Graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization

Abstract

The invention provides a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization, and relates to the technical field of automation control, and the graphene antibacterial polyester yarn production control platform comprises the steps that a quality control analysis module obtains a production operation parameter set of graphene antibacterial polyester yarns for quality control analysis, and sets product performance indexes; the optimization model building module is used for building an objective function and root and building a multi-objective optimization model; the process optimization module performs process optimization by using a multi-objective optimization model to obtain production optimization parameters; the operation change analysis module performs operation control monitoring based on the production optimization parameters and draws a production parameter change trend chart; the product quality prediction module performs product quality prediction through reinforcement learning; the control strategy making module makes a production control strategy based on the product quality prediction data set. The technical problems of product quality fluctuation and cost control difficulty caused by lack of real-time accurate adjustment means for production parameters in the prior art are solved, and the production cost is reduced while the product quality is improved.

Description

Technical Field

[0001] The present application relates to the field of automated control technology, and in particular to a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization. Background Art

[0002] As a new type of functional fiber, graphene antibacterial polyester yarn is widely used in medical, clothing and other fields due to its excellent antibacterial properties, good biocompatibility and high strength.

[0003] The production process of graphene antibacterial polyester yarn involves a variety of complex process parameters, such as temperature, tension, heating rate, etc. These factors directly affect the antibacterial performance and quality stability of the product. Existing graphene antibacterial polyester yarn production is mostly based on static production parameter settings, lacking real-time and accurate data monitoring and feedback mechanisms, and unable to respond promptly to changes that may occur in the production process, such as equipment failures or external environmental fluctuations. Secondly, due to the lack of a flexible optimization mechanism, traditional methods are not efficient and accurate enough when dealing with a changing production environment. The adjustment of production parameters often lags behind actual demand, resulting in fluctuations in product quality, as well as waste of resources and excessively high production costs. Summary of the invention

[0004] The present application provides a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization, which solves the technical problems in the prior art that the graphene antibacterial polyester yarn production process lacks real-time monitoring and dynamic adjustment of production parameters, resulting in product quality fluctuations and difficulty in cost control, and achieves the technical effect of improving the quality of graphene antibacterial polyester yarn products while reducing production costs.

[0005] In view of the above problems, the present application provides a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization, and the platform includes: a quality control analysis module, which is used to obtain a production operation parameter set by operating sensing of the graphene antibacterial polyester yarn production line, perform quality control analysis based on the production operation parameter set, and set multiple product performance indicators; an optimization model construction module, which is used to perform multi-objective optimization according to the multiple product performance indicators in combination with the production operation parameter set, establish an objective function, and construct a multi-objective optimization model based on the objective function; a process optimization module, which is used to synchronize the production operation parameter set to the multi-objective optimization model for process optimization and obtain multiple production optimization parameters; an operation change analysis module , used to perform operation control and monitoring of the production line of graphene antibacterial polyester yarn based on the multiple production optimization parameters, generate control feedback information, perform operation change analysis of the production line according to the control feedback information, and draw a production parameter change trend chart; a product quality prediction module, used to perform reinforcement learning according to the production parameter change trend chart to obtain learning effect, perform product quality prediction for the multiple production optimization parameters according to the learning effect, and generate a product quality prediction data set; a control strategy formulation module, used to dynamically adjust the multiple production optimization parameters based on the product quality prediction data set, formulate a production control strategy, and execute the production control strategy to intelligently control the production of graphene antibacterial polyester yarn.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The quality control analysis module obtains the production operation parameter set through the operation sensing of the graphene antibacterial polyester yarn production line, and performs quality control analysis to set product performance indicators, providing a basic standard for subsequent optimization and production control. The optimization model construction module establishes the objective function and constructs a multi-objective optimization model according to the performance indicators obtained from the quality control analysis and the production operation parameter set, providing a theoretical framework for process optimization. The process optimization module synchronizes the production operation parameter set to the multi-objective optimization model for process optimization, obtains multiple production optimization parameters, and realizes the preliminary optimization of real-time process parameters. The operation change analysis module monitors the operation control of the graphene antibacterial polyester yarn production line based on multiple production optimization parameters, generates control feedback information and analyzes operation changes, draws a production parameter change trend chart, monitors the production process in real time and feedbacks the production status, and provides data support for product quality prediction. The product quality prediction module predicts product quality after reinforcement learning based on the production parameter change trend chart, generates a product quality prediction data set, and provides a basis for the formulation of control strategies, so as to better cope with dynamic changes in production and reduce quality fluctuations. The control strategy formulation module dynamically adjusts multiple production optimization parameters based on the product quality prediction data set, formulates production control strategies, and executes production control strategies to achieve intelligent control of graphene antibacterial polyester yarn production and optimize product quality and production costs.

[0007] In summary, this application, through the synergy of each module, starts from setting product performance indicators, constructs an optimization model and obtains optimization parameters, then monitors, analyzes and predicts the quality of the production process, and finally formulates a control strategy based on the prediction results to intelligently control the production, thereby realizing closed-loop control of the production of graphene antibacterial polyester yarn. This solution comprehensively considers multiple product performance indicators, makes full use of production data, and realizes the comprehensive optimization of the production process of graphene antibacterial polyester yarn. It not only improves the stability and consistency of product quality and reduces the defective rate, but also reduces waste in the production process and reduces production costs, which effectively promotes the production of graphene antibacterial polyester yarn towards intelligence and efficiency.

[0008] 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. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic diagram of the structure of a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization provided in an embodiment of the present application.

[0010] Figure 2 A schematic diagram of the process of establishing an objective function in a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization provided in an embodiment of the present application.

[0011] Figure 3 A schematic diagram of the process of obtaining multiple production optimization parameters in the graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization provided in an embodiment of the present application.

[0012] Explanation of the reference numerals: quality control analysis module 10 , optimization model building module 20 , process optimization module 30 , operation change analysis module 40 , product quality prediction module 50 , control strategy formulation module 60 . DETAILED DESCRIPTION

[0013] The embodiment of the present application solves the technical problems in the prior art of lacking real-time monitoring and dynamic adjustment of production parameters in the production process of graphene antibacterial polyester yarn, resulting in fluctuations in product quality and difficulty in cost control, by providing a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization. The technical effect of improving the quality of graphene antibacterial polyester yarn products while reducing production costs is achieved.

[0014] like Figure 1 As shown, the embodiment of the present application provides a graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization, and the platform includes: The quality control analysis module 10 is used to obtain a production operation parameter set by sensing the operation of the production line of graphene antibacterial polyester yarn, perform quality control analysis based on the production operation parameter set, and set multiple product performance indicators.

[0015] Specifically, various sensors, such as temperature sensors, tension sensors, humidity sensors, etc., are installed on the production line of graphene antibacterial polyester yarn to collect production operation parameters in the production process and summarize them to generate a set of production operation parameters. These parameter sets may include temperature, humidity, machine speed, temperature and pressure of the coating material, etc. in the production process. The collected data are analyzed, for example, the impact of different parameters such as temperature and humidity in the production process on the quality of the final product (such as antibacterial properties and tensile strength) is calculated, the product quality measurement standard is determined, and multiple product performance indicators are set, including the antibacterial rate and tensile strength of graphene polyester yarn. Setting product performance indicators through quality control analysis provides optimization direction and evaluation criteria for subsequent parameter optimization.

[0016] The optimization model building module 20 is used to perform multi-objective optimization according to the multiple product performance indicators in combination with the production operation parameter set, establish an objective function, and build a multi-objective optimization model according to the objective function.

[0017] Specifically, the objective function is a mathematical expression in multi-objective optimization that is used to quantify and measure the quality of the objectives. The multi-objective optimization model is a mathematical model composed of objective functions and constraints, which is used to balance and optimize multiple objectives. Through this model, the best compromise solution between multiple objectives can be found.

[0018] In the production process, we often not only focus on the optimization of a certain indicator, but also need to comprehensively consider multiple factors to achieve the overall optimization of the production process. First, the objective function is constructed based on product performance indicators (such as antibacterial rate, tensile strength) and production parameters (such as graphene concentration, coating temperature, etc.). For example, the optimization objectives include cost optimization objectives (such as minimizing energy consumption) and performance optimization objectives (such as maximizing antibacterial rate and tensile strength), etc., and multiple objectives are integrated using a weighted average method to generate a total objective function. According to the constructed objective function combined with the constraints in the production process, such as equipment capacity limitations, production time limitations, raw material supply limitations, etc., a multi-objective optimization model is constructed using a multi-objective optimization algorithm (such as particle swarm optimization algorithm, genetic algorithm, etc.), to achieve comprehensive optimization in a complex production process and ensure high-quality and efficient production of products.

[0019] The process optimization module 30 is used to synchronize the production operation parameter set to the multi-objective optimization model to perform process optimization and obtain multiple production optimization parameters.

[0020] Specifically, the production operation parameter set is input into the constructed multi-objective optimization model. The multi-objective optimization model continuously adjusts the various production operation parameters through the internal optimization algorithm to find a set of parameters that best suits the current production conditions so that the value of the objective function is maximized. This set of production operation parameters is output as the production optimization parameters.

[0021] The operation change analysis module 40 is used to perform operation control monitoring on the production line of graphene antibacterial polyester yarn based on the multiple production optimization parameters, generate control feedback information, perform operation change analysis of the production line according to the control feedback information, and draw a production parameter change trend chart.

[0022] Specifically, multiple production optimization parameters are applied to the production line of graphene antibacterial polyester yarn. The production process is monitored in real time through the monitoring equipment on the production line, the changes in various parameters during the production process are recorded, and feedback information is generated. This feedback information includes production operation parameters at each moment (coating temperature, graphene concentration, energy consumption, etc.) and product performance parameters (antibacterial performance and physical strength).

[0023] Perform data analysis on the collected control feedback information, and by comparing the current parameters with multiple production optimization parameters, find out the fluctuation trend of the operating parameters in the production process, and provide a scientific basis for subsequent parameter adjustments. Draw a production parameter change trend chart based on the results of the operation change analysis. The production parameter change trend chart is a way to intuitively present data changes. It uses graphics (such as line charts, bar charts, etc.) to show the changes of various key production parameters at different time points and under different production conditions. For example, by drawing a line chart with time as the horizontal axis and spinning temperature as the vertical axis, you can clearly see the fluctuation of spinning temperature in different time periods.

[0024] By drawing a production parameter change trend chart, the fluctuation of operating status in the production process can be intuitively displayed, providing an analytical basis for subsequent product quality prediction and control strategy formulation.

[0025] The product quality prediction module 50 is used to perform reinforcement learning according to the production parameter change trend diagram to obtain a learning effect, perform product quality prediction on the multiple production optimization parameters according to the learning effect, and generate a product quality prediction data set.

[0026] Specifically, the reinforcement learning algorithm is used to build a learning environment based on the production parameter change trend chart. In the learning environment, the module continuously adjusts the parameters and observes the corresponding feedback (i.e., the change in product quality), and evaluates the quality of a certain parameter combination based on the feedback signal. For example, if a certain parameter combination leads to a higher antibacterial rate, positive feedback (reward) is given; conversely, if the combination leads to a decrease in product quality, negative feedback (penalty) is given. Through multiple trial and error learning, the prediction accuracy of the algorithm for product quality is continuously improved, thereby obtaining a good learning effect. The learning effect includes the influence of production parameters on product quality. The learning effect is used to predict the corresponding product quality based on multiple production optimization parameters to generate a product quality prediction data set. This data set is a compilation and summary of the product quality prediction results. For example, the data set may contain the corresponding predicted product quality indicators at different times (such as predicted antibacterial rate, tensile strength, etc.), as well as the production optimization parameters at that time (such as the temperature, pressure, raw material ratio, etc. at that time). These data can provide an important basis for the formulation of subsequent production control strategies.

[0027] The control strategy formulation module 60 is used to dynamically adjust the multiple production optimization parameters based on the product quality prediction data set, formulate a production control strategy, and execute the production control strategy to intelligently control the production of graphene antibacterial polyester yarn.

[0028] Specifically, the product quality prediction information and corresponding production optimization parameters under different production conditions are extracted from the product quality prediction data set, and the production optimization parameters and product quality prediction data in the data set are mined for association rules to find the association between the production optimization parameters and the corresponding product quality results at a specific moment. According to the results of the association rule analysis, multiple production optimization parameters are adaptively adjusted at different production moments. For example, if it is found that the change of a certain production optimization parameter (such as temperature) within a certain range will lead to a significant decrease in the antibacterial rate of the product, then the temperature parameter will be adjusted. Through the adjustment of production optimization parameters, a specific production control strategy is generated. The control strategy includes: the optimal parameter value in the production process, the adjustment cycle (how long or how many batches are produced to adjust the parameters), the adjustment range, etc. After the control strategy is formulated, it is sent to the automation control equipment (such as PLC, SCADA system, etc.) to perform the adjustment operation. At the same time, during the production process, the various parameters on the production line are continuously monitored. If some key parameters are found to deviate from the target value, they are adjusted in time. By formulating a control strategy, the key parameters in the production process are adjusted in real time, thereby improving the stability of product quality.

[0029] Further, such as Figure 2 As shown, the optimization model building module 20 of the embodiment of the present application is also used to perform the following steps: Step P21: Analyze the production operation parameter set to obtain graphene concentration parameters, coating temperature parameters, and energy consumption parameters, perform cost analysis based on the graphene concentration parameters, the coating temperature parameters, and the energy consumption parameters, and set a cost optimization target.

[0030] Step P22: Analyze the multiple product performance indicators to obtain an antibacterial rate indicator and a tensile strength indicator, perform performance analysis based on the antibacterial rate indicator and the tensile strength indicator, and set a performance optimization target.

[0031] Step P23: Calculate the graphene distribution uniformity deviation value according to the graphene concentration parameter in combination with the multiple product performance indicators according to the performance standard, and set optimization constraints according to the graphene distribution uniformity deviation value.

[0032] Step P24: performing weight calculation on the graphene concentration parameter, the coating temperature parameter, the energy consumption parameter, the antibacterial rate index, and the tensile strength index to obtain a plurality of weight coefficients.

[0033] Step P25: According to the optimization constraints, the multiple weight coefficients are matched with the performance optimization target and the cost optimization target to construct the objective function.

[0034] Specifically, the set of production operation parameters collected during the production process is parsed to extract key data such as graphene concentration, coating temperature and energy consumption. Among them, graphene concentration refers to the amount of graphene added to the polyester yarn, usually expressed in percentage. Coating temperature refers to the temperature used during the coating process. Different temperatures will affect the uniformity of graphene distribution on the surface of the polyester yarn. Energy consumption refers to the energy consumed in the production process, usually expressed in electricity or other energy consumption units. Cost analysis is performed based on the obtained graphene concentration parameters, coating temperature parameters and energy consumption parameters. Graphene concentration affects the cost of raw materials, coating temperature affects the cost of energy consumption and equipment loss, and energy consumption parameters directly reflect the energy cost. By comprehensively considering these factors, a cost calculation model is established to calculate the production cost. For example, by calculating the unit energy consumption cost, the procurement cost of graphene and the impact of coating temperature on production cost, a comprehensive cost analysis model is established, and then the cost optimization goal is set, that is, to reduce the production cost as much as possible without affecting product quality.

[0035] Based on multiple product performance indicators, the antibacterial rate indicator and tensile strength indicator are obtained. Performance analysis is performed based on the antibacterial rate indicator and tensile strength indicator. The antibacterial rate indicator reflects the antibacterial function of the product. The higher the antibacterial rate, the stronger the antibacterial performance of the polyester yarn. The tensile strength indicator reflects the physical strength characteristics of the product. Too low tensile strength will affect the service life and stability of the product. Set performance optimization goals by comparing with market demand, product standards or internal quality requirements of the enterprise. For example, if the market requires the antibacterial rate of graphene antibacterial polyester yarn to be no less than 90%, the optimization target of the antibacterial rate indicator is set to A⩾90%, where A represents the antibacterial rate; for the tensile strength indicator, if the company requires the tensile strength of the product to reach S⩾S0, where S is the tensile strength and S0 is the preset tensile strength target value.

[0036] Graphene distribution uniformity reflects the uniformity of graphene coating on the surface of polyester yarn. If graphene is not evenly distributed, it will lead to inconsistent product performance. According to the product performance indicators and production operation parameter set, the allowable deviation range of graphene distribution uniformity is determined when the product performance meets the standard, that is, the graphene distribution uniformity deviation value, and the deviation value is set as the optimization constraint to ensure that the product quality requirements are always met in the subsequent parameter optimization process.

[0037] The graphene concentration parameters, coating temperature parameters, energy consumption parameters, antibacterial rate index, and tensile strength index are weighted to determine the relative importance of each parameter and index in multi-objective optimization. This can be achieved through the analytic hierarchy process, expert scoring method, or data-based statistical analysis methods. For example, when using the analytic hierarchy process, a hierarchical model is constructed, and the target layer (multi-objective optimization) and the criterion layer (various parameters and indicators) are layered. The weight coefficients of each parameter and index are obtained by constructing a judgment matrix and calculating the eigenvector. After weight calculation, multiple weight coefficients corresponding to multiple production operation parameters are obtained.

[0038] According to the optimization constraints, multiple weight coefficients are matched with the performance optimization goals and cost optimization goals to construct the objective function. In this way, in the subsequent optimization process, the optimization model can seek the optimal production parameters under multiple constraints and goals, thereby reducing costs while ensuring that product quality meets the predetermined standards.

[0039] Further, such as Figure 3 As shown, the process optimization module 30 in the embodiment of the present application is also used to perform the following steps: Step P31: Synchronize the production operation parameter set to the multi-objective optimization model, and generate an initial population based on random combinations of the production operation parameter set.

[0040] Step P32: introducing a fitness function, calculating the initial population through the fitness function, and obtaining a plurality of fitnesses, wherein the plurality of fitnesses correspond to the initial population.

[0041] Step P33: sorting the initial population in descending order according to the multiple fitnesses to generate a production parameter sequence, performing multiple extractions according to the production parameter sequence until there is a population whose fitness is less than a preset threshold, and obtaining multiple production parameter combinations.

[0042] Step P34: extract the first-order production parameter combination and the second-order production parameter combination based on the multiple production parameter combinations, cross-operate the first-order production parameter combination and the second-order production parameter combination to generate a first production parameter combination, and iterate until the multiple production parameter combinations are traversed to generate a second production parameter combination...the Nth production parameter combination, where N is an integer greater than 1.

[0043] Step P35: Perform fitness analysis based on the first production parameter combination, the second production parameter combination...the Nth production parameter combination, determine the ith production parameter combination according to the fitness analysis result, and add the ith production parameter combination to the multiple production optimization parameters, where i is an integer greater than 0.

[0044] Specifically, the set of production operation parameters collected from the production process is synchronized into the multi-objective optimization model. Based on these parameters, random combinations are made to generate an initial population. Each individual in the initial population represents a combination of production parameters, and this set of initial solutions will serve as the starting point for the parameter optimization process. Exemplarily, there are three parameters: graphene concentration (2%, 3%, 4%), coating temperature (180°C, 200°C), and energy consumption (50 kW, 60 kW). Then the initial population may contain 8 random combinations, for example, combination 1: 2%, 180°C, 50 kW; combination 2: 3%, 200°C, 60 kW, etc.

[0045] The fitness function is introduced to calculate the fitness value of each production parameter combination in the initial population. The fitness function is used to evaluate the pros and cons of each combination. The higher the value of the fitness function, the better the production parameter combination. The fitness function can be defined according to the production goal, for example, calculated based on the weights of production cost and antibacterial performance. Each combination in the initial population is substituted into each combination for calculation to obtain multiple fitness values ​​corresponding to multiple parameter combinations.

[0046] This step is to arrange the combinations in the initial population from high to low according to fitness and generate a production parameter sequence. In this way, priority can be given to combinations with high fitness, which are more likely to be solutions close to the optimal solution. The preset threshold is a pre-set fitness standard value. When the fitness of a combination is lower than this standard value, it is considered that the combination does not meet the requirements and the extraction is stopped. Production parameter combinations are extracted from them in the order of the production parameter sequence. Each time the extraction is performed, the fitness of the extracted parameter combination is compared with the preset threshold. When the fitness of the extracted combination is lower than the preset threshold, the extraction is stopped. At this time, the multiple production parameter combinations obtained are relatively optimal production parameter combinations.

[0047] Extract the first-order production parameter combination and the second-order production parameter combination (i.e., the production parameter combination with the largest fitness value and the production parameter combination with the second largest fitness value) from multiple production parameter combinations, and perform a crossover operation on the two combinations, that is, exchange certain parameters of the two production parameter combinations to generate a new parameter combination, which is recorded as the first production parameter combination. Traverse multiple production parameter combinations, repeat the crossover operation, and generate new and possibly better combinations by continuously combining different production parameter combinations to obtain the second production parameter combination, the third production parameter combination, and the Nth production parameter combination, where N is an integer greater than 1.

[0048] The fitness function is used to perform fitness analysis on these newly generated production parameter combinations. According to the analysis results, the i-th production parameter combination (i is an integer greater than 0) with the highest fitness value is selected as the final production optimization parameter and added to multiple production optimization parameters.

[0049] The above steps use genetic algorithms to optimize and evaluate the production parameter combinations, find the best balance between multiple objectives, and ensure the optimization of production costs and product quality.

[0050] Furthermore, step P35 also includes: Step P351: Sort the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination in descending order according to the fitness analysis result to determine a fitness sequence.

[0051] Step P352: performing adjacent clustering on the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination according to the fitness sequence to generate a plurality of production parameter clusters.

[0052] Step P353: Perform a multi-objective balance analysis based on the multiple production parameter clusters, and screen the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination based on the balance analysis results to determine the i-th production parameter combination.

[0053] Specifically, according to the fitness analysis result of each production parameter combination (i.e., the fitness value calculated according to the fitness function), the first production parameter combination, the second production parameter combination ... the Nth production parameter combination are sorted in descending order according to the fitness value to obtain a fitness sequence.

[0054] Based on the fitness sequence, these production parameter combinations are clustered adjacently, and the production parameter combinations in the fitness sequence are grouped according to certain similarity standards (such as fitness value or parameter similarity) to form multiple production parameter clusters. The production parameter combinations in each cluster have certain similarities, representing different optimization directions or local optimal solutions.

[0055] Perform multi-objective balance analysis on multiple production parameter clusters, and judge the pros and cons of each production parameter cluster by analyzing the relationship between multiple objectives (such as cost, quality, etc.). Through balance analysis, it can be ensured that multiple objectives can both reduce costs and improve product performance during the optimization process, avoiding over-optimization of a certain objective that leads to the deterioration of other objectives.

[0056] According to the balance analysis results, multiple production parameter clusters are screened, and the optimization effects of each target are comprehensively considered to select the production parameter cluster that is more balanced among multiple targets. Finally, the optimal production parameter combination (i.e., the i-th production parameter combination) is screened out, and these combinations will be used as optimized production parameters to guide the actual production process.

[0057] Furthermore, step P352 also includes: P352-1: judging whether the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination are within the expected difference range according to the fitness sequence, P352-2: If there are production parameter combinations among the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination that are in the expected difference range, then the production parameter combinations in the expected difference range are extracted for correlation analysis to obtain multiple correlation coefficients and determine the initial cluster center.

[0058] P352-3: Calculate according to the multiple correlation coefficients combined with the production parameter combination in the expected difference range, and define multiple clustering distance information.

[0059] P352-4: Adjacent parameter combination determination is performed based on the plurality of clustering distance information, and production parameter combinations in the expected difference interval are associated and classified according to the initial clustering centers according to the adjacent determination results to generate the plurality of production parameter clusters.

[0060] Specifically, the expected difference interval is a pre-set range used to screen out production parameter combinations with a specific degree of difference in fitness. The setting of this interval can be based on actual needs in the production process, empirical data, or specific requirements for optimization goals, such as a fitness value difference of less than 3. Find a production parameter combination that meets the set expected difference interval in the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination.

[0061] If there are production parameter combinations in the expected difference range, these combinations are enumerated in pairs and correlation analysis is performed. The correlation coefficients between the enumerated production parameter combinations are calculated using the Pearson correlation coefficient or the Spearman rank correlation coefficient to obtain multiple correlation coefficients. According to the correlation coefficient, a production parameter combination with a strong correlation with other combinations is selected as the initial clustering center. This initial clustering center will guide subsequent clustering operations so that similar production parameter combinations can be classified around it.

[0062] According to multiple correlation coefficients combined with production parameter combinations in the expected difference interval, multiple cluster distance information is defined. These cluster distance information are important bases for judging whether the production parameter combinations are adjacent and whether they should be classified into the same category. By defining cluster distance information, the production parameter combinations can be classified more accurately, so that the clustering results are more in line with the actual relationship and optimization requirements in the production process. Exemplarily, the distance between the two pairs of parameter combinations with the highest and lowest correlation in the production parameter combinations in the expected difference interval can be calculated, and then the interval distance can be divided according to the size of the two distances to determine multiple cluster distances. For example, using the Euclidean distance, Manhattan distance, etc., the distance between the two pairs of parameter combinations with the highest and lowest correlation in the production parameter combinations in the expected difference interval is calculated to be 2 and 10 respectively, and the number of clusters is set to 4, then the interval distance is (10-2) / 4=2, and multiple cluster distances are defined as 2, 4, 6, 8, and 10 respectively.

[0063] According to the calculated cluster distance information, the adjacent parameter combination of the production parameter combination is judged, and the production parameter combination in the expected difference range is associated and classified according to the initial cluster center according to the adjacent judgment result to generate multiple production parameter clusters. Through clustering, production parameter combinations with similar characteristics can be classified into one category to form multiple production parameter clusters, so as to carry out targeted processing in the subsequent multi-objective balance analysis.

[0064] Furthermore, step P353 also includes: Step P353-1: Based on the first production parameter combination, the second production parameter combination...the Nth production parameter combination, balance calculation is performed on the multiple production parameter clusters to obtain multiple balance scores, and the multiple balance scores correspond to the multiple production parameter clusters.

[0065] Step P353-2: Preliminary screening is performed on the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination according to the multiple balance scores to obtain a preliminary screening result.

[0066] Step P353-3: Determine whether the number of candidate combinations in the preliminary screening results is greater than or equal to the preset number of parameter groups. If the number of candidate combinations in the preliminary screening results is greater than or equal to the preset number of parameter groups, perform a balance deviation analysis according to the multiple balance scores, perform secondary screening on the candidate combinations in the preliminary screening results according to the balance deviation values, and determine the i-th production parameter combination based on the secondary screening results.

[0067] Step P353-4: If the number of candidate combinations in the initial screening result is less than the preset number of parameter groups, the production parameter combination corresponding to the maximum balance score in the candidate combination is recorded as the i-th production parameter combination for output.

[0068] Specifically, a balance calculation is performed for each production parameter cluster, and multiple balance scores are obtained based on the calculation results. The balance score is a metric used to judge the stability and balance of a production parameter cluster. In production, different production parameters may have different effects on product quality. The purpose of the balance score is to determine the combined effect of these parameter combinations. The balance calculation can be completed by calculating indicators such as the average deviation, fluctuation range, or standard deviation of production results of all production parameter combinations in the cluster.

[0069] Based on the multiple balance scores obtained in the previous step, all production parameter combinations are preliminarily screened. Specifically, a minimum balance score threshold can be set, all parameter combinations can be sorted according to the balance score, and those production parameter combinations with scores higher than this threshold can be selected as candidate parameters. Preliminary screening can narrow the range of candidate parameters, thus preparing for subsequent screening and optimization.

[0070] Determine whether the number of candidate combinations in the initial screening results is greater than or equal to the preset number of parameter groups. The preset number of parameter groups is a pre-set quantitative standard, which can be set based on factors such as computing resources, production requirements, or experience values. If there are many candidate combinations, it means that there is room for further screening. If the number of candidate combinations is greater than or equal to the preset number of parameter groups, balance deviation analysis is performed according to multiple balance scores. Balance deviation analysis is a more detailed analysis of the degree of deviation of the candidate combinations in the initial screening results in terms of multi-objective balance. For example, the balance deviation value between each candidate combination in terms of product quality, production cost and other objectives is calculated. According to the balance deviation value, the candidate combinations in the initial screening results are secondary screened, the number of candidate combinations is reduced to less than the preset number of parameter groups, and then the i-th production parameter combination is determined based on the secondary screening results. Secondary screening is based on the preliminary screening, and further excludes those candidate combinations that have high balance scores but have large deviations in some aspects, so as to determine the i-th production parameter combination that best meets the multi-objective balance requirements.

[0071] If the number of candidate combinations in the initial screening results is less than the preset number of parameter groups, it means that the number of candidate combinations remaining after the initial screening is small. In this case, the production parameter combination corresponding to the maximum balance score in the candidate combination is recorded as the i-th production parameter combination for output to meet the optimization requirements in the production process.

[0072] Furthermore, the product quality prediction module 50 of the embodiment of the present application is also used to perform the following steps: Step P51: traverse the production parameter change trend chart to perform production status analysis and construct a state learning space.

[0073] Step P52: traverse the production parameter change trend chart to perform production action analysis and construct an action learning space.

[0074] Step P53: Calculate based on the multiple product performance indicators and the multiple production optimization parameters to construct a reward function.

[0075] Step P54: Reinforcement learning is performed based on the state learning space, the action learning space and the reward function to obtain the learning effect.

[0076] Specifically, the production state analysis is performed by traversing the production parameter change trend chart. The production state represents the production parameters at a certain moment and the actual situation of the production process. The state learning space is constructed by extracting the historical change data of the production parameters. The state learning space contains the production conditions at each moment, such as temperature, humidity, coating dosage, etc., providing state information for subsequent reinforcement learning.

[0077] Similarly, the production parameter change trend chart is traversed to analyze the production action. Production action refers to the operation or decision that can be performed in the production process. For example, operations such as adjusting the graphene concentration and changing the coating temperature can be regarded as production actions. This includes equipment operation adjustment, raw material addition adjustment, etc. The action learning space is constructed based on the production action analysis results. The action learning space is a space that contains all possible production actions, indicating all possible operations or decisions that may be taken under a certain production state, providing a basis for action selection in reinforcement learning.

[0078] A reward function is calculated based on multiple product performance indicators and multiple production optimization parameters. The reward function is used to measure the effect of taking an action under a specific state. Specifically, the reward function can evaluate the quality of the current decision based on the actual performance of the product. If the action taken leads to an improvement in product quality, the reward function will give a positive reward value; conversely, if the product quality decreases, a negative reward value will be given.

[0079] Reinforcement learning is performed based on the previously constructed state learning space, action learning space, and reward function. After each action is selected, the decision effect is evaluated based on the feedback of the reward function. If the reward obtained is large, the choice of the action is strengthened; if the reward is small, the strategy is adjusted. By continuously adjusting the action, the reinforcement learning algorithm continuously adjusts its strategy to maximize the cumulative value of the reward function and finally obtains the learning effect. This learning effect can reflect how to improve product quality by optimizing production actions under the current production parameters and product performance indicators.

[0080] Furthermore, step P54 also includes: Step P541: Based on the mapping of the multiple production optimization parameters to the state learning space, a reinforcement learning state is obtained, and the action learning space is analyzed according to the reinforcement learning state in combination with the production parameter change trend chart to determine the reinforcement learning action.

[0081] Step P542: Utilize the reward function to calculate the reinforcement learning state in combination with the reinforcement learning action to obtain a state reward value and an action reward value.

[0082] Step P543: Perform a reward growth evaluation based on the state reward value and the action reward value, obtain a reward growth trend, and add the reward growth trend to the learning effect.

[0083] Specifically, a reinforcement learning state is obtained by mapping multiple production optimization parameters to the state learning space. This reinforcement learning state represents the current state of the production process under the control of multiple production optimization parameters. Combined with the production parameter change trend chart, the corresponding action learning space is further analyzed to determine the reinforcement learning action. The production parameter change trend chart provides historical information on parameter changes in the production process. Combined with the reinforcement learning state, the appropriate action can be selected from the action learning space more accurately.

[0084] The reward function is used to evaluate and calculate the effect of executing reinforcement learning actions in the reinforcement learning state, and determine the corresponding state reward value and action reward value. The state reward value and action reward value represent the quality of the current state and the action taken, respectively. If the product performance indicators are improved after executing a reinforcement learning action to adjust the operating parameters of the equipment in a certain reinforcement learning state, then according to the reward function, a higher state reward value and action reward value will be obtained; otherwise, a lower value will be obtained.

[0085] According to the current state reward value and action reward value, the effect of each action is evaluated in the long term, and the state reward value and action reward value are analyzed through reward growth evaluation to see whether and how fast they grow over time. If the reward value gradually increases, it means that the action decision is getting closer to the optimal solution. Otherwise, the decision strategy needs to be adjusted. The reward growth trend is added to the learning effect to reflect the effectiveness and stability of the reinforcement learning algorithm and provide a basis for future action selection. The reward growth trend reflects the changes in the cumulative rewards obtained by selecting different actions over a period of time. If certain action strategies bring continuous positive feedback over a period of time (such as improved antibacterial performance or reduced production costs), the probability of selecting these actions is increased.

[0086] Finally, the learning effect can realize the dynamic adjustment of production control parameters in the production process of graphene antibacterial polyester yarn, ensuring the continuous improvement of product quality and the continuous optimization of production costs.

[0087] In summary, the graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization provided in the embodiments of the present application has the following technical effects: The quality control analysis module 10 obtains the production operation parameter set by sensing the operation of the graphene antibacterial polyester yarn production line, and performs quality control analysis to set product performance indicators, providing basic standards for subsequent optimization and production control. The optimization model construction module 20 analyzes the production operation parameter set and product performance indicators to set cost optimization goals, performance optimization goals, and optimization constraints, and calculates the weight coefficient to construct the objective function, associates various parameters in production with product performance, cost and other factors, and builds a theoretical framework for multi-objective optimization. The process optimization module 30 synchronizes the production operation parameter set to the multi-objective optimization model for process optimization, uses the optimization algorithm to select a better combination from many possible production parameter combinations, obtains multiple production optimization parameters, and realizes the preliminary optimization of real-time process parameters. The operation change analysis module 40 performs operation control monitoring of the graphene antibacterial polyester yarn production line based on multiple production optimization parameters, generates control feedback information and analyzes operation changes, draws a production parameter change trend chart, monitors the production process in real time and feedbacks the production status, and provides data support for product quality prediction. The product quality prediction module 50 performs reinforcement learning by constructing a state learning space, an action learning space and a reward function, and uses the information in the production parameter change trend diagram to deeply analyze the production process to obtain learning effects, thereby laying the foundation for product quality prediction. The control strategy formulation module 60 dynamically adjusts multiple production optimization parameters based on the product quality prediction data set, formulates a production control strategy, and executes the production control strategy to achieve intelligent control of the production of graphene antibacterial polyester yarn, thereby optimizing product quality and production costs.

[0088] In general, the embodiments of the present application use the synergy of each module, the optimization algorithm to find the best combination of production parameters, monitor the changes in the production process in real time and predict the product quality, and finally realize the dynamic intelligent control of the production process. The embodiments of the present application comprehensively consider multiple product performance indicators, make full use of production data, and realize the comprehensive optimization of the production process of graphene antibacterial polyester yarn. It not only improves the stability and consistency of product quality, but also reduces the defective rate and production cost, reduces the waste in the production process, and effectively promotes the production of graphene antibacterial polyester yarn to develop intelligently and efficiently.

[0089] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The production control platform of graphene antibacterial polyester yarn based on intelligent algorithm optimization is characterized by: The platform includes: A quality control analysis module is used to obtain a production operation parameter set by sensing the operation of the production line of graphene antibacterial polyester yarn, perform quality control analysis based on the production operation parameter set, and set multiple product performance indicators; An optimization model building module, used to perform multi-objective optimization according to the multiple product performance indicators combined with the production operation parameter set, establish an objective function, and build a multi-objective optimization model according to the objective function; A process optimization module, used to synchronize the production operation parameter set to the multi-objective optimization model to perform process optimization and obtain multiple production optimization parameters; An operation change analysis module, used to perform operation control monitoring on the production line of graphene antibacterial polyester yarn based on the multiple production optimization parameters, generate control feedback information, perform operation change analysis of the production line according to the control feedback information, and draw a production parameter change trend chart; A product quality prediction module, used to perform reinforcement learning according to the production parameter change trend diagram to obtain a learning effect, perform product quality prediction on the multiple production optimization parameters according to the learning effect, and generate a product quality prediction data set; A control strategy formulation module is used to dynamically adjust the multiple production optimization parameters based on the product quality prediction data set, formulate a production control strategy, and execute the production control strategy to intelligently control the production of graphene antibacterial polyester yarn.

2. The graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization as claimed in claim 1, characterized in that: The optimization model building module is used to perform multi-objective optimization according to the multiple product performance indicators combined with the production operation parameter set to establish an objective function, and the execution steps include: Parsing the production operation parameter set to obtain graphene concentration parameters, coating temperature parameters, and energy consumption parameters, performing cost analysis based on the graphene concentration parameters, the coating temperature parameters, and the energy consumption parameters, and setting a cost optimization target; Analyze the multiple product performance indicators to obtain an antibacterial rate indicator and a tensile strength indicator, perform performance analysis based on the antibacterial rate indicator and the tensile strength indicator, and set a performance optimization goal; Calculating according to the performance standard based on the graphene concentration parameter in combination with the multiple product performance indicators to determine the graphene distribution uniformity deviation value, and setting optimization constraints based on the graphene distribution uniformity deviation value; Performing weight calculation on the graphene concentration parameter, the coating temperature parameter, the energy consumption parameter, the antibacterial rate index, and the tensile strength index to obtain a plurality of weight coefficients; According to the optimization constraints, the multiple weight coefficients are matched with the performance optimization target and the cost optimization target to construct the objective function.

3. The graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization as claimed in claim 1, characterized in that: The process optimization module is used to synchronize the production operation parameter set to the multi-objective optimization model for process optimization to obtain multiple production optimization parameters. The execution steps include: Synchronizing the production operation parameter set to the multi-objective optimization model, and generating an initial population by performing random combinations based on the production operation parameter set; Introducing a fitness function, calculating the initial population by using the fitness function to obtain multiple fitnesses, wherein the multiple fitnesses have a corresponding relationship with the initial population; The initial population is sorted in descending order according to the multiple fitnesses to generate a production parameter sequence, and extraction is performed multiple times according to the production parameter sequence until a population with a fitness less than a preset threshold exists, thereby obtaining multiple production parameter combinations; Extracting a first-order production parameter combination and a second-order production parameter combination based on the multiple production parameter combinations, performing a cross operation on the first-order production parameter combination and the second-order production parameter combination to generate a first production parameter combination, and iterating until the multiple production parameter combinations are traversed to generate a second production parameter combination... an Nth production parameter combination, where N is an integer greater than 1; Perform fitness analysis based on the first production parameter combination, the second production parameter combination...the Nth production parameter combination, determine the ith production parameter combination according to the fitness analysis result, and add the ith production parameter combination to the multiple production optimization parameters, where i is an integer greater than 0.

4. The graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization as claimed in claim 3, characterized in that: Based on the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination, fitness analysis is performed, the i-th production parameter combination is determined according to the fitness analysis result, and the i-th production parameter combination is added to the multiple production optimization parameters. The execution steps include: Sorting the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination in descending order according to the fitness analysis result to determine a fitness sequence; performing adjacent clustering on the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination according to the fitness sequence to generate a plurality of production parameter clusters; A multi-objective balance analysis is performed based on the multiple production parameter clusters, and the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination are screened based on the balance analysis results to determine the i-th production parameter combination.

5. The graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization as claimed in claim 4, characterized in that: The first production parameter combination, the second production parameter combination, ... the Nth production parameter combination are clustered according to the fitness sequence to generate a plurality of production parameter clusters, and the execution steps include: According to the fitness sequence, it is determined whether the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination exist in the expected difference range. If there is a production parameter combination in the expected difference range among the first production parameter combination, the second production parameter combination ... the Nth production parameter combination, the production parameter combination in the expected difference range is extracted for correlation analysis to obtain multiple correlation coefficients, and an initial clustering center is determined; Calculating according to the multiple correlation coefficients combined with the production parameter combinations in the expected difference range, and defining multiple cluster distance information; Adjacent parameter combination determination is performed based on the plurality of cluster distance information, and production parameter combinations in the expected difference interval are associated and classified according to the initial cluster centers according to the adjacent determination results to generate the plurality of production parameter clusters.

6. The graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization as claimed in claim 4, characterized in that: Performing a multi-objective balance analysis according to the multiple production parameter clusters, screening the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination according to the balance analysis result, and determining the i-th production parameter combination, the execution steps include: Based on the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination, balance calculation is performed on the multiple production parameter clusters to obtain multiple balance scores, where the multiple balance scores correspond to the multiple production parameter clusters; Preliminarily screening the first production parameter combination, the second production parameter combination, ... the Nth production parameter combination according to the multiple balance scores to obtain a preliminary screening result; Determine whether the number of candidate combinations in the preliminary screening results is greater than or equal to the preset number of parameter groups, if the number of candidate combinations in the preliminary screening results is greater than or equal to the preset number of parameter groups, perform a balance deviation analysis according to the multiple balance scores, perform secondary screening on the candidate combinations in the preliminary screening results according to the balance deviation values, and determine the i-th production parameter combination according to the secondary screening results; If the number of candidate combinations in the initial screening result is less than the preset number of parameter groups, the production parameter combination corresponding to the maximum balance score in the candidate combination is recorded as the i-th production parameter combination for output.

7. The graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization as claimed in claim 1, characterized in that: The product quality prediction module is used to perform reinforcement learning according to the production parameter change trend diagram to obtain a learning effect, and the execution steps include: Traversing the production parameter change trend graph to perform production status analysis and construct a state learning space; Traversing the production parameter change trend graph to perform production action analysis and construct an action learning space; Calculating according to the multiple product performance indicators and the multiple production optimization parameters to construct a reward function; Reinforcement learning is performed based on the state learning space, the action learning space and the reward function to obtain the learning effect.

8. The graphene antibacterial polyester yarn production control platform based on intelligent algorithm optimization as claimed in claim 7, characterized in that: Reinforcement learning is performed based on the state learning space, the action learning space and the reward function to obtain the learning effect. The execution steps include: Based on mapping the multiple production optimization parameters to the state learning space, a reinforcement learning state is obtained, and according to the reinforcement learning state and the production parameter change trend diagram, the action learning space is analyzed to determine the reinforcement learning action; Calculating the reinforcement learning state in combination with the reinforcement learning action using the reward function to obtain a state reward value and an action reward value; A reward growth evaluation is performed based on the state reward value and the action reward value to obtain a reward growth trend, and the reward growth trend is added to the learning effect.

Images