and for the production of ultra-low dielectric resins, neural networks provide a way to make decisions based on actual production data rather than just experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of a method flow of a software control system and method for resin material production and processing proposed by the present invention;

[0047] Figure 2 This is a schematic diagram of the structure of a software-controlled terminal device for resin material production and processing proposed by the present invention;

[0048] Figure 3 A schematic diagram of the structure of a computer-readable storage medium for a software control method for resin material production and processing proposed by the present invention. DETAILED DESCRIPTION

[0049] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0050] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0051] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative work are within the scope of protection of the present invention. It should be noted that relational terms such as the terms "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0052] Moreover, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a list of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus. In the absence of more restrictions, an element defined by the phrase "comprises a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0053] The features and performance of the present invention are further described in detail below in conjunction with the embodiments.

[0054] A software control system for resin material production and processing, comprising:

[0055] A data acquisition module, used to collect raw material composition data, production parameter data and target output data, wherein the target output data is dielectric constant data;

[0056] The neural network module is used to predict new data points based on the MLP architecture and obtain the predicted dielectric constant by giving a new raw material ratio and production parameters;

[0057] The result prediction module is used to make decisions on production parameters or raw material ratios based on the predicted results;

[0058] Wherein, the neural network module specifically includes:

[0059] A data preprocessing unit is used to normalize raw material composition data, production parameter data and target output data, and scale the range of all numerical features to [0, 1] or convert their distribution using standardization;

[0060] A neural network building unit is used to build an MLP neural network, including: an input layer, a hidden layer and an output layer, wherein the input layer includes four features and four neurons, and the four features include the proportion of raw material A , Raw material B ratio ,temperature and pressure ; The output layer includes a neuron for outputting the regression task of predicting the dielectric constant;

[0061] The model training unit is used to train the neural network using the data in the dataset and adopt the mean square error as the loss function.

[0062] Furthermore, as a preferred implementation scheme of this embodiment, a control process for the above system is proposed, which is as follows:

[0063] 1. Collect data related to the production process, including raw material composition, production parameters and properties of the final product, as shown in Table 1 below:

[0064]

[0065] Table 1 Data example table

[0066] Preprocessing: Data normalization, scaling all numerical features to the [0,1] interval or using standardization to convert them to a mean of 0 and a standard deviation of 1;

[0067] 2. Define the model architecture and choose a neural network architecture. For most production control tasks, a simple feedforward network, such as a multilayer perceptron (MLP), is usually sufficient. MLP consists of an input layer, one or more hidden layers, and an output layer. Specifically:

[0068] Input layer: The number of neurons in this layer is the same as the number of features. For this embodiment, there are four features (ratio of raw material A, ratio of raw material B, temperature, and pressure), so the input layer has four neurons.

[0069] Hidden layer: The number of neurons in the hidden layer is a hyperparameter, which is selected through cross-validation. Adding more hidden layers and / or neurons makes the network more complex and can capture more complex functional relationships, but it may also lead to overfitting.

[0070] Output layer: For regression tasks (such as predicting dielectric constant in this example), the output layer usually has only one neuron, which outputs a continuous value. For classification tasks, the number of neurons in the output layer is the same as the number of categories.

[0071] 3. The activation function adds nonlinearity to the neural network, enabling it to approximate complex functions. Common activation functions include ReLU, Sigmoid, and Tanh. For example, this embodiment uses ReLU as the activation function, which is commonly used in the hidden layer. The specific calculation formula is:

[0072] ;

[0073] For hidden layers, ReLU speeds up training and helps alleviate the vanishing gradient problem.

[0074] 4. Define the loss function and optimizer, and use the loss function to measure the difference between the model's prediction and the true value. In this embodiment, the mean square error is used as the loss function; the optimizer is an algorithm for adjusting the network weights to minimize the loss function. Common optimizers include SGD, Adam, and RMSprop.

[0075] Prediction and control: After the model is trained, it is used to predict new data points. Given a new raw material ratio and production parameters, the predicted dielectric constant can be obtained, and then based on the prediction, it is decided whether the production parameters or the raw material ratio need to be adjusted to achieve the desired dielectric constant.

[0076] Furthermore, the neural network establishment unit also includes:

[0077] The input layer-hidden layer calculation subunit is used to calculate the output of the hidden layer. The specific calculation method is as follows:

[0078] ;

[0079] Among them, the represents the output of the hidden layer, represents the activation function, represents a column vector, represents the weight matrix from the input layer to the hidden layer, represents the bias vector; the column vector Including the values ​​of raw material A percentage, raw material B percentage, temperature and pressure, that is, ;

[0080] The hidden layer-output layer calculation subunit is used to calculate the output of the output layer. The specific calculation method is as follows:

[0081] ;

[0082] Among them, the represents the dielectric constant predicted by the model, represents the weight matrix from the hidden layer to the output layer, Represents the bias vector.

[0083] Furthermore, the model training unit also includes:

[0084] The loss value calculation subunit averages the errors of all data points according to the mean square error loss function to obtain a single loss value. The specific calculation method is as follows:

[0085] ;

[0086] Among them, the Indicates loss, represents the number of data points, Indicates The true value of the data point, Indicates The model prediction value of data points, Represents the square of the difference between the true value and the predicted value;

[0087] The neural network training subunit trains the neural network according to the back propagation algorithm.

[0088] A software control method for resin material production and processing, the method is based on any one of the software control systems for resin material production and processing, such as Figure 1 As shown, the following steps are included:

[0089] S1. Collecting raw material composition data, production parameter data and target output data, wherein the target output data is dielectric constant data;

[0090] S2. According to the MLP architecture, new data points are predicted, and the predicted dielectric constant is obtained by giving a new raw material ratio and production parameters;

[0091] S3. Make decisions on production parameters or raw material ratios based on the predicted results;

[0092] Wherein, the step S2 specifically includes the following sub-steps:

[0093] S201. Normalize the raw material composition data, production parameter data, and target output data, scaling the range of all numerical features to [0, 1] or converting their distribution using standardization;

[0094] S202. Establish an MLP neural network, including: an input layer, a hidden layer and an output layer, wherein the input layer includes four features and four neurons, and the four features include the proportion of raw material A , Raw material B ratio ,temperature and pressure ; The output layer includes a neuron for outputting the regression task of predicting the dielectric constant;

[0095] S203. Train the neural network using the data in the dataset and use mean square error as the loss function.

[0096] Furthermore, the step S202 specifically includes the following sub-steps:

[0097] S2021. Calculate the output of the hidden layer. The specific calculation method is as follows:

[0098] ;

[0099] Among them, the represents the output of the hidden layer, represents the activation function, represents a column vector, represents the weight matrix from the input layer to the hidden layer, represents the bias vector; the column vector Including the values ​​of raw material A percentage, raw material B percentage, temperature and pressure, that is, ;

[0100] S2022. Calculate the output of the output layer. The specific calculation method is as follows:

[0101] ;

[0102] Among them, the represents the dielectric constant predicted by the model, represents the weight matrix from the hidden layer to the output layer, Represents the bias vector.

[0103] Furthermore, the step S203 specifically includes the following sub-steps:

[0104] S2031. Average the errors of all data points according to the mean square error loss function to obtain a single loss value. The specific calculation method is as follows:

[0105] ;

[0106] Among them, the Indicates loss, represents the number of data points, Indicates The true value of the data point, Indicates The model prediction value of data points, Represents the square of the difference between the true value and the predicted value;

[0107] S2032. Training a neural network based on a back-propagation algorithm.

[0108] Furthermore, as a preferred technical solution of this embodiment, specific steps of training a neural network using a back propagation algorithm are proposed, including:

[0109] S20321. Calculating losses by the chain rule About Output The specific calculation method of the gradient is as follows:

[0110] ;

[0111] Among them, the Represents the derivative formula of the mean square error (MSE) loss function for a single data point, Represents the loss function About the model's prediction output The partial derivative of

[0112] S20322. Calculate the gradient of the previous layer using the chain rule based on the gradient of the current layer and the derivative of the activation function;

[0113] S20323. Update weights and biases using the calculated gradient and learning rate parameters. The specific calculation method is as follows:

[0114] ;

[0115] ;

[0116] Among them, the represents the learning rate, Representation weight The new value of , that is, the updated value of the weight after an optimization step, is Representation weight The current value of , that is, the updated value of the weight after one optimization step, is Indicates bias The new value of , that is, the updated value of the bias after an optimization step, is Indicates bias The current value of , that is, the bias value before the current step, Represents the loss function About weight The partial derivative of Represents the loss function About Bias The partial derivative of .

[0117] S20324. Iteration, repeating the above process multiple times until the loss reaches an acceptably low value or no longer decreases significantly; specifically, in each iteration, the weights and biases are slightly adjusted to reduce the difference between the model's predicted output and the actual output.

[0118] Furthermore, in step S3, for making a decision based on the predicted value, illustratively, a decision is made based on the predicted value: when the predicted value is close to or equal to the target value, the production process is normal and the current parameter settings are normal; otherwise, adjustments are required.

[0119] As a further preferred embodiment, a software-controlled terminal device for resin material production and processing is proposed, such as Figure 2 The terminal device 200 includes at least one memory 210, at least one processor 220, and a bus 230 connecting different platform systems.

[0120] The memory 210 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 211 and / or a cache memory 212 , and may further include a read-only memory (ROM) 213 .

[0121] The memory 210 also stores a computer program, which can be executed by the processor 220, so that the processor 220 executes any one of the software control methods for resin material production and processing in the embodiments of the present application, and its specific implementation is consistent with the implementation method and the technical effect achieved in the embodiments of the above method, and some of the contents are not repeated. The memory 210 can also include a program / utility 214 having a group (at least one) of program modules 215, such program modules include but are not limited to: an operating system, one or more application programs, other program modules and program data, each of these examples or some combination may include the implementation of a network environment.

[0122] Accordingly, the processor 220 may execute the above-mentioned computer programs, and may execute the program / utility 214 .

[0123] The bus 230 may be a bus representing one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.

[0124] The terminal device 200 may also communicate with one or more external devices 240 such as keyboards, pointing devices, Bluetooth devices, etc., and may also communicate with one or more devices that can interact with the terminal device 200, and / or communicate with any device (such as a router, a modem, etc.) that enables the terminal device 200 to communicate with one or more other computing devices. Such communication may be performed through an input / output (I / O) interface 250. In addition, the terminal device 200 may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and / or a public network, such as the Internet) through a network adapter 260. The network adapter 260 may communicate with other modules of the terminal device 200 through a bus 230. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the terminal device 200, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0125] As a further preferred embodiment, a computer-readable storage medium for software control of resin material production and processing is proposed, wherein instructions are stored on the computer-readable storage medium, and when the instructions are executed by the processor, any of the software control methods for resin material production and processing described above is implemented. The specific implementation method is consistent with the implementation method and the technical effect achieved in the embodiment of the above method, and some contents are not repeated here.

[0126] Figure 3 The program product 300 provided in this embodiment for implementing the above method is shown, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product 300 of the present invention is not limited to this. In this embodiment, the readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it. The program product 300 can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0127] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium may also be any readable medium other than the readable storage medium, which may send, propagate, or transmit a program used by or in combination with an instruction execution system, an apparatus, or a device. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above. The program code for performing the operation of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and also conventional procedural programming languages ​​such as "C" language or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, as an independent software package, partially on a user computing device, partially on a remote computing device, or entirely on a remote computing device or server. Where a remote computing device is involved, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0128] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A software control system for resin material production and processing, characterized in that: The resin is an ultra-low dielectric resin, and the system comprises: A data acquisition module, used to collect raw material composition data, production parameter data and target output data, wherein the target output data is dielectric constant data; The neural network module is used to predict new data points based on the MLP architecture and obtain the predicted dielectric constant by giving a new raw material ratio and production parameters; The result prediction module is used to make decisions on production parameters or raw material ratios based on the predicted results; Wherein, the neural network module specifically includes: A data preprocessing unit is used to normalize raw material composition data, production parameter data and target output data, and scale the range of all numerical features to [0, 1] or convert their distribution using standardization; A neural network building unit is used to build an MLP neural network, including: an input layer, a hidden layer and an output layer, wherein the input layer includes four features and four neurons, and the four features include the proportion of raw material A , Raw material B ratio ,temperature and pressure ; The output layer includes a neuron for outputting the regression task of predicting the dielectric constant; A model training unit, used to train a neural network using data in a data set and using mean square error as a loss function; The neural network building unit also includes: The input layer-hidden layer calculation subunit is used to calculate the output of the hidden layer. The specific calculation method is as follows: ； Among them, the represents the output of the hidden layer, represents the activation function, represents a column vector, Represents the weight matrix from the input layer to the hidden layer, represents the bias vector; the column vector Including the proportion of raw material A, the proportion of raw material B, temperature and pressure, that is, ; The hidden layer-output layer calculation subunit is used to calculate the output of the output layer. The specific calculation method is as follows: ； Among them, the represents the dielectric constant predicted by the model, represents the weight matrix from the hidden layer to the output layer, Represents the bias vector.

2. A software control system for resin material production and processing as claimed in claim 1, characterized in that: The model training unit also includes: The loss value calculation subunit averages the errors of all data points according to the mean square error loss function to obtain a single loss value. The specific calculation method is as follows: ； Among them, the Indicates loss, represents the number of data points, Indicates The true value of the data point, Indicates The model prediction value of data points, Represents the square of the difference between the true value and the predicted value; The neural network training subunit trains the neural network according to the back propagation algorithm.

3. A software control method for resin material production and processing, the method is based on a software control system for resin material production and processing according to any one of claims 1-2, characterized in that: The following steps are involved: S1. Collecting raw material composition data, production parameter data and target output data, wherein the target output data is dielectric constant data; S2. According to the MLP architecture, new data points are predicted, and the predicted dielectric constant is obtained by giving a new raw material ratio and production parameters; S3. Make decisions on production parameters or raw material ratios based on the predicted results; Wherein, the step S2 specifically includes the following sub-steps: S201. Normalize the raw material composition data, production parameter data, and target output data, scaling the range of all numerical features to [0, 1] or converting their distribution using standardization; S202. Establish an MLP neural network, including: an input layer, a hidden layer and an output layer, wherein the input layer includes four features and four neurons, and the four features include the proportion of raw material A , Raw material B accounts for ,temperature and pressure ; The output layer includes a neuron for outputting the regression task of predicting the dielectric constant; S203. Train the neural network using the data in the data set and use the mean square error as the loss function; The step S202 specifically includes the following sub-steps: S2021. Calculate the output of the hidden layer. The specific calculation method is as follows: ； Among them, the represents the output of the hidden layer, represents the activation function, represents a column vector, Represents the weight matrix from the input layer to the hidden layer, represents the bias vector; the column vector Including the proportion of raw material A, the proportion of raw material B, temperature and pressure, that is, ; S2022. Calculate the output of the output layer. The specific calculation method is as follows: ； Among them, the represents the dielectric constant predicted by the model, represents the weight matrix from the hidden layer to the output layer, Represents the bias vector.

4. A software control method for resin material production and processing as claimed in claim 3, characterized in that: The step S203 specifically includes the following sub-steps: S2031. , according to the mean square error loss function, the errors of all data points are averaged to obtain a single loss value, which is calculated as follows: ； Among them, the Indicates loss, represents the number of data points, Indicates The true value of the data point, Indicates The model prediction value of data points, Represents the square of the difference between the true value and the predicted value; S2032. Training a neural network based on a back-propagation algorithm.

Images