Method for controlling temperature and humidity of lutein steam eyeshade

Abstract

The invention provides a temperature and humidity control method of a lutein steam eye mask. Belongs to the technical field of functional health care material and intelligent material controlled release. The method comprises the following steps: carrying out dynamic environment partitioning on a xanthophyll steam eye patch production workshop, and generating multi-region temperature and humidity baseline data; performing simulation prediction on the iron powder oxidation rate of each partition based on the reaction kinetic model, and generating a feed-forward compensation control instruction set in combination with raw material batch parameters; by means of the temperature and humidity control method of the lutein steam eye mask, the stability of the production process is improved, when the relative humidity of a workshop fluctuates by 40%-80%, the heat release curve variable coefficient can still be controlled within 3%, and it is ensured that the iron powder oxidation reaction is conducted stably.

Description

Technical Field

[0001] This invention proposes a method for controlling the temperature and humidity of a lutein steam eye mask, belonging to the field of functional health care products and intelligent material controlled release technology. Background Technology

[0002] In the production of lutein steam eye masks, traditional processes face numerous challenges that urgently need to be addressed. Currently, most production methods are highly dependent on the environment, and fluctuations in workshop temperature and humidity can severely impact product quality. In summer, when relative humidity exceeds 70%, the iron powder oxidation reaction starts prematurely, while in winter, when the temperature is below 15℃, the activation of the reaction is delayed, making it difficult to maintain stable control over the production process.

[0003] Meanwhile, the monitoring of active ingredients is severely lacking. There is a lack of effective online detection methods for lutein after its addition, and it is extremely prone to degradation under high temperature and humidity conditions, resulting in fluctuations of over 30% in the lutein content of the final product, leading to inconsistent product quality.

[0004] In terms of process control, temperature and humidity regulation relies on fixed setpoints and cannot be flexibly adjusted according to batch differences in raw materials. For example, changes in the particle size distribution of iron powder will affect the oxidation reaction rate, but traditional processes cannot respond accordingly.

[0005] Quality traceability is also rather rudimentary, only recording batch numbers and failing to link the microenvironmental parameters at specific production moments with the performance of individual products. Furthermore, existing patents mostly remain at the level of general measures such as "temperature-controlled workshops" and "humidity regulation," failing to deeply integrate reaction kinetics, active ingredient stability, and intelligent manufacturing to build a cognitive-level production operating system with "reverse modeling—feedforward compensation—digital twins," thus hindering the leap to an intelligent pharmaceutical-grade manufacturing paradigm. Therefore, the development of a novel temperature and humidity control method is urgently needed. Summary of the Invention

[0006] This invention provides a method for controlling the temperature and humidity of a lutein steam eye mask, in order to solve the problems mentioned in the background art above:

[0007] This invention proposes a method for controlling the temperature and humidity of a lutein steam eye mask, the method comprising:

[0008] S1. Dynamically partition the lutein steam eye mask production workshop to generate baseline data of temperature and humidity in multiple areas; simulate and predict the iron powder oxidation rate of each zone based on the reaction kinetic model, and generate a feedforward compensation control instruction set in combination with raw material batch parameters.

[0009] S2. Deploy a near-infrared online monitoring array according to the feedforward compensation control instruction set to collect the iron powder oxidation exothermic curve and lutein characteristic absorption peak data in real time during the production process; use a multivariate statistical process control model to remove outliers and reconstruct signals from the collected data to generate dynamically corrected temperature, humidity and component coupled data streams.

[0010] S3. Construct a digital twin based on the dynamic correction data stream, and synchronously simulate the temperature and humidity field and reaction heat field distribution in the virtual space; use the particle swarm optimization algorithm to iteratively optimize the parameters of the digital twin and generate a feedforward and feedback composite control strategy.

[0011] S4. Based on blockchain distributed ledger technology, the data of the entire production process is encrypted and stored, and the NIR monitoring data, temperature and humidity control records and lutein retention rate test results of each eye mask unit are bound to generate a unique digital fingerprint; the quality traceability chain is automatically triggered through smart contracts.

[0012] S5. Based on the optimization results of the digital twin and the traceability data of the blockchain, dynamically update the reaction dynamics model parameter library; transfer historical batch control experience to new batch production through transfer learning algorithm to generate an adaptive control knowledge graph; build a cognitive-level production operating system based on the knowledge graph.

[0013] The beneficial effects of this invention are as follows: By employing a temperature and humidity control method for lutein steam eye masks, the stability of the production process is improved. Even when the relative humidity in the workshop fluctuates between 40-80%, the coefficient of variation of the exothermic curve can still be controlled within 3%, ensuring the stable progress of the iron powder oxidation reaction. The risk of lutein degradation is reduced. Through online monitoring and precise control, the lutein retention rate reaches over 95%, guaranteeing product efficacy. Quality traceability is enhanced. Blockchain technology is used to link production microenvironment parameters with individual mask performance, achieving precise traceability. Production anomalies caused by environmental fluctuations and raw material differences are reduced, avoiding excessive batch-to-batch quality fluctuations. The reliance on fixed setpoints in traditional processes is avoided, allowing for flexible adjustment of control strategies based on batch-to-batch raw material differences. This invention adapts to production needs in different seasons and environments while meeting high product quality standards, providing a new generation of pharmaceutical-grade intelligent manufacturing core technology for functional steam eye masks and enhancing the product's market competitiveness. Attached Figure Description

[0014] Figure 1 This is a diagram illustrating the steps of the method described in this invention;

[0015] Figure 2 This is a flowchart of step S1 of the present invention. Detailed Implementation

[0016] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0017] One embodiment of the present invention, such as Figure 1 As shown, a method for controlling the temperature and humidity of a lutein vapor eye mask includes:

[0018] S1. Dynamically partition the lutein steam eye mask production workshop to generate baseline data of temperature and humidity in multiple areas; simulate and predict the oxidation rate of iron powder in each partition based on the reaction kinetic model, and generate a feedforward compensation control instruction set in combination with raw material batch parameters (such as iron powder particle size distribution and initial concentration of lutein).

[0019] S2. Deploy a near-infrared (NIR) online monitoring array according to the feedforward compensation control instruction set to collect the iron powder oxidation exothermic curve and lutein characteristic absorption peak data in real time during the production process; use a multivariate statistical process control (MSPC) model to remove outliers and reconstruct signals from the collected data to generate dynamically corrected temperature, humidity and composition coupled data streams.

[0020] S3. Construct a digital twin based on the dynamic correction data stream, and synchronously simulate the temperature and humidity field and reaction heat field distribution in the virtual space; use the particle swarm optimization algorithm to iteratively optimize the parameters of the digital twin, generate a feedforward and feedback composite control strategy, and achieve precise control of the exothermic curve coefficient of variation (CV) <3%.

[0021] S4. Based on blockchain distributed ledger technology, the data of the entire production process is encrypted and stored, and the NIR monitoring data, temperature and humidity control records and lutein retention rate test results of each eye mask unit are bound to generate a unique digital fingerprint; the quality traceability chain is automatically triggered through smart contracts to realize the two-way mapping between the performance of a single product and the micro-environment parameters at the moment of production.

[0022] S5. Based on the optimization results of the digital twin and the traceability data of the blockchain, dynamically update the parameter library of the reaction dynamics model; transfer historical batch control experience to new batch production through transfer learning algorithm to generate an adaptive control knowledge graph; build a cognitive-level production operating system based on the knowledge graph to achieve continuous and stable control with lutein retention rate ≥95%.

[0023] The working principle and effects of the above technical solution are as follows: By precisely controlling the workshop temperature and humidity and the iron powder oxidation rate, fluctuations in the exothermic curve are reduced, effectively improving lutein retention rate and reducing raw material waste caused by temperature and humidity imbalances. Multi-dimensional data monitoring and correction reduce data anomalies and improve the accuracy of production data, avoiding control errors caused by data deviations. Digital twins and iterative optimization enhance control flexibility, adapting to different production batches and environmental changes, and reducing production parameter debugging costs. Blockchain evidence storage strengthens the traceability capability of the entire production process, avoiding gaps in quality problem tracing and reducing the outflow of unqualified products. Transfer learning facilitates rapid adaptation to new batches, shortening the production debugging cycle and improving production efficiency. This not only ensures stable product quality consistency but also improves the level of production intelligence, avoiding production risks caused by lagging manual control and data chaos, further reducing production losses and enhancing product market competitiveness.

[0024] One embodiment of the present invention, such as Figure 2 As shown, S1 includes:

[0025] S11. Collect the spatial layout, equipment distribution and environmental influencing factors of the lutein steam eye mask production workshop, and simultaneously record the initial temperature and humidity data of each production area to generate a basic environmental dataset for the workshop.

[0026] S12. Based on the workshop basic environment dataset, dynamic environment zones are divided according to the raw material reaction characteristics. Initial temperature and humidity data of each zone are continuously sampled and analyzed to generate multi-region temperature and humidity baseline data.

[0027] S13. Input the baseline data of temperature and humidity in multiple regions and the parameters of raw material batches into the reaction kinetic model, perform simulation calculations on the iron powder oxidation process in each zone, and generate time-series data of iron powder oxidation rate in each zone.

[0028] S14. Integrate baseline data of temperature and humidity in multiple regions with time-series data of iron powder oxidation rate in each zone, quantify the correlation between temperature and humidity deviation and oxidation rate fluctuation, and generate a set of feedforward compensation control instructions.

[0029] The working principle and effects of the above technical solution are as follows: By accurately collecting workshop environment and initial temperature and humidity data, the basic environmental dataset of the workshop is improved, enhancing the completeness of basic production data and reducing control deviations caused by data omissions. Dynamic environmental zones are divided based on the reaction characteristics of raw materials, refining temperature and humidity baseline control and effectively improving the targeting of multi-region temperature and humidity baseline data, avoiding the problem of insufficient adaptability of unified control. Simulation calculations are used to understand the variation law of iron powder oxidation rate in each zone, quantifying the correlation between temperature and humidity and oxidation rate, enhancing the foresight of feedforward control, and reducing raw material losses caused by blind control. A feedforward compensation control instruction set adapted to each zone is generated, which not only lays a solid foundation for subsequent precise temperature and humidity control but also avoids the risk of lutein degradation caused by oxidation rate fluctuations in advance, reducing production debugging costs and further improving the scientific and efficient nature of production control.

[0030] In one embodiment of the present invention, S13 includes:

[0031] Integrate baseline temperature and humidity data from multiple regions with raw material batch parameters, sort out the correlation dimensions between the two types of data, standardize data format and numerical accuracy, and generate input dataset for reaction kinetics model;

[0032] The model input dataset is normalized to eliminate differences in data units, filter out invalid data and extreme fluctuation values, and generate standardized model input data.

[0033] The standardized model input data is imported into the reaction kinetics model, the simulation boundary conditions and iteration step size of the oxidation reaction in each zone are set, and multiple rounds of oxidation process simulation calculations are carried out to generate the original data of iron powder oxidation rate in each zone.

[0034] The raw data of iron powder oxidation rate in each zone were time-series aligned and smoothed to analyze the variation pattern of oxidation rate at different time points and generate time-series data of iron powder oxidation rate in each zone.

[0035] The working principle and effects of the above technical solution are as follows: By integrating baseline temperature and humidity data with raw material batch data, the data format and accuracy are standardized, improving the uniformity of input data for the reaction kinetic model and reducing simulation errors caused by inconsistent data formats. Normalization eliminates dimensional differences and filters invalid data, further improving the purity of input data and preventing extreme values ​​from interfering with the accuracy of simulation results. Multi-round oxidation process simulation calculations, combined with boundary conditions and iteration step size settings, enhance the comprehensiveness of the original iron powder oxidation rate data and reduce biases caused by single calculations. Time-series alignment and smoothing processes clarify the variation patterns of oxidation rates, generating accurate oxidation rate time-series data. This provides reliable data support for subsequent temperature and humidity control, avoids lutein degradation due to inaccurate oxidation rate predictions, reduces production losses, and improves the accuracy and scientific rigor of production control.

[0036] In one embodiment of the present invention, step S14 includes:

[0037] Import baseline data of temperature and humidity from multiple regions and time-series data of iron powder oxidation rate from each zone, complete the spatial dimension alignment and time axis synchronization of the two types of data, and generate an integrated associated dataset.

[0038] Data cleaning is performed on the integrated associated dataset to remove redundant and abnormally fluctuating data, calibrate deviations during data transmission, and generate a standardized associated dataset.

[0039] Quantitative analysis is performed on the standardized correlation dataset to calculate the deviation between the actual temperature and humidity values ​​and the baseline values, correlate the deviation with the oxidation rate fluctuation, and generate correlation quantitative results.

[0040] Based on the correlation quantification results, the production process requirements of each zone are matched, oxidation rate control rules are formulated to adapt to different temperature and humidity deviation scenarios, and a feedforward compensation control instruction set is generated.

[0041] The working principle and effects of the above technical solution are as follows: By achieving spatiotemporal synchronization and alignment of temperature and humidity baselines with oxidation rate time-series data, an integrated correlated dataset is generated, improving the correlation between the two types of data and avoiding analytical biases caused by spatiotemporal misalignment. Data cleaning removes redundant and abnormal data and calibrates transmission biases, improving the purity of correlated data and reducing interference from invalid data in subsequent analyses. Quantitative analysis clarifies the correlation between temperature and humidity deviations and oxidation rate fluctuations, making control more precise and avoiding blindly formulating control rules. A feedforward compensation control instruction set is generated based on the needs of zoned processes, providing targeted guidance for subsequent precise temperature and humidity control, and proactively avoiding lutein degradation caused by abnormal oxidation rate fluctuations, reducing production losses, further improving the targeting and reliability of production control, and laying a solid foundation for stable production throughout the entire process.

[0042] In one embodiment of the present invention, S2 includes:

[0043] S21. Based on the feedforward compensation control instruction set, determine the monitoring points and monitoring frequencies of each dynamic environment zone, complete the deployment and debugging of the near-infrared online monitoring array, and generate a real-time monitoring network covering the entire production process.

[0044] S22. Start the real-time monitoring network, collect the iron powder oxidation exothermic curve and lutein characteristic absorption peak data during the production process, and simultaneously record the real-time temperature and humidity data of each monitoring point to generate a multi-dimensional raw monitoring dataset.

[0045] S23. Input the multi-dimensional raw monitoring dataset into the multivariate statistical process control model, identify and remove outliers in the data, complete signal denoising and reconstruction, and generate a standardized monitoring data stream.

[0046] S24. Correlate the temperature and humidity data, oxidation exothermic data and lutein component data in the standardized monitoring data stream, construct a data coupling model, and generate a dynamically corrected temperature, humidity and component coupled data stream.

[0047] The working principle and effects of the above technical solution are as follows: By combining feedforward commands with the deployment and debugging of near-infrared monitoring arrays, a fully covered real-time monitoring network is established to fill local monitoring gaps, improve the comprehensiveness and real-time performance of monitoring throughout the entire production process, and avoid parameter malfunctions caused by monitoring blind spots. Multi-dimensional data collection of oxidation exothermics, lutein composition, and temperature and humidity data enriches the monitoring data dimensions and reduces the loss of key reaction data. Outliers are eliminated and signals are reconstructed through a multivariate statistical model, improving the purity and accuracy of monitoring data and avoiding interference from noise and invalid data in subsequent analysis. A data coupling model is constructed to generate a corrected coupled data stream, strengthening the correlation between multiple types of data and making the matching of temperature and humidity with reaction processes and component changes more accurate. This provides high-quality data support for the construction of digital twins, avoids control errors caused by data deviations in advance, reduces lutein degradation and loss, and further enhances the reliability and data support capabilities of production process monitoring.

[0048] In one embodiment of the present invention, step S23 includes:

[0049] Import the multi-dimensional raw monitoring dataset into the multivariate statistical process control model, adapt the data format to the model input requirements, and generate a model-adapted monitoring dataset.

[0050] Based on the model-adapted monitoring dataset, anomaly feature identification is carried out to locate abnormal fluctuation nodes in the data and generate anomaly data identifier set;

[0051] Based on the abnormal data identifier set, corresponding abnormal data is removed, and valid data fragments are retained to generate a valid monitoring dataset;

[0052] The effective monitoring dataset is subjected to signal denoising processing to eliminate the noise impact caused by external interference, and then the signal is reconstructed to generate a standardized monitoring data stream.

[0053] The working principle and effects of the above technical solution are as follows: By adapting the formats of multi-dimensional raw monitoring data and models, an adapted dataset is generated, improving the compatibility between data and models and avoiding data analysis failures or result deviations caused by format inconsistencies. Abnormal data fluctuation nodes are accurately identified, abnormal data is removed, and valid data fragments are retained, reducing interference from invalid data in subsequent processing and improving the effectiveness of monitoring data. External interference is eliminated through signal denoising, and signal reconstruction is completed, optimizing data quality and making the monitoring data more consistent with actual production scenarios, avoiding data analysis errors caused by noise. Finally, a standardized monitoring data stream is generated, which not only provides high-quality data support for subsequent data coupling modeling but also avoids deviations in temperature, humidity, and component regulation caused by data distortion, reduces lutein degradation loss, and further enhances the reliability and accuracy of data processing in the production process.

[0054] In one embodiment of the present invention, S3 includes:

[0055] S31. Import the dynamically corrected temperature, humidity and composition coupled data stream, combine it with the workshop space layout and equipment parameters, build a digital twin that matches the actual production scenario, and generate a virtual simulation model of the workshop.

[0056] S32. Based on the workshop virtual simulation model, the temperature and humidity field and reaction heat field distribution of each dynamic environmental zone are simulated synchronously to capture the dynamic change trend of the field distribution and generate a dynamic simulation dataset of the field distribution.

[0057] S33. Input the dynamic simulation dataset of field distribution into the particle swarm optimization algorithm, perform multiple rounds of iterative optimization on the control parameters in the digital twin, and output the optimal combination of control parameters.

[0058] S34. By combining the optimal control parameter combination, integrating the feedforward compensation control logic and the real-time feedback control logic, a feedforward and feedback composite control strategy is generated to achieve precise control of the exothermic curve.

[0059] The working principle and effects of the above technical solution are as follows: By combining the corrected coupled data stream with actual workshop parameters, a matching digital twin and virtual simulation model are built to recreate the entire production scenario, avoiding control deviations caused by the disconnect between virtual and reality, and improving the realism of the simulation. Simultaneous simulation of temperature and humidity fields and reaction heat field distribution captures dynamic trends, enhancing control over the production reaction process and reducing control lags caused by unpredictable field distribution changes. Iterative optimization of control parameters using a particle swarm optimization algorithm outputs the optimal combination, reducing losses from blind debugging and improving the rationality of parameter settings. Integration of feedforward and feedback control logic generates a composite strategy, achieving precise control of the exothermic curve and reducing curve fluctuations. This provides scientific guidance for subsequent production control, avoids lutein degradation caused by temperature, humidity, and reaction heat runaway, enhances the flexibility and accuracy of production control, further improves product quality stability, and reduces production management costs.

[0060] In one embodiment of the present invention, S32 includes:

[0061] Load the workshop virtual simulation model, activate the simulation calculation modules of each dynamic environment partition, and generate the partitioned simulation calculation environment;

[0062] In the partitioned simulation computing environment, the simulation calculations of temperature and humidity fields and reaction thermal fields are carried out simultaneously to generate real-time simulation data of field distribution in each partition.

[0063] The real-time simulation data of field distribution in each zone is time-series tracked to record the field distribution change characteristics at different time points and generate a dynamic change feature set of field distribution.

[0064] By integrating the dynamic change feature set of field distribution, the temporal splicing and spatial integration of the data are completed to generate a dynamic simulation dataset of field distribution.

[0065] The working principle and effects of the above technical solution are as follows: By activating the simulation calculation modules of each zone, a dedicated zone simulation environment is constructed, improving the relevance of temperature and humidity field and reaction heat field simulations, avoiding omissions of zone details caused by uniform simulation across the entire domain, and restoring the real reaction scenarios of each zone. Simultaneous simulation calculations of both fields generate real-time field distribution data, reducing the disconnect between temperature and humidity and reaction heat data and improving data correlation. Time-series tracking records the characteristics of field distribution changes, accurately capturing dynamic fluctuation patterns and avoiding control lags caused by missing key change nodes. Data integration completes time-series stitching and spatial integration, generating a complete dynamic simulation dataset of field distribution, improving data integrity and usability. This provides comprehensive and accurate data support for subsequent particle swarm optimization algorithm parameter optimization, enhances control over field distribution changes during production, avoids abnormal iron powder oxidation caused by uncontrolled field distribution, reduces lutein degradation loss, and further improves the reliability of production simulation and the scientific nature of subsequent control.

[0066] In one embodiment of the present invention, S33 includes:

[0067] Import the field distribution dynamic simulation dataset, complete the input adaptation of the dataset and the particle swarm optimization algorithm, calibrate the data dimensions and operation format, and generate an algorithm-adapted simulation dataset.

[0068] Define the optimization objectives and parameter constraints, focus on temperature, humidity and reaction heat control parameters, set the control threshold for the coefficient of variation of the exothermic curve, and generate parameter optimization baseline conditions;

[0069] The particle swarm optimization algorithm is started, and multiple rounds of iterative calculations are carried out on the control parameters based on the optimization baseline conditions. In each round of iteration, the parameter combination is updated and the fitness is calculated to generate a parameter iteration process dataset.

[0070] The optimal parameter combination is selected during the iteration process, the stability and effectiveness of parameter control are verified, invalid parameter combinations are eliminated, and the optimal control parameter combination is output.

[0071] The working principle and effects of the above technical solution are as follows: By calibrating the dimension and operation format of the field distribution simulation data, it achieves compatibility with the particle swarm optimization algorithm, generates a compatible dataset, improves the compatibility between data and algorithm, and avoids iterative calculation failures caused by format inconsistencies or dimensional deviations. It defines the optimization objective and parameter constraint range, sets the threshold for the coefficient of variation of the exothermic curve, clarifies the direction of parameter optimization, reduces invalid calculations caused by blind iteration, and improves optimization efficiency. Multiple rounds of iterative updates of parameter combinations and calculation of fit comprehensively cover the potential optimal parameter range, reducing the problem of missing optimal parameters. Screening and verifying the optimal parameter combination and eliminating invalid combinations ensures the stability and effectiveness of parameter control, avoiding temperature, humidity, and reaction heat control deviations caused by unreasonable parameters. This provides accurate and reliable parameter support for the subsequent construction of composite control strategies, and also helps in the precise control of the exothermic curve, reducing oxidation rate fluctuations, reducing lutein degradation losses, and further improving the scientific nature and accuracy of production control.

[0072] In one embodiment of the present invention, step S4 includes:

[0073] S41. Activate the blockchain distributed ledger system, import temperature and humidity control, oxidation monitoring, component detection and parameter optimization data from the entire production process, complete data encryption and distributed storage, and generate an immutable production data ledger.

[0074] S42. Divide the production unit into individual eye mask units, associate and bind the near-infrared monitoring data, temperature and humidity control records and lutein retention rate test results of each unit, and extract data features to generate a unique digital fingerprint.

[0075] S43. Based on blockchain technology, trigger smart contracts to automatically build a quality traceability chain by linking micro-environmental parameters and quality inspection data throughout the entire production process with digital fingerprints as the core.

[0076] S44. Establish a correlation model between the performance of a single product and the micro-environment parameters at the time of production through the quality traceability chain, and complete the two-way mapping.

[0077] The working principle and effects of the above technical solution are as follows: By encrypting and storing production data throughout the entire process using blockchain, the security and authenticity of production data are improved, preventing data tampering or loss and generating an immutable production data ledger. Individual product production units are divided and unique digital fingerprints are generated, binding exclusive monitoring and testing data to enhance the uniqueness of individual product traceability and reduce traceability confusion caused by mixed batches. Smart contracts are triggered to automatically build a quality traceability chain, linking the entire process micro-environment with quality data, improving quality traceability efficiency and avoiding gaps in tracing the source of quality problems. A two-way mapping between product performance and the production micro-environment is established, facilitating rapid identification of potential quality issues and reducing situations where quality problems cannot be located. This approach not only ensures the credibility of production data but also achieves full-process traceability for individual products, strengthening quality control, reducing the risk of substandard products leaving the market, and further improving product quality controllability and market acceptance.

[0078] In one embodiment of the present invention, S41 includes:

[0079] Launch the blockchain distributed ledger system, complete the calibration of system operating parameters and node deployment, and generate a secure and stable ledger operating environment;

[0080] Import various data from the entire production process, including temperature and humidity control, oxidation monitoring, component detection and parameter optimization data, complete data classification and format adaptation, and generate standardized ledger data;

[0081] Standardized ledger data is encrypted by using encryption algorithms to transform the data format and generate encrypted data fragments, thus ensuring the security of data storage and transmission.

[0082] The encrypted data fragments are synchronized to each blockchain node to complete multi-node evidence verification and data synchronization, generating an immutable production data ledger.

[0083] The working principle and effects of the above technical solution are as follows: By calibrating the operating parameters of the blockchain system and completing node deployment, a secure and stable ledger operating environment is constructed, improving system reliability and avoiding interruptions in production data processing or failures in evidence storage due to system failures. Production data throughout the entire process is categorized, organized, and adapted, standardizing data formats, reducing evidence storage deviations caused by data chaos, and improving the standardization of ledger data. Data is encrypted, transforming its form and enhancing the security of data storage and transmission, preventing data leakage or illegal tampering. Multi-node synchronous evidence storage and verification strengthen data consistency and generate an immutable production data ledger. This not only strengthens the security defense of production data evidence storage but also enhances data credibility, providing reliable data support for subsequent single-product quality traceability and reducing traceability failures caused by unreliable data.

[0084] In one embodiment of the present invention, step S5 includes:

[0085] S51. Summarize the optimization results of digital twin parameters and blockchain traceability data, extract key parameters that affect oxidation reaction and lutein stability, dynamically update the reaction kinetic model parameter library, and improve the model simulation accuracy.

[0086] S52. Import historical production batch control data and optimization experience, and use transfer learning algorithms to transfer historical control experience to new batch production scenarios, explore common production patterns and differentiated characteristics, and generate an adaptive control knowledge graph.

[0087] S53. Based on the adaptive control knowledge graph, integrate data monitoring, parameter optimization, and quality traceability modules to build a cognitive-level production operating system and realize intelligent decision-making in the production process.

[0088] S54. Through the cognitive-level production operating system, an adaptive control strategy is executed to optimize temperature, humidity and reaction parameters in real time, continuously monitor lutein retention rate, and ensure that lutein retention rate remains stable at the target level.

[0089] The working principle and effects of the above technical solution are as follows: By summarizing optimization results and traceability data, key parameters are extracted to update the reaction kinetic model parameter library, improving model simulation accuracy, avoiding control deviations caused by model lag, and providing more reliable support for subsequent simulation calculations. Historical production experience is migrated to new batches, common and different characteristics of production are identified, reducing the debugging cycle and trial production losses of new batches, and enhancing the adaptability to production scenarios. A cognitive-level production operating system is built by integrating multi-functional modules, enabling intelligent production decision-making, reducing errors in manual control, and improving production management efficiency. Through adaptive control and real-time parameter optimization, and monitoring of lutein retention rate, it is ensured to remain stable at the target level, avoiding the impact of component fluctuations on product quality. This not only promotes the intelligent upgrading of the production process but also continuously ensures product quality consistency, reduces long-term production management costs, and further enhances the stability and sustainability of production.

[0090] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for controlling the temperature and humidity of a lutein vapor eye mask, characterized in that, The method includes: S1. Dynamically partition the lutein steam eye mask production workshop to generate baseline data of temperature and humidity in multiple areas; simulate and predict the iron powder oxidation rate of each zone based on the reaction kinetic model, and generate a feedforward compensation control instruction set in combination with raw material batch parameters. S2. Deploy a near-infrared online monitoring array according to the feedforward compensation control instruction set to collect the iron powder oxidation exothermic curve and lutein characteristic absorption peak data in real time during the production process; use a multivariate statistical process control model to remove outliers and reconstruct signals from the collected data to generate dynamically corrected temperature, humidity and component coupled data streams. S3. Construct a digital twin based on the dynamic correction data stream, and synchronously simulate the temperature and humidity field and reaction heat field distribution in the virtual space; use the particle swarm optimization algorithm to iteratively optimize the parameters of the digital twin and generate a feedforward and feedback composite control strategy. S4. Based on blockchain distributed ledger technology, the data of the entire production process is encrypted and stored, and the NIR monitoring data, temperature and humidity control records and lutein retention rate test results of each eye mask unit are bound to generate a unique digital fingerprint; the quality traceability chain is automatically triggered through smart contracts. S5. Based on the optimization results of the digital twin and the traceability data of the blockchain, dynamically update the reaction dynamics model parameter library; transfer historical batch control experience to new batch production through transfer learning algorithm to generate an adaptive control knowledge graph; build a cognitive-level production operating system based on the knowledge graph.

2. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 1, characterized in that, S1 includes: S11. Collect the spatial layout, equipment distribution and environmental influencing factors of the lutein steam eye mask production workshop, and simultaneously record the initial temperature and humidity data of each production area to generate a basic environmental dataset for the workshop. S12. Based on the workshop basic environment dataset, dynamic environment zones are divided according to the raw material reaction characteristics. Initial temperature and humidity data of each zone are continuously sampled and analyzed to generate multi-region temperature and humidity baseline data. S13. Input the baseline data of temperature and humidity in multiple regions and the parameters of raw material batches into the reaction kinetic model, perform simulation calculations on the iron powder oxidation process in each zone, and generate time-series data of iron powder oxidation rate in each zone. S14. Integrate baseline data of temperature and humidity in multiple regions with time-series data of iron powder oxidation rate in each zone, quantify the correlation between temperature and humidity deviation and oxidation rate fluctuation, and generate a set of feedforward compensation control instructions.

3. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 2, characterized in that, S13 includes: Integrate baseline temperature and humidity data from multiple regions with raw material batch parameters, sort out the correlation dimensions between the two types of data, standardize data format and numerical accuracy, and generate input dataset for reaction kinetics model; The model input dataset is normalized to generate standardized model input data; The standardized model input data is imported into the reaction kinetics model, the simulation boundary conditions and iteration step size of the oxidation reaction in each zone are set, and multiple rounds of oxidation process simulation calculations are carried out to generate the original data of iron powder oxidation rate in each zone. The raw data of iron powder oxidation rate in each zone were time-series aligned and smoothed to analyze the variation pattern of oxidation rate at different time points and generate time-series data of iron powder oxidation rate in each zone.

4. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 1, characterized in that, S2 includes: S21. Based on the feedforward compensation control instruction set, determine the monitoring points and monitoring frequencies of each dynamic environment zone, complete the deployment and debugging of the near-infrared online monitoring array, and generate a real-time monitoring network covering the entire production process. S22. Start the real-time monitoring network, collect the iron powder oxidation exothermic curve and lutein characteristic absorption peak data during the production process, and simultaneously record the real-time temperature and humidity data of each monitoring point to generate a multi-dimensional raw monitoring dataset. S23. Input the multi-dimensional raw monitoring dataset into the multivariate statistical process control model, identify and remove outliers in the data, complete signal denoising and reconstruction, and generate a standardized monitoring data stream. S24. Correlate the temperature and humidity data, oxidation exothermic data and lutein component data in the standardized monitoring data stream, construct a data coupling model, and generate a dynamically corrected temperature, humidity and component coupled data stream.

5. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 4, characterized in that, S23 includes: Import the multi-dimensional raw monitoring dataset into the multivariate statistical process control model, adapt the data format to the model input requirements, and generate a model-adapted monitoring dataset. Based on the model-adapted monitoring dataset, anomaly feature identification is carried out to locate abnormal fluctuation nodes in the data and generate anomaly data identifier set; Based on the abnormal data identifier set, corresponding abnormal data is removed, and valid data fragments are retained to generate a valid monitoring dataset; The effective monitoring dataset is subjected to signal denoising processing to eliminate the noise impact caused by external interference, and then the signal is reconstructed to generate a standardized monitoring data stream.

6. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 1, characterized in that, The S3 includes: S31. Import the dynamically corrected temperature, humidity and composition coupled data stream, combine it with the workshop space layout and equipment parameters, build a digital twin that matches the actual production scenario, and generate a virtual simulation model of the workshop. S32. Based on the workshop virtual simulation model, the temperature and humidity field and reaction heat field distribution of each dynamic environmental zone are simulated synchronously to capture the dynamic change trend of the field distribution and generate a dynamic simulation dataset of the field distribution. S33. Input the dynamic simulation dataset of field distribution into the particle swarm optimization algorithm, perform multiple rounds of iterative optimization on the control parameters in the digital twin, and output the optimal combination of control parameters. S34. By combining the optimal control parameter combination, integrating the feedforward compensation control logic and the real-time feedback control logic, a feedforward and feedback composite control strategy is generated to achieve precise control of the exothermic curve.

7. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 6, characterized in that, S32 includes: Load the workshop virtual simulation model, activate the simulation calculation modules of each dynamic environment partition, and generate the partitioned simulation calculation environment; In the partitioned simulation computing environment, the simulation calculations of temperature and humidity fields and reaction thermal fields are carried out simultaneously to generate real-time simulation data of field distribution in each partition. The real-time simulation data of field distribution in each zone is time-series tracked to record the field distribution change characteristics at different time points and generate a dynamic change feature set of field distribution. By integrating the dynamic change feature set of field distribution, the temporal splicing and spatial integration of the data are completed to generate a dynamic simulation dataset of field distribution.

8. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 1, characterized in that, The S4 includes: S41. Activate the blockchain distributed ledger system, import temperature and humidity control, oxidation monitoring, component detection and parameter optimization data from the entire production process, complete data encryption and distributed storage, and generate an immutable production data ledger. S42. Divide the production unit into individual eye mask units, associate and bind the near-infrared monitoring data, temperature and humidity control records and lutein retention rate test results of each unit, and extract data features to generate a unique digital fingerprint. S43. Based on blockchain technology, trigger smart contracts to automatically build a quality traceability chain by linking micro-environmental parameters and quality inspection data throughout the entire production process with digital fingerprints as the core. S44. Establish a correlation model between the performance of a single product and the micro-environment parameters at the time of production through the quality traceability chain, and complete the two-way mapping.

9. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 8, characterized in that, S41 includes: Launch the blockchain distributed ledger system, complete the calibration of system operating parameters and node deployment, and generate a secure and stable ledger operating environment; Import various data from the entire production process, including temperature and humidity control, oxidation monitoring, component detection and parameter optimization data, complete data classification and format adaptation, and generate standardized ledger data; Standardized ledger data is encrypted by using encryption algorithms to transform the data format and generate encrypted data fragments, thus ensuring the security of data storage and transmission. The encrypted data fragments are synchronized to each blockchain node to complete multi-node evidence verification and data synchronization, generating an immutable production data ledger.

10. The method for controlling the temperature and humidity of the lutein steam eye mask according to claim 1, characterized in that, The S5 includes: S51. Summarize the optimization results of digital twin parameters and blockchain traceability data, extract key parameters affecting oxidation reaction and lutein stability, and dynamically update the reaction kinetic model parameter library; S52. Import historical production batch control data and optimization experience, and use transfer learning algorithms to transfer historical control experience to new batch production scenarios, explore common production patterns and differentiated characteristics, and generate an adaptive control knowledge graph. S53. Based on the adaptive control knowledge graph, integrate data monitoring, parameter optimization, and quality traceability modules to build a cognitive-level production operating system; S54. Through the cognitive-level production operating system, an adaptive control strategy is executed to optimize temperature, humidity and reaction parameters in real time and continuously monitor lutein retention rate.

Images