Healthcare product production control method and system based on twinborn simulation and medium

Abstract

The invention discloses a health care product production control method and system based on twinborn simulation and a medium, and relates to the related field of industrial process control, and the method comprises the steps: carrying out the twinborn simulation of technological parameters and production quality on a production line of a target health care product, and constructing a twinborn production model; configuring a raw material detection network at a raw material input end of the production line, wherein the network specifically comprises effective component detection equipment and a detection mechanism; and performing dynamic detection on effective components of the raw materials through the raw material detection network, inputting a detection result and current process parameters of the production line into the twin production model for production quality simulation, and when a production quality index does not meet a preset quality range, adjusting and optimizing the current process parameters by using the twin production model. The technical problems that production factor changes cannot be mastered in real time, technological parameters are difficult to accurately adjust and quality defects cannot be effectively prevented in existing health care product production control are solved, and the technical effects of accurately controlling the production process in real time and improving the product quality stability are achieved.

Description

Technical Field

[0001] This application relates to the field of industrial process control, and in particular to a method, system and medium for controlling the production of health products based on twin simulation. Background Technology

[0002] In the health supplement manufacturing industry, ensuring stable production quality and compliance with high standards is crucial for protecting consumer health, enhancing corporate market competitiveness, and promoting the healthy development of the entire industry. Currently, the main methods for addressing quality assurance in health supplement production involve adhering to predetermined process parameters during production and conducting random sampling inspections of finished products after production. The results are used to determine if the product quality meets standards; if not, the entire batch is processed, and production parameters are adjusted retrospectively. However, this current method, relying primarily on post-production sampling, struggles to comprehensively and in real-time monitor changes in various factors affecting product quality during production. For example, fluctuations in the active ingredients of raw materials cannot be promptly reported to the production stage. Consequently, by the time quality issues are discovered, a significant number of substandard products have already been produced, leading to raw material waste, reduced production efficiency, and difficulty in accurately pinpointing the root cause for precise adjustments to process parameters, thus failing to effectively prevent recurrence of quality problems.

[0003] At present, the production control of health products faces technical problems such as the inability to monitor changes in production factors in real time, difficulty in accurately adjusting process parameters, and inability to effectively prevent quality defects. Summary of the Invention

[0004] This application provides a method, system, and medium for controlling the production of health products based on twin simulation. It employs twin simulation of process parameters and production quality for a target health product production line to construct a twin production model. A raw material detection network is set up at the raw material input end of the production line. This network includes active ingredient detection equipment and corresponding detection mechanisms. The active ingredients in the raw materials are dynamically detected using the raw material detection network. The detection results, along with the current process parameters of the production line, are input into the twin production model for production quality simulation. If the production quality indicators do not meet the preset quality range, the model is used to optimize the current process parameters. These technical means solve the technical problems of existing health product production control, such as the inability to monitor changes in production factors in real time, the difficulty in accurately adjusting process parameters, and the inability to effectively prevent quality defects. This achieves the technical effect of real-time and accurate control of the production process and improved product quality stability.

[0005] This application provides a method for controlling the production of health products based on twin simulation, including: performing twin simulation of process parameters and production quality of the production line of the target health product to construct a twin production model; configuring a raw material detection network at the raw material input end of the production line, specifically including effective ingredient detection equipment and detection mechanism; performing dynamic detection of the effective ingredients of the raw materials through the raw material detection network, inputting the detection results and the current process parameters of the production line into the twin production model for production quality simulation, and optimizing the current process parameters using the twin production model when the production quality indicators do not meet the preset quality range.

[0006] In a possible implementation, a raw material detection network is configured at the raw material input end of the production line, specifically including effective component detection equipment and a detection mechanism, and the following processes are performed: collecting multiple raw materials at the raw material input end, and multiple sets of effective components corresponding to each of the multiple raw materials; collecting the storage area of ​​the multiple raw materials upon entering the raw material input end, and performing effective component attenuation analysis in combination with the multiple sets of effective components to establish a first detection trigger mechanism based on component attenuation values; configuring the effective component detection equipment with the multiple sets of effective components, and configuring the detection mechanism with the first detection trigger mechanism.

[0007] In a possible implementation, the following processing is performed: a supply batch traceability channel for the various raw materials is constructed, a second detection triggering mechanism is configured with a process batch change node, and the second detection triggering mechanism is added to the detection mechanism.

[0008] In a possible implementation, the storage area of ​​the various raw materials upon entering the raw material input end is collected, and the decay analysis of the effective components is performed in conjunction with the multiple sets of effective components. A first detection trigger mechanism based on the component decay value is established, and the following processing is performed: collecting any storage environment information of any storage area of ​​any raw material; extracting each effective component from any set of effective components of any raw material; analyzing the content decay curves of each effective component over time based on the storage environment information; and configuring the first detection trigger mechanism according to a preset decay threshold based on the content decay curves.

[0009] In a possible implementation, when the production quality index does not meet the preset quality range, the twin production model is used to optimize the current process parameters by performing the following processing: using the content of effective components in raw materials as variables and production quality as a quantifier, several sets of process parameters are collected and analyzed to determine the relevant parameter set between the process parameter type and the content of effective components; when the production quality index is greater than the maximum boundary value of the preset quality range, the difference between the production quality index and the median value of the preset quality range is calculated to generate a quality reduction index; starting from the current process parameters, with the quality reduction index as the target, and with the relevant parameter set as the adjustment object, iterative simulation is performed to generate parameter optimization results.

[0010] In a possible implementation, the following processing is performed: when the production quality index is less than the minimum boundary value of the preset quality range, the difference between the median of the preset quality range and the production quality index is calculated to generate a quality improvement index; taking the current process parameters as the starting point, the quality improvement index as the target, and the relevant parameter set as the adjustment object, iterative simulation is performed to generate parameter optimization results.

[0011] In one possible implementation, a twin simulation of the process parameters and production quality of the target health product production line is performed to construct a twin production model. The following processes are then performed: physical modeling of each production node on the target health product production line is collected to construct a twin production line; continuous production parameters and quality inspection samples of each production node are collected, and the twin production line is simulated and trained to obtain a converged twin production model.

[0012] In a possible implementation, the following processing is performed: the raw material input end of the production line also includes a raw material property detection network; the correlation between the physical properties of the various raw materials and the process parameters of the relevant process nodes is analyzed, and a raw material property correlation network is trained; the raw material property parameters are detected in real time through the raw material property detection network, and the process parameters of the relevant process nodes are optimized based on the raw material property correlation network.

[0013] This application also provides a health product production control system based on twin simulation, including: a twin simulation module for performing twin simulation of process parameters and production quality of the target health product production line, and constructing a twin production model; a raw material detection network configuration module for configuring a raw material detection network at the raw material input end of the production line, specifically including effective ingredient detection equipment and detection mechanism; and a process parameter optimization module for dynamically detecting the effective ingredients of raw materials through the raw material detection network, inputting the detection results and the current process parameters of the production line into the twin production model for production quality simulation, and optimizing the current process parameters using the twin production model when the production quality indicators do not meet the preset quality range.

[0014] This application also provides a computer-readable storage medium, including: a computer program stored thereon, which, when executed by a processor, implements a health product production control method based on twin simulation.

[0015] The proposed method, system, and medium for controlling the production of health supplements based on twin simulation firstly involves twin simulation of the process parameters and production quality of the target health supplement production line, constructing a twin production model. Then, a raw material detection network is configured at the raw material input end of the production line, specifically including active ingredient detection equipment and mechanisms. Finally, the active ingredients of the raw materials are dynamically detected through the raw material detection network. The detection results, along with the current process parameters of the production line, are input into the twin production model for production quality simulation. When the production quality indicators do not meet the preset quality range, the twin production model is used to optimize the current process parameters. This achieves the technical effect of real-time and precise control of the production process and improved product quality stability. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 A schematic flowchart of a health product production control method based on twin simulation provided in this application embodiment.

[0018] Figure 2 This is a schematic diagram of the structure of a health product production control system based on twin simulation, provided in an embodiment of this application.

[0019] Figure labeling: Twin simulation module 10, raw material detection network configuration module 20, process parameter optimization module 30. Detailed Implementation

[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. 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 explicitly listed, but may include other steps or modules not explicitly 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 commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0023] This application provides a method for controlling the production of health products based on twin simulation, such as... Figure 1 As shown, the method includes:

[0024] Step S100: Perform twin simulation of process parameters and production quality of the target health product production line to construct a twin production model.

[0025] Specifically, computer simulation technology is used, combined with information such as the actual process flow and equipment parameters of the target health supplement production line, to simulate and model the relationship between process parameters and production quality. Twin simulation refers to creating a digital model in virtual space that corresponds to the real physical system, simulating the behavior and characteristics of the physical system to achieve real-time monitoring, prediction, and optimization. In this application, twin simulation is used to simulate the relationship between process parameters and production quality of the health supplement production line. By collecting historical production data from the production line, including production quality indicators such as product purity, active ingredient content, and stability under different process parameter settings, machine learning algorithms are used to train and analyze this data to construct a twin production model that accurately reflects the mapping relationship between process parameters and production quality. This model can simulate the production situation of the production line under different process parameters in real time and predict production quality indicators.

[0026] In one possible implementation, a twin simulation of the process parameters and production quality of the target health product production line is performed to construct a twin production model. Step S100 further includes step S110, which involves collecting data on each production node on the target health product production line for physical modeling to construct a twin production line. Specifically, the production line of the target health product is analyzed, and each production node is identified based on the logical sequence and functional characteristics of the production process. For example, a vitamin health product production line may include raw material storage nodes, raw material pretreatment nodes, mixing nodes, granulation nodes, drying nodes, tableting nodes, and packaging nodes. For each production node, relevant physical parameters are collected, including equipment dimensions, equipment materials, equipment operating speed range, and temperature control range. Data can be collected by consulting equipment manuals, on-site measurements, and communication with equipment suppliers. Using 3D modeling software, each production node is 3D modeled based on the collected physical parameters. The 3D models of each production node are assembled and built in a virtual environment according to the layout and connection relationships of the actual production line, constructing a twin production line that is completely consistent with the actual production line in terms of geometry, spatial location, and connection method. For example, a 3D model of the raw material storage node can be placed at the starting position of the virtual production line, and then connected to the raw material pretreatment node through a conveyor belt model to complete the virtual construction of the entire production line.

[0027] Step S120 involves collecting continuous production parameters and quality inspection samples from each production node, and performing simulation training on the twin production line to obtain a converged twin production model. Specifically, in the actual production process, various sensors and data acquisition systems are used to collect production parameters from each production node in real time and continuously. For example, torque sensors are installed at the mixing node to collect torque change data during the mixing process, reflecting the uniformity of material mixing; temperature and humidity sensors are installed at the drying node to collect changes in drying temperature and material humidity, respectively; and pressure sensors are installed at the tableting node to collect pressure data during the tableting process, etc. These continuous production parameters reflect the actual operating status of the production process.

[0028] Quality inspection samples are collected from the production line according to a certain sampling frequency and method. For example, samples are collected at certain intervals after the tableting stage. The collected samples are then subjected to quality testing using equipment such as high-performance liquid chromatography (HPLC) and laser particle size analyzer. The test indicators include active ingredient content, particle size distribution, tablet weight variation, hardness, and solubility, and the test results are recorded.

[0029] The collected continuous production parameters and quality inspection results are used as input data and fed into the twin production line constructed in step S110 for simulation training. Using machine learning algorithms, the twin production line continuously learns and adjusts the model parameters to make the model's output as close as possible to the actual collected quality inspection results. Through multiple iterations of training, when the model's prediction error gradually decreases and reaches a pre-set threshold, or when the model's performance indicators no longer significantly improve, the model is considered to have converged. At this point, the twin production model can reflect the relationship between process parameters and production quality.

[0030] This approach involves constructing a twin production line that mirrors the actual production line and collecting a large number of real continuous production parameters and quality inspection samples for simulation training. The twin production model can accurately simulate various physical and chemical changes during the production process, thereby precisely predicting production quality indicators.

[0031] Step S200: Configure a raw material testing network at the raw material input end of the production line, specifically including effective ingredient testing equipment and testing mechanisms.

[0032] Specifically, the raw material testing network is a system composed of active ingredient detection equipment and a detection mechanism, used to detect the active ingredients in raw materials fed into the production line. Specifically, active ingredient detection equipment, such as high-performance liquid chromatographs (HPLC) and gas chromatographs (GC), is installed at the raw material input end of the production line. This equipment can analyze and quantify the active ingredients in the raw materials. Simultaneously, a detection mechanism is established, including sampling methods, detection frequency, and data recording and transmission. Sampling methods ensure that the collected raw material samples are representative and accurately reflect the quality status of the entire batch of raw materials. The detection frequency is determined based on production needs and the stability of raw material quality; for example, a comprehensive test is performed at the start of each batch of raw materials, or samples are taken periodically during production. Data recording and transmission are carried out through an automated data acquisition system, which records the data output from the detection equipment and transmits it to the central control system.

[0033] In one possible implementation, a raw material testing network is configured at the raw material input end of the production line, specifically including active ingredient detection equipment and mechanisms. Step S200 further includes step S210, collecting multiple raw materials at the raw material input end, as well as multiple sets of active ingredients corresponding to each raw material. Specifically, at the raw material input end of the health supplement production line, raw materials are collected according to a predetermined sampling plan. For example, for the same vitamin raw material purchased from different suppliers, samples are collected from different locations in the raw material piles of each supplier; for raw materials from the same batch but stored in different areas, multiple sampling points are also conducted. Based on the health supplement formula and production requirements, the active ingredients contained in each raw material are obtained by reviewing the raw material quality inspection reports, relevant literature, and communicating with raw material suppliers. For example, for a health supplement with ginseng as the main raw material, the active ingredients in ginseng include ginsenosides, ginseng polysaccharides, etc., and these active ingredients are combined into a set of active ingredients for the ginseng raw material. The same operation is performed on each collected raw material, ultimately obtaining multiple sets of active ingredients corresponding to each raw material.

[0034] Step S220 involves collecting data from the storage area where the various raw materials enter the raw material input end, performing attenuation analysis of the effective components based on the multiple sets of effective components, and establishing a first detection trigger mechanism based on the component attenuation values. Specifically, various environmental monitoring devices, such as temperature and humidity sensors and light sensors, are installed in the raw material storage area to monitor parameters such as temperature, humidity, and light intensity in the storage environment in real time. Simultaneously, raw material samples are periodically collected from the storage area, and the content of each effective component in the samples is detected according to the effective component detection method determined in step S210. The time of each collection, storage environment parameters, and effective component content data are recorded.

[0035] Using the collected data and data analysis methods, such as time series analysis, the variation patterns of the content of effective components in each raw material over time were analyzed. Taking ginseng as an example, the changes in the content of ginsenosides and ginseng polysaccharides with storage time under different temperature, humidity, and light conditions were analyzed, and the decay rate of each effective component was calculated. For example, the analysis found that under conditions of 25℃, 60% relative humidity, and 500 lux light intensity, the monthly decay rate of ginsenosides was 2%.

[0036] Based on the results of the effective component decay analysis, a first detection trigger mechanism based on the component decay value is established. For example, it is stipulated that when the content decay value of a certain effective component in a raw material reaches 5% of the initial content, a second detection of the raw material is triggered. At the same time, combined with storage environment parameters, if the storage environment exceeds the preset suitable range, detection is also triggered to ensure that the quality of the raw material is effectively monitored during storage.

[0037] Step S230: Configure an effective component detection device with the multiple sets of effective components, and configure the detection mechanism with the first detection triggering mechanism. Specifically, based on the multiple sets of effective components of various raw materials determined in step S210, select and configure an effective component detection device. For example, for detecting ginsenosides in ginseng, a high-performance liquid chromatograph (HPLC) can be selected, which can separate and quantitatively analyze the various components of ginsenosides. According to the detection requirements of different effective components, set the parameters of the detection device, such as the selection of the chromatographic column, the ratio of the mobile phase, and the setting of the detection wavelength, to ensure that the device can accurately detect the effective components in each raw material.

[0038] The first detection triggering mechanism established in step S220 is integrated into the control system of the raw material detection network. The control system receives data from the environmental monitoring equipment and the detection results from the effective component detection equipment in real time, and makes a judgment based on the first detection triggering mechanism. When the triggering conditions are met, the control system automatically issues a detection command to start the effective component detection equipment to re-detect the corresponding raw material.

[0039] This approach, by establishing a detection trigger mechanism based on component decay values, can monitor the changes in the effective components of raw materials during storage in real time, promptly detect fluctuations in raw material quality, and take corresponding measures to ensure the consistency of the final health product quality.

[0040] In one possible implementation, step S200 further includes step S240, constructing a supply batch traceability channel for the various raw materials, configuring a second detection triggering mechanism with a process batch change node, and adding the second detection triggering mechanism to the detection mechanism.

[0041] Specifically, at the very beginning of the raw material production line, each batch is assigned a unique batch identification code. This code can be presented in the form of a barcode, QR code, or RFID tag. Throughout the procurement, transportation, and warehousing stages of the raw materials, corresponding data acquisition equipment records the batch identification code along with the corresponding time, location, and operator information. Simultaneously, the raw material quality inspection reports, supplier information, and other relevant data are linked to this batch identification code and stored in the company's database, thus integrating data related to the raw material supply batches.

[0042] Based on the integrated data, a digital supply batch traceability system is built. This system has data query, analysis and visualization functions. By entering the batch identification code, information such as the source of the raw materials, purchase time, transportation process, warehousing and storage status, and the results of previous quality inspections can be found.

[0043] During the raw material input process, there are several critical batch change nodes. For example, when the same raw material from different suppliers is mixed, the properties of the mixed raw material will change due to differences in quality and composition. This is an important batch change node. Based on the identified batch change nodes in the raw material input process, a corresponding secondary detection trigger mechanism is set up. That is, when production reaches these nodes, the effective components of the raw material are detected.

[0044] The established supply batch traceability system and the set second detection trigger mechanism are integrated with the raw material detection network control system at the raw material input end. Data interaction and signal transmission between the traceability system, detection equipment, and production control system are achieved through corresponding program code and development interfaces. When production reaches the process batch change node at the raw material input stage, the production control system sends a specific signal to the detection network control system to trigger the second detection trigger mechanism, which then detects the effective components of the raw materials. This implementation, combined with the configuration of the second detection trigger mechanism at the process batch change node, can promptly detect changes in raw material quality during the input process.

[0045] In one possible implementation, data is collected from the storage areas of the various raw materials entering the raw material input end. Attenuation analysis of the effective components is performed using the multiple sets of effective components, and a first detection trigger mechanism based on the component attenuation value is established. Step S220 further includes step S221, collecting storage environment information from any storage area of ​​any raw material. Specifically, various types of sensors are installed in each storage area at the raw material input end to comprehensively collect storage environment information. For example, in a warehouse storing Chinese medicinal materials, temperature and humidity sensors are installed to monitor temperature and humidity changes in real time; light sensors are installed to detect light intensity, as some Chinese medicinal materials are sensitive to light, and excessive light may cause decomposition of effective components; gas sensors are installed to monitor the concentration of gases such as oxygen and carbon dioxide in the warehouse, as some raw materials may undergo oxidation or fermentation reactions under specific gas environments. These sensors continuously collect storage environment information at set time intervals and transmit the data to the enterprise's central control system or data management platform via wired or wireless communication.

[0046] Step S222: Extract each effective component from any set of effective components of any raw material. Specifically, when it is necessary to analyze a batch of raw materials, extract the set of effective components corresponding to that raw material. For example, when processing a batch of newly received ginseng raw materials, retrieve the effective component information of ginseng to determine the effective components that need to be focused on and analyzed.

[0047] Step S223: Analyze the decay curves of the content of each active ingredient over time based on the storage environment information. Specifically, through experimental research, data on the changes in the content of various active ingredients of raw materials over time under different storage environments are accumulated. Mathematical models are established using this data to describe the relationship between the content of active ingredients and storage environment factors and time. For example, by conducting storage experiments on ginseng under different temperature and humidity conditions and periodically detecting the content of ginsenosides, a mathematical model of the changes in ginsenoside content with temperature, humidity, and time is established. Based on the real-time storage environment information collected in step S221, combined with the mathematical model established above, the decay curve of the content of each active ingredient over time under the current storage environment is analyzed.

[0048] Step S224: Based on the content decay curves, configure the first detection trigger mechanism according to a preset decay threshold. Specifically, a reasonable decay threshold is set for each active ingredient based on factors such as raw material quality standards, production process requirements, and product shelf life. For example, for ginsenosides in ginseng, the decay threshold is set to ensure that the content does not fall below 80% of the initial content during storage. Setting a preset decay threshold ensures that the raw material meets product quality requirements in subsequent production while avoiding overly strict thresholds that lead to frequent detection and increased production costs. Based on the content decay curves of each active ingredient obtained in step S223, monitor the decay of the active ingredients in real time. When the decay value of any active ingredient reaches the preset threshold, trigger the detection device to determine whether the raw material still meets production requirements.

[0049] Compared with the traditional periodic testing method, this approach can flexibly trigger testing based on the actual quality degradation of the raw materials. Testing is only triggered when there is a possibility of problems with the quality of the raw materials, thus reducing the number of tests and improving the utilization efficiency of testing resources.

[0050] Step S300: Dynamically detect the effective components of the raw materials through the raw material detection network, input the detection results and the current process parameters of the production line into the twin production model for production quality simulation, and optimize the current process parameters using the twin production model when the production quality indicators do not meet the preset quality range.

[0051] Specifically, an established raw material testing network is used to dynamically detect the effective components of raw materials according to a set detection mechanism. The testing equipment transmits the detected effective component data, along with current process parameter data obtained from the production line control system, to the twin production model via a data interface. The twin production model performs production quality simulation calculations based on the input data, deriving predicted values ​​for production quality indicators under the current production conditions. These predicted values ​​are compared with a preset quality range, which is determined in advance based on product quality standards and production process requirements. If the predicted value does not meet the preset quality range, the twin production model uses its built-in optimization algorithm to adjust and optimize the current process parameters, finding a combination of process parameters that allows the production quality indicators to meet the preset range. The optimized process parameters are then fed back to the production line control system for real-time adjustment and control of the production process.

[0052] In one possible implementation, when the production quality indicators do not meet the preset quality range, the twin production model is used to optimize the current process parameters. Step S300 further includes step S310, which involves collecting several sets of process parameters and analyzing the correlation set of process parameter types and effective component content, using the content of effective components in the raw materials as a variable and production quality as a quantifier. Specifically, while keeping other conditions relatively stable, the content of effective components in the raw materials is treated as a variable, and production quality is set as a quantifier, i.e., the product quality that meets the preset quality range is used as the target reference. For different effective component contents in the raw materials, multiple sets of corresponding process parameter data are collected. For example, when producing a health product using Chinese medicinal herbs as raw materials, the effective component content of the same Chinese medicinal herb from different origins and with different growth years will vary. For each type of Chinese medicinal herb, production is carried out according to the same production process, but some key process parameters, such as extraction temperature, extraction time, and concentration factor, are adjusted, and the corresponding process parameter values ​​for each set of effective component contents are recorded.

[0053] Multiple sets of collected process parameter data were analyzed to identify the relationship between process parameter types and the content of active ingredients. Statistical methods, such as correlation analysis and regression analysis, were used to determine which process parameters have a significant impact on the content of active ingredients in raw materials, as well as the direction and extent of this impact. For example, analysis revealed a positive correlation between extraction temperature and the extraction rate of a certain active ingredient in traditional Chinese medicine; that is, the higher the extraction temperature, the higher the extraction rate of that active ingredient. Extraction time affects the content of active ingredients within a certain range, but the effect becomes insignificant beyond a certain time. These significantly influential process parameters and their relationships with the content of active ingredients were compiled into a set of relevant parameters to provide a basis for optimizing process parameters.

[0054] Step S320: When the production quality index exceeds the maximum boundary value of the preset quality range, calculate the difference between the production quality index and the median value of the preset quality range to generate a quality reduction index. Specifically, monitor the predicted value of the production quality index in real time and compare it with the preset quality range. When the predicted value of the production quality index is found to be greater than the maximum boundary value of the preset range, it indicates that the production quality is abnormally high, which may indicate over-processing, resource waste, or potential impact on product quality stability. To adjust the production quality index to a reasonable range, calculate the difference between the predicted value of the production quality index and the median value of the preset quality range. The median value of the preset quality range is the middle value of the range. The calculated quality reduction index reflects the degree to which the production quality index needs to be reduced and is the target for process parameter optimization, used to reduce the production quality index to a reasonable level.

[0055] Step S330: Starting with the current process parameters, targeting the quality reduction index, and using the relevant parameter set as the adjustment object, iterative simulation is performed to generate parameter optimization results. Specifically, the currently used process parameters are used as the initial parameters of the twin production model for initialization, ensuring the model is in the same state as actual production. Using the quality reduction index as the target and the relevant parameter set as the adjustment object, iterative simulation is performed in the twin production model. In each iteration, the current process parameters are fine-tuned based on the relationship between process parameters and effective ingredient content in the relevant parameter set. The adjusted process parameters are then input into the twin production model to simulate the production process and calculate new production quality indicators. The new production quality indicators are compared with a preset quality range. If the difference is still significant, the process parameters are adjusted again, and the next round of iterative simulation is performed until the production quality indicators meet the preset quality range. After multiple iterative simulations, a set of process parameters that allows the production quality indicators to meet the requirements is obtained; this set of parameters is the parameter optimization result.

[0056] In one possible implementation, step S300 further includes step S340, where when the production quality index is less than the minimum boundary value of the preset quality range, the difference between the median of the preset quality range and the production quality index is calculated to generate a quality improvement index; step S350, taking the current process parameters as the starting point, the quality improvement index as the target, and the relevant parameter set as the adjustment object, an iterative simulation is performed to generate parameter optimization results.

[0057] Specifically, this implementation method addresses situations where the production quality index is below the minimum boundary value of a preset quality range. Similar to steps 320-330, the difference between the median value of the preset quality range and the predicted value of the production quality index is first calculated to generate a quality improvement index, which reflects the degree to which production quality needs to be improved. Then, starting with the current process parameters and targeting the quality improvement index, and using the previously analyzed set of process parameter types and effective ingredient content-related parameters as adjustment objects, iterative simulations are performed in a twin production model. Finally, parameter optimization results that enable the production quality index to meet the requirements are generated to guide production adjustments.

[0058] In one possible implementation, the method further includes: the raw material input end of the production line also includes a raw material property detection network; analyzing the correlation between the physical properties of the various raw materials and the process parameters of relevant process nodes, and training a raw material property correlation network; detecting raw material property parameters in real time through the raw material property detection network, and optimizing the process parameters of relevant process nodes based on the raw material property correlation network.

[0059] Specifically, the newly added raw material property detection network is dedicated to detecting the physical properties of raw materials, including key indicators such as particle size, flowability, moisture content, and density. For example, in particle size detection, a laser particle size analyzer is used. A laser beam is emitted to illuminate the raw material particles, and by analyzing the angle and intensity distribution of the scattered light, the size distribution of the raw material particles is measured, thus obtaining detailed particle size data. For flowability detection, a Hall effect flowmeter is used. This instrument assesses the flowability of the raw material by measuring the flow rate through a small orifice within a specific time period. Moisture detection uses a high-precision capacitive humidity sensor, which can quickly and accurately detect the moisture content in the raw material. Density detection employs a hydrometer based on the buoyancy principle. The raw material sample is placed in a liquid of known density, and its density is calculated based on the sample's floating or sinking state in the liquid.

[0060] A raw material property detection network is deployed at the raw material input end of the production line to ensure timely acquisition of property information at the initial stage of the production process. The various detection devices are connected via high-speed data transmission lines, forming an integrated system that transmits detected data to the data processing and analysis system.

[0061] A large amount of historical production data was collected, including process parameters at relevant process nodes under different combinations of raw material properties, as well as the quality indicators of the final product. Statistical methods and data mining techniques were used to analyze the intrinsic relationships and correlations between raw material properties and process parameters. For example, correlation analysis revealed that when the raw material particle size is small, the mixing time and stirring speed need to be appropriately extended in the mixing process to ensure that the raw materials are fully and uniformly mixed; while when the raw material has poor flowability, the molding pressure and mold temperature need to be adjusted in the molding process to avoid molding defects.

[0062] Based on the correlation analysis results, a raw material attribute correlation network was constructed. This network employs a neural network algorithm, using raw material attributes as input variables and process parameters of relevant process nodes as output variables. Historical data was collected for network training, and by continuously adjusting the network's weights and bias parameters, the network was able to accurately predict the optimal process parameters for different combinations of raw material attributes. During training, techniques such as cross-validation and early stopping were used to prevent overfitting and improve the network's generalization ability and prediction accuracy.

[0063] During the production process, the raw material property detection network detects the particle size, flowability, moisture content, and density of the raw materials, and transmits this data to the raw material property-related network. Simultaneously, the existing raw material detection network continues to detect the content of active ingredients in the raw materials and transmits the active ingredient data to the subsequent systems.

[0064] After receiving raw material attribute data, the raw material attribute correlation network combines it with the trained model to calculate the optimized process parameters for relevant process nodes. These optimized parameters are promptly fed back to the production line control system, which automatically adjusts the operating status of each process device based on these parameters to optimize the process parameters.

[0065] This application employs a twin simulation of process parameters and production quality for a target health supplement production line to construct a twin production model. A raw material detection network is set up at the raw material input end of the production line. This network includes active ingredient detection equipment and corresponding detection mechanisms. The active ingredients in the raw materials are dynamically detected using the raw material detection network. The detection results, along with the current process parameters of the production line, are input into the twin production model for production quality simulation. If the production quality indicators do not meet the preset quality range, the model is used to optimize the current process parameters. These techniques solve the technical problems of existing health supplement production control, such as the inability to monitor changes in production factors in real time, the difficulty in accurately adjusting process parameters, and the inability to effectively prevent quality defects. This achieves the technical effect of real-time and accurate control of the production process and improved product quality stability.

[0066] In the above text, refer to Figure 1 A method for controlling the production of health supplements based on twin simulation, according to an embodiment of the present invention, is described in detail. Next, reference will be made to... Figure 2 A health product production control system based on twin simulation according to an embodiment of the present invention is described.

[0067] The twin simulation-based health product production control system according to embodiments of the present invention addresses the technical problems of existing health product production control systems, such as the inability to monitor changes in production factors in real time, difficulty in accurately adjusting process parameters, and inability to effectively prevent quality defects. It achieves the technical effect of real-time and precise control of the production process and improved product quality stability. The twin simulation-based health product production control system includes: a twin simulation module 10, a raw material detection network configuration module 20, and a process parameter optimization module 30.

[0068] The twin simulation module 10 is used to perform twin simulation of process parameters and production quality of the target health product production line, and construct a twin production model; the raw material detection network configuration module 20 is used to configure a raw material detection network at the raw material input end of the production line, specifically including effective ingredient detection equipment and detection mechanism; the process parameter optimization module 30 is used to perform dynamic detection of the effective ingredients of raw materials through the raw material detection network, input the detection results and the current process parameters of the production line into the twin production model for production quality simulation, and when the production quality indicators do not meet the preset quality range, the twin production model is used to optimize the current process parameters.

[0069] The detailed description of the specific configuration of the raw material detection network configuration module 20 is explained as follows: As mentioned above, a raw material detection network is configured at the raw material input end of the production line, specifically including effective component detection equipment and detection mechanism. The raw material detection network configuration module 20 may further include: a raw material acquisition unit for acquiring multiple raw materials at the raw material input end, and multiple effective component sets corresponding to each of the multiple raw materials; an attenuation analysis unit for acquiring the storage area of ​​the multiple raw materials entering the raw material input end, combining the multiple effective component sets to perform attenuation analysis of the effective components, and establishing a first detection trigger mechanism based on the component attenuation value; and a detection mechanism configuration unit for configuring the effective component detection equipment with the multiple effective component sets and configuring the detection mechanism with the first detection trigger mechanism.

[0070] The raw material testing network configuration module 20 may further include: a second testing trigger mechanism configuration unit for constructing a supply batch traceability channel for the various raw materials, configuring a second testing trigger mechanism with process batch change nodes, and adding the second testing trigger mechanism to the testing mechanism.

[0071] The process involves collecting data from the storage areas of the various raw materials upon entering the raw material input end, performing attenuation analysis of the effective components based on the multiple sets of effective components, and establishing a first detection trigger mechanism based on the component attenuation values. The attenuation analysis unit may further include: a storage environment information acquisition subunit for acquiring storage environment information of any storage area for any raw material; an effective component extraction subunit for extracting each effective component from any set of effective components of any raw material; a content attenuation curve analysis subunit for analyzing the content attenuation curves of each effective component over time based on the storage environment information; and a first detection trigger mechanism configuration subunit for configuring the first detection trigger mechanism based on the content attenuation curves and a preset attenuation threshold.

[0072] The detailed description of the specific configuration of the process parameter optimization module 30 is as follows: As mentioned above, when the production quality index does not meet the preset quality range, the twin production model is used to optimize the current process parameters. The process parameter optimization module 30 may further include: a relevant parameter set analysis unit, which uses the content of effective components in raw materials as a variable and production quality as a quantitative measure, to collect several sets of process parameters and analyze the relevant parameter sets of process parameter types and effective component contents; a quality reduction index generation unit, which calculates the difference between the production quality index and the median of the preset quality range when the production quality index is greater than the maximum boundary value of the preset quality range, and generates a quality reduction index; and an iterative simulation unit, which uses the current process parameters as the starting point, the quality reduction index as the target, and the relevant parameter set as the adjustment object to perform iterative simulation and generate parameter optimization results.

[0073] The process parameter optimization module 30 may further include: a quality improvement index generation unit for calculating the difference between the median of the preset quality range and the production quality index when the production quality index is less than the minimum boundary value of the preset quality range, and generating a quality improvement index; and an iterative simulation unit for performing iterative simulation with the current process parameters as the starting point, the quality improvement index as the target, and the relevant parameter set as the adjustment object, and generating parameter optimization results.

[0074] The detailed description of the specific configuration of the twin simulation module 10 is explained as follows: As mentioned above, the twin simulation of process parameters and production quality of the target health product production line is performed to construct a twin production model. The twin simulation module 10 may further include: a physical modeling unit for collecting data on each production node on the target health product production line to perform physical modeling and construct a twin production line; and a simulation training unit for collecting continuous production parameters and quality inspection samples of each production node to perform simulation training on the twin production line to obtain a converged twin production model.

[0075] The system may further include: a raw material property detection network construction module for building a raw material property detection network at the raw material input end of the production line; a raw material property correlation network training module for analyzing the correlation between the physical properties of the various raw materials and the process parameters of the relevant process nodes, and training the raw material property correlation network; and a process parameter optimization module for detecting raw material property parameters in real time through the raw material property detection network, and optimizing the process parameters of the relevant process nodes based on the raw material property correlation network.

[0076] The health product production control system based on twin simulation provided in this invention can execute the health product production control method based on twin simulation provided in any embodiment of this invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0077] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. 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 each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0078] Based on the foregoing embodiments, this application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor of an electronic device, it can implement the health product production control method based on twin simulation as described in any of the foregoing embodiments.

[0079] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand 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 within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for controlling the production of health products based on twin simulation, characterized in that, include: Perform twin simulations of process parameters and production quality on the production line of the target health product to construct a twin production model; A raw material testing network is configured at the raw material input end of the production line, specifically including active ingredient detection equipment and mechanisms, including: Collect multiple raw materials from the raw material input end, as well as multiple sets of effective components corresponding to each raw material; The storage area of ​​the various raw materials is collected when they enter the raw material input end, and the decay analysis of the effective components is performed in combination with the multiple sets of effective components to establish a first detection trigger mechanism based on the component decay value. The effective ingredient detection device is configured with the plurality of effective ingredient sets, and the detection mechanism is configured with the first detection triggering mechanism; The raw material detection network is used to dynamically detect the effective components of the raw materials. The detection results and the current process parameters of the production line are input into the twin production model for production quality simulation. When the production quality indicators do not meet the preset quality range, the twin production model is used to optimize the current process parameters.

2. The method for controlling the production of health products based on twin simulation as described in claim 1, characterized in that, A supply batch traceability channel for the aforementioned raw materials is established, a second detection triggering mechanism is configured with process batch change nodes, and the second detection triggering mechanism is added to the detection mechanism.

3. The method for controlling the production of health products based on twin simulation as described in claim 1, characterized in that, Collect data from the storage area of ​​the various raw materials upon entering the raw material input end, perform attenuation analysis of the effective components based on the multiple sets of effective components, and establish a first detection trigger mechanism based on the component attenuation value, including: Collect storage environment information for any storage area of ​​any raw material; Extract each effective component from any set of effective components of any raw material; Based on the storage environment information, analyze the content decay curves of each active ingredient over time; Based on the various content decay curves, the first detection triggering mechanism is configured according to a preset decay threshold.

4. The method for controlling the production of health products based on twin simulation as described in claim 1, characterized in that, When production quality indicators do not meet the preset quality range, the twin production model is used to optimize the current process parameters, including: Using the content of effective components in raw materials as variables and production quality as a quantitative measure, several sets of process parameters were collected to analyze the correlation between process parameter types and the content of effective components. When the production quality index is greater than the maximum boundary value of the preset quality range, the difference between the production quality index and the median value of the preset quality range is calculated to generate a quality reduction index. Starting with the current process parameters, taking the quality reduction index as the target, and using the relevant parameter set as the adjustment object, iterative simulation is performed to generate parameter optimization results.

5. The health product production control method based on twin simulation as described in claim 4, characterized in that, When the production quality index is less than the minimum boundary value of the preset quality range, the difference between the median value of the preset quality range and the production quality index is calculated to generate a quality improvement index. Starting with the current process parameters, taking the quality improvement index as the target, and using the relevant parameter set as the adjustment object, iterative simulation is performed to generate parameter optimization results.

6. The method for controlling the production of health products based on twin simulation as described in claim 1, characterized in that, Perform twin simulations of process parameters and production quality for the target health product production line to construct a twin production model, including: Physical modeling of each production node on the target health product's production line is performed to construct a twin production line; Continuous production parameters and quality inspection samples are collected from each production node, and the twin production line is simulated and trained to obtain a converged twin production model.

7. The method for controlling the production of health products based on twin simulation as described in claim 1, characterized in that, The raw material input end of the production line also includes a raw material property detection network; The correlation between the physical properties of the various raw materials and the process parameters of relevant process nodes is analyzed, and a raw material property correlation network is trained. The raw material property detection network detects raw material property parameters in real time, and optimizes process parameters of relevant process nodes based on the raw material property correlation network.

8. A health product production control system based on twin simulation, characterized in that, The system is used to implement the health product production control method based on twin simulation as described in any one of claims 1-7, and the system comprises: The twin simulation module is used to perform twin simulations of process parameters and production quality of the target health product production line, and to build a twin production model; The raw material testing network configuration module is used to configure the raw material testing network at the raw material input end of the production line, specifically including effective component testing equipment and testing mechanisms; The process parameter optimization module is used to dynamically detect the effective components of raw materials through the raw material detection network, input the detection results and the current process parameters of the production line into the twin production model for production quality simulation, and optimize the current process parameters using the twin production model when the production quality indicators do not meet the preset quality range.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the health product production control method based on twin simulation as described in any one of claims 1-7.

Images