Intelligent regulation and control method and system for food-grade carbon dioxide purification process

Abstract

The invention discloses an intelligent regulation and control method and system for a food-grade carbon dioxide purification process, and relates to the technical field of purification control, and the method comprises the steps: obtaining M process nodes of a purification production line, extracting historical anomaly characterization to carry out node anomaly degree recognition, and configuring a credible window for each process node according to the anomaly degree, and continuously monitoring each node, analyzing an operation risk, generating operation monitoring data and a risk coefficient sequence of the node, combining a risk coefficient of each node and a corresponding monitoring data set with a credible time period identifier, and performing risk regulation and control on the M process nodes. The method solves the technical problems that the existing carbon dioxide purification process lacks a real-time monitoring and adjusting mechanism, so that the risk cannot be recognized and processed in time, and the stability and safety of the purification process are influenced, and achieves the purposes of recognizing and processing the risk in time through a real-time monitoring and intelligent regulation mechanism, and improving the safety of the purification process. And the stability and safety of the purification process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of purification control, and particularly to an intelligent regulation method and system for the purification process of food-grade carbon dioxide. Background Art

[0002] In the process of carbon dioxide purification, the operating status of process nodes directly affects the purification effect and product quality. Traditional purification processes usually rely on manual experience and simple monitoring systems, making it difficult to achieve precise regulation and real-time monitoring of each process node. Most of the monitoring systems in the prior art focus on parameter monitoring of a single node, lacking a comprehensive analysis of the mutual relationships between multiple process nodes, and failing to effectively conduct real-time risk identification and intelligent regulation. This results in the failure to detect abnormal situations that may occur during the process in a timely manner, thereby affecting the purification effect and production safety. Summary of the Invention

[0003] This application provides an intelligent regulation method and system for the purification process of food-grade carbon dioxide, aiming to solve the technical problems that the existing carbon dioxide purification process lacks a real-time monitoring and adjustment mechanism, resulting in the inability to identify and handle risks in a timely manner, and affecting the stability and safety of the purification process.

[0004] In the first aspect of this application, an intelligent regulation method for the purification process of food-grade carbon dioxide is provided. The method includes: obtaining M process nodes of a purification production line, traversing the M process nodes to extract historical abnormal characteristics, and obtaining a set of historical abnormal characteristics of the M process nodes; identifying the abnormal degree of the M process nodes for the set of historical abnormal characteristics of the M process nodes, obtaining the abnormal degrees of the M process nodes, and configuring M confidence windows for the M process nodes according to the abnormal degrees of the M process nodes; respectively using the M confidence windows as the monitoring periods, continuously monitoring the operation of the M process nodes and conducting operation risk analysis, obtaining a sequence of sets of window operation monitoring data of the M process nodes and a sequence of window risk coefficients of the M process nodes, where each window risk coefficient of a process node and the corresponding set of window operation monitoring data of the process node have a reliable time period identifier; combining the reliable time period identifier, and conducting risk regulation on the M process nodes according to the sequence of window risk coefficients of the M process nodes and the sequence of sets of window operation monitoring data of the M process nodes.

[0005] In the second aspect of the present application, an intelligent control system for the purification process of food-grade carbon dioxide is provided. The system includes: an abnormal characterization extraction module, which is used to obtain M process nodes of the purification production line, traverse the M process nodes to extract historical abnormal characterizations, and obtain a set of historical abnormal characterizations of the M process nodes; a credible window configuration module, which is used to identify the node abnormality degree for the set of historical abnormal characterizations of the M process nodes, obtain the abnormality degrees of the M process nodes, and configure M credible windows for the M process nodes according to the abnormality degrees of the M process nodes; an operation risk analysis module, which is used to respectively use the M credible windows as the monitoring period, continuously monitor the operation of the M process nodes and conduct operation risk analysis, and obtain a sequence of monitoring data sets for the window operation of the M process nodes and a sequence of window risk coefficients for the M process nodes. Among them, each process node window risk coefficient and the corresponding process node window operation monitoring data set have a credible time period identifier; a risk control module, which is used to combine the credible time period identifier and perform risk control on the M process nodes according to the sequence of window risk coefficients of the M process nodes and the sequence of window operation monitoring data sets of the M process nodes.

[0006] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The intelligent control method and system for the purification process of food-grade carbon dioxide provided by the present application relate to the technical field of purification control. By obtaining M process nodes of the purification production line, extracting historical abnormal data and identifying the abnormality degree, configuring a credible window for each node, based on the credible window, continuously monitoring the operation status of the node, analyzing risks, generating a sequence of monitoring data and risk coefficients, and performing control in combination with the credible time period identifier, it ensures the stability and safety of the purification process, solves the technical problem that the existing carbon dioxide purification process lacks a real-time monitoring and adjustment mechanism, resulting in risks that cannot be identified and processed in time, affecting the stability and safety of the purification process, and realizes the technical effect of timely identifying and processing risks through a real-time monitoring and intelligent control mechanism, improving the stability and safety of the purification process. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0008] Figure 1 It is a schematic flowchart of the intelligent control method for the purification process of food-grade carbon dioxide provided by the embodiment of the present application; Figure 2 Schematic diagram of the intelligent control system for the food-grade carbon dioxide purification process provided by the embodiment of the present application.

[0009] Explanation of reference numerals: Abnormal characterization extraction module 11, credible window configuration module 12, operation risk analysis module 13, risk control module 14. Detailed implementation manners

[0010] The present application provides an intelligent control method and system for the food-grade carbon dioxide purification process, which is used to solve the technical problems that the existing carbon dioxide purification process lacks a real-time monitoring and adjustment mechanism, resulting in risks that cannot be identified and processed in time, and affecting the stability and safety of the purification process.

[0011] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above accompanying drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such used data may be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices.

[0013] Embodiment 1, as Figure 1 shown, the present application provides an intelligent control method for the food-grade carbon dioxide purification process, and the method includes: P10: Obtain M process nodes of the purification production line, traverse the M process nodes to extract historical abnormal characterizations, and obtain a set of historical abnormal characterizations of the M process nodes.

[0014] Specifically, it is first necessary to comprehensively sort out and analyze the entire purification production line to clarify all the process nodes included in the production line. These process nodes are the key links in the process of carbon dioxide purification, covering multiple steps such as pretreatment of the raw material gas, separation of carbon dioxide, removal of impurities, refining, and finally product detection. Each process node has an independent function and measurable parameters, and its operating status and parameter settings have a crucial impact on the quality of the final product and production efficiency. For example, in the pretreatment stage, it may include nodes such as gas compression, cooling, and drying; in the separation stage, it may involve nodes such as membrane separation or adsorption separation; and in the refining stage, it may include nodes such as condensation and evaporation. The operating parameters and status data of these nodes are the basis for realizing intelligent control.

[0015] After obtaining all the process nodes, it is necessary to extract the historical anomaly characteristics of each node. Historical anomaly characteristics refer to the characteristic manifestations when each process node has abnormal situations during the past production process, which may include equipment failures, parameter fluctuations exceeding the normal range, and excessive impurity content. By analyzing historical data, the characteristics of these abnormal situations can be extracted to form a set of historical anomaly characteristics. Exemplarily, first, collect the historical operation data of each process node from the production control system, which includes equipment parameters, sensor data, operation logs, etc. Then, clean and normalize the collected data to remove noise data and invalid data to ensure the accuracy and consistency of the data. Next, use anomaly detection algorithms, such as statistics-based anomaly detection and machine learning-based anomaly detection, to identify the abnormal points in the data and extract the characteristics of the identified abnormal points to form a set of historical anomaly characteristics for each process node. These characteristics may include the time when the anomaly occurred, the duration of the anomaly, and the changes in key parameters during the anomaly.

[0016] Traverse the M process nodes and extract their historical anomaly characteristics from each node respectively to form a set of historical anomaly characteristics for the M process nodes. This set is a data set that contains detailed records of various abnormal situations that occurred in each process node during past production. This historical data set is the basis for subsequent analysis and risk prediction and provides a basis for node anomaly degree evaluation. Through data mining and pattern recognition methods, representative abnormal characteristics can be extracted, thus providing accurate data support for subsequent anomaly degree identification and risk control.

[0017] Through the above steps, a set of historical anomaly characteristics for each process node can be obtained. This set provides important data support for subsequent node anomaly degree identification and credible window configuration and is the key basis for realizing intelligent control.

[0018] P20: Identify the node abnormality degrees for the set of historical abnormality characterizations of the M process nodes, obtain the M process node abnormality degrees, and configure M confidence windows for the M process nodes according to the M process node abnormality degrees.

[0019] Further, step P20 of the embodiment of the present application further includes: P21: Perform clustering of the same-type abnormality characterizations within the process nodes on the set of historical abnormality characterizations of the M process nodes respectively to determine M clusters of historical abnormality characterizations of the clustered process nodes; P22: Traverse the M clusters of historical abnormality characterizations of the clustered process nodes for weighted analysis to determine the M process node abnormality degrees.

[0020] It should be understood that after the extraction of the set of historical abnormality characterizations of the M process nodes of the purification production line is completed, it is necessary to conduct in-depth analysis on these abnormality characterizations next to identify the abnormality degree of each process node, that is, the node abnormality degree. This process is crucial for the subsequent confidence window configuration because the node abnormality degree can reflect the frequency and severity of abnormalities occurring in each process node during historical operation.

[0021] Specifically, first, perform clustering analysis on the set of historical abnormality characterizations of each process node. Clustering of the same-type abnormality characterizations within a process node means that for a specific process node, all abnormality characterizations (such as too high temperature, too low pressure, etc.) will be classified according to the abnormality degree. For example, if the abnormality characterization of a certain process node is "too high pressure", then it can be clustered according to different degrees of pressure exceeding the range (such as slightly exceeding the standard, severely exceeding the standard, etc.) to form multiple subsets. Each subset represents a specific type of abnormal situation. This process can utilize clustering algorithms, such as common data clustering methods like the K-means algorithm, DBSCAN, etc. The goal of clustering is to group abnormal characterizations with similar properties into one category to ensure the targeted and accurate identification of abnormal situations.

[0022] After such clustering analysis for each process node, M clusters of historical abnormality characterizations of the clustered process nodes will be generated. These clusters contain characterization data similar to the abnormal situations of the node, helping us to more accurately understand the abnormal patterns of the node and the frequency and severity of their occurrence. Different clusters can represent different types of abnormal performances of the process node during the historical process, such as the difference between minor abnormalities and severe abnormalities.

[0023] Next, weighted analysis is performed on the historical abnormal characterization clusters of the M clustering process nodes to determine the abnormality degree of each process node. Specifically, different types of abnormalities have different degrees of impact on process nodes. For example, some abnormalities may have a greater impact on product quality or equipment safety, while some have a smaller impact. Therefore, in the weighted analysis, a weight coefficient needs to be assigned to each clustering cluster according to the severity and occurrence frequency of each type of abnormality. The assignment of weights can be determined based on factors such as the severity of abnormal characterization, occurrence frequency, and the impact on product quality and production efficiency. For example, for those abnormal characterization clusters that occur frequently and have a greater impact on product quality, a higher weight can be assigned; while for those abnormal characterization clusters that occur occasionally and have a smaller impact, a lower weight can be assigned. Through the weighted analysis of each cluster, the impacts of different abnormal characterizations can be comprehensively considered, thereby obtaining the abnormality degree of each process node. This abnormality degree is a quantitative indicator that can intuitively reflect the abnormal situation of each process node during historical operation. The higher the value of the abnormality degree, the more frequent or severe the abnormal situation of the process node, and the higher the risk. The abnormality degree of each process node is dynamically adjusted, and as the process node operates and the monitoring data changes, the abnormality degree will be updated accordingly.

[0024] Finally, based on the abnormality degrees of the M process nodes calculated above, a credible window is configured for each process node. The credible window refers to the range of parameter fluctuations allowed for the process node during normal operation. Under normal circumstances, the operating parameters of the process node should be maintained within this window range, and behaviors outside this range will be regarded as abnormal and trigger the risk control mechanism. The configuration of the credible window not only depends on the abnormality degree but also takes into account the characteristics and historical performance of the process node, so as to ensure its safety and efficiency in actual production. In this way, a suitable monitoring and control strategy can be customized for each process node, thereby improving the stability of the entire purification process and product quality.

[0025] Furthermore, step P20 of the embodiment of the present application further includes: P23: Divide the abnormality degrees of the M process nodes by the sum of the abnormality degrees of the M process nodes respectively to obtain M process node credibility coefficients; P24: Multiply the M process node credibility coefficients by the standard credible window to obtain M credible windows.

[0026] Optionally, in a further implementation process, the configuration of the credible window can be further optimized and adjusted based on the calculated abnormality degree of the process node.

[0027] First, by calculating the ratio of the abnormality degree of each process node to the sum of the abnormality degrees of all process nodes, the credibility coefficient of each process node is obtained. Specifically, the purpose of this process is to transform the abnormality degree of each node into a relative value through normalization, so as to measure the relative risk degree of different nodes in the whole purification process. The credibility coefficient calculated in this way can reflect the proportion of the abnormal risk of each process node in the overall purification process. The lower the credibility coefficient, the fewer abnormal situations of the process node in historical operation, and the relatively stable its operating state; on the contrary, the higher the credibility coefficient, the relatively higher the abnormal risk of the process node, and more strict monitoring and control are required.

[0028] Next, multiply the credibility coefficient of each process node by the standard credibility window to obtain M credibility windows. The standard credibility window is usually a fixed reference value, which can be set according to actual production requirements and process requirements, and represents the range of operating parameters that each process node should maintain under normal operation. In this way, the credibility window of each process node can be dynamically adjusted according to its abnormal risk. For process nodes with lower abnormal risks, their credibility windows are relatively wide, which means that during this time period, the operating state of the node is relatively stable, and the monitoring frequency can be appropriately relaxed; while for process nodes with higher abnormal risks, their credibility windows are relatively narrow, and more frequent monitoring and control are required to ensure that their operating states are always within the controllable range.

[0029] Through the above steps, the embodiment of the present application can not only identify the abnormality degree of each process node, but also further calculate the credibility coefficient according to the abnormality degree, and configure the credibility window suitable for each process node accordingly, so as to realize the refined and intelligent control of the purification process, effectively improve the stability of the entire purification system and the product quality, and at the same time reduce the risks in the production process.

[0030] P30: Respectively use the M credibility windows as the monitoring periods, continuously monitor the operation of the M process nodes and conduct operation risk analysis to obtain a sequence of M process node window operation monitoring data sets and a sequence of M process node window risk coefficients, where each process node window risk coefficient and the corresponding process node window operation monitoring data set have a credible time period identifier.

[0031] Furthermore, step P30 of the embodiment of the present application further includes: P31: When the purification production line is started, obtain the first monitoring time point; P32: At the first monitoring time point, perform node operation monitoring on the M process nodes to obtain M sets of first process node window operation monitoring data, and perform operation risk analysis on the M sets of first process node window operation monitoring data to obtain M first process node window risk coefficients; P33: Using the M confidence windows as the monitoring period, determine M second monitoring time points in combination with the first monitoring time point; P34: Based on the M second monitoring time points, perform node operation monitoring on the M process nodes to obtain M sets of second process node window operation monitoring data, and perform operation risk analysis on the M sets of second process node window operation monitoring data to obtain M second process node window risk coefficients; P35: By analogy, using the M confidence windows as the monitoring period, continuously monitor the M process nodes and perform operation risk analysis to obtain the sequence of M sets of process node window operation monitoring data and the sequence of M process node window risk coefficients.

[0032] It should be understood that after the confidence window configuration of the M process nodes of the purification production line is completed, the next step is to continuously monitor the node operation and perform operation risk analysis on these process nodes. Using the M confidence windows as the monitoring period, each node is monitored in real time to ensure that it operates within the normal working range.

[0033] When the purification production line is started, the system will automatically record the start time and use it as the first monitoring time point, that is, the starting monitoring point. Traverse the M process nodes for comprehensive node operation monitoring, collect the operation parameters and status information of each process node, such as key indicators like temperature, pressure, flow rate, gas composition, etc., to form M sets of first process node window operation monitoring data. Subsequently, use a pre-set risk assessment model to perform operation risk analysis on these monitoring data sets, calculate the risk coefficient of each process node at this time point, and thus obtain M first process node window risk coefficients. The calculation of the risk coefficient will be based on multiple factors, such as the degree of parameter deviation from the normal range, the stability of equipment operation, etc., to quantitatively reflect the operation risk level of each process node.

[0034] Next, use the completion point of the first round of monitoring starting from the first monitoring time point as the starting point of the second round of monitoring, and use the M confidence windows as the monitoring period to calculate and determine M second monitoring time points. The monitoring period is set according to the confidence windows of each process node to ensure that each node is monitored at an appropriate frequency during operation to promptly capture any potential risks. The determination of the second monitoring time point not only depends on the setting of the confidence window but also takes into account the actual operation conditions of the process nodes to ensure that the monitoring of each node is flexible and meets the actual requirements.

[0035] Next, at M second monitoring time points, a second round of node operation monitoring is performed on M process nodes. This monitoring process is similar to step P32. At the second monitoring time point, the system again performs node operation monitoring on M process nodes, collects new operation parameters and status information, forms a set of M second process node window operation monitoring data, and performs operation risk analysis on these data sets to obtain M second process node window risk coefficients. This process is similar to the monitoring and analysis at the first monitoring time point, but the time points are determined according to the trusted window period, ensuring the periodicity and continuity of the monitoring.

[0036] And so on, steps P33 and P34 will continue to repeat. The system will continue to perform continuous monitoring and risk analysis on each process node in each subsequent monitoring cycle based on the completion time point of the previous monitoring cycle. Each new monitoring cycle will generate a new data set and a sequence of risk coefficients, ensuring that the system can track the operation status of each process node in real time, evaluate its potential risks, and adjust the monitoring and risk control strategies according to the changes.

[0037] Moreover, the window risk coefficient of each process node and the corresponding set of window operation monitoring data have a trusted time period identifier. This identifier can ensure that the monitoring data and risk analysis results correspond to the corresponding trusted window, thus providing an accurate time reference for subsequent risk regulation. In this way, the system can grasp the operation status of each process node in real time, discover potential risk points in a timely manner, and provide important data support for subsequent risk regulation, ensuring the stability of the entire purification process and the product quality.

[0038] Furthermore, step P30 of the embodiment of the present application further includes: <> P30a: Obtain M first trusted time periods, and identify M first process node window risk coefficients and the corresponding M first process node window operation monitoring data sets, where the start time points of the M first trusted time periods are the first monitoring time points, and the durations of the M first trusted time periods are M trusted windows.

[0039] In a possible embodiment of the present application, the system will obtain M first trusted time periods, and identify M first process node window risk coefficients and the corresponding M first process node window operation monitoring data sets.

[0040] First, it is necessary to obtain M first credible time periods. The start time points of these credible time periods are determined by the first monitoring time point, that is, starting from the moment when the purification production line is started. The duration of these time periods is the same as the time range of M credible windows, that is, the monitoring period of each process node. Specifically, the credible windows of each process node have been configured with a suitable operating parameter range according to their abnormality degree and credibility coefficient, and these ranges determine the monitoring period of each process node.

[0041] Next, the M first process node window risk coefficients and the M first process node window operation monitoring data sets of each process node are identified. The purpose of this identification is to associate the monitoring data and risk coefficients of each process node within the specified first credible time period with that time period for subsequent analysis and processing. These data and risk coefficients represent the operating conditions and potential risks of each process node within the first credible time period.

[0042] Specifically, the system binds the monitoring data set of each process node and its corresponding risk coefficient to the corresponding first credible time period. The data and risk coefficients of each process node will be sorted according to the time period, so that the subsequent monitoring system can clearly identify which process nodes have abnormal operating states and which nodes have relatively high risk coefficients within a specific time period, thereby providing a basis for subsequent process regulation and risk warning.

[0043] This identification process enables the system to clearly distinguish the operating states of different process nodes in different time periods and enables dynamic monitoring and comparison based on historical data. When the subsequent monitoring period continues, the system can use these historical data as a reference to judge the change trend of the process nodes in different monitoring periods, so as to more accurately evaluate the risks and operating states of the process nodes.

[0044] P40: Combining the credible time period identification, risk regulation is performed on the M process nodes according to the M process node window risk coefficient sequences and the M process node window operation monitoring data set sequences.

[0045] Furthermore, step P40 of the embodiment of the present application further includes: P41: Extract the maximum value of the process node window risk coefficients in the sequence of M process node window risk coefficients, and use the corresponding process node as the risk process node; P42: According to the reliable time period identifier corresponding to the maximum value of the process node window risk coefficients, extract the M - 1 sets of operation monitoring data of the M - 1 cooperating process nodes with the same reliable time period in the sequence of M process node window operation monitoring data sets, and the operation monitoring data set of the risk process node window of the risk process node; P43: Perform risk regulation based on the operation monitoring data set of the risk process node window of the risk process node and the M - 1 sets of operation monitoring data of the M - 1 cooperating process nodes of the M - 1 cooperating process nodes.

[0046] It should be understood that, based on the previously determined reliable time period identifier, the sequence of M process node window risk coefficients, and the sequence of M process node window operation monitoring data sets, precise risk regulation is performed on the M process nodes to ensure that all process nodes can maintain a safe operating state during operation, and by comprehensively analyzing the mutual relationships between the nodes, the operating parameters are adjusted in a timely manner to avoid the impact of potential risks on the purification process.

[0047] First, compare the risks of each process node and select the process node with the highest risk in the current monitoring period. To this end, first extract the maximum value of the risk coefficients from the sequence of M process node window risk coefficients, which represents which process node has the highest risk level in the current monitoring period. The risk coefficient of each process node reflects the degree of deviation between the operating state of the node and the normal parameter range. The larger the value, the higher the risk. When the system identifies a risk process node, this node is used as the node that needs to be prioritized for attention and regulation in the entire purification process. The node with the largest risk coefficient may have a large deviation or potential failure, directly affecting product quality or system stability, so it needs to be adjusted with emphasis.

[0048] Next, according to the reliable time period identifier corresponding to the maximum value of the risk coefficient, extract the operation monitoring data sets of the other M - 1 process nodes within the same reliable time period from the sequence of M process node window operation monitoring data sets, as well as the operation monitoring data set of the risk process node itself. The purpose of this screening process is to obtain the operating conditions of other process nodes when the process node with the highest risk appears abnormal, so as to provide comprehensive data support for subsequent risk regulation. In this way, the system can understand whether there are potential abnormalities or parameters that need to be adjusted in other process nodes when the risk process node has problems, so as to perform coordinated regulation.

[0049] Finally, comprehensively considering the abnormal conditions of the risk process nodes and the operating states of other process nodes, by adjusting relevant process parameters, optimizing equipment operating conditions, or taking other necessary measures, the risk level of the risk process nodes is reduced, and the stable operation of the entire purification process is ensured. Specifically, the system will evaluate the interrelationships between process nodes based on this monitoring data and risk analysis. For example, if the risk coefficient of a risk process node increases due to excessive pressure, the system may adjust the operating parameters of relevant equipment, such as reducing the rotation speed of the compressor or increasing the power of the cooling system, and at the same time check whether the other cooperating process nodes need corresponding adjustments to ensure the coordinated operation of the entire system.

[0050] In this way, not only can the current risk nodes be identified, but also through considering the associations with other nodes, linkage control can be implemented to ensure the safety and stability of the purification process. This precise control method can effectively avoid the negative impact of a single node abnormality on the entire system, thereby improving the efficiency of the purification process and the quality of the product.

[0051] Furthermore, step P43 of the embodiment of the present application further includes: P43-1: Adjust the risk parameters of the risk process node for the risk process node window operation monitoring data set to obtain risk process node adjustment parameters; P43-2: Based on the connection relationship between the M process nodes, identify the adjustment parameters of the M-1 cooperating process nodes according to the risk process node adjustment parameters and the M-1 cooperating process node window operation monitoring data sets to obtain M-1 cooperating process node adjustment parameters; P43-3: Perform risk control according to the risk process node adjustment parameters and the M-1 cooperating process node adjustment parameters. Among them, adjusting the risk parameters of the risk process node for the risk process node window operation monitoring data set includes: pre-constructing an adjustment parameter identifier, and using the adjustment parameter identifier to identify the parameters of the risk process node window operation monitoring data set to obtain risk process node adjustment parameters.

[0052] Optionally, to achieve precise risk control, the stability and safety of the entire purification process can be ensured by adjusting the parameters of the risk process nodes and the cooperating process nodes.

[0053] First, perform risk parameter regulation on the operation monitoring data set of the risk process node to obtain the regulation parameters of the risk process node. By analyzing various parameters in the monitoring data set (such as pressure, temperature, flow rate, etc.), identify the main factors affecting the risk, and propose corresponding regulation measures. Exemplarily, use a pre-constructed regulation parameter identifier to conduct in-depth analysis of the monitoring data. This regulation parameter identifier can be constructed based on machine learning or data mining techniques, and can automatically identify the key parameters associated with the risk node, and predict the parameter values that need to be adjusted based on historical data. For example, if the risk process node is a gas pretreatment node, and the risk manifestation is poor water treatment effect, such as excessive moisture, the regulation parameter identifier may identify parameter adjustment suggestions such as increasing the cooling efficiency or enhancing the adsorption capacity of the molecular sieve.

[0054] After completing the identification of the regulation parameters of the risk process node, next, according to the connection relationship between the M process nodes, identify the regulation parameters of the M - 1 cooperating process nodes. The core of this operation is to analyze the relationship between the regulation parameters of the risk process node and the operation monitoring data set of the M - 1 cooperating process node windows, and identify which parameters of the cooperating process nodes need to be adjusted. Since there are usually close interactions between various process nodes, when a process node has a large risk, the operating states of other nodes may also be affected, or they may work together to mitigate the risk. For example, if the temperature of the risk process node is too high, the operations of cooling, filtering, adsorption, etc. of other nodes may need to be increased or adjusted. Therefore, based on the regulation parameters of the risk node, combined with the operation data of other cooperating nodes, calculate the regulation parameters of the M - 1 cooperating process nodes. These regulation parameters of the cooperating nodes will provide a basis for subsequent coordinated regulation to ensure the stability of the entire process chain.

[0055] In the middle, perform comprehensive regulation based on the regulation parameters of the risk process node and the regulation parameters of the M - 1 cooperating process nodes obtained in the first two steps to ensure that when an abnormality occurs in the risk process node, by adjusting the parameters of the cooperating process nodes, the impact of the risk can be minimized to the greatest extent. The specific regulation strategy will consider the mutual influence and dependency relationships between the nodes to ensure that all process nodes can cooperate together to achieve the best operating effect. For example, when an abnormality occurs in a certain process node, the operating conditions (such as temperature, pressure, flow rate, etc.) of its downstream or related nodes may need to be synchronously adjusted according to the adjusted regulation parameters to avoid an abnormality in one place causing failures in other nodes.

[0056] Specific examples of the coordination adjustment strategy include: when the risk manifestation at the gas pretreatment node is poor water treatment effect or ineffective removal of particulate matter during the coarse filtration process, the coordination adjustment strategy is to improve the cooling efficiency of the compression and cooling node, reduce the gas temperature, ensure that the condensed water is effectively removed, and prevent it from entering the downstream equipment; increase the adsorption capacity of the molecular sieve adsorption node or extend the adsorption cycle to ensure further removal of moisture and impurities in the gas; improve the filtration accuracy of the precision filtration and adsorption node to ensure effective removal of smaller particulate matter and dissolved substances. When the risk manifestation at the separation node is low separation efficiency, resulting in insufficient carbon dioxide purity, the coordination adjustment strategy is to optimize the pretreatment process to ensure that the gas quality entering the separation node meets the requirements; adjust the parameters of the refining process, such as increasing the condensation temperature or optimizing the evaporation process, to further improve the purity of carbon dioxide. When the risk manifestation at the refining node is impurity residue or purity fluctuation during the refining process, the coordination adjustment strategy is to optimize the separation process to ensure that the gas purity entering the refining node meets the requirements; strengthen the detection frequency, promptly detect and adjust the parameters during the refining process to ensure the quality of the final product.

[0057] Through this comprehensive regulation strategy, precise adjustment of each process node is achieved, ensuring that in high-risk situations, the coordinated adjustment of other process nodes can effectively guarantee the safe and stable operation of the entire purification process.

[0058] Furthermore, the embodiment of the present application further includes step P50: Determine the risk regulation feedback window, perform anomaly identification on M process nodes within the risk regulation feedback window, and generate a warning instruction according to the identification result.

[0059] Specifically, after completing the risk regulation of the M process nodes of the purification production line, anomaly identification of all M process nodes can be further performed through the risk regulation feedback window to ensure that the system can capture potential anomalies in real time during operation and issue an alarm in a timely manner when a risk occurs, so as to take corresponding measures for intervention and adjustment.

[0060] Specifically, first determine the risk regulation feedback window. This window refers to the time period used to evaluate the regulation effect and monitor new anomalies after the implementation of the risk regulation measures. The length of the feedback window can be set according to actual production requirements and process characteristics, and generally should cover the key operation cycle after the implementation of the regulation measures to ensure that the regulation effect can be comprehensively evaluated.

[0061] Within the risk control feedback window, the system will comprehensively identify anomalies in M process nodes. Exemplarily, a pre-set anomaly identification model can be utilized to analyze the operating status of each node in real time in combination with the operation monitoring data of each process node. The anomaly identification model can be constructed based on historical data, statistical analysis, or machine learning algorithms, and can identify anomalies that deviate from the normal operating status. For example, if the temperature or pressure parameters of a certain process node still exceed the normal range after regulation, or new abnormal fluctuations occur, these will be identified as abnormal situations.

[0062] According to the results of anomaly identification, corresponding warning instructions are generated. The warning instructions are response measures for the identified abnormal situations, aiming to remind the operators or the automated control system to take further inspection or regulation measures in a timely manner. The warning instructions can include, but are not limited to: issuing an alarm signal, prompting the operator to conduct an inspection, automatically adjusting relevant process parameters, or starting standby equipment, etc. These warning instructions can ensure that after risk control, the system can promptly detect and handle new abnormal situations, thereby maintaining the stability and safety of the entire purification process.

[0063] In addition, in some cases, if multiple process nodes simultaneously present anomalies or interrelated abnormal situations, the system may generate more complex warning instructions to coordinate the regulation actions of multiple nodes. This coordinated regulation can ensure that the entire system responds comprehensively when facing risks, thereby minimizing the impact of risks on the purification process to the greatest extent and improving the stability and safety of the system. Through this real-time monitoring and response mechanism, the system can detect and handle anomalies in the shortest possible time, ensuring the smooth progress of the purification process.

[0064] In summary, the embodiments of the present application at least have the following technical effects: The present application obtains M process nodes of the purification production line, extracts historical anomaly characteristics, identifies the anomaly degree of the nodes, and configures a credible window for each process node based on the anomaly degree. Based on the credible window, continuous monitoring is performed on each node, the operation risk is analyzed, and the operation monitoring data and risk coefficient sequence of the nodes are generated. The risk coefficient of each node and the corresponding monitoring data set are combined with the credible time period identifier to precisely control the risk of the M process nodes, ensuring the stable and safe operation of the purification process.

[0065] It achieves the technical effect of timely identifying and handling risks through a real-time monitoring and intelligent regulation mechanism, improving the stability and safety of the purification process.

[0066] Embodiment 2, based on the same inventive concept as the intelligent regulation method for the food-grade carbon dioxide purification process in the foregoing embodiment, such as Figure 2As shown in the figure, the present application provides an intelligent control system for the purification process of food-grade carbon dioxide. The system and method embodiments in the embodiments of the present application are based on the same inventive concept. Among them, the system includes: An abnormal characterization extraction module 11, which is used to obtain M process nodes of the purification production line, traverse the M process nodes to extract historical abnormal characterizations, and obtain a set of historical abnormal characterizations of the M process nodes.

[0067] A credible window configuration module 12, which is used to identify the node abnormality degree of the set of historical abnormal characterizations of the M process nodes, obtain the abnormality degrees of the M process nodes, and configure M credible windows for the M process nodes according to the abnormality degrees of the M process nodes.

[0068] An operation risk analysis module 13, which is used to respectively use the M credible windows as the monitoring period, continuously monitor the operation of the M process nodes and perform operation risk analysis, and obtain a sequence of window operation monitoring data sets of the M process nodes and a sequence of window risk coefficients of the M process nodes. Among them, each process node window risk coefficient and the corresponding process node window operation monitoring data set have a credible time period identifier.

[0069] A risk regulation module 14, which is used to combine the credible time period identifier and perform risk regulation on the M process nodes according to the sequence of window risk coefficients of the M process nodes and the sequence of window operation monitoring data sets of the M process nodes.

[0070] Furthermore, the credible window configuration module 12 is further used to perform the following steps: Cluster the abnormal characterizations of the same type within the process nodes for the set of historical abnormal characterizations of the M process nodes respectively to determine M clusters of historical abnormal characterizations of the process nodes; traverse the M clusters of historical abnormal characterizations of the process nodes for weighted analysis to determine the abnormality degrees of the M process nodes.

[0071] Furthermore, the credible window configuration module 12 is further used to perform the following steps: Divide the abnormality degrees of the M process nodes by the sum of the abnormality degrees of the M process nodes respectively to obtain the credible coefficients of the M process nodes; multiply the credible coefficients of the M process nodes by the standard credible window to obtain M credible windows.

[0072] Furthermore, the operation risk analysis module 13 is further used to perform the following steps: When the purification production line is started, a first monitoring time point is obtained; at the first monitoring time point, node operation monitoring is performed on the M process nodes to obtain M sets of first process node window operation monitoring data, and operation risk analysis is performed on the M sets of first process node window operation monitoring data to obtain M first process node window risk coefficients; with the M credible windows as the monitoring period, M second monitoring time points are determined in combination with the first monitoring time point; based on the M second monitoring time points, node operation monitoring is performed on the M process nodes to obtain M sets of second process node window operation monitoring data, and operation risk analysis is performed on the M sets of second process node window operation monitoring data to obtain M second process node window risk coefficients; and so on, with the M credible windows as the monitoring period, continuous monitoring and operation risk analysis are performed on the M process nodes to obtain the sequence of M sets of process node window operation monitoring data and the sequence of M process node window risk coefficients.

[0073] Further, the operation risk analysis module 13 is further configured to perform the following steps: Obtain M first credible time periods, and label the M first process node window risk coefficients and the corresponding M sets of first process node window operation monitoring data, where the start time points of the M first credible time periods are the first monitoring time point, and the duration of the M first credible time periods is the M credible windows.

[0074] Further, the risk regulation module 14 is further configured to perform the following steps: Extract the maximum value of the process node window risk coefficients in the sequence of M process node window risk coefficients, and use the corresponding process node as the risk process node; according to the credible time period label corresponding to the maximum value of the process node window risk coefficients, extract the M - 1 sets of M - 1 cooperating process node window operation monitoring data of the M - 1 cooperating process nodes in the same credible time period in the sequence of M sets of process node window operation monitoring data, and the risk process node window operation monitoring data set of the risk process node; perform risk regulation based on the risk process node window operation monitoring data set of the risk process node and the M - 1 sets of M - 1 cooperating process node window operation monitoring data of the M - 1 cooperating process nodes.

[0075] Further, the risk regulation module 14 is further configured to perform the following steps: Perform risk parameter regulation on the risk process node window operation monitoring data set to obtain risk process node regulation parameters; based on the connection relationship between M process nodes, identify regulation parameters for the M - 1 cooperating process nodes according to the risk process node regulation parameters and the M - 1 cooperating process node window operation monitoring data sets to obtain M - 1 cooperating process node regulation parameters; perform risk regulation according to the risk process node regulation parameters and the M - 1 cooperating process node regulation parameters.

[0076] Further, the risk regulation module 14 is further configured to perform the following steps: Pre - construct a regulation parameter identifier, and use the regulation parameter identifier to perform parameter identification on the risk process node window operation monitoring data set to obtain risk process node regulation parameters.

[0077] Further, the system further includes a risk regulation feedback module, configured to determine a risk regulation feedback window, perform anomaly identification on M process nodes in the risk regulation feedback window, and generate a warning instruction according to the identification result.

[0078] It should be noted that the above - mentioned sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above - mentioned specific embodiments of this specification are described. Additionally, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0079] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0080] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications 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 equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. An intelligent control method for the purification process of food-grade carbon dioxide, characterized in that, The method includes: Obtain M process nodes of the purification production line, traverse the M process nodes to extract historical anomaly characterizations, and obtain a set of historical anomaly characterizations of the M process nodes; Identify the node anomaly degrees of the set of historical anomaly characterizations of the M process nodes, obtain the anomaly degrees of the M process nodes, and configure M confidence windows for the M process nodes according to the anomaly degrees of the M process nodes; Taking the M confidence windows as monitoring periods respectively, continuously monitor the operation of the M process nodes and conduct operation risk analysis to obtain a sequence of sets of window operation monitoring data of the M process nodes and a sequence of window risk coefficients of the M process nodes. Among them, each window risk coefficient of a process node and the corresponding set of window operation monitoring data of the process node have a confidence time period identifier; Combined with the confidence time period identifier, conduct risk regulation on the M process nodes according to the sequence of window risk coefficients of the M process nodes and the sequence of sets of window operation monitoring data of the M process nodes.

2. The intelligent regulation method for the food-grade carbon dioxide purification process according to claim 1, characterized in that, Identify the node anomaly degrees of the set of historical anomaly characterizations of the M process nodes, including: Cluster the historical anomaly characterizations of the same type within the process nodes for the set of historical anomaly characterizations of the M process nodes respectively to determine M clusters of historical anomaly characterizations of the process nodes; Traverse the M clusters of historical anomaly characterizations of the process nodes for weighted analysis to determine the anomaly degrees of the M process nodes.

3. The intelligent control method for the food-grade carbon dioxide purification process according to claim 1, wherein, Configure M confidence windows for the M process nodes according to the anomaly degrees of the M process nodes, including: Divide the anomaly degrees of the M process nodes by the sum of the anomaly degrees of the M process nodes respectively to obtain the confidence coefficients of the M process nodes; Multiply the confidence coefficients of the M process nodes by the standard confidence window to obtain M confidence windows.

4. The intelligent regulation method for the food-grade carbon dioxide purification process according to claim 1, characterized in that, Taking the M confidence windows as monitoring periods respectively, continuously monitor the operation of the M process nodes and conduct operation risk analysis, including: When the purification production line starts, obtain the first monitoring time point; Monitor the operation of the M process nodes at the first monitoring time point to obtain M sets of first window operation monitoring data of the process nodes, and conduct operation risk analysis on the M sets of first window operation monitoring data of the process nodes to obtain M first window risk coefficients of the process nodes; Taking the M confidence windows as monitoring periods, determine M second monitoring time points in combination with the first monitoring time point; Based on the M second monitoring time points, monitor the operation of the M process nodes to obtain M sets of second window operation monitoring data of the process nodes, and conduct operation risk analysis on the M sets of second window operation monitoring data of the process nodes to obtain M second window risk coefficients of the process nodes; And so on, taking the M confidence windows as monitoring periods, continuously monitor the M process nodes and conduct operation risk analysis to obtain the sequence of sets of window operation monitoring data of the M process nodes and the sequence of window risk coefficients of the M process nodes.

5. The intelligent regulation method for the food-grade carbon dioxide purification process according to claim 4, characterized in that, Obtain M first trusted time periods, and identify the risk coefficients of M first process node windows and the corresponding M sets of operation monitoring data of the first process node windows. Among them, the starting time point of the M first trusted time periods is the first monitoring time point, and the duration of the M first trusted time periods is M trusted windows.

6. The intelligent regulation method for the food-grade carbon dioxide purification process according to claim 1, characterized in that Combined with the trusted time period identification, perform risk regulation on M process nodes according to the sequence of risk coefficients of M process node windows and the sequence of M sets of operation monitoring data of process node windows, including: Extract the maximum risk coefficient of the process node window in the sequence of risk coefficients of the M process node windows, and use the corresponding process node as the risk process node; According to the trusted time period identification corresponding to the maximum risk coefficient of the process node window, extract M - 1 sets of operation monitoring data of M - 1 cooperating process nodes in the same trusted time period in the sequence of M sets of operation monitoring data of process node windows, and the set of operation monitoring data of the risk process node window of the risk process node; Perform risk regulation based on the set of operation monitoring data of the risk process node window of the risk process node and the M - 1 sets of operation monitoring data of M - 1 cooperating process nodes.

7. The intelligent regulation method for the food-grade carbon dioxide purification process according to claim 6, characterized in that Performing risk regulation based on the set of operation monitoring data of the risk process node window of the risk process node and the M - 1 sets of operation monitoring data of M - 1 cooperating process nodes includes: Perform risk parameter regulation on the set of operation monitoring data of the risk process node window to obtain risk process node regulation parameters; Based on the connection relationship between M process nodes, identify regulation parameters for the M - 1 cooperating process nodes according to the risk process node regulation parameters and the M - 1 sets of operation monitoring data of the cooperating process nodes, to obtain M - 1 sets of cooperating process node regulation parameters; Perform risk regulation according to the risk process node regulation parameters and the M - 1 sets of cooperating process node regulation parameters.

8. The intelligent regulation method for the food-grade carbon dioxide purification process according to claim 7, characterized in that, Pre - construct a regulation parameter identifier, and use the regulation parameter identifier to perform parameter identification on the set of operation monitoring data of the risk process node window to obtain risk process node regulation parameters.

9. The intelligent regulation method for the food-grade carbon dioxide purification process according to claim 1, wherein, Including: Determine a risk regulation feedback window, perform anomaly identification on M process nodes in the risk regulation feedback window, and generate a warning instruction according to the identification result.

10. An intelligent control system for the purification process of food-grade carbon dioxide, characterized in that, The system includes: Anomaly characterization extraction module, which is used to obtain M process nodes of the purification production line, traverse the M process nodes to extract historical anomaly characterizations, and obtain a set of historical anomaly characterizations of M process nodes; Trusted window configuration module, which is used to identify node anomaly degrees for the set of historical anomaly characterizations of M process nodes, obtain M process node anomaly degrees, and configure M trusted windows for M process nodes according to the M process node anomaly degrees; An operating risk analysis module, which is used to respectively use the M trusted windows as the monitoring period to continuously monitor the operation of M process nodes and conduct operating risk analysis, so as to obtain an M-process-node window operation monitoring data set sequence and an M-process-node window risk coefficient sequence. Among them, each process-node window risk coefficient and the corresponding process-node window operation monitoring data set have a trusted time period identifier; A risk regulation module, which is used to combine the trusted time period identifier and perform risk regulation on the M process nodes according to the M process-node window risk coefficient sequence and the M process-node window operation monitoring data set sequence.

Images