A green production method and system for cable materials

Abstract

The invention discloses a green production method and system for cable materials, and relates to the technical field of data processing. The method comprises: obtaining the component ratio constraint and preparation environment constraint of the cable material formula; constructing a cable material degradation state prediction network; performing green correlation analysis on the component ratio constraint and preparation environment constraint; optimizing the cable material formula based on the cable material degradation state prediction network according to the forward attribute set and the reverse attribute set, with the component ratio constraint and preparation environment constraint as restrictions, generating the cable material formula optimization result; and executing the cable material production management according to the cable material formula optimization result. The technical problem that the existing cable materials are difficult to degrade in the natural environment, resulting in long-term pollution of the ecological environment, is solved, and the technical effect of improving the degradability of the cable materials and realizing green production is achieved.

Description

Technical Field

[0001] The present application relates to data processing technology-related fields, and specifically to a green production method and system for cable materials. Background Art

[0002] As global environmental problems become increasingly serious, sustainable development and green production have become the focus of attention in all walks of life. As the basic material for power transmission and communication, the environmental impact generated during the production and use of cable materials cannot be ignored. Traditional cable materials, such as polyvinyl chloride (PVC) and polyethylene (PE), rely on fossil energy as a power source during production and use a large amount of chemical raw materials for synthesis and processing. Although they have good electrical and physical properties, they are difficult to degrade in the natural environment, and the environmental problems and resource consumption problems they bring are gradually becoming prominent. They accumulate in the soil and water for a long time, posing a potential threat to the ecological environment.

[0003] Therefore, in the current technologies related to cable material production, there are technical problems such as difficulty in degradation in the natural environment, which leads to long-term pollution of the ecological environment. Summary of the invention

[0004] The present application provides a green production method and system for cable materials, and adopts technical means such as correlation analysis and construction of a degradation state prediction network to solve the technical problem that existing cable materials are difficult to degrade in the natural environment, resulting in long-term pollution of the ecological environment, thereby achieving the technical effect of improving the degradability of cable materials and realizing green production.

[0005] The present application provides a green production method for cable materials, the method comprising: obtaining component proportion constraints and preparation environment constraints of a cable material formula; constructing a cable material degradation state prediction network according to the component proportion constraints and the preparation environment constraints; performing green correlation analysis on the component proportion constraints and the preparation environment constraints to generate a forward attribute set and a reverse attribute set; optimizing the cable material formula based on the cable material degradation state prediction network according to the forward attribute set and the reverse attribute set, with the component proportion constraints and the preparation environment constraints as restrictions, to generate a cable material formula optimization result; and executing cable material production management according to the cable material formula optimization result.

[0006] In a possible implementation, a cable material degradation state prediction network is constructed according to the component ratio constraint and the preparation environment constraint, and the following processing is performed: subject to the component ratio constraint and the preparation environment constraint, cable material test record data is collected, wherein the cable material test record data includes component ratio record information, preparation environment record information and preset time degradation ratio record information; the cable material test record data is divided into three equal parts, which are set as a first data set, a second data set and a third data set; based on the first data set, a first front channel is configured with the preset time degradation ratio record information as supervision and the component ratio record information and the preparation environment record information as input; based on the second data set, The preset time degradation ratio record information is used as supervision, and the component ratio record information and the preparation environment record information are used as input to configure the second front channel; based on the third data set, the preset time degradation ratio record information is used as supervision, and the component ratio record information and the preparation environment record information are used as input to configure the third front channel; the first output value of the first front channel, the second output value of the second front channel and the third output value of the third front channel are used as input, and the preset time degradation ratio record information is used as supervision to configure an integrated fusion channel; the first front channel, the second front channel, the third front channel and the integrated fusion channel are merged to generate the cable material degradation state prediction network.

[0007] In a possible implementation, based on the first data set, with the preset time degradation ratio record information as supervision, with the component ratio record information and the preparation environment record information as input, a first front channel is configured, and the following processing is performed: when the first front channel meets the preset number of training times, a fourth data set of the preset data volume of the second data set or / and the third data set is selected to verify the first front channel to obtain a first verification mean residual; when the first verification mean residual modulus is less than or equal to the residual modulus threshold, the configuration of the first front channel is completed; when the first verification mean residual modulus is greater than the residual modulus threshold, a first residual fitting subchannel of the first front channel is built according to the first verification mean residual, wherein the first residual fitting subchannel is used to fit the output of the first front channel; according to the first data set, the first front channel and the first residual fitting subchannel are continuously trained until the Nth verification mean residual modulus is less than or equal to the residual modulus threshold, and the first residual fitting subchannel, the second residual fitting subchannel, and the Nth residual fitting subchannel are sequentially connected in series to the output layer of the first front channel to complete the configuration of the first front channel.

[0008] In a possible implementation, a green correlation analysis is performed on the component ratio constraint and the preparation environment constraint to generate a forward attribute set and a reverse attribute set, and the following processing is also performed: based on big data, a benchmark sequence is constructed with a preset time degradation ratio, and a comparison sequence is constructed with the attribute sets of the component ratio constraint and the preparation environment constraint to perform a grey correlation analysis, and a first attribute set with a correlation greater than or equal to a correlation threshold is extracted; any one attribute of the first attribute set is used as a unique variable, and an attribute progressive sequence and a preset time degradation ratio sequence are collected; a correlation analysis is performed based on the preset time degradation ratio sequence and the attribute progressive sequence to generate a Pearson correlation coefficient; based on the Pearson correlation coefficient, the forward attribute set and the reverse attribute set are extracted from the first attribute set.

[0009] In a possible implementation, according to the forward attribute set and the reverse attribute set, with the component ratio constraint and the preparation environment constraint as restrictions, the cable material formula is optimized based on the cable material degradation state prediction network to generate a cable material formula optimization result, and the following processing is also performed: according to the component ratio constraint and the preparation environment constraint, an initial solution set of cable materials is generated through a random uniform distribution function; the initial solution set of cable materials is analyzed through the cable material degradation state prediction network to generate an initial solution fitness set, wherein the solution fitness refers to a preset time degradation ratio; according to the initial solution fitness set, a first number of optimal solutions are selected from the initial solution set of cable materials, and set as a first selected solution; according to the initial solution fitness set, a second number of worst solutions are selected from the initial solution set of cable materials, and set as a second selected solution; according to the forward attribute set and the reverse attribute set, the second selected solution is mutated and optimized with the first selected solution to generate the cable material formula optimization result.

[0010] In a possible implementation, according to the forward attribute set and the reverse attribute set, the second selected solution is mutated and optimized with the first selected solution to generate the cable material formula optimization result, and the following processing is also performed: a solution to be mutated is randomly selected from the second selected solution, and a mutated reference solution is randomly selected from the first selected solution; according to the solution to be mutated and the mutated reference solution, a mutable attribute set and a mutable direction set are selected based on the forward attribute set and the reverse attribute set; according to the mutable direction set and the solution to be mutated, a variable step length standard is performed based on the component ratio constraint and the preparation environment constraint. identification, generating a variable asynchronous length threshold set; based on the variable attribute set, the variable direction set and the variable asynchronous length threshold set, performing several random perturbations on the solution to be mutated to generate a cable material variation solution set; when the cable material variation solution set is greater than or equal to a preset number, analyzing the cable material variation solution set through the cable material degradation state prediction network to generate a variation solution fitness set; updating the cable material initial solution set according to the variation solution fitness set and then executing a loop; when the number of loops meets the preset number, outputting the optimal solution, which is set as the cable material formula optimization result.

[0011] In a possible implementation, when the number of cycles meets the preset number of times, the optimal solution is output and set as the cable material formula optimization result, and the following processing is also performed: a convergence fitness threshold is configured; when the number of cycles does not meet the preset number of times, before each cycle starts, it is determined whether there is a solution that meets the convergence fitness threshold, and if so, it is set as the cable material formula optimization result.

[0012] The present application also provides a green production system for cable materials, comprising:

[0013] A cable material constraint obtaining module, wherein the cable material constraint obtaining module is used to obtain component ratio constraints and preparation environment constraints of a cable material formula;

[0014] A cable material degradation state prediction network construction module, wherein the cable material degradation state prediction network construction module is used to construct a cable material degradation state prediction network according to the component ratio constraint and the preparation environment constraint;

[0015] A green correlation analysis module, the green correlation analysis module is used to perform green correlation analysis on the component ratio constraint and the preparation environment constraint to generate a positive attribute set and a negative attribute set;

[0016] A cable material formula optimization module, wherein the cable material formula optimization module is used to optimize the cable material formula based on the cable material degradation state prediction network according to the forward attribute set and the reverse attribute set, with the component ratio constraint and the preparation environment constraint as constraints, and generate a cable material formula optimization result;

[0017] A cable material production management module, wherein the cable material production management module is used to perform cable material production management according to the cable material formula optimization result.

[0018] Through the green production method and system of cable materials proposed in this application, it is intended to obtain the component ratio constraints and preparation environment constraints of the cable material formula; based on the component ratio constraints and preparation environment constraints, a cable material degradation state prediction network is constructed; green correlation analysis is performed on the component ratio constraints and preparation environment constraints to generate a forward attribute set and a reverse attribute set; based on the forward attribute set and the reverse attribute set, the cable material formula is optimized based on the cable material degradation state prediction network with the component ratio constraints and preparation environment constraints as restrictions, and the cable material formula optimization results are generated; the cable material production management is performed according to the cable material formula optimization results. The technical problem that the existing cable materials are difficult to degrade in the natural environment, resulting in long-term ecological environmental pollution, is solved, and the technical effect of improving the degradability of cable materials and realizing green production is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0020] Figure 1 A schematic diagram of a green production method for cable materials provided in an embodiment of the present application;

[0021] Figure 2 A schematic diagram of the structure of a green production system for cable materials provided in an embodiment of the present application.

[0022] Explanation of the reference numerals: cable material constraint acquisition module 10, cable material degradation state prediction network construction module 20, green correlation analysis module 30, cable material formula optimization module 40, cable material production management module 50. DETAILED DESCRIPTION

[0023] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0024] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0025] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0026] The present application embodiment provides a green production method of cable materials, such as Figure 1 As shown, the method includes:

[0027] Step S100, obtain the ingredient ratio constraints and preparation environment constraints of the cable material formula. The ingredient ratio constraints of cable materials mean that when designing the cable material formula, the composition ratio of various raw materials must meet certain requirements and restrictions, usually based on factors such as performance requirements, cost considerations, and environmental friendliness of the cable material. For example, according to the use environment and functional requirements of the cable, it is necessary to determine the material's conductivity, heat resistance, cold resistance, corrosion resistance and other performance parameters to determine the composition ratio of each raw material; on the premise of meeting performance requirements, it is necessary to reduce the cost of materials as much as possible; cable materials should minimize the impact on the environment during production and disposal, and the environmental friendliness of the raw materials needs to be considered during design to avoid the use of substances harmful to the environment, or reduce the use of these substances; Preparation environment constraints of cable materials It refers to the environmental conditions that cable materials need to meet during the production process, mainly including temperature control, humidity control, ventilation control, dust-free and static-free, etc. Specifically, the production process of cable materials needs to be carried out within a certain range of temperature, humidity and light. Too high or too low temperature and humidity will affect the performance and quality of the materials. For example, too high humidity may cause the material to absorb moisture and affect its insulation performance; harmful gases or dust may be generated during the production process of cable materials, so the production environment needs to be well ventilated; a dust-free and static-free environment needs to be maintained during the production process of cable materials to avoid contamination or damage to the materials, such as using a dust-free workshop, wearing anti-static clothing, etc., to ensure that the cable material production environment is dust-free and static-free.

[0028] Step S200, constructing a cable material degradation state prediction network according to the component ratio constraint and the preparation environment constraint. According to the component ratio constraint and the preparation environment constraint of the cable material formula, a network model capable of predicting the degradation state of the cable material is constructed, and the degradation degree or degradation trend of the cable material under different time and different environmental conditions is predicted by analyzing the components, preparation environment and other factors of the cable material. Specifically, features related to the degradation state of the cable material are selected from the formula data and the preparation environment data, and a prediction model is constructed based on a neural network. The model is trained using the collected data, and the model parameters are adjusted to optimize the prediction performance. Finally, a cable material degradation state prediction network is constructed to achieve real-time monitoring and prediction of the degradation state of the cable material.

[0029] In a possible implementation, step S200 further includes step S210, collecting cable material test record data subject to the component ratio constraint and the preparation environment constraint, wherein the cable material test record data includes component ratio record information, preparation environment record information and preset time degradation ratio record information. When performing cable material performance testing, according to specific component ratios and preparation environment requirements, the relevant data of the cable material is recorded, including component ratio record information, preparation environment record information and preset time degradation ratio record information. Specifically, the component ratio record information refers to the specific ratio of each component in the cable material. According to the formulation optimization result of the cable material, the content of each raw material in the cable material is accurately calculated and recorded; the preparation environment record information includes the environmental parameters involved in the preparation process of the cable material, such as temperature, humidity, pressure, ventilation conditions, etc.; the preset time degradation ratio record information refers to the degradation ratio of the cable material within a specific time period. The degradation ratio is an important indicator for evaluating the durability and stability of the cable material. The preset time can be a time period set according to the use environment and life requirements of the cable material. It also includes step S220, dividing the cable material test record data into three equal parts, which are set as the first data set, the second data set and the third data set. It also includes step S230, based on the first data set, taking the preset time degradation ratio record information as supervision, taking the component ratio record information and the preparation environment record information as input, configuring the first front channel. According to the given first data set and the specific supervision information (preset time degradation ratio record information), a specific processing path or channel is set for the input data (component ratio record information and preparation environment record information), that is, the first front channel. It also includes step S240, based on the second data set, taking the preset time degradation ratio record information as supervision, taking the component ratio record information and the preparation environment record information as input, configuring the second front channel. According to the given second data set and the specific supervision information (preset time degradation ratio record information), a specific processing path or channel is set for the input data (component ratio record information and preparation environment record information), that is, the second front channel. The method also includes step S250, which configures a third front channel based on the third data set, with the preset time degradation ratio record information as supervision, and with the component ratio record information and the preparation environment record information as input. According to the given third data set and specific supervision information (preset time degradation ratio record information), a specific processing path or channel, i.e., the third front channel, is set for the input data (component ratio record information and preparation environment record information). The method also includes step S260, which configures an integrated fusion channel with the first output value of the first front channel, the second output value of the second front channel, and the third output value of the third front channel as input, and with the preset time degradation ratio record information as supervision.The method further includes step S270, merging the first front channel, the second front channel, the third front channel and the integrated fusion channel to generate the cable material degradation state prediction network. The generated cable material degradation state prediction network is a complex network structure integrating multiple front channels and integrated fusion channels, which can simultaneously process multiple types of data, extract key features related to the cable material degradation state prediction, and improve the accuracy and reliability of the prediction by fusing these features.

[0030] In a possible implementation, step S230 further includes step S231, when the first front channel meets the preset number of training times, the fourth data set of the preset data volume of the second data set or / and the third data set is selected to verify the first front channel, and the first verification mean residual is obtained. When the model reaches this preset number of training times, the second data set or / and a part of the third data set (the fourth data set of the preset data volume) is selected as the verification data set, and the trained first front channel is tested on the verification data set to calculate the difference between the predicted value and the actual value of the model. This difference is usually called the residual. By calculating the residuals of all verification data samples and taking the average, the first verification mean residual can be obtained, which reflects the prediction ability of the model on the unseen data (i.e., the verification data set). If the first verification mean residual is small, it means that the prediction performance of the model on the verification data set is good. It also includes step S232, when the first verification mean residual modulus is less than or equal to the residual modulus threshold, the first front channel configuration is completed. The residual modulus threshold is a pre-set value used to determine whether the prediction performance of the model on the validation data set meets the requirements. When the first validation mean residual modulus (i.e., the absolute value of the first validation mean residual) is less than or equal to the preset residual modulus threshold, this indicates that the configuration process of the first front channel has been completed, and the performance of the channel on the validation data set has reached the expected standard. It also includes step S233, when the first validation mean residual modulus is greater than the residual modulus threshold, according to the first validation mean residual, the first residual fitting subchannel of the first front channel is built, wherein the first residual fitting subchannel is used to fit the output of the first front channel. When the first verification mean residual modulus is greater than the preset residual modulus threshold, it means that the performance of the first front channel on the verification data set has not yet reached the expected standard, that is, there is a certain error in the model when predicting. The first residual fitting subchannel of the first front channel can be built according to the first verification mean residual. Specifically, the function of the first residual fitting subchannel is to further adjust or correct the output of the first front channel to reduce the prediction error. For example, if the residual is +2, it means that the error is greater than 2 each time, then the residual fitting channel is the output of the original model minus 2. It also includes step S234, according to the first data set, the first front channel and the first residual fitting subchannel continue to be trained until the Nth verification mean residual modulus is less than or equal to the residual modulus threshold, and the first residual fitting subchannel, the second residual fitting subchannel until the Nth residual fitting subchannel are sequentially connected in series to the output layer of the first front channel to complete the configuration of the first front channel.If the first verification mean residual modulus is greater than the preset residual modulus threshold, it means that the prediction performance of the first front channel still has room for improvement. According to the first verification mean residual, a first residual fitting sub-channel is built, and the first front channel and the first residual fitting sub-channel are jointly trained based on the first data set. After the training is completed, the fourth data set is used for verification again, and the second verification mean residual modulus is calculated. This process is repeated until the Nth verification mean residual modulus is less than or equal to the preset residual modulus threshold. The first residual fitting sub-channel, the second residual fitting sub-channel, and the Nth residual fitting sub-channel are sequentially connected in series to the output layer of the first front channel to form a first front channel containing multiple residual fitting sub-channels.

[0031] Step S300, performing green correlation analysis on the ingredient ratio constraint and the preparation environment constraint to generate a positive attribute set and a negative attribute set. Performing green correlation analysis on the ingredient ratio constraint and the preparation environment constraint of the cable material formula refers to a systematic investigation of these two types of constraints to identify the degree of correlation between them and the green performance (i.e., environmental friendliness) of the cable material, aiming to find out which factors (attributes) have a positive impact on the green performance of the cable material (i.e., positive attributes) and which factors have a negative impact (i.e., negative attributes). Specifically, the positive attribute set refers to those factors or attributes that can enhance the green performance of the cable material, which are usually related to the recyclability, degradability, low toxicity, low energy consumption production, etc. of the cable material, for example, using renewable or bio-based raw materials as The main components of cable materials, the reduction or avoidance of toxic and harmful chemicals in the formula, the use of energy-saving technologies such as low temperature and low pressure in the preparation process, and the selection of additives or auxiliaries with less impact on the environment; the reverse attribute set refers to those factors or attributes that may reduce the green performance of cable materials, which are usually related to the difficulty of recycling, degradation, high toxicity, and high energy consumption in production of cable materials. For example, the formula uses a large amount of non-renewable resources such as petroleum-based plastics, uses additives containing heavy metals or other toxic substances, requires high temperature and high pressure and other energy-intensive processes in the preparation process, and the cable materials are difficult to degrade after being discarded, causing long-term pollution to the environment.

[0032] In a possible implementation, step S300 further includes step S310, which constructs a benchmark sequence based on big data with a preset degradation ratio, constructs a comparison sequence with the attribute set of the component ratio constraint and the preparation environment constraint, performs grey correlation analysis, and extracts the first attribute set with a correlation greater than or equal to a correlation threshold. The reference sequence is a sequence composed of these preset time degradation ratios, representing the baseline of the degradation performance of cable materials over time. The comparison sequence is a sequence composed of the values ​​of these attributes at different times or under different conditions. Each attribute corresponds to a subsequence in the comparison sequence, which together constitute a complete comparison sequence. Grey correlation analysis is a data analysis method based on grey system theory, which is used to analyze the degree of association between different factors in the system and to evaluate the degree of association between various attributes (such as component ratio, preparation environment, etc.) and degradation ratio. In the cable material degradation performance analysis, the grey correlation between the reference sequence and each comparison subsequence is calculated, which is a value between 0 and 1, indicating the degree of association between the two. The greater the correlation, the greater the influence of the attribute on the degradation ratio. After calculating the correlation between all comparison subsequences and the reference sequence, according to the preset correlation threshold (usually a positive number close to 1), the attributes with a correlation greater than or equal to the threshold are screened out to form the first attribute set. It also includes step S320, taking any attribute of the first attribute set as the only variable, collecting the attribute progressive sequence and the preset time degradation ratio sequence. An attribute is selected from the first attribute set as the only variable for study, and a series of different values ​​are designed for the selected attribute, which constitute a progressive sequence of the attribute. For example, if the selected attribute is the ratio of a certain component, the progressive sequence may include different ratio values ​​from low to high. For each value in the attribute progressive sequence, a corresponding cable material sample is prepared, which differs only in the selected attribute, while other factors (such as preparation environment, other component ratios, etc.) should remain consistent. It also includes step S330, performing correlation analysis based on the preset time degradation ratio sequence and the attribute progressive sequence to generate a Pearson correlation coefficient. According to the preset time degradation ratio sequence and attribute progressive sequence, correlation analysis is performed to generate the Pearson correlation coefficient to quantitatively evaluate the linear correlation between the two sequences, wherein the Pearson correlation coefficient is used to measure the strength and direction of the linear relationship between two variables, and the value range is between -1 and 1. When the coefficient is close to 1, it indicates that the two sequences are completely positively correlated, that is, as one sequence increases, the other sequence will also increase, and the magnitude of the change is basically the same; when the coefficient is close to -1, it indicates that the two sequences are completely negatively correlated; when r is close to 0, it indicates that there is no linear relationship between the two sequences. It also includes step S340, according to the Pearson correlation coefficient, extracting the positive attribute set and the reverse attribute set from the first attribute set.All attributes that are positively correlated with the degradation ratio (i.e., the Pearson correlation coefficient is positive) are extracted from the first attribute set to form a positive attribute set, which means that the attributes whose increasing attribute values ​​tend to increase the degradation ratio; all attributes that are negatively correlated with the degradation ratio (i.e., the Pearson correlation coefficient is negative) are extracted from the first attribute set to form a reverse attribute set, which means that the attributes whose increasing attribute values ​​tend to reduce the degradation ratio.

[0033] Step S400, according to the forward attribute set and the reverse attribute set, with the component ratio constraint and the preparation environment constraint as constraints, the cable material formula is optimized based on the cable material degradation state prediction network to generate a cable material formula optimization result. According to the forward attribute set and the reverse attribute set, the cable material formula is optimized based on the cable material degradation state prediction network with the component proportion constraint and the preparation environment constraint of the cable material formula as the constraint conditions, and the cable material formula optimization result is generated, aiming to improve the green performance and environmental friendliness of the cable material. Specifically, based on the forward attribute set, the goal of cable material formula optimization is clarified, and according to the component proportion constraint and the preparation environment constraint, the composition ratio range of various raw materials in the cable material and the environmental conditions that need to be met in the production process are set, and the cable material degradation state prediction network is integrated into the optimization model. The formula parameters of the cable material (such as the raw material composition ratio) are used as decision variables, and the green performance of the cable material (such as the degradation state prediction result) is used as the objective function. Different combinations of formula parameters are continuously tried, and the green performance of these combinations is evaluated through the cable material degradation state prediction network. The formula parameters are continuously adjusted according to the evaluation results to find a formula that meets the constraints and has the best green performance, and finally converge to a formula that meets the constraints and has the best green performance, that is, the cable material formula optimization result.

[0034] In a possible implementation, step S400 further includes step S410, generating an initial solution set of cable materials through a random uniform distribution function according to the component ratio constraint and the preparation environment constraint. According to the component ratio constraint and the preparation environment constraint, random sampling is performed through a random uniform distribution function to generate a set of possible cable material formulas or preparation conditions as an initial solution set. Specifically, the random uniform distribution function is used to generate random numbers uniformly distributed within a specified range, that is, the value range of these random numbers is determined according to the component ratio constraint and the preparation environment constraint. Each random number represents a parameter of the cable material formula or preparation condition, and they are combined to form an initial solution of the cable material. By repeating this process multiple times, an initial solution set containing multiple initial solutions can be generated. It also includes step S420, analyzing the initial solution set of the cable material through the cable material degradation state prediction network to generate an initial solution fitness set, wherein the solution fitness refers to the preset time degradation ratio. Solution fitness refers to the degradation ratio of each initial solution (i.e., the combination of cable material formula or preparation conditions) within a preset time length. Each solution in the cable material initial solution set is input into the cable material degradation state prediction network to obtain the degradation ratio of each solution within a preset time length. The fitness of all solutions is combined to form an initial solution fitness set. It also includes step S430, selecting the first number of optimal solutions from the cable material initial solution set according to the initial solution fitness set, and setting it as the first selected solution. Sort each fitness value in the initial solution fitness set (i.e., the preset time length degradation ratio corresponding to each initial solution), select the first number of optimal solutions from the cable material initial solution set, and set it as the first selected solution. It also includes step S440, selecting the second number of worst solutions from the cable material initial solution set according to the initial solution fitness set, and setting the second selected solution. Select the second number of worst solutions from the cable material initial solution set, set the second selected solution, and have relatively poor performance in the initial solution set. The method further includes step S450, wherein, according to the forward attribute set and the reverse attribute set, the first selected solution is used to perform mutation optimization on the second selected solution to generate the cable material formula optimization result. Based on the forward attribute set and the reverse attribute set, the poor solution (the second selected solution) is improved and optimized with the existing excellent solution (the first selected solution). Specifically, according to the forward attribute set and the reverse attribute set, the attribute values ​​in the second selected solution are fine-tuned to make it closer to the attribute value distribution in the first selected solution. After the mutation optimization, the performance of the second selected solution may be improved, and a new candidate solution for the cable material formula is generated, and the cable material formula with the best performance is screened out as the final optimization result.

[0035] In a possible implementation, step S450 further includes step S451, randomly selecting a solution to be mutated from the second selected solution, and randomly selecting a mutated reference solution from the first selected solution. Randomly select one or more solutions from the second selected solution as the solution to be mutated, and randomly select one or more solutions from the first selected solution as the mutated reference solution, and the mutated reference solution provides a direction and basis for improvement for the solution to be mutated. It also includes step S452, according to the solution to be mutated and the mutated reference solution, select a mutable attribute set and a mutable direction set based on the forward attribute set and the reverse attribute set. By comparing the differences between the solution to be mutated and the mutated benchmark solution in the forward attribute set and the reverse attribute set, it is determined which attributes are likely to improve performance through mutation, and a set of mutable attributes is formed; for each selected mutable attribute, it is necessary to determine the direction of its mutation (i.e., increase or decrease the attribute value). If the value of the attribute in the mutated benchmark solution is better than the solution to be mutated, then the direction of mutation is usually close to the direction of the mutated benchmark solution (i.e., increase or decrease the attribute value to approach the value of the mutated benchmark solution). For example, the a attribute becomes larger and the degradation time becomes shorter, and the a attribute of the mutated benchmark solution is already greater than the a attribute of the solution to be mutated, then its mutation direction is to decrease. It also includes step S453, according to the set of mutable directions and the solution to be mutated, the variable step length is marked based on the component ratio constraint and the preparation environment constraint, and a variable step length threshold set is generated. According to the attribute value of the solution to be mutated, the variable direction set and the preset constraint conditions, a suitable variable step length or change amplitude is determined for each variable attribute. The variable step length threshold set is a set that contains the maximum and minimum variable step lengths (or change amplitudes) of each variable attribute, which is used to limit the change range of the attribute in the subsequent mutation process. It also includes step S454, based on the variable attribute set, the variable direction set and the variable step length threshold set, the solution to be mutated is subjected to several random perturbations to generate a cable material variation solution set. For each variable attribute, the algorithm randomly selects a step length within the allowed range for change according to its variable direction and variable step length threshold set. After several random perturbations, a new set of cable material formula candidate solutions, i.e., a cable material variation solution set, is generated. It also includes step S455, when the cable material variation solution set is greater than or equal to the preset number, the cable material degradation state prediction network is used to analyze the cable material variation solution set to generate a variation solution fitness set.Each variation solution in the cable material variation solution set is used as input and passed to the cable material degradation state prediction network. The cable material degradation state prediction network processes and analyzes the variation solution, predicts the degradation state or performance of the cable material under the formula, and generates a fitness value for each variation solution according to the output result of the cable material degradation state prediction network, which represents the performance of the variation solution under the optimization target. The higher the fitness value, the better the performance of the variation solution, and the closer to the optimization target. The variation solution fitness set is a set of fitness values ​​of all variation solutions in the cable material variation solution set. It also includes step S456, and executes a loop after updating the cable material initial solution set according to the variation solution fitness set. Based on the variation solution fitness set, the algorithm will screen out variation solutions with higher fitness values ​​and add them to the cable material initial solution set. At the same time, some solutions with lower fitness values ​​may be eliminated. The cable material initial solution set is updated to a new set containing more excellent candidate solutions, and then it is repeatedly updated. It also includes step S457, when the number of cycles meets the preset number of times, the optimal solution is output and set as the cable material formula optimization result.

[0036] In a possible implementation, step S457 further includes step S4571, configuring a convergence fitness threshold. A parameter is set to determine when to stop updating the initial solution set of the cable material. Specifically, when the algorithm cannot find a new solution with a fitness value lower than the threshold in multiple consecutive iterations, the algorithm will consider that it has converged to the vicinity of the optimal solution and stop searching. It also includes step S4572, when the number of cycles does not meet the preset number, before each cycle starts, it is determined whether there is a solution that meets the convergence fitness threshold, and if so, it is set as the cable material formula optimization result. When the number of cycles has not reached the preset number, in order to ensure that the algorithm does not continue to iterate meaninglessly, and in order to capture possible excellent solutions in time, it is usually determined before each cycle starts whether there is a solution that meets the convergence fitness threshold in the current solution set. If there is a solution that meets the convergence fitness threshold, this solution is set as the cable material formula optimization result, and terminated in advance before the preset number of cycles is reached, thereby saving computing resources and improving optimization efficiency.

[0037] Step S500, executing cable material production management according to the cable material formula optimization result. Apply the optimized cable material formula to the actual production process, and comprehensively manage the entire production process to ensure that the quality and performance of the cable material meet the optimization target, while achieving green production, improving production efficiency and economic benefits, and meeting market demand and environmental protection requirements.

[0038] In the above, refer to Figure 1 A green production method for cable materials according to an embodiment of the present invention is described in detail. Figure 2A green production system for cable materials according to an embodiment of the present invention is described.

[0039] A green production system for cable materials according to an embodiment of the present invention is used to solve the technical problem that existing cable materials are difficult to degrade in the natural environment, resulting in long-term ecological pollution, and achieves the technical effect of improving the degradability of cable materials and realizing green production. A green production system for cable materials includes: a cable material constraint acquisition module 10, a cable material degradation state prediction network construction module 20, a green correlation analysis module 30, a cable material formula optimization module 40, and a cable material production management module 50.

[0040] A cable material constraint obtaining module 10, wherein the cable material constraint obtaining module 10 is used to obtain component ratio constraints and preparation environment constraints of a cable material formula;

[0041] A cable material degradation state prediction network construction module 20, wherein the cable material degradation state prediction network construction module 20 is used to construct a cable material degradation state prediction network according to the component ratio constraint and the preparation environment constraint;

[0042] A green correlation analysis module 30, the green correlation analysis module 30 is used to perform green correlation analysis on the component ratio constraint and the preparation environment constraint to generate a positive attribute set and a negative attribute set;

[0043] A cable material formula optimization module 40, wherein the cable material formula optimization module 40 is used to optimize the cable material formula based on the cable material degradation state prediction network according to the forward attribute set and the reverse attribute set, with the component ratio constraint and the preparation environment constraint as constraints, and generate a cable material formula optimization result;

[0044] The cable material production management module 50 is used to perform cable material production management according to the cable material formula optimization result.

[0045] The specific configuration of the cable material degradation state prediction network construction module 20 will be described in detail below. The cable material degradation state prediction network construction module 20 further includes: collecting cable material test record data subject to the component ratio constraint and the preparation environment constraint, wherein the cable material test record data includes component ratio record information, preparation environment record information and preset time degradation ratio record information; dividing the cable material test record data into three equal parts, set as the first data set, the second data set and the third data set; based on the first data set, with the preset time degradation ratio record information as supervision, with the component ratio record information and the preparation environment record information as input, configuring the first front channel; based on the second data set, with the preset time degradation ratio record information The first front channel, the second front channel and the preparation environment record information are used as supervision, and the second front channel is configured with the component ratio record information and the preparation environment record information as input; based on the third data set, the preset time degradation ratio record information is used as supervision, and the third front channel is configured with the component ratio record information and the preparation environment record information as input; the first output value of the first front channel, the second output value of the second front channel and the third output value of the third front channel are used as input, and the preset time degradation ratio record information is used as supervision to configure an integrated fusion channel; the first front channel, the second front channel, the third front channel and the integrated fusion channel are combined to generate the cable material degradation state prediction network.

[0046] The specific configuration of the cable material degradation state prediction network construction module 20 will be described in detail below. The cable material degradation state prediction network construction module 20 may further include: when the first front channel meets the preset number of training times, the first front channel is verified by selecting the fourth data set of the preset data volume of the second data set or / and the third data set to obtain the first verification mean residual; when the first verification mean residual modulus is less than or equal to the residual modulus threshold, the first front channel configuration is completed; when the first verification mean residual modulus is greater than the residual modulus threshold, according to the first verification mean residual, the first residual fitting subchannel of the first front channel is built, wherein the first residual fitting subchannel is used to fit the output of the first front channel; according to the first data set, the first front channel and the first residual fitting subchannel are continuously trained until the Nth verification mean residual modulus is less than or equal to the residual modulus threshold, and the first residual fitting subchannel, the second residual fitting subchannel, and the Nth residual fitting subchannel are sequentially connected in series to the output layer of the first front channel to complete the configuration of the first front channel.

[0047] The specific configuration of the green correlation analysis module 30 will be described in detail below. The green correlation analysis module 30 may further include: based on big data, constructing a benchmark sequence with a preset time degradation ratio, constructing a comparison sequence with the attribute set of the component ratio constraint and the preparation environment constraint, performing grey correlation analysis, and extracting a first attribute set with a correlation greater than or equal to a correlation threshold; taking any attribute of the first attribute set as a unique variable, collecting an attribute progressive sequence and a preset time degradation ratio sequence; performing correlation analysis based on the preset time degradation ratio sequence and the attribute progressive sequence to generate a Pearson correlation coefficient; extracting the forward attribute set and the reverse attribute set from the first attribute set based on the Pearson correlation coefficient.

[0048] The specific configuration of the cable material formula optimization module 40 will be described in detail below. The cable material formula optimization module 40 further includes: generating a cable material initial solution set through a random uniform distribution function according to the component ratio constraint and the preparation environment constraint; analyzing the cable material initial solution set through the cable material degradation state prediction network to generate an initial solution fitness set, wherein the solution fitness refers to a preset time degradation ratio; selecting a first number of optimal solutions from the cable material initial solution set according to the initial solution fitness set, and setting it as a first selected solution; selecting a second number of worst solutions from the cable material initial solution set according to the initial solution fitness set, and setting a second selected solution; performing mutation optimization on the second selected solution with the first selected solution according to the forward attribute set and the reverse attribute set, and generating the cable material formula optimization result.

[0049] The specific configuration of the cable material formula optimization module 40 will be described in detail below. The cable material formula optimization module 40 further includes: randomly selecting a solution to be mutated from the second selected solution, and randomly selecting a mutated reference solution from the first selected solution; according to the solution to be mutated and the mutated reference solution, selecting a mutable attribute set and a mutable direction set based on the forward attribute set and the reverse attribute set; according to the mutable direction set and the solution to be mutated, performing variable asymmetry identification based on the component ratio constraint and the preparation environment constraint, and generating a variable asymmetry threshold set; based on the mutable attribute set, the mutable direction set and the variable asymmetry threshold set, performing several random perturbations on the solution to be mutated to generate a cable material mutated solution set; when the cable material mutated solution set is greater than or equal to a preset number, analyzing the cable material mutated solution set through the cable material degradation state prediction network to generate a mutated solution fitness set; updating the cable material initial solution set according to the mutated solution fitness set and then executing a loop; when the number of loops meets the preset number, outputting the optimal solution, which is set as the cable material formula optimization result.

[0050] The specific configuration of the cable material formula optimization module 40 will be described in detail below. The cable material formula optimization module 40 may further include: configuring a convergence fitness threshold; when the number of cycles does not meet the preset number, before each cycle starts, determining whether there is a solution that meets the convergence fitness threshold, and if so, setting it as the cable material formula optimization result.

[0051] A green production system for cable materials provided in an embodiment of the present invention can execute a green production method for cable materials provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0052] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0053] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A green production method for cable materials, characterized in that: include: Obtaining ingredient ratio constraints and preparation environment constraints for cable material formulations; Constructing a cable material degradation state prediction network according to the component ratio constraint and the preparation environment constraint; Performing green relevance analysis on the component ratio constraint and the preparation environment constraint to generate a positive attribute set and a negative attribute set; According to the forward attribute set and the reverse attribute set, with the component ratio constraint and the preparation environment constraint as constraints, the cable material formula is optimized based on the cable material degradation state prediction network to generate a cable material formula optimization result; Execute cable material production management according to the cable material formula optimization result; According to the component ratio constraint and the preparation environment constraint, a cable material degradation state prediction network is constructed, including: Subject to the component ratio constraint and the preparation environment constraint, collecting cable material test record data, wherein the cable material test record data includes component ratio record information, preparation environment record information and preset time degradation ratio record information; Divide the cable material test record data into three equal parts, setting them as a first data set, a second data set and a third data set; Based on the first data set, the first front channel is configured with the preset time degradation ratio record information as supervision and the component ratio record information and the preparation environment record information as input; Based on the second data set, the second front channel is configured with the preset time degradation ratio record information as supervision and the component ratio record information and the preparation environment record information as input; Based on the third data set, with the preset time degradation ratio record information as supervision, and with the component ratio record information and the preparation environment record information as input, a third front channel is configured; Taking the first output value of the first front channel, the second output value of the second front channel and the third output value of the third front channel as input, and taking the preset time degradation ratio record information as supervision, configuring an integrated fusion channel; Combining the first front channel, the second front channel, the third front channel and the integrated fusion channel to generate the cable material degradation state prediction network; Based on the first data set, with the preset time degradation ratio record information as supervision, and with the component ratio record information and the preparation environment record information as input, a first front channel is configured, including: When the first front channel meets the preset number of training times, a fourth data set with a preset data amount of the second data set or / and the third data set is selected to verify the first front channel to obtain a first verification mean residual; When the first verification mean residual modulus value is less than or equal to the residual modulus value threshold, the first front channel configuration is completed; When the first verification mean residual modulus is greater than the residual modulus threshold, building a first residual fitting subchannel of the first front channel according to the first verification mean residual, wherein the first residual fitting subchannel is used to fit the output of the first front channel; According to the first data set, the first front channel and the first residual fitting sub-channel continue to be trained until the Nth verification mean residual modulus value is less than or equal to the residual modulus value threshold, and the first residual fitting sub-channel, the second residual fitting sub-channel until the Nth residual fitting sub-channel are sequentially connected in series to the output layer of the first front channel to complete the configuration of the first front channel.

2. A green production method for cable materials according to claim 1, characterized in that: A green correlation analysis is performed on the ingredient ratio constraint and the preparation environment constraint to generate a positive attribute set and a reverse attribute set, including: Based on big data, a benchmark sequence is constructed with a preset time degradation ratio, a comparison sequence is constructed with the attribute set of the component ratio constraint and the preparation environment constraint, and a grey correlation analysis is performed to extract the first attribute set with a correlation greater than or equal to a correlation threshold; Taking any one attribute of the first attribute set as a unique variable, collecting an attribute progressive sequence and a preset time-length degradation ratio sequence; Performing correlation analysis based on the preset time degradation ratio sequence and the attribute progressive sequence to generate a Pearson correlation coefficient; The forward attribute set and the reverse attribute set are extracted from the first attribute set according to the Pearson correlation coefficient.

3. A green production method for cable materials according to claim 1, characterized in that: According to the forward attribute set and the reverse attribute set, with the component ratio constraint and the preparation environment constraint as constraints, the cable material formula is optimized based on the cable material degradation state prediction network to generate a cable material formula optimization result, including: According to the component ratio constraint and the preparation environment constraint, generating an initial solution set of cable materials through a random uniform distribution function; The cable material initial solution set is analyzed by the cable material degradation state prediction network to generate an initial solution fitness set, wherein the solution fitness refers to a preset time degradation ratio; Selecting a first number of optimal solutions from the cable material initial solution set according to the initial solution fitness set, and setting them as first selected solutions; Selecting a second number of worst solutions from the cable material initial solution set according to the initial solution fitness set, and setting a second selected solution; According to the forward attribute set and the reverse attribute set, the second selected solution is mutated and optimized with the first selected solution to generate the cable material formula optimization result.

4. A green production method for cable materials as claimed in claim 3, characterized in that: According to the forward attribute set and the reverse attribute set, the second selected solution is mutated and optimized with the first selected solution to generate the cable material formula optimization result, including: Randomly select a solution to be mutated from the second selected solution, and randomly select a mutated reference solution from the first selected solution; According to the solution to be mutated and the mutated benchmark solution, selecting a mutable attribute set and a mutable direction set based on the forward attribute set and the reverse attribute set; According to the set of variable directions and the solution to be mutated, variable step length identification is performed based on the component ratio constraint and the preparation environment constraint to generate a variable step length threshold set; Based on the variable attribute set, the variable direction set and the variable asynchronous length threshold set, the to-be-varied solution is subjected to several random perturbations to generate a cable material variation solution set; When the cable material variation solution set is greater than or equal to a preset number, the cable material variation solution set is analyzed by the cable material degradation state prediction network to generate a variation solution fitness set; The cable material initial solution set is updated according to the variation solution fitness set and then a loop is executed; When the number of cycles meets the preset number, the optimal solution is output and set as the cable material formula optimization result.

5. A green production method for cable materials as claimed in claim 4, characterized in that: When the number of cycles meets the preset number, the optimal solution is output and set as the cable material formula optimization result, including: Configure the convergence fitness threshold; When the number of cycles does not meet the preset number, before each cycle starts, it is determined whether there is a solution that meets the convergence fitness threshold, and if so, it is set as the cable material formula optimization result.

6. A green production system for cable materials, characterized in that: The system is used to implement a green production method for cable materials according to any one of claims 1 to 5, and the system comprises: A cable material constraint obtaining module, wherein the cable material constraint obtaining module is used to obtain component ratio constraints and preparation environment constraints of a cable material formula; A cable material degradation state prediction network construction module, wherein the cable material degradation state prediction network construction module is used to construct a cable material degradation state prediction network based on the component ratio constraint and the preparation environment constraint; A green correlation analysis module, the green correlation analysis module is used to perform green correlation analysis on the component ratio constraint and the preparation environment constraint to generate a positive attribute set and a negative attribute set; A cable material formula optimization module, wherein the cable material formula optimization module is used to optimize the cable material formula based on the cable material degradation state prediction network according to the forward attribute set and the reverse attribute set, with the component ratio constraint and the preparation environment constraint as constraints, and generate a cable material formula optimization result; A cable material production management module, wherein the cable material production management module is used to perform cable material production management according to the cable material formula optimization result.

Images