Joint purification monitoring method and joint purification monitoring system for food-grade carbon dioxide

Abstract

The invention discloses a combined purification monitoring method and a combined purification monitoring system for food-grade carbon dioxide, and relates to the technical field of purification monitoring, and the combined purification monitoring method comprises the following steps: carrying out purification analysis on a purification process, carrying out sequential decomposition on purification steps, and establishing step association of the purification steps; configuring an integrated monitoring platform, and configuring a monitoring sensor on each purification step; step monitoring of the purification process is executed, and a real-time feedback data stream is established and returned to the integrated monitoring platform; carrying out independent verification analysis on the returned data based on an independent verification network, and establishing a purification early warning result; performing combined purification fitting analysis on the step purification early warning result and the step association, and establishing a pre-analysis result; and completing joint purification monitoring according to a pre-analysis result and a step purification early warning result. The technical problems of insufficient real-time monitoring and weak early warning capability in the prior art are solved, and the technical effects of dynamic monitoring and abnormal early warning of the food-grade carbon dioxide purification process are realized.

Description

Technical Field

[0001] The invention relates to the technical field of purification monitoring, and in particular to a combined purification monitoring method and a combined purification monitoring system for food-grade carbon dioxide. Background Art

[0002] The purification of food-grade carbon dioxide is a critical process that is widely used in beverage manufacturing, food preservation, and other food processing fields. The purity of carbon dioxide directly affects the quality and safety of the final product, so it is crucial to ensure efficient monitoring during the purification process. Traditional purification monitoring methods mostly rely on regular testing, which has the problem of insufficient monitoring accuracy. Summary of the invention

[0003] The present application provides a combined purification monitoring method and a combined purification monitoring system for food-grade carbon dioxide, which are used to solve the technical problems of insufficient real-time monitoring and weak early warning capabilities in the prior art.

[0004] In view of the above problems, the present application provides a combined purification monitoring method and a combined purification monitoring system for food-grade carbon dioxide.

[0005] In a first aspect of the present application, a method for monitoring and purifying food-grade carbon dioxide is provided, the method comprising: Perform purification analysis on the purification process, decompose the purification steps in sequence based on the purification analysis results, and establish step associations for the purification steps; configure an integrated monitoring platform, configure monitoring sensors on each purification step, and establish digital communication between the monitoring sensors and the integrated monitoring platform; use the monitoring sensors to perform step monitoring of the purification process, establish a real-time feedback data stream, and transmit the real-time feedback data stream back to the integrated monitoring platform via digital communication; activate an independent verification network, perform independent verification analysis on the feedback data of the integrated monitoring platform based on the independent verification network, and establish step purification early warning results; perform joint purification fitting analysis on the step purification early warning results and step associations to establish pre-analysis results; and complete joint purification monitoring based on the pre-analysis results and step purification early warning results.

[0006] The second aspect of the present application provides a combined purification monitoring system for food-grade carbon dioxide, the system comprising: A purification analysis module, wherein the purification analysis module performs purification analysis on the purification process, decomposes the purification steps in sequence based on the purification analysis results, and establishes step associations for the purification steps; a monitoring platform configuration module, wherein the monitoring platform configuration module configures an integrated monitoring platform, configures monitoring sensors on each purification step, and establishes digital communication between the monitoring sensors and the integrated monitoring platform; a monitoring module, wherein the monitoring module uses monitoring sensors to perform step monitoring of the purification process, establishes a real-time feedback data stream, and transmits the real-time feedback data stream back to the integrated monitoring platform via digital communication; an independent verification analysis module, wherein the independent verification analysis module activates an independent verification network, performs independent verification analysis on the return data of the integrated monitoring platform based on the independent verification network, and establishes a step purification warning result; a joint purification fitting analysis module, wherein the joint purification fitting analysis module performs a joint purification fitting analysis on the step purification warning result and the step association, and establishes a pre-analysis result; and a joint purification monitoring module, wherein the joint purification monitoring module completes joint purification monitoring based on the pre-analysis result and the step purification warning result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application performs purification analysis on the purification process, decomposes the purification step sequence with the purification analysis result, and establishes the step association of the purification step; configures the integrated monitoring platform, configures the monitoring sensor on each purification step, and establishes the digital communication between the monitoring sensor and the integrated monitoring platform; uses the monitoring sensor to perform the step monitoring of the purification process, establishes the real-time feedback data stream, and transmits the real-time feedback data stream back to the integrated monitoring platform through digital communication; activates the independent verification network, and performs independent verification analysis on the return data of the integrated monitoring platform based on the independent verification network, and establishes the step purification early warning result; performs joint purification fitting analysis on the step purification early warning result and the step association, and establishes the pre-analysis result; completes the joint purification monitoring according to the pre-analysis result and the step purification early warning result. The present invention solves the technical problems that the existing technology has insufficient real-time monitoring and weak early warning ability, analyzes the purification process, decomposes the steps and establishes the association, configures the integrated monitoring platform and the sensor, realizes the real-time data collection and feedback, activates the independent verification network to verify the data, generates the step purification early warning result, and performs the joint purification fitting analysis, and finally forms the pre-analysis result, so as to realize the technical effect of dynamic monitoring and abnormal early warning of the food-grade carbon dioxide purification process. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic flow chart of a combined purification and monitoring method for food-grade carbon dioxide provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a combined purification monitoring system for food-grade carbon dioxide provided in an embodiment of the present application.

[0010] Explanation of the accompanying drawings: purification analysis module 11, monitoring platform configuration module 12, monitoring module 13, independent verification analysis module 14, joint purification fitting analysis module 15, joint purification monitoring module 16. DETAILED DESCRIPTION

[0011] The present application provides a joint purification monitoring method and a joint purification monitoring system for food-grade carbon dioxide, aiming to solve the technical problems of insufficient real-time monitoring and weak early warning capabilities in the prior art. By analyzing the purification process, decomposing the steps and establishing associations, configuring an integrated monitoring platform and sensors, realizing real-time data collection and feedback, activating an independent verification network to verify the data, generating step purification early warning results, and performing joint purification fitting analysis, the present application finally forms a pre-analysis result, thereby realizing the technical effect of dynamic monitoring of the food-grade carbon dioxide purification process and abnormal early warning.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a combined purification monitoring method for food-grade carbon dioxide, the method comprising: Step S100: Perform a purification analysis on the purification process, decompose the purification steps in sequence based on the purification analysis results, and establish step associations for the purification steps.

[0015] In the present application embodiment, the purification process is first subjected to purification analysis. Specifically, the purification process is first subjected to systematic sequential analysis. The process of the entire purification process is modeled and simulated using process simulation and analysis software, such as AspenPlus. In the simulation process, the dependency of each step is considered. For example, preliminary cooling is a basic step because subsequent separation and purification operations need to be performed under low temperature conditions. If the cooling step is not completed in a predetermined order, the subsequent purification purity or efficiency will be affected. Therefore, through this simulation method, the position of each step in the entire process is determined, thereby determining a scientific order. After determining the order of each step, the process parameters involved in each step are analyzed. This step is to clarify which parameters are key factors in each step through the analysis of important parameters such as temperature, pressure, and flow, which directly affect the purification effect. The key parameters in each step are identified and extracted by adopting the principal component analysis method. For example, in the cooling step, it may be found that the influence of temperature on the final purification purity is more significant than other parameters. Therefore, temperature is used as a key parameter. Through the above process, the purification analysis of the purification process is completed.

[0016] Next, based on the results of the above purification analysis, the purification steps are sequentially decomposed, and the purification process is divided into several independent steps and numbered. Then the parameters in each step are further analyzed, and the key parameters are screened out based on the principal component analysis threshold, and the key parameter set associated with the step is established.

[0017] Finally, the step association is established. Specifically, first, the association constraints are established according to the step number and the function and key parameters of each step. Then, the association constraints are used to generate a step dependency matrix to reflect the dependencies between different steps. For example, each value of the dependency matrix represents the strength of the association between the previous and next steps. Finally, the dependency matrix is ​​used to clarify the role and relevance of each step in the entire purification process, so that each step can be executed efficiently and logically consistently on the monitoring platform.

[0018] Through the above process, the establishment of step association is completed.

[0019] Furthermore, in the method provided in the embodiment of the application, the sequential decomposition of the purification steps based on the purification analysis results and the establishment of step associations of the purification steps also include: The steps are numbered based on the results of the purified step sequence decomposition; the step parameters of each step are parsed, and the criticality analysis of the step parameters of each step is performed based on the principal component threshold, and a key parameter set mapped to each step is established; the step numbers are used to generate association constraints, and a step dependency matrix is ​​established based on the association constraints and the execution function and key parameter set of each step; step associations are generated based on the step dependency matrix.

[0020] In the embodiment of the present application, the results are first decomposed according to the order of the purification steps, and a unique number is assigned to each step to ensure that the logical position of each step in the process is clear.

[0021] Then, the operating parameters of each step, such as temperature, pressure, flow rate, etc., are analyzed to identify the key parameters that have a significant impact on the purification effect. This process is achieved through principal component analysis (PCA). PCA sets the principal component threshold (usually when the cumulative explained variance reaches more than 90%) to screen out the parameters with significant influence and form the key parameter set for each step.

[0022] After determining the step number and key parameter set, use this information to generate association constraints, that is, the dependency of the preceding step on the succeeding step. The association constraint stipulates that only the preceding numbered step has an impact on the succeeding numbered step, and the succeeding step cannot affect the preceding step. For example, the separation step (post-numbered) will only be started after the temperature of the cooling step (previous number) reaches the set value, ensuring the one-way dependency of the process. Next, based on the association constraints, the execution function of each step and the key parameter set, a step dependency matrix is ​​established. The dependency matrix represents the dependency strength of the previous and next steps in numerical form. Specifically, the dependency value is calculated by a formula to reflect the degree of influence of each preceding step on its subsequent steps, providing a quantitative basis for process control.

[0023] Finally, the step association relationship is generated through the step dependency matrix.

[0024] Furthermore, in the method provided in the embodiment of the application, the step dependency matrix is ​​established based on the associated constraints and the execution function and key parameter set of each step, and further includes: The step dependency value is established by the formula: in, Characterization steps Steps The dependent value of is the correlation coefficient, which is determined by the correlation constraint. ,but , The number of critical parameters in the characterization step, For any key parameter, Characterization steps Middle The weight of the key parameters, Characterization steps Middle Key parameters for the preceding steps The demand for Characterization steps No. The actual values ​​of the key parameters; generate a step dependency matrix based on the step dependency values.

[0025] In the embodiment of the present application, the step dependency value is established by the formula: ;in, Characterization steps Steps The dependent value of is the correlation coefficient, which is determined by the correlation constraint and represents whether there is a dependency relationship between the steps. If step i depends on step j, that is, i>j, then 1. When ,but , The number of critical parameters in the characterization step, For any key parameter, Characterization steps Middle The weights of the key parameters are pre-set by technical experts. Characterization steps Middle Key parameters for the preceding steps The demand is determined by technical experts based on experimental data. Characterization steps No. The actual value of the key parameters is obtained by real-time acquisition by the sensor. Calculate in and get the step-dependent value.

[0026] Finally, fill the dependency values ​​of each step on other steps into the corresponding positions of the dependency matrix in turn. The dependency matrix is ​​a lower triangular matrix, that is, it only contains the dependency values ​​of the predecessor step on the successor step. The upper triangular elements are set to 0, indicating that the successor step will not affect the predecessor step.

[0027] Through the above process, the step dependency matrix is ​​obtained.

[0028] Step S200: configure an integrated monitoring platform, configure monitoring sensors at each purification step, and establish digital communication between the monitoring sensors and the integrated monitoring platform.

[0029] In the embodiment of the present application, an integrated monitoring platform is first configured, and corresponding monitoring sensors are installed at each purification step to collect data of key parameters and material content data in real time. The configured integrated monitoring platform is based on SCADA or DCS technology and has multi-point data collection and centralized management functions.

[0030] Different types of monitoring sensors are configured at each purification step, such as temperature sensors installed at the cooling and heating steps, pressure sensors installed at the compression step, and flow sensors installed at the separation and purification steps. Each sensor collects key parameters in the corresponding step to ensure that each step operates within the set range. In addition, in order to obtain the substance content, analytical equipment such as gas chromatographs or liquid chromatographs are used to analyze the chemical composition of the sample in real time.

[0031] After configuration is completed, the monitoring sensors are connected to the integrated monitoring platform through digital communication technology. Industrial network protocols such as Modbus are used to achieve data transmission. Each sensor converts the collected parameters into digital signals and sends them to the monitoring platform through the network protocol.

[0032] Step S300: Utilize monitoring sensors to perform step monitoring of the purification process, establish a real-time feedback data stream, and transmit the real-time feedback data stream back to the integrated monitoring platform via digital communication.

[0033] In the embodiment of the present application, the monitoring sensors configured in each step collect key parameters (such as temperature, pressure, flow, etc.) in real time to generate a real-time feedback data stream. The analog signal collected by each sensor is converted into a digital signal through an A / D (analog-to-digital) converter to form a digital real-time feedback data stream. Subsequently, these data streams are transmitted back to the integrated monitoring platform through the digital communication channel established by the industrial network protocol.

[0034] Step S400: Activate the independent verification network, perform independent verification and analysis on the return data of the integrated monitoring platform based on the independent verification network, and establish a step purification warning result.

[0035] In an embodiment of the present application, an independent verification network is first activated. After the independent verification network is activated, the independent verification network performs independent verification and analysis on the return data of the integrated monitoring platform item by item to identify deviations and abnormal situations.

[0036] Specifically, the independent verification network receives the key parameter data (such as temperature, pressure, etc.) sent back and compares these data with the preset benchmark range. The benchmark range is set by technical experts based on process standards and historical operation data, representing the range of parameter values ​​under normal operating conditions. For each key parameter, the independent verification network immediately compares the current value with the benchmark range when receiving the data to determine whether the data is within the normal fluctuation range. When a parameter data exceeds the benchmark range, the independent verification network calculates the deviation value of the parameter, that is, the difference between the current value and the benchmark value. For example, if the standard range of temperature is 40℃ to 60℃, and the current data is 65℃, the deviation value is +5℃. The size and direction of the deviation (positive or negative deviation) are recorded to determine the impact of the deviation on the purification process. After that, based on the deviation value and the fluctuation frequency of the parameter, the independent verification network further determines the type of abnormality. Common types of abnormalities include instantaneous deviation (single deviation in a short period of time) and continuous deviation (parameters are in the abnormal range for a long time). For example, if the pressure is continuously higher than the benchmark range in multiple return data, it is marked as "continuous deviation", indicating potential process instability.

[0037] To improve accuracy, the independent verification network sets thresholds for screening and judging abnormal situations. Parameter deviations will only be marked as abnormal when they reach a certain threshold. For example, the temperature deviation is set to ±3°C. If it exceeds this range, an early warning will be triggered, while slight fluctuations below this range will be ignored. By setting the threshold, insignificant small fluctuations are filtered out, and only meaningful abnormal situations are retained.

[0038] Once an abnormal situation is identified, the network generation steps are independently verified to purify the warning results, including information such as deviation value, deviation type, abnormal steps and parameters.

[0039] Furthermore, in the method provided in the embodiment of the application, the activating the independent verification network further includes: Perform a self-inspection of the network database within the independent verification network, and establish a self-inspection result, wherein the self-inspection result includes a data richness result and a data credibility result; establish data requirements based on the self-inspection result; call the adversarial generation network within the integrated monitoring platform, obtain the generation evaluation of the adversarial generation network, and establish a first segmentation preference based on the generation evaluation; obtain the scheduling evaluation of the federal scheduling, and establish a second segmentation preference based on the scheduling evaluation; perform data requirement segmentation based on the first segmentation preference and the second segmentation preference, perform data trust expansion based on the data requirement segmentation result, and after supplementing the network database with the data trust expansion result, complete the activation of the independent verification network.

[0040] In the embodiment of the present application, the independent verification network first performs a self-check on its internal network database, evaluates the quality of the data, and generates data richness results and data credibility results. The data richness results are generated by data coverage analysis and data distribution evaluation, that is, the data entries of each key parameter are counted one by one, and the data density under different operating conditions is ensured to meet the preset process requirements. For example, the data entries of the temperature parameters are counted according to a specific time period to ensure that the data are sufficient in each interval. If the data is insufficient under a certain condition, the gap is marked in the data richness result for subsequent data supplementation reference. At the same time, the data credibility results are generated by consistency check and historical trend comparison to evaluate the accuracy and consistency of the data. The independent verification network uses historical data for comparative detection, uses statistical methods such as mean and variance analysis to mark abnormal data that are not within the normal range, or identifies data that deviates from the historical trend through a rule matching algorithm. For example, for pressure parameters, if historical data shows that its normal range is 1.0 to 1.5MPa, but the current data exceeds 1.6MPa many times, these data are marked as low-credibility data. During the self-inspection process, the data richness results and data credibility results respectively revealed the deficiencies in data coverage and consistency, providing direction for subsequent data expansion operations.

[0041] Based on the self-inspection results, the network generates data needs independently verified, key parameters with insufficient data or low data quality are identified, and the data types and time periods that need to be supplemented are clarified to provide a basis for subsequent data supplementation.

[0042] Then the independent verification network calls the adversarial generative network in the integrated monitoring platform to generate samples that meet the data requirements. The GAN network consists of a generator and a discriminator. The generator generates simulated data according to the data requirements, and the discriminator evaluates the similarity and authenticity of these generated data. Each generated sample is scored and checked by the discriminator to obtain a generated evaluation. The evaluation is carried out from the two aspects of data similarity and representativeness, including dimensions such as parameter range, statistical distribution, and trend matching to ensure that the generated data conforms to the characteristics of real data. The first segmentation preference is generated according to the score of the generated evaluation, and data samples with higher evaluations are preferentially selected for supplementation to meet the standards of data requirements.

[0043] At the same time, the independent verification network obtains scheduling evaluations from the federal scheduling system to coordinate data supplementation needs from multiple data sources. The federal scheduling system uses methods such as the federal averaging algorithm to analyze the data stability and consistency of each data source. Scheduling evaluation covers the availability and data consistency of data sources. The availability of data sources reflects the online rate and data transmission rate of the data source, ensuring the continuity and stability of transmission; data consistency ensures the synergy of different data sources by comparing the parameter ranges, trends and distributions of multiple data sources. For example, if the temperature data of a data source deviates significantly from the average of other data sources, the consistency evaluation of the data source will be lower. Based on the scheduling evaluation, the independent verification network generates a second segmentation preference, giving priority to data sources with higher scheduling evaluations for supplementation to ensure the reliability and consistency of data transmission.

[0044] After obtaining the first segmentation preference and the second segmentation preference, the independent verification network segments the data requirements and prioritizes the order of data supplementation to ensure the priority of key data items. After the segmentation is completed, the independent verification network performs data trust expansion on the data, and further improves the quality of supplementary data through data cleaning, redundancy removal and credibility scoring. Data cleaning removes noise data to ensure the accuracy of supplementary data, and credibility scoring further screens out data from reliable sources.

[0045] The data after data trust expansion is used to supplement the network database, so that it can reach a high standard in terms of data richness and data credibility. After the data supplement is completed, the independent verification network enters the activation state.

[0046] Step S500: performing joint purification fitting analysis on the step purification warning result and step association to establish a pre-analysis result.

[0047] In the embodiment of the present application, in the process of associating the step purification warning result with the step for joint purification fitting analysis, each deviation point in the purification warning result is first mapped one by one with the step association relationship to form a "deviation-association table". This table clearly lists the abnormal deviations and influence chains of each previous and subsequent steps such as the cooling step and the separation step. For example, if a temperature deviation of +5°C is detected in the cooling step, the parameter changes that may be caused by the deviation in the separation step and the purification step are marked.

[0048] Next, based on the correlation between the deviation point and the subsequent steps, the influence relationship of each deviation point in the associated steps is gradually established. To this end, the technical experts set the influence coefficient to indicate how the deviation of the previous step is transmitted in the subsequent steps. The influence coefficient is determined by technical experts based on historical data. For example, the influence coefficient of cooling temperature on the separation pressure may be 0.1, indicating that the temperature change in the cooling step has a 10% effect on the pressure of the separation step. The fitting calculation formula is the impact value of the subsequent step = the deviation of the previous step × the influence coefficient. Assuming that the temperature deviation of the cooling step is +5°C, multiply this deviation by the influence coefficient of 0.1 to obtain a pressure increment of +0.5MPa. This result indicates that in the separation step, the deviation of the cooling temperature will cause the pressure of the separation step to increase by 0.5MPa. After that, the influence value of each deviation point in the subsequent associated steps is calculated in this way, so that the influence relationship of each deviation point can be quantified.

[0049] After completing the single-step fitting, the deviation impact value is analyzed in multiple steps, and the impact is gradually accumulated along the path of the deviation point in the association table. For example, the temperature deviation in the cooling step will first affect the pressure in the separation step, and then the pressure deviation will further affect the flow in the purification step. When accumulating the impact of each step, the step-by-step superposition effect of the deviation is considered to ensure that the chain effect of each deviation point in the process can be fully tracked in the end. The cumulative impact of each level is recorded to identify potential multi-step chain reactions.

[0050] After completing the fitting accumulation of all steps, apply risk threshold screening to identify high-risk steps. A safety threshold is set for each step in the fitting analysis. If the cumulative deviation exceeds the threshold in a certain step, the step is marked as "high risk". For example, if the pressure deviation accumulation of the separation step exceeds 1MPa, it will be marked as a high-risk step and clearly marked in the results, so that operators can give priority attention and take countermeasures to prevent the deviation from continuing to spread.

[0051] Finally, the pre-analysis results are generated, which list in detail the initial position of each deviation point, the fitting impact value of each step, the cumulative impact value, and the steps marked as high risk.

[0052] Step S600: completing joint purification monitoring according to the pre-analysis results and the step purification warning results.

[0053] In the embodiment of the present application, in the process of completing the joint purification monitoring according to the pre-analysis results and the step purification warning results, a joint response plan is first generated according to the pre-analysis results, that is, adjustment measures formulated for possible deviations. Then, the correction effect evaluation of the joint response plan is performed to evaluate the effect of the plan in actual application and generate an evaluation result of the correction effect. At the same time, an additional cost analysis of the execution of the plan is performed to determine the additional costs that may be incurred when implementing the plan, and to form an additional cost analysis result.

[0054] After obtaining the correction effect evaluation results and additional cost analysis results, the joint response plan is analyzed and an effective and economical plan is selected to ensure the feasibility and efficiency of the adjustment. Finally, the joint purification monitoring is completed based on the results of the execution analysis to achieve dynamic adjustment and precise control of the entire purification process.

[0055] Furthermore, in the method provided in the embodiment of the application, the combined purification monitoring is completed according to the pre-analysis results and the step purification warning results, and further includes: Generate a joint response plan based on the preliminary analysis results; evaluate the effect of plan revision based on the joint response plan and establish the revised effect evaluation results; analyze the additional cost of plan execution based on the joint response plan and establish the additional cost analysis results; perform execution analysis of the joint response plan using the revised effect evaluation results and additional cost analysis results, and perform early warning response management of the preliminary analysis results based on the execution analysis results.

[0056] In the embodiment of the present application, a joint response plan is first formulated based on the deviation points and their propagation paths identified in the pre-analysis results. The plan generation adopts a rule-matching method to match each deviation point with the preset process standards and operating rules in the historical data to select appropriate adjustment measures for each possible deviation scenario. For example, if the pre-analysis shows that the temperature deviation in the cooling step will cause the pressure fluctuation in the separation step, the appropriate solution is found through rule matching, such as reducing the cooling temperature or fine-tuning the pressure in the separation step.

[0057] After generating the response plan, the effect of the plan correction is evaluated. This step uses simulation technology to test the effect of the plan in a virtual environment to evaluate its actual correction ability for the deviation. In the simulation environment, each joint response plan is applied, and the changes in key parameters are monitored in real time to observe whether the deviation under simulation returns to the normal range. For example, after simulating the adjustment of the cooling temperature, observe whether the pressure change in the separation step can gradually return to the standard value to confirm the effect of the plan. The reduction in deviation is the basis for quantifying the correction effect, and the correction effect evaluation result is finally generated through this process.

[0058] After completing the evaluation of the correction effect, the joint response plan will continue to conduct additional cost analysis, that is, to evaluate the additional resource expenditure that may be generated by the implementation of the plan through cost simulation. By analyzing factors such as energy consumption, material consumption and time cost of each plan, ensure that the implementation of the plan is economically feasible. For example, by simulating the power consumption after adjusting the cooling temperature, the energy consumption cost that may increase when the plan is implemented is calculated. After the simulation and summary of various resource consumption data, the additional cost analysis results are generated.

[0059] Then, the execution analysis of the joint response plan is carried out using the correction effect evaluation results and additional cost analysis results. At this time, the correction effect evaluation results and additional cost analysis results of all joint response plans are evaluated item by item to finally select the optimal execution plan. Specifically, all plans are first sorted according to the correction effect to give priority to plans with significant correction effects, so as to ensure that the plans can effectively reduce deviations and stabilize the process. On this basis, a detailed comparative analysis of the correction effect and cost is carried out to ensure that the final selected plan is both effective and economical. First, the plan with significant correction effect is given priority, but the premise is that the execution cost of the plan must be within the set cost threshold. For example, if plan A can significantly reduce the pressure deviation of the separation step and its cost does not exceed the threshold, while plan B has a lower cost but the correction effect is not as good as plan A, then plan A with a more significant effect will be given priority to ensure that the process has priority in terms of the stability of key parameters. If the two plans are similar or even the same in terms of correction effect, the plan with lower execution cost will be given priority to optimize resource utilization and reduce operating costs. For example, if both Scheme C and Scheme D can effectively control the deviation within the set range in terms of temperature adjustment in the cooling step, the scheme with the lowest execution cost is selected, thereby achieving efficient resource management and cost savings.

[0060] Finally, based on the optimal solution selected by the execution analysis, the early warning response management stage is entered. In this stage, the changes in key parameters are monitored in real time through closed-loop control to ensure that the adjustment effect is consistent with the simulation expectations. For example, if the temperature adjustment of the cooling step gradually eliminates the pressure deviation of the separation step, the adjustment frequency is gradually reduced to stabilize the process. Through early warning response management, efficient response and precise control of deviations can be achieved in actual operations to ensure the stable operation and resource optimization of the purification process.

[0061] Furthermore, the method provided in the application embodiment also includes: Perform time-series step verification on the step purification warning results to establish time-series step anomalies; perform cumulative anomaly identification of steps through time-series step anomalies to establish cumulative anomaly identification results; and report anomalies based on the cumulative anomaly identification results.

[0062] In an embodiment of the present application, the step purification warning results are first verified by sequential step verification to identify and mark continuous abnormal changes, thereby gradually establishing sequential step anomalies. Specifically, a time series analysis method is used to arrange the key parameters of each step in chronological order, construct a time series, and compare the data at each time point with its preset standard value pre-set by technical experts. When a parameter continues to deviate from the standard value at multiple time points, the step is marked as a sequential step anomaly, for example, the temperature of the cooling step exceeds the standard multiple times within an hour. By establishing sequential step anomalies, occasional deviations and continuous deviations are distinguished, which helps to identify abnormal stability trends.

[0063] Based on the abnormality of the sequential steps, the cumulative abnormality identification of the steps is further performed, and the cumulative abnormality identification results are gradually established. The abnormality cumulative counting method is used to accumulate the abnormal conditions of each step within a specified time period. For example, the number of temperature exceeding the standard in the cooling step is counted within an hour, and these abnormal times are summarized to generate a cumulative abnormality value. Through this abnormality accumulation method, it is possible to identify which steps have the most frequent abnormal conditions, so that these high-frequency abnormal steps can be marked in the cumulative abnormality identification results for centralized monitoring by operators.

[0064] Finally, based on the cumulative abnormality identification results, the high-frequency abnormal steps are reported abnormally. In this step, the threshold trigger mechanism is used to automatically trigger an alarm when the cumulative abnormal value exceeds the preset threshold. For example, if the temperature abnormality in the cooling step exceeds ten times, an abnormal alarm is generated to remind the operator to intervene and check in time.

[0065] Furthermore, the method provided in the application embodiment also includes: Establish a set of food hazard substances and establish a material warning ratio identification for the food hazard substance set; perform data verification on the real-time feedback data stream based on the food hazard substance set and establish data verification results; when any real-time feedback data stream meets the material warning ratio, an exception is reported.

[0066] In the embodiment of the present application, a food hazard substance set is first established, that is, a list of potentially hazardous substances is created, and a substance warning ratio mark is set for each substance. This set includes key hazardous substances that may appear in the purification process (such as heavy metals, residual chemicals, etc.), and a safety limit is assigned to each substance, indicating that an early warning is triggered when the content ratio of the substance exceeds this limit. These ratio marks are set by technical experts according to process standards, making them a benchmark for judging food safety.

[0067] Next, the real-time feedback data stream is verified based on the food hazardous substance set. The real-time feedback data stream refers to the key parameters and substance content data collected in real time during the purification process. By comparing the real-time data with the warning ratio mark in the hazardous substance set, the data verification results are gradually established, that is, to determine in real time whether the current data is within the safe range. For example, if the data stream shows that the content of a certain heavy metal is close to the warning ratio in the set, the substance is marked as risky in the data verification result.

[0068] Finally, when the content of any substance in the real-time feedback data stream reaches or exceeds the preset material warning ratio, an abnormality is reported. This step uses a threshold trigger mechanism, that is, when the content of a certain substance is detected to be excessive, an abnormal alarm is automatically triggered. For example, if the content of a certain type of chemical residue exceeds the warning ratio mark, an abnormal notification is immediately issued to prompt the operator to check and intervene to ensure that there will be no food safety problems in the process.

[0069] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: The present application performs purification analysis on the purification process, decomposes the purification step sequence with the purification analysis result, and establishes the step association of the purification step; configures the integrated monitoring platform, configures the monitoring sensor on each purification step, and establishes the digital communication between the monitoring sensor and the integrated monitoring platform; uses the monitoring sensor to perform the step monitoring of the purification process, establishes the real-time feedback data stream, and transmits the real-time feedback data stream back to the integrated monitoring platform through digital communication; activates the independent verification network, and performs independent verification analysis on the return data of the integrated monitoring platform based on the independent verification network, and establishes the step purification early warning result; performs joint purification fitting analysis on the step purification early warning result and the step association, and establishes the pre-analysis result; completes the joint purification monitoring according to the pre-analysis result and the step purification early warning result. The present invention solves the technical problems that the existing technology has insufficient real-time monitoring and weak early warning ability, analyzes the purification process, decomposes the steps and establishes the association, configures the integrated monitoring platform and the sensor, realizes the real-time data collection and feedback, activates the independent verification network to verify the data, generates the step purification early warning result, and performs the joint purification fitting analysis, and finally forms the pre-analysis result, so as to realize the technical effect of dynamic monitoring and abnormal early warning of the food-grade carbon dioxide purification process.

[0070] Embodiment 2, based on the same inventive concept as the combined purification and monitoring method of food-grade carbon dioxide in the previous embodiment, Figure 2 As shown, the present application provides a food-grade carbon dioxide combined purification monitoring system, and the system and method embodiments in the present application embodiments are based on the same inventive concept. The system includes: a purification analysis module 11, wherein the purification analysis module 11 performs purification analysis on the purification process, decomposes the purification steps in sequence according to the purification analysis results, and establishes step associations of the purification steps; a monitoring platform configuration module 12, wherein the monitoring platform configuration module 12 configures an integrated monitoring platform, configures monitoring sensors on each purification step, and establishes digital communication between the monitoring sensors and the integrated monitoring platform; a monitoring module 13, wherein the monitoring module 13 uses monitoring sensors to perform step monitoring of the purification process, establishes a real-time feedback data stream, and transmits the real-time feedback data stream back to the integrated monitoring platform through digital communication; an independent verification analysis module 14, wherein the independent verification analysis module 14 activates an independent verification network, performs independent verification analysis on the return data of the integrated monitoring platform based on the independent verification network, and establishes a step purification warning result; a joint purification fitting analysis module 15, wherein the joint purification fitting analysis module 15 performs a joint purification fitting analysis on the step purification warning result and the step association, and establishes a pre-analysis result; a joint purification monitoring module 16, wherein the joint purification monitoring module 16 completes joint purification monitoring according to the pre-analysis result and the step purification warning result.

[0071] Furthermore, the system is also used to implement the following functions: The steps are numbered based on the results of the purified step sequence decomposition; the step parameters of each step are parsed, and the criticality analysis of the step parameters of each step is performed based on the principal component threshold, and a key parameter set mapped to each step is established; the step numbers are used to generate association constraints, and a step dependency matrix is ​​established based on the association constraints and the execution function and key parameter set of each step; step associations are generated based on the step dependency matrix.

[0072] Furthermore, the system is also used to implement the following functions: The step dependency value is established by the formula: ;in, Characterization steps Steps The dependent value of is the correlation coefficient, which is determined by the correlation constraint. ,but , The number of critical parameters in the characterization step, For any key parameter, Characterization steps Middle The weight of the key parameters, Characterization steps Middle Key parameters for the preceding steps The demand for Characterization steps No. The actual values ​​of the key parameters; generate a step dependency matrix based on the step dependency values.

[0073] Furthermore, the system is also used to implement the following functions: Perform a self-inspection of the network database within the independent verification network, and establish a self-inspection result, wherein the self-inspection result includes a data richness result and a data credibility result; establish data requirements based on the self-inspection result; call the adversarial generation network within the integrated monitoring platform, obtain the generation evaluation of the adversarial generation network, and establish a first segmentation preference based on the generation evaluation; obtain the scheduling evaluation of the federal scheduling, and establish a second segmentation preference based on the scheduling evaluation; perform data requirement segmentation based on the first segmentation preference and the second segmentation preference, perform data trust expansion based on the data requirement segmentation result, and after supplementing the network database with the data trust expansion result, complete the activation of the independent verification network.

[0074] Furthermore, the system is also used to implement the following functions: Generate a joint response plan based on the preliminary analysis results; evaluate the effect of plan revision based on the joint response plan and establish the revised effect evaluation results; analyze the additional cost of plan execution based on the joint response plan and establish the additional cost analysis results; perform execution analysis of the joint response plan using the revised effect evaluation results and additional cost analysis results, and perform early warning response management of the preliminary analysis results based on the execution analysis results.

[0075] Furthermore, the system is also used to implement the following functions: Perform time-series step verification on the step purification warning results to establish time-series step anomalies; perform cumulative anomaly identification of steps through time-series step anomalies to establish cumulative anomaly identification results; and report anomalies based on the cumulative anomaly identification results.

[0076] Furthermore, the system is also used to implement the following functions: Establish a set of food hazard substances and establish a material warning ratio identification for the food hazard substance set; perform data verification on the real-time feedback data stream based on the food hazard substance set and establish data verification results; when any real-time feedback data stream meets the material warning ratio, an exception is reported.

[0077] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0078] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0079] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A combined purification and monitoring method for food-grade carbon dioxide, characterized in that: The method comprises: Performing purification analysis on the purification process, decomposing the purification steps in sequence based on the purification analysis results, and establishing step associations of the purification steps; Configure an integrated monitoring platform, configure monitoring sensors at each purification step, and establish digital communication between the monitoring sensors and the integrated monitoring platform; Using monitoring sensors to perform step monitoring of the purification process, establish a real-time feedback data stream, and transmit the real-time feedback data stream back to the integrated monitoring platform via digital communication; Activate the independent verification network, conduct independent verification and analysis on the return data of the integrated monitoring platform based on the independent verification network, and establish steps to purify the early warning results; The purification warning results of the steps are associated with the steps to perform joint purification fitting analysis to establish pre-analysis results; Complete joint purification monitoring based on the pre-analysis results and step purification warning results.

2. The combined purification and monitoring method of food-grade carbon dioxide according to claim 1, characterized in that: The step of sequentially decomposing the purification steps based on the purification analysis results and establishing step associations of the purification steps also includes: Decompose the results in the order of purification steps and number the steps; Analyze the step parameters of each step, perform criticality analysis of the step parameters of each step based on the principal component threshold, and establish a key parameter set mapped to each step; Generate association constraints using step numbers, and build a step dependency matrix based on the association constraints and the execution function and key parameter set of each step; Generate step associations based on the step dependency matrix.

3. The combined purification and monitoring method of food-grade carbon dioxide according to claim 2, characterized in that: The step dependency matrix is ​​established based on the associated constraints and the execution function and key parameter set of each step, and further includes: The step dependency value is established by formula: in, Characterization steps Steps The dependent value of is the correlation coefficient, which is determined by the correlation constraint. ,but , The number of critical parameters in the characterization step, For any key parameter, Characterization steps Middle The weight of the key parameters, Characterization steps Middle Key parameters for the preceding steps The demand for Characterization steps No. The actual values ​​of the key parameters; Generates a step dependency matrix based on the step dependency values.

4. The combined purification and monitoring method of food-grade carbon dioxide according to claim 1, characterized in that: The activating the independent verification network further includes: Performing a self-check of a network database within an independent verification network and establishing a self-check result, wherein the self-check result includes a data richness result and a data credibility result; Establish data requirements based on self-inspection results; Calling a generative adversarial network in the integrated monitoring platform, obtaining a generation evaluation of the generative adversarial network, and establishing a first segmentation preference based on the generation evaluation; Obtaining a scheduling evaluation of the federated scheduling, and establishing a second segmentation preference based on the scheduling evaluation; The data demand segmentation is performed based on the first segmentation preference and the second segmentation preference, and the data trust is expanded based on the data demand segmentation result. After the network database is supplemented with the data trust expansion result, the independent verification network activation is completed.

5. The combined purification and monitoring method of food-grade carbon dioxide according to claim 1, characterized in that: The combined purification monitoring is completed according to the pre-analysis results and the step purification warning results, and further includes: Generate joint response plans based on pre-analysis results; Conduct evaluation of the effect of program revision based on the joint response program and establish evaluation results of the revision effect; Conduct additional cost analysis on the plan execution based on the joint response plan and establish additional cost analysis results; Use the modified effect evaluation results and additional cost analysis results to conduct execution analysis of the joint response plan, and conduct early warning response management based on the pre-analysis results.

6. The combined purification and monitoring method of food-grade carbon dioxide according to claim 1, characterized in that: The method further comprises: Perform time sequence step verification on the step purification warning results and establish time sequence step anomalies; The cumulative abnormality identification of the steps is performed through the abnormality of the time sequence steps, and the cumulative abnormality identification result is established; Exceptions are reported based on the accumulated exception recognition results.

7. The combined purification and monitoring method of food-grade carbon dioxide according to claim 1, characterized in that: The method further comprises: Establish a food hazard substance collection and establish a material warning ratio label for the food hazard substance collection; Conduct data verification of real-time feedback data stream based on food hazard collection and establish data verification results; When any real-time feedback data stream meets the material warning ratio, an abnormality is reported.

8. A combined purification monitoring system for food-grade carbon dioxide, characterized in that: The system comprises: A purification analysis module, which performs purification analysis on the purification process, decomposes the purification steps in sequence according to the purification analysis results, and establishes step associations of the purification steps; A monitoring platform configuration module, wherein the monitoring platform configuration module configures an integrated monitoring platform, configures monitoring sensors at each purification step, and establishes digital communication between the monitoring sensors and the integrated monitoring platform; A monitoring module, wherein the monitoring module uses monitoring sensors to perform step monitoring of the purification process, establishes a real-time feedback data stream, and transmits the real-time feedback data stream back to the integrated monitoring platform via digital communication; An independent verification and analysis module, wherein the independent verification and analysis module activates an independent verification network, performs independent verification and analysis on the return data of the integrated monitoring platform based on the independent verification network, and establishes a step-purification warning result; A joint purification fitting analysis module, wherein the joint purification fitting analysis module associates the step purification warning result with the step to perform joint purification fitting analysis and establish a pre-analysis result; A combined purification monitoring module is provided, wherein the combined purification monitoring module completes combined purification monitoring according to the pre-analysis results and the step purification early warning results.

Images