A production quality control method, equipment and medium for a plant protein beverage

Abstract

The embodiments of this specification disclose a production quality control method, device and medium for plant protein beverages, which relate to the technical field of plant protein beverages. The method includes: obtaining the production line parameters of the current beverage production line, based on which an allergen cross-contact risk analysis is carried out to determine at least one target allergen quality inspection node in the current beverage production line; collecting in real time the allergen quality inspection data of each target allergen quality inspection node to determine the real-time beverage quality index parameters corresponding to each target allergen quality inspection node; when the preset conditions are met, based on the real-time beverage quality index parameters and the previously obtained real-time production line parameters, a traceability analysis of the reasons for allergen exceeding the standard is carried out to determine the exceeding-standard evaluation data of the target allergen quality inspection node, and through a pre-constructed allergen component control knowledge base, corresponding production control strategies are determined according to the exceeding-standard evaluation data, so as to optimize the production quality through the production control strategies.

Description

Technical Field

[0001] This specification relates to the technical field of plant protein beverages, and particularly relates to a production quality control method, equipment and medium for plant protein beverages. Background Art

[0002] Plant protein beverages are milk-like beverages prepared from plant kernels, fruits, and soybeans (such as soybeans, peanuts, almonds, walnuts, coconuts, etc.) through processing, blending, and then high-pressure sterilization or aseptic packaging. Plant protein beverages are typical beverage products based on allergenic raw materials. In plant protein beverage processing enterprises, different types of beverages are produced in the same factory. During the production process of different types of plant protein beverages, there is a risk of cross-contact of allergens. For example, when the production lines of allergenic raw materials and non-allergenic raw materials are close, allergenic raw materials may fall into the production line of non-allergenic raw materials, increasing the risk of cross-contact of allergens. Since some production processes are common in the processing technologies of plant protein beverages with different types of raw materials, the common production processes are produced using the same production line. For example, almond raw materials and oat raw materials are processed using the same grinding equipment and blending tank, etc., which has the characteristics of flexible production lines, high daily production capacity per single product, and low equipment investment. However, it will also cause problems of cross-contact of allergens due to the residual allergenic raw materials in the equipment. Therefore, in the processing of plant protein beverages, in addition to the situation where production lines are close, there will also be a situation where some processes share production lines. In the production scenario of producing different types of plant protein beverages in the same factory, the risk of cross-contact of allergens is increased. Even non-allergenic raw materials will be affected by cross-contact of allergens during the processing, and there is a risk of accidentally introducing allergic components into other types of food and beverages during the processing.

[0003] In the traditional beverage production process, after the beverage is produced, a unified quality inspection is carried out on the beverage, and only the finally unqualified products are removed. In the complex scenario of producing multiple plant protein beverages in the same factory, it is difficult to detect the problem of cross-contact of allergens in a timely manner by the traditional method. Once it is found that the allergen exceeds the standard in the final inspection link, a large number of products may have been contaminated before, resulting in the inability to intercept products containing allergens in a timely manner during the production process, increasing the number of unqualified products; in addition, the particularity of allergen exceeding the standard is not considered. Different from other quality problems, if it is found that the allergen exceeds the standard in the previous process, the allergen activity can be eliminated or reduced without affecting the product stability and taste.

[0004] Therefore, in the production scenario of producing different types of plant protein beverages in the same factory, due to the influence of cross-contact of allergens during the processing, a unified quality inspection is carried out on the beverage after it is produced, which increases the number of unqualified products and affects the product quality of beverage products made from non-allergenic raw materials. Summary of the Invention

[0005] One or more embodiments of this specification provide a production quality control method, device and medium for plant protein beverages to solve the following technical problems: In the production scenario of producing different types of plant protein beverages in the same factory, due to the influence of cross-contact of allergens during the processing, after the beverages are produced, unified quality inspection is carried out on the beverages, which increases the number of unqualified products and affects the product quality of beverage products made from non-allergen raw materials.

[0006] One or more embodiments of this specification adopt the following technical solutions:

[0007] One or more embodiments of this specification provide a production quality control method for plant protein beverages. The method includes: obtaining the production line parameters of the current beverage production line, based on the production line parameters, analyzing the risk of cross-contact of allergens for the current beverage production line, and determining at least one target allergen quality inspection node in the current beverage production line; collecting in real time the allergen quality inspection data of each target allergen quality inspection node, and determining the real-time beverage quality index parameters corresponding to each target allergen quality inspection node through the allergen quality inspection data; when the real-time beverage quality index parameters meet the preset conditions, based on the real-time beverage quality index parameters and the previously obtained real-time production line parameters, tracing and analyzing the reasons for allergen over-standard, and determining the over-standard evaluation data of the target allergen quality inspection node, where the over-standard evaluation data includes at least one reason for allergen over-standard and the predicted amount of allergen over-standard corresponding to each over-standard reason; through a pre-constructed allergen component control knowledge base, based on the over-standard evaluation data of the target allergen quality inspection node, determining the corresponding production control strategy to optimize the production quality through the production control strategy, where the production control strategy includes a preventive node control strategy and a remedial node control strategy.

[0008] One or more embodiments of this specification provide a production quality control device for plant protein beverages, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0009] A non-volatile computer storage medium provided by one or more embodiments of this specification stores computer-executable instructions, and the computer-executable instructions are set to: execute the above method.

[0010] The above at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects: Through the above technical solutions, production line parameters are obtained and allergen cross-contact risk analysis is carried out. Based on the risk analysis, at least one target allergen quality inspection node is determined. Based on the quantified cross-contact risk index, risk production nodes are determined. By setting clear preset conditions, production nodes with actual risks are screened out, avoiding over-concern about the entire production line or missing high-risk areas, and providing accurate targets for subsequent quality control and preventive measures. When the real-time beverage quality index parameters meet the preset conditions, traceability analysis is carried out by combining the real-time beverage quality index parameters and real-time production line parameters, which can accurately find out the specific reasons for allergen over-standard. The over-standard evaluation data not only clarifies the reasons for allergen over-standard, but also gives the predicted amount of allergen over-standard corresponding to each reason. By obtaining real-time production line parameters, real-time production risk identification is carried out on the previous production node of the target allergen quality inspection node, comprehensively considering the three main sources of allergen risks: cross-contact in the production line, cross-contact at nodes, and introduction during the production process. In the production scenario of producing different types of plant protein beverages in the same factory, not only the allergen cross-contact risks that may occur between different production lines and between different batches at the same production node are concerned, but also the risk that non-allergen raw materials may produce allergen components due to the processing technology itself is considered, realizing a full coverage of various potential allergen risks in the production process. The preventive node control strategy in the production control strategy takes measures in advance against potential factors that may lead to allergen over-standard based on the risk analysis of the production process. The remedial node control strategy provides effective solutions for products with allergen over-standard conditions, can save the problem beverages, reduce the economic losses caused by non-conformities, and at the same time ensure that the beverage quality, taste, etc. meet the standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings:

[0012] Figure 1 is a schematic flowchart of a production quality control method for a plant protein beverage provided by an embodiment of this specification;

[0013] Figure 2 is a schematic structural diagram of a production quality control device for a plant protein beverage provided by an embodiment of this specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] To enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.

[0015] The embodiments of this specification provide a method for controlling the production quality of a plant protein beverage. It should be noted that the execution subject in the embodiments of this specification can be a server or any device with data processing capabilities. Figure 1 It is a schematic flowchart of a method for controlling the production quality of a plant protein beverage provided by the embodiments of this specification, as Figure 1 shown, mainly including the following steps:

[0016] Step S101, obtain the production line parameters of the current beverage production line, and based on the production line parameters, conduct an analysis of the risk of allergen cross-contact for the current beverage production line to determine at least one target allergen quality inspection node in the current beverage production line.

[0017] In an embodiment of this specification, when the current beverage production line is put into operation, the production line parameters of the current beverage production line are obtained. The production line parameters here include the production log corresponding to the current beverage production line, and the production records of each production node are stored in the production log; in addition, in addition to the production log, it also includes the node positions of each production node in the current beverage production line.

[0018] In traditional beverage production lines or other types of product production lines, quality inspection processes are set during the finished product production stage or after each production node. When setting the quality inspection process during the finished product production stage, in the complex scenario of producing multiple plant protein beverages in the same factory, it is difficult to detect allergen cross-contact problems in a timely manner. Once it is found that the allergen exceeds the standard in the final inspection link, a large number of products may have been contaminated before, resulting in the inability to intercept products containing allergens in a timely manner during the production process and increasing the number of unqualified products. When setting the quality inspection process after each production node, the quality inspection workload is increased. Equalization processing is carried out for each production node. Not all production nodes have the same risk of allergen cross-contact. Detecting production nodes without the risk of allergen cross-contact increases additional production losses. Therefore, in the complex scenario of producing multiple types of plant protein beverages in a factory, target allergen quality inspection nodes should be set according to the actual situation of each beverage production line. The allergen cross-contact risks of different production lines and production nodes are different. By setting target allergen quality inspection nodes, it is possible to focus on detecting the key links where allergen cross-contact is most likely to occur.

[0019] In one embodiment of this specification, through the production log in the production line parameters, the previous batch production information corresponding to each production node in the current beverage production line is obtained. Among them, the previous batch production information includes the previous batch of beverage raw material information and the previous batch production time. During the beverage production process, the production log is data that records production activities, which details the information of each link from raw material input to product output. An enterprise resource planning (ERP) system or a dedicated production management software can be used to record and manage the production log. According to the layout and process of the production line, the identifier of each production node is determined. Here, the production node can be a specific production device (such as a mixing tank, sterilizer, filling machine, etc.). In the production log, each production node corresponds to a corresponding record identifier for accurate identification and query. According to the identifier of the production node, the corresponding previous batch production record is searched in the production log, that is, the previous production record of the production line. Through index information such as chronological order or batch number, the record of the previous production of each production node is located, and the previous batch of beverage raw material information is extracted from these records, including detailed information such as the type, source, batch number, and usage amount of the raw materials. At the same time, the previous batch production time is obtained. Here, the production time includes the specific date and time.

[0020] Obtain the node location information of each production node in the production line parameters within the production park. In a modern production park, Geographic Information System (GIS) technology or a production management system is usually used to record and manage the location information of production equipment and production nodes. For example, in a production management system, each production node has corresponding coordinate information (X, Y) indicating its specific location within the production park. Based on this node location information, determine the allergen flow assessment area. It should be noted that the scope of the assessment area is determined according to the understanding of the allergen transmission characteristics. When the production line of allergen raw materials is close to that of non-allergen raw materials, the allergen raw materials may fall into the non-allergenic raw material production line. Generally, in a well-ventilated condition and in the direction of air flow, the transmission distance of allergens is the largest. Therefore, the maximum transmission distance of allergens can be used as the assessment radius R1 of the allergen flow assessment area through empirical data or test data. Set a circular area with the production node as the center and a radius of R1 as the allergen flow assessment area. Determine the adjacent production nodes of the adjacent production lines located within this allergen flow assessment area through the production line distribution in the production park. And determine the node distance between the two through the node locations of the adjacent production nodes and the production nodes in the current beverage production line. When determining the node distance between the two, the Euclidean distance formula can be used. Generally, the smaller the node distance, the greater the risk of allergen raw materials falling into the non-allergenic raw material production line. If there are no adjacent production nodes of other production lines within the allergen flow assessment area, it indicates that there is almost no risk of allergen raw materials falling into this production node. Determine the risk production nodes in the current beverage production line based on the previous batch production information corresponding to each production node and the node distance, and determine the target allergen quality inspection node with the quality inspection node after the risk production node.

[0021] Through the above technical solutions, by obtaining the previous batch production information and node location information, it is possible to comprehensively and accurately evaluate the risk of allergen cross-contamination during the production process, targeting the specific situation of each production node, including the previously used raw materials and the potential risks related to its location; using the quality inspection node after the risk production node as the target allergen quality inspection node avoids comprehensive and unfocused quality inspections of the entire production line. The targeted method of determining the quality inspection node can concentrate limited quality inspection resources on the links most likely to have problems, improving the quality inspection efficiency; the method of comprehensively considering the production history (previous batch production information) and spatial layout (node location information) can more effectively control the cross-contamination of allergens between different production lines. By determining and monitoring the risk production nodes and target allergen quality inspection nodes, it is possible to detect and handle the situation of allergen exceeding the standard in a timely manner, ensuring that the products meet the allergen safety standards.

[0022] Determine the risk production nodes in the current beverage production line according to the production information of the previous batch corresponding to each production node and the node distance, specifically including: evaluating the cross-contact risk of adjacent production lines of the production node through the node distance and the regional radius corresponding to the allergen flow evaluation area, and determining the corresponding cross-contact risk index of the production line of the production node; determining the node cross-contact risk coefficient of the production node according to the raw material attributes of the beverage raw material information of the previous batch corresponding to each production node, where the raw material attributes include allergenic raw materials and non-allergenic raw materials; using the production time of the previous batch corresponding to each production node to determine the current node processing duration corresponding to the production node, and based on the current node processing duration and the preset allergen component processing duration threshold corresponding to the production node, determining the corresponding node cross-contact risk quantification parameter of the production node; based on the node cross-contact risk coefficient of the production node and the corresponding node cross-contact risk quantification parameter of the production node, determining the corresponding node cross-contact risk index of the production node; generating a cross-contact risk index according to the cross-contact risk index of the production line and the node cross-contact risk index, and determining the risk production nodes in the current beverage production line where the cross-contact risk index meets the preset conditions.

[0023] In an embodiment of the present specification, the smaller the node distance between the production node and the adjacent production node, the greater the risk that the allergenic raw material falls into the non-allergenic raw material production line. Therefore, the cross-contact risk of adjacent production lines of the production node is evaluated by the ratio of the node distance to the regional radius corresponding to the allergen flow evaluation area, and 1 - ratio is determined as the cross-contact risk index of the production line corresponding to this production node. That is to say, the cross-contact risk index of the production line is used to represent the influence of the adjacent production line on this production node in the current production line. Assuming that the adjacent production line produces with allergenic raw materials, the higher the cross-contact risk index of the production line, the easier it is for the allergenic raw materials in the adjacent production line to have a cross-contact impact on this production node in the current production line.

[0024] In addition, the raw material attributes are included in the beverage raw material information of the previous batch corresponding to each production node. Here, the raw material attributes include two types: allergenic raw materials and non-allergenic raw materials. The node cross-contact risk coefficient of the production node is determined through the raw material attributes of this production node during the previous batch production process. When the raw material attribute is an allergenic raw material, it indicates that allergen residues will be generated during the previous batch production process, resulting in a node cross-contact risk in the previous and subsequent production processes of this node. Therefore, the corresponding node cross-contact risk coefficient is 1; when the raw material attribute is a non-allergenic raw material, since the raw material is a non-allergenic raw material, the previous batch production process has no impact on the current production node. Therefore, the corresponding node cross-contact risk coefficient is 0.

[0025] Determine the current node processing duration corresponding to each production node based on the previous batch production time of the production node. That is to say, it is the duration between the previous batch production time of this production node during the previous batch production process and the current production process of this production node. Generally, if the allergen raw material was processed by equipment A during the previous batch production, in order to avoid obvious cross - contamination during the subsequent processing of non - allergen raw materials by equipment A, usually a long - time equipment cleaning operation is carried out on equipment A during production change. The longer the equipment cleaning operation duration, the better the corresponding cleaning effect, that is, the less the residue of the allergen raw material.

[0026] Obtain the minimum duration threshold corresponding to the equipment cleaning operation. Here, the minimum duration threshold is the minimum value under the condition of ensuring the cleaning effect, and determine the threshold of the allergen component processing duration corresponding to the production node. Obtain the current cumulative production duration of the production node during the current batch production process, and calculate the difference between the current node processing duration and the current cumulative production duration. Here, the difference can represent the equipment cleaning operation duration corresponding to the production change after the previous batch production process; then, calculate the ratio of the difference to the threshold of the allergen component processing duration, and determine the node cross - contact risk quantification parameter corresponding to the production node in the way of 1 - ratio. It should be noted that the smaller the current node processing duration here, the larger the corresponding node cross - contact risk quantification parameter, and the higher the corresponding node cross - contact risk, indicating that the production change cleaning time is short and the risk of allergen residue is high. Determine the node cross - contact risk index corresponding to the production node through the product of the node cross - contact risk coefficient of the production node and the node cross - contact risk quantification parameter corresponding to the production node.

[0027] Generate a cross - contact risk index based on the sum of the production line cross - contact risk index and the node cross - contact risk index, and determine the risk production nodes in the current beverage production line where the cross - contact risk index meets the preset conditions. It should be noted that the preset condition here is the risk production nodes where the cross - contact risk index is greater than 0. After determining the risk production nodes, determine the target allergen quality inspection node with the quality inspection node located after this risk production node.

[0028] Through the above technical solution, by comprehensively generating the cross-contact risk index based on the cross-contact risk index of the integrated production line and the cross-contact risk index of the nodes, it is possible to comprehensively consider the cross-contact risk of allergens between adjacent production lines and different batches on the same production line during the production process. It not only considers the possibility of cross-contamination of allergens caused by the distance factor between adjacent production lines but also takes into account the risk of residual allergens due to incomplete cleaning during the production of different batches on the same equipment. By determining the risk production nodes based on the quantified cross-contact risk index and screening out the production nodes with actual risks through setting clear preset conditions, it avoids over-focusing on the entire production line or missing high-risk areas, providing a precise target for subsequent quality control and preventive measures.

[0029] Step S102: Real-time collect the allergen quality inspection data of each target allergen quality inspection node to determine the real-time beverage quality index parameters corresponding to each target allergen quality inspection node through the allergen quality inspection data.

[0030] In the prior art, there are various methods for quality inspection of allergens. For example, through immunochromatographic test strips, enzyme-linked immunosorbent assay (ELISA) kits, etc. However, the immunochromatographic test strip is based on the specific reaction of antigen-antibody. The specific antibody is fixed in the detection area of the test strip. When a sample containing an allergen is dropped onto the test strip, the allergen (antigen) in the sample will bind to the antibody, and the presence of the allergen is indicated by the color development of a label (such as colloidal gold). In addition, ELISA is based on the specific binding of antigen-antibody and amplifies the detection signal through an enzyme-labeled secondary antibody. Although ELISA kits have high sensitivity and accuracy, the operation is relatively complex and requires multiple steps of incubation, washing, and sample addition processes. The entire detection process usually takes several hours (about 2 - 4 hours), and professional equipment such as an enzyme-labeled immunosorbent assay reader is needed to read the results. Therefore, the traditional means of allergen quality inspection cannot be compatible in terms of allergen content and real-time performance.

[0031] In an embodiment of this specification, through near-infrared spectroscopy technology, the allergen quality inspection data of each target allergen quality inspection node is collected. Near-infrared spectroscopy mainly reflects the vibration information of hydrogen-containing groups (such as C-H, N-H, O-H). Components such as allergen proteins in plant protein beverages contain these groups, and different allergen components have specific absorption peaks in the near-infrared region. Therefore, the effect of real-time obtaining the allergen content can be achieved through near-infrared spectroscopy technology.

[0032] Based on the allergen quality inspection data, determine the real-time beverage quality index parameters corresponding to each target allergen quality inspection node, specifically including: obtaining the allergen quality inspection data, where the allergen quality inspection data is beverage spectral data; performing data cleaning and preprocessing operations on the beverage spectral data to determine the spectral processed data; using a pre-constructed regression prediction model to determine the corresponding real-time beverage quality index parameters based on the spectral processed data, where the real-time beverage quality index parameters include the allergen quality inspection content.

[0033] In one embodiment of the present specification, obtain the allergen quality inspection data, where the allergen quality inspection data is beverage spectral data; perform data cleaning and preprocessing operations on the beverage spectral data to determine the spectral processed data. Data cleaning of the beverage spectral data removes outliers and noise. Outliers may be data points that significantly deviate from the normal spectral range due to spectrometer failures or other factors, and noise can be removed through methods such as smoothing filtering, for example, using the moving average method, to improve the data quality. In order to convert the beverage spectral data into the allergen quality inspection content, a regression prediction model needs to be pre-constructed.

[0034] First, collect a large number of plant protein beverage samples with known allergen content. The allergen content of the samples can be accurately measured by enzyme-linked immunosorbent assay, covering different raw material sources, production batches, and possible allergen concentration ranges to ensure that the calibration model has good representativeness and generality. Use a near-infrared spectrometer to perform spectral scans on each sample to obtain near-infrared spectral data with a wavelength range usually between 780 - 2500 nm. During this process, ensure that the spectral acquisition conditions (such as optical path, temperature, number of scans, etc.) remain consistent to reduce the influence of external factors on the spectrum. Clean the collected original spectral data to remove outliers and noise to improve the data quality. To eliminate the influence of factors such as baseline drift and scattering effects on the spectrum, spectral preprocessing is usually required. Common preprocessing methods include standardization (such as mean centering, standard deviation standardization), normalization (such as maximum-minimum normalization), derivative processing (first derivative or second derivative), and multiplicative scatter correction (MSC), etc. Through the above preprocessing methods, the correlation between spectral features and allergen content can be enhanced, improving the accuracy of the model.

[0035] A partial least squares (PLS) regression model is established, and the collected sample data is divided into a training set and a validation set. The training set is used to construct the PLS model, usually accounting for 70%-80% of the total number of samples, and the validation set is used to evaluate the performance of the model, accounting for 20%-30% of the total number of samples. During this process, it is ensured that the sample distributions of the training set and the validation set can represent the characteristics of the entire data set. For example, they should have similar ranges and proportions in the distribution of allergen content. The core of PLS regression is to extract principal components to explain the relationship between spectral data (X variable) and allergen content (Y variable). In the training set, the covariance matrix of the spectral data matrix X is calculated and eigen-decomposed to obtain a set of principal components. These principal components are linear combinations of the original spectral variables and can maximize the explanation of the covariance between X and Y. To ensure the selection of an appropriate number of principal components, methods such as cross-validation are usually used to avoid overfitting or underfitting. Using the spectral data and corresponding allergen content data in the training set, the PLS model is fitted by the least squares method. That is, a regression equation between the principal components and the allergen content is established to determine the coefficients of the model. In this process, the model learns the internal relationship between spectral features and allergen content, so as to be able to predict the allergen content based on new spectral data. The established PLS model is evaluated using the validation set. Common evaluation indicators include root mean square error (RMSE), coefficient of determination (R²), and relative analysis error (RPD), etc. RMSE reflects the average error between the predicted value and the true value of the model, R² measures the fitting degree of the model to the data, and RPD is used to evaluate the prediction ability of the model. According to the evaluation results, the model is optimized, such as adjusting the number of principal components, trying different spectral preprocessing methods, or increasing the number of samples, etc., until satisfactory model performance is obtained.

[0036] In a small data processing unit beside the production line, the established PLS calibration model is embedded in the chemometrics software. When the near-infrared spectrometer obtains new beverage spectral data, first, the data is preprocessed in the same way as in the model training stage, and then the preprocessed spectral data is input into the PLS model. The model will predict the allergen content in the beverage according to the relationship it has learned to determine the allergen quality inspection content. As the production process progresses and sample data accumulates, the model is periodically verified and updated. Due to factors such as production raw materials, process parameters, or spectrometer performance, the accuracy of the model may decline. By continuously adding new sample data, re-evaluating and optimizing the model, it can be ensured that the model maintains good prediction performance over a long period and accurately converts spectral data into allergen content values.

[0037] Through the above technical solution, the near-infrared spectroscopy technology can be used to collect the allergen quality inspection data of each target allergen quality inspection node in real time. Compared with traditional detection methods such as ELISA kits that take several hours to obtain results, the near-infrared spectroscopy technology can obtain data immediately, which enables the allergen content to be grasped in real time during the production process; perform data cleaning and preprocessing operations on the collected beverage spectral data to remove noise and interference information and improve data quality; then, through a pre-constructed regression prediction model, accurately determine the real-time beverage quality index parameters based on the processed spectral data, including the allergen quality inspection content. A series of operations from data collection to processing ensure the accuracy of the detection results.

[0038] Step S103, when the real-time beverage quality index parameters meet the preset conditions, trace the source of the allergen exceeding the standard according to the real-time beverage quality index parameters and the previously obtained real-time production line parameters, and determine the exceeding-standard evaluation data of the target allergen quality inspection node.

[0039] Among them, the exceeding-standard evaluation data includes at least one allergen exceeding-standard reason and the allergen exceeding-standard prediction amount corresponding to each such exceeding-standard reason;

[0040] In an embodiment of this specification, if the allergen quality inspection content in the real-time beverage quality index parameters of the target allergen quality inspection node is a non-zero value, it indicates that during the processing of non-allergen raw materials at the target allergen quality inspection node of the current beverage production line, the situation of allergen exceeding the standard has occurred. In this case, if the allergen quality inspection content is higher than the preset allergen content threshold, it indicates that there are relatively large allergen introduction problems at the production node corresponding to the target allergen quality inspection node. It should be noted that the preset allergen content threshold here can be obtained through experimental data or set according to empirical data, and is used to represent the maximum allergen content that can be eliminated or reduced in activity through subsequent production processes. That is to say, beyond this allergen content threshold, it is impossible to eliminate or reduce the activity through subsequent production processes. Therefore, when the allergen quality inspection content is greater than 0 and lower than the preset allergen content threshold, trace the source of the allergen exceeding the standard according to the real-time beverage quality index parameters and the previously obtained real-time production line parameters, and determine the exceeding-standard evaluation data of the target allergen quality inspection node. The exceeding-standard evaluation data includes at least one allergen exceeding-standard reason and the allergen exceeding-standard prediction amount corresponding to each such exceeding-standard reason.

[0041] Through the above technical solution, by real-time monitoring of the allergen quality inspection content in the real-time beverage quality index parameters of the target allergen quality inspection node, once this value is non-zero and higher than the preset allergen content threshold, it can quickly determine that there are significant allergen introduction problems at the production node, be able to detect abnormalities in a timely manner during the production process, prevent problem products from further flowing into subsequent production links, reduce potential losses. On the beverage production line, when the allergen content of a certain target allergen quality inspection node is detected to be higher than the preset allergen content threshold, the relevant production process can be immediately stopped to prevent the production of more unqualified products; the preset allergen content threshold, as a measurement standard, can not only judge whether it exceeds the standard, but also distinguish the severity of allergen introduction. When the allergen quality inspection content is greater than 0 but lower than this threshold, it indicates that the allergen exceeding the standard problem at this time can be controlled through subsequent quality control processes.

[0042] Based on the real-time beverage quality index parameters and the pre-acquired real-time production line parameters, conduct a traceability analysis of the reasons for allergen exceeding the standard, and determine the evaluation data for allergen exceeding the standard of the target allergen quality inspection node, specifically including: obtaining the real-time production line parameters, where the real-time production line parameters include the real-time processing technology data of each production node; according to the real-time production line parameters of the current beverage production line, conduct real-time production risk identification on the previous production node of the target allergen quality inspection node, and determine the production risk index set corresponding to the previous production node, where the production risk index set includes the production line cross-contact risk index, the node cross-contact risk index, and the production process introduction risk index; through the production risk index set and the real-time beverage quality index parameters, assign weights to each reason for allergen exceeding the standard to determine the predicted amount of allergen exceeding the standard corresponding to each reason for allergen exceeding the standard, where the reasons for allergen exceeding the standard include production line cross-contact, node cross-contact, and production process introduction.

[0043] In an embodiment of this specification, obtain the real-time production line parameters. The real-time production line parameters include the real-time processing technology data of each production node, such as heat treatment process parameters, chemical treatment process parameters, enzyme treatment process parameters. In addition to the real-time processing technology data, the real-time production line parameters also include the real-time production logs of each production node and the adjacent production line nodes of each production node within the allergen flow evaluation area. According to the real-time production line parameters of the current beverage production line, conduct real-time production risk identification on the previous production node of the target allergen quality inspection node, and determine the production risk index set corresponding to the previous production node. The production risk index set includes the production line cross-contact risk index, the node cross-contact risk index, and the production process introduction risk index; it should be noted that the previous production node of the target allergen quality inspection node refers to the output node of the inspection object of the quality inspection node.

[0044] In addition to the cross-contact of allergens between adjacent production lines and the cross-contact of allergens in the production process of adjacent batches at the same production node, in the processing process of certain production nodes, even non-allergen raw materials will have allergen components during the processing process, so the introduction of risk indicators in the production process is used to measure this risk. For example, heat treatment processes such as high-temperature sterilization and baking may change the protein structure in non-allergen raw materials and produce new allergen epitopes; in addition, when adding chemical reagents for processing, the chemical properties of the raw materials may be changed, resulting in the production of allergens. For example, in the process of adjusting the pH value of the beverage, some proteins may undergo structural changes due to changes in pH and generate allergenic components; in addition, enzymes are used in the processing of plant protein beverages to improve the functional properties of proteins, but if the type of enzyme or the processing conditions are inappropriate, allergens may be produced. For example, using certain proteases to hydrolyze plant proteins may produce some peptides with allergenic activity. Through the production risk indicator set and real-time beverage quality indicator parameters, weights are assigned to each cause of allergen exceeding the standard to determine the predicted amount of allergen exceeding the standard corresponding to each cause of allergen exceeding the standard. The causes of allergen exceeding the standard include cross-contact of production lines, cross-contact of nodes and introduction into the production process.

[0045] Through the above technical solution, by obtaining real-time production line parameters, real-time production risk identification is performed on the previous production node of the target allergen quality inspection node, and the three main allergen risk sources of production line cross-contact, node cross-contact and production process introduction are fully considered. Not only the risk of allergen cross-contact between different production lines and different batches at the same production node is paid attention to, but also the risk that the processing technology itself may cause non-allergen raw materials to produce allergen components. All-round coverage of various potential allergen risks in the production process is achieved; based on real-time processing technology data, real-time production logs and adjacent production line production nodes and other information, the risk status of specific production nodes can be accurately located; a set of production risk indicators is determined, and different types of risks are converted into specific risk indicators, such as production line cross-contact risk indicators, node cross-contact risk indicators and production process introduction risk indicators, so that the risk level can be quantified, which is convenient for intuitive comparison and analysis of different risks. After clarifying the cause of each allergen exceeding the standard and its corresponding predicted quantity of exceeding the standard, prevention and control strategies can be formulated in a targeted manner according to the severity and possibility of the risk.

[0046] In one embodiment of this specification, according to the real-time production line parameters of the current beverage production line, real-time production risk identification is performed on the previous production node of the target allergen quality inspection node, and a set of production risk indicators corresponding to the previous production node is determined, specifically including: According to the beverage raw material information in the current beverage production line, the standard values of process parameters corresponding to each production node are matched, so as to determine the overflow ratio of processing parameters corresponding to the production node through the real-time processing process data of the previous production node of the target allergen quality inspection node and the corresponding standard values of process parameters, so as to determine the corresponding production process introduction risk indicators. First, collect the ideal process parameter standard values of different plant protein beverage raw materials at each production node, which can come from the standard specifications of the beverage industry, the long-term accumulated production experience of enterprises, the research results of scientific research institutions, and the raw material processing guides provided by suppliers, etc. For example, for soy protein beverages, collect the appropriate grinding speed and temperature in the grinding link, the standard addition amount and pH value range of various additives in the blending link, the appropriate temperature and time in the sterilization link, etc., and build a database through the collected data. Through sensors and data acquisition devices installed on the production line, the processing process data of the previous production node of the target allergen quality inspection node are obtained in real time. The devices can include temperature sensors, pressure sensors, flow meters, component analyzers, etc. Matching the real-time collected processing process data with the corresponding standard values of process parameters in the database can be achieved through data processing software. According to the production node and raw material information, the corresponding standard values are searched in the database and associated with the real-time data. Processing parameter overflow includes positive overflow (real-time data is greater than the standard value) and negative overflow (real-time data is less than the standard value). For different process parameters, the impact of overflow may be different. For example, in the sterilization link, a temperature lower than the standard value may result in incomplete sterilization, while in the addition link of chemical additives, an addition amount higher than the standard value may cause abnormal chemical reactions and increase the risk of generating allergens. Therefore, here, it is determined by what kind of overflow (positive overflow or negative overflow) of a certain process parameter value will have an impact on the generation of allergens, that is, the overflow ratio can measure the risk of allergen introduction in the processing process. According to the nature of the process parameters, a suitable calculation method is determined to calculate the overflow ratio. For positive overflow, the overflow ratio = (real-time data - standard value) / standard value × 100%; for negative overflow, the overflow ratio = (standard value - real-time data) / standard value × 100%. For example, the standard addition amount of citric acid in the blending link of a certain protein beverage is 0.5% (mass fraction), and the real-time detected addition amount is 0.6%, then the positive overflow ratio = (0.6% - 0.5%) / 0.5% × 100% = 20%. The processing parameter overflow ratio is determined as the corresponding production process introduction risk indicator.

[0047] In addition, based on the real-time production logs of each production node in the real-time production line parameters, determine the first production timestamp of the previous allergen raw material corresponding to the previous production node of the target allergen quality inspection node, so as to determine the allergen processing duration of this production node. According to this allergen processing duration and the allergen component processing duration threshold corresponding to the previous allergen raw material, determine the corresponding real-time node cross-contact risk indicator. The real-time production logs record the detailed production activity information of each production node. First, it is necessary to parse the production logs to extract the records related to the previous production node of the target allergen quality inspection node. Among the extracted relevant records, determine the production record of the previous use of the allergen raw material by searching and identifying the raw material information. For example, there may be a record like "[Timestamp 1] - Start production with peanut raw material" in the log. By identifying "peanut raw material", it is determined that this is the production activity of the previous use of the allergen raw material, and the corresponding timestamp, that is, the first production timestamp, is recorded. After determining the first production timestamp, obtain the current production time, which can be achieved through the clock or timestamp recording function in the production system. The current time represents the time point when the current production process advances to the previous production node of the target allergen quality inspection node. Obtain the current cumulative production duration of the production node during the current batch production. According to the current time and the first production timestamp, the time difference is the sum of the allergen processing duration and the working duration of the current production line in processing the current raw material. Therefore, subtract the current cumulative production duration from the time difference to obtain the allergen processing duration. Calculate the difference between the current node processing duration and the current cumulative production duration. This difference can represent the equipment cleaning operation duration corresponding to the production change after the previous batch production. For example: Suppose the time when the previous use of the allergen raw material (such as peanut raw material) started production in this mixing tank was 9:00 am on November 10, 2024, that is, the first production timestamp was 2024-11-10 09:00:00. When the current production process advances to the previous production node of the target allergen quality inspection node (still this mixing tank), the time is 2:00 pm on November 10, 2024, that is, 2024-11-10 14:00:00. During the current batch production, from the start of the current batch production to this moment (2024-11-10 14:00:00), the cumulative production duration of this mixing tank in processing the current non-allergenic raw material is 3 hours. Calculate the time difference between the current time and the first production timestamp, that is, the time difference from 2024-11-10 09:00:00 to 2024-11-10 14:00:00, which is 5 hours. This 5 hours is actually the total duration experienced since the end of the previous production using peanut raw material until now, including the processing duration of the peanut allergen and the working duration of the current production line in processing the current non-allergenic raw material. Given that the current cumulative production duration is 3 hours, subtract the current cumulative production duration of 3 hours from the previously obtained time difference of 5 hours, that is, 2 hours.Therefore, the processing duration of allergens at this production node is 2 hours.

[0048] The threshold of the processing duration for allergen components is determined based on past production experience, relevant experimental studies, and understanding of allergen characteristics. Different allergen raw materials require different processing durations to effectively reduce or eliminate the risk of allergen residues due to their different physicochemical properties. For example, for peanut allergens, through multiple experiments and actual production verification, it is found that using a specific cleaning process, at least 3 hours of processing time is required to reduce the allergen residues to a safe level. Then 3 hours is the threshold of the processing duration for peanut allergen components. With the improvement of production processes, the introduction of new equipment, or in-depth research on allergens, the threshold of the processing duration for allergen components may need to be updated. When more efficient cleaning equipment or cleaning agents are adopted, the processing duration threshold for peanut allergens may be shortened to 2 hours, and this threshold data needs to be updated in a timely manner to ensure the accuracy of risk assessment. Subtract the threshold of the processing duration for allergen components from the processing duration of allergens, calculate the ratio of this difference to the threshold of the processing duration for allergen components, and determine the corresponding real-time node cross-contact risk indicator of the production node in the way of 1 - ratio. The higher the real-time node cross-contact risk indicator, the higher the node cross-contact risk.

[0049] In addition, determine the adjacent production line nodes of each production node in the real-time production line parameters within the pre-set allergen flow assessment area. Through the real-time production log of this production node, determine the current production time interval corresponding to this production node, so as to determine the start-up status of the adjacent production line node within this current production time interval. Based on this start-up status, determine the node cross-contact risk coefficient corresponding to this production node. For example, taking a circular area with a radius of 5 meters centered on each production node as the allergen flow assessment area. Within the allergen flow assessment area of node X, there is an adjacent production line (designated as production line B) producing almond protein beverage. The mixing tank production node (designated as node Y) of this production line is within the assessment area. Therefore, node Y is the adjacent production line node of node X. By checking the real-time production log of node X, it is determined that the current production time interval corresponding to node X is from 10:00 to 11:00 on December 15, 2024. Obtain the real-time production log of production line B again to determine that node Y is in the start-up production state within this time interval. Since node Y is in the start-up state within the current production time interval, it means that there is a possibility of allergen cross-contact with node X through air transmission, equipment connection, etc.; if the adjacent production line node is not started, the risk is relatively low, and the coefficient can be set to a value close to 0. According to the preset rules, when the adjacent production line node is started within the current production time interval, the node cross-contact risk coefficient is set to 0.8, 0.9 or 1; if it is not started, it is set to 0.2, 0.1 or 0, and the setting rules can be selected according to requirements. Here, 0.8 and 0.2 are taken as examples for illustration, and the node cross-contact risk coefficient corresponding to node X is 0.8.

[0050] By using the pre-acquired node distance between adjacent production line nodes and the production node, and the regional radius corresponding to the allergen flow assessment area, the risk quantification parameter of cross-contact between production lines corresponding to the production node is determined. Continuing with the previous example, the node distance between node X and node Y is 3 meters, and the radius of the allergen flow assessment area is 5 meters. Then the ratio of the node distance to the regional radius is 3 / 5 = 0.6. The larger this data is, it indicates that under the same startup state, the greater the node distance. Since the greater the node distance, the lower the corresponding cross-contact risk between production lines. Therefore, the risk quantification parameter of cross-contact between production lines of the risk production line is determined by the method of 1 - ratio, that is, 1 - 0.6 = 0.4. According to the product of the risk quantification parameter of cross-contact between production lines and the cross-contact risk coefficient of the node, 0.4 * 0.8 = 0.32 is obtained, and the real-time cross-contact risk index corresponding to node X is determined, thereby determining the real-time cross-contact risk index of the production line; through this index, the risk degree of the production node affected by cross-contact with adjacent production lines can be intuitively evaluated. That is to say, the smaller the node distance, the smaller the ratio of the corresponding node distance to the regional radius, and the larger the corresponding risk quantification parameter of cross-contact between production lines of the risk production line, and the higher the risk of falling into; in this case, if the adjacent production line production node is in the startup state, compared with the non-startup state, the cross-contact risk coefficient of the node is large, and the corresponding real-time cross-contact risk index of the production line is large.

[0051] Through the production process introduction risk index, real-time node cross-contact risk index, and the real-time cross-contact risk index of the production line obtained through the above process, the production risk index set corresponding to the previous production node is determined.

[0052] In an embodiment of the present specification, weights are assigned to each allergen exceeding standard cause through the production risk index set and the real-time beverage quality index parameters to determine the predicted amount of allergen exceeding standard corresponding to each allergen exceeding standard cause, specifically including: establishing a mapping relationship between each allergen exceeding standard cause and the corresponding risk index. The mapping relationship in the corresponding Embodiment 1 here is that the risk index corresponding to cross-contact between production lines is the real-time cross-contact risk index of the production line, the risk index corresponding to node cross-contact is the real-time node cross-contact risk index, and the risk index corresponding to production process introduction is the production process introduction risk index.

[0053] In addition to the above methods, it can also be achieved through Embodiment 2: Collect detailed data on the situation of allergen exceeding the standard in the historical production process, including the reasons for allergen exceeding the standard of different types (cross-contact between production lines, cross-contact at nodes, introduction during the production process, etc.) and the corresponding production risk indicators (cross-contact risk indicator between production lines, cross-contact risk indicator at nodes, introduction risk indicator during the production process). Conduct in-depth analysis on the above data to determine the internal relationship between each reason for allergen exceeding the standard and each risk indicator. Use statistical analysis methods, such as correlation analysis, to quantify the relationship between the reason for allergen exceeding the standard and each risk indicator. For example, through data analysis, it is found that there is a positive correlation between the allergen exceeding the standard caused by cross-contact between production lines and the cross-contact risk indicator between production lines, and it generally shows a certain proportional relationship. Based on the above analysis results, establish a mapping relationship model between each reason for allergen exceeding the standard and the corresponding risk indicator. Establish a functional relationship, such as the predicted value of allergen exceeding the standard caused by cross-contact between production lines = f (cross-contact risk indicator between production lines). This function can be linear (such as y = kx + b) or non-linear, and the specific form depends on the analysis results of the actual data.

[0054] According to the set of production risk indicators, process multiple risk indicators in the set of production risk indicators to determine the risk proportion of each such risk indicator in the set of production risk indicators. Since the risk indicators in the above embodiments are all values greater than 0 and less than 1, here, the direct ratio method can be used. Let the cross-contact risk indicator value between production lines be x1, the cross-contact risk indicator value at nodes be x2, and the introduction risk indicator value during the production process be x3. Calculate the sum of the cross-contact risk indicator value between production lines being x1, the cross-contact risk indicator value at nodes being x2, and the introduction risk indicator value during the production process being x3. The corresponding proportion of the cross-contact risk indicator between production lines is x1 / (x1 + x2 + x3), and so on, to obtain the risk proportion of each such risk indicator in the set of production risk indicators. If the risk indicators in the above embodiments adopt other methods, then in this step, after normalizing the risk indicators, calculate the risk proportion again.

[0055] Based on the risk proportion of each such risk indicator and the real-time beverage quality indicator parameter, determine the predicted excess amount corresponding to each such risk indicator, and allocate the total excess amount of the allergen content in the real-time beverage quality indicator. First, calculate the difference between the real-time beverage quality indicator parameter and the preset standard of the allowable amount of allergen to determine the total excess amount of the allergen content, and calculate the product of the risk proportion of each such risk indicator and the total excess amount of the allergen content to obtain the predicted excess amount corresponding to each such risk indicator. And based on this mapping relationship and the predicted excess amount corresponding to each such risk indicator, determine the predicted excess amount of the allergen corresponding to each cause of the allergen exceeding the standard. In the case of the above-mentioned Embodiment 1, according to the corresponding relationship between the cause of the allergen exceeding the standard and the risk indicator in the mapping relationship, determine the predicted excess amount of the allergen corresponding to each cause of the allergen exceeding the standard as the predicted excess amount of the allergen corresponding to each such risk indicator.

[0056] In the case corresponding to the above-mentioned Embodiment 2, according to the mapping relationship established previously between each cause of the allergen exceeding the standard and the corresponding risk indicator, and the predicted excess amount corresponding to each risk indicator, determine the predicted excess amount of the allergen corresponding to each cause of the allergen exceeding the standard. For example, it is known that the predicted excess amount corresponding to the risk indicator of cross-contact on the production line is Q1. Through the mapping relationship "the predicted excess amount of the allergen caused by cross-contact on the production line = f(risk indicator of cross-contact on the production line)", substitute Q1 into the function f to obtain the predicted excess amount of the allergen P1 caused by cross-contact on the production line. Similarly, according to the predicted excess amount Q2 corresponding to the risk indicator of cross-contact at the node and the corresponding mapping relationship, obtain the predicted excess amount of the allergen P2 caused by cross-contact at the node; according to the predicted excess amount Q3 corresponding to the risk indicator of introduction during the production process and the corresponding mapping relationship, obtain the predicted excess amount of the allergen P3 caused by introduction during the production process.

[0057] Through the above technical solution, by establishing the mapping relationship between each cause of the allergen exceeding the standard and the corresponding risk indicator, determining the connection between different risk factors and the cause of exceeding the standard, and calculating the risk proportion of each risk indicator in the set of production risk indicators, the contribution degree of different risk factors to the overall risk can be quantified; by calculating the total excess amount of the allergen content and allocating it to each risk indicator according to the risk proportion, the predicted excess amount can better fit the risk situation in actual production; combining the mapping relationship and the predicted excess amount to determine the predicted excess amount of the allergen corresponding to each cause of the allergen exceeding the standard further refines the prediction accuracy; after clarifying the predicted excess amount corresponding to each cause of the allergen exceeding the standard, it is convenient to formulate more targeted production control strategies.

[0058] Step S104, through the pre-constructed knowledge base for controlling the allergen composition, determine the corresponding production control strategy according to the over-standard assessment data of the target allergen quality inspection node, so as to optimize the production quality through the production control strategy.

[0059] Based on the pre - constructed knowledge base for allergen component control, and according to the over - standard evaluation data of this target allergen quality inspection node, determine the corresponding production control strategy, which specifically includes: determining the allergen over - standard cause with an allergen over - standard prediction higher than the preset over - standard threshold in the over - standard evaluation data as the target control over - standard cause; determining the corresponding target control object, where the target control object includes the production node corresponding to this target allergen quality inspection node and the next production node corresponding to this target allergen quality inspection node; through the target control over - standard cause, using the allergen component control knowledge base, determine the preventive node control strategy corresponding to the next production process of this production node in the target control object, where the preventive node control strategy includes any one or more of equipment cleaning operations, equipment isolation operations, and process parameter adjustment operations; according to the allergen over - standard prediction corresponding to the target control over - standard cause and the real - time beverage quality index parameters, using the allergen component control knowledge base, determine the remedial node control strategy corresponding to the current production process of this next production node in the target control object, where the remedial node control strategy includes any one or more of physical treatment and chemical treatment.

[0060] In an embodiment of this specification, extract the allergen over - standard prediction corresponding to each allergen over - standard cause from the over - standard evaluation data of the target allergen quality inspection node, and compare the prediction with the preset over - standard threshold one by one. The preset over - standard threshold can be obtained after processing according to the product quality standard, past production experience, or relevant regulatory requirements. For example, 60% of the standard value can be set. If the standard value is that the allergen protein content in each milliliter of beverage does not exceed 10 micrograms, then the allergen over - standard threshold is set to 6 micrograms. Since there may be multiple reasons for cross - contact, even if the prediction generated by each reason is less than 10 micrograms, the final cumulative over - standard amount may be greater than 10 micrograms after the superposition of multiple allergen over - standard causes. Therefore, the preset over - standard threshold should be set to a value lower than the standard value. Compare the prediction with the preset over - standard threshold one by one to determine the target control over - standard cause, and there is at least one target control over - standard cause here. For example, in the production of plant - protein beverages, if the allergen over - standard threshold is set to that the allergen protein content in each milliliter of beverage does not exceed 6 micrograms, when the allergen over - standard prediction caused by cross - contact on the production line is 8 micrograms per milliliter, then cross - contact on the production line is determined as the target control over - standard cause.

[0061] By using the associated information between production process records and quality inspection nodes, the production node corresponding to the target allergen quality inspection node can be determined through the process mapping table in the production management system. Similarly, according to the production process sequence, the next production node corresponding to the target allergen quality inspection node can be determined by judging the logical relationship of the production process. For example, if the target allergen quality inspection node is in the middle of the processing link, the next production node may be the production nodes of subsequent sterilization and filling links.

[0062] Pre-construct a knowledge base for allergen ingredient control. Collect internal and external data of the enterprise. Internal data of the enterprise includes the historical production data of the enterprise, including production logs, equipment operation records, raw material procurement records, etc., and extract data on production processes, equipment usage, raw material batch information, and past quality problems. For example, by analyzing production logs, understand the impact of changes in raw material suppliers in different seasons on the allergen risk of products. Obtain the internal quality inspection reports of the enterprise, including raw material inspection reports, semi-finished product and finished product quality inspection reports, and obtain information on allergen detection results, exceeding standards, and corresponding production links. For example, in the finished product quality inspection report, if it is found that a certain allergen in a batch of beverages exceeds the standard, combined with production records, the links where problems may occur during the production process of this batch of products can be traced back. In addition, expert experience sharing and communication can also be carried out. For example, front-line operators may feedback that during the equipment cleaning process, certain parts are prone to allergen residues and need special attention to cleaning. External data of the enterprise includes the latest research results, technological development trends, and typical case analyses on allergen control released by industry research institutions. Such as the research report points out that by optimizing the temperature and time parameters in the production process, the activity of potential allergens in certain plant protein beverages can be effectively reduced. It can also include thesis literature on food allergen detection, control, prevention, etc.; in addition, due to the existence of food standards, the definition, detection methods, limit requirements, and labeling regulations of allergens are also clarified.

[0063] Determine the storage architecture of the allergen ingredient control knowledge base, including the allergen basic information base, the production environment knowledge base, the control strategy base, and the case knowledge base. The allergen basic information base details various common allergens, such as peanuts, nuts, wheat, soybeans, etc., including the chemical composition, molecular structure, sensitization mechanism of each allergen, etc., and covers plant raw materials, additives, excipients, etc. involved in the production process. The production link knowledge base includes production process and node information and equipment information. Sort out the complete production process and divide it into different production nodes, such as raw material receiving, storage, pretreatment, blending, processing (such as grinding, sterilization, fermentation, etc.), packaging, etc. Clearly define the specific operation content, equipment used, and the connection relationship between the front and back production nodes for each production node. For the equipment used in each production node, record the model, specifications, functions, applicable scope, and cleaning and maintenance methods of the equipment. For example, the material, capacity, stirring speed adjustment range of the blending tank, and the recommended cleaning steps and types of cleaning agents after each use. In the control strategy base, formulate corresponding preventive measures for different production nodes and possible allergen risks, including equipment cleaning operation specifications (such as cleaning frequency, cleaning agent concentration and usage methods), equipment isolation measures (such as the material and installation position of the isolation device), process parameter adjustment strategies (such as appropriate temperature, pressure, time range), etc. When the allergen exceeds the standard, provide corresponding remedial measures. It is divided into physical treatment methods (such as filtration, centrifugation, adsorption, etc., including the equipment model and operation parameters used) and chemical treatment methods (such as the types, dosages, and reaction conditions of chemical reagents added). For example, when a small amount of allergen particles in the beverage are detected to exceed the standard, a microfiltration membrane with a pore size of 0.2 microns can be used for filtration, and the filtration pressure is controlled at 0.3 - 0.5 MPa. The case knowledge base can include historical over-standard cases and successful prevention and control cases. It should be noted that a regular review mechanism is established to conduct a comprehensive review of the knowledge base at regular intervals. With the emergence of new allergen detection technologies, the content of allergen detection methods in the knowledge base needs to be updated in a timely manner. When new allergen over-standard cases or successful prevention and control cases occur within the enterprise, new research results are released in the industry, or regulatory standards change, the knowledge base is updated in a timely manner to ensure the availability of the knowledge base.

[0064] Using the allergen component control knowledge base, a query mechanism is established. It should be noted that the knowledge base here is a database containing various information, such as the associated data table of "allergen type - over-standard reason - preventive strategy". When the target control over-standard reason (such as cross-contact in the production line) and the production node are determined, query in the knowledge base using this information as an index. According to the records in the knowledge base, select the appropriate preventive node control strategy. For example, if the knowledge base has suggestions for equipment cleaning operations in the next production process corresponding to the production node for allergen problems caused by cross-contact in the production line, such as information on the use of specific cleaning agents, cleaning time, and cleaning process, these equipment cleaning operations will be used as part of the preventive strategy. At the same time, it may also combine relevant suggestions for equipment isolation operations (such as setting the specifications and positions of isolation baffles) and process parameter adjustment operations (such as adjusting parameters such as temperature, pressure, and time) to form a complete preventive node control strategy. That is to say, the purpose of the preventive node control strategy here is to reduce the impact of residual allergen components on subsequent production processes.

[0065] According to the predicted over-standard amount of allergen corresponding to the target control over-standard reason and the real-time beverage quality index parameters, query again using the allergen component control knowledge base. The real-time beverage quality index parameters can include information such as the allergen content in the beverage, the acidity and alkalinity of the beverage, and the temperature. The knowledge base can store associated information such as "allergen over-standard degree - real-time beverage parameters - remedial strategy". By querying the knowledge base, determine the remedial node control strategy applicable to the current production process corresponding to the next production node. For example, if the predicted over-standard amount of allergen is relatively high and the real-time beverage quality index parameters show that the acidity and alkalinity of the beverage are within a certain range, the knowledge base may recommend using a chemical treatment method, such as adding a specific enzyme to decompose the allergen protein. Or if the allergen in the beverage is particulate and the predicted over-standard amount is relatively low, it may recommend using a physical treatment method, such as filtering with a microfiltration device, and at the same time give parameters such as the pore size and filtration rate of the microfiltration device to determine the specific remedial node control strategy. That is to say, the remedial node control strategy here is for products that have already exceeded the standard of allergen components, and add steps to eliminate or reduce the activity of allergen components in subsequent production processes to achieve the remediation of over-standard products.

[0066] Based on the association of allergen type - over - standard reason - preventive strategy, when the target control over - standard reason (such as cross - contact in the production line) and the production node are determined, the preventive node control strategy can be accurately obtained from the knowledge base. For example, for cross - contact in the production line, specific cleaning operation suggestions such as using a specific cleaner, cleaning time, and process for equipment cleaning, as well as measures such as equipment isolation and process parameter adjustment are clearly given. Effective preventive measures can be taken in advance for specific risks, reducing the risk of allergen residue and ensuring that subsequent production is not affected. Combining the predicted amount of allergen over - standard corresponding to the target control over - standard reason and real - time beverage quality index parameters (such as pH, temperature, allergen content, etc.), the remedial node control strategy can be accurately matched. For example, when the predicted amount of allergen over - standard is high and the pH is within a specific range, it is recommended to add a specific enzyme to decompose the allergen protein; when the allergen is granular and the predicted amount of over - standard is low, it is recommended to filter with a microfiltration device with a specific pore size and filtration speed to achieve precise remediation for different over - standard situations. The preventive node control strategy can avoid potential allergen over - standard problems in advance. By taking appropriate measures such as equipment cleaning, isolation, and process parameter adjustment, the probability of allergen - related quality problems in subsequent production is reduced, and production interruptions, product scrapping, etc. caused by problems are decreased, improving production efficiency. For the situation where allergen over - standard has already occurred, the remedial node control strategy provides targeted solutions. By adding steps to eliminate or reduce the allergen activity in the subsequent production process, the over - standard products can be remedied, reducing the economic losses caused by unqualified beverages, and at the same time ensuring that the quality and taste of the beverages meet the standards.

[0067] Through the above technical solutions, production line parameters are obtained and the risk of allergen cross-contact is analyzed. Based on the risk analysis, at least one target allergen quality inspection node is determined. Based on the quantified cross-contact risk index, risk production nodes are determined. By setting clear preset conditions, production nodes with actual risks are screened out, avoiding excessive attention to the entire production line or omission of high-risk areas, and providing accurate targets for subsequent quality control and preventive measures. When the real-time beverage quality index parameters meet the preset conditions, traceability analysis is carried out by combining the real-time beverage quality index parameters and the real-time production line parameters, which can accurately find out the specific reasons for allergen over-standard. The over-standard evaluation data not only clarifies the reasons for allergen over-standard, but also gives the predicted amount of allergen over-standard corresponding to each reason. By obtaining the real-time production line parameters, real-time production risk identification is carried out on the previous production node of the target allergen quality inspection node, comprehensively considering the three main sources of allergen risk, namely cross-contact between production lines, cross-contact between nodes, and introduction during the production process. In the production scenario of producing different types of plant protein beverages in the same factory, not only the risk of allergen cross-contact that may occur between different production lines and between different batches at the same production node is concerned, but also the risk that the processing technology itself may cause allergen components to be generated from non-allergen raw materials is considered, realizing a full coverage of various potential allergen risks in the production process. The preventive node control strategy in the production control strategy takes measures in advance against potential factors that may cause allergen over-standard based on the risk analysis of the production process. The remedial node control strategy provides effective solutions for products that have already exceeded the allergen standard, can save the beverages with problems, reduce the economic losses caused by non-conformities, and at the same time ensure that the quality, taste, etc. of the beverages meet the standards.

[0068] The embodiments of this specification also provide a production quality control device for plant protein beverages, as Figure 2 shown. The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0069] The embodiments of this specification also provide a non-volatile computer storage medium storing computer-executable instructions, and the computer-executable instructions are set to: execute the above method.

[0070] The embodiments in this specification are all described in a progressive manner. The same or similar parts between the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments. In particular, for the embodiments of the device, equipment, and non-volatile computer storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0071] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0072] The devices and media provided by the embodiments of this specification correspond one-to-one with the methods. Therefore, the devices and media also have beneficial technical effects similar to those of their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be elaborated here.

[0073] Those skilled in the art should understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification can take the form of an all-hardware embodiment, an all-software embodiment, or an embodiment combining software and hardware aspects. Moreover, this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0074] This specification is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0075] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 steps of the functions specified in one block or multiple blocks.

[0077] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0078] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0079] Computer-readable media includes permanent and non-permanent, removable and non-removable media and can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0080] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of another identical element in the process, method, commodity or device comprising the said element.

[0081] The above are only one or more embodiments of this specification and are not intended to limit this specification. For those skilled in the art, various changes and modifications can be made to one or more embodiments of this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included within the scope of the claims of this specification.

Claims

1. A method for controlling the production quality of a plant protein beverage, characterized in that: The method comprises: Acquire production line parameters of a current beverage production line, perform allergen cross-contact risk analysis on the current beverage production line based on the production line parameters, and determine at least one target allergen quality inspection node in the current beverage production line; Collecting allergen quality inspection data of each target allergen quality inspection node in real time, so as to determine the real-time beverage quality index parameter corresponding to each target allergen quality inspection node through the allergen quality inspection data; When the real-time beverage quality index parameter meets the preset conditions, the cause of the allergen exceeding the standard is traced and analyzed according to the real-time beverage quality index parameter and the real-time production line parameters obtained in advance, and the exceeding standard assessment data of the target allergen quality inspection node is determined, wherein the exceeding standard assessment data includes at least one allergen exceeding standard cause and the predicted amount of allergen exceeding standard corresponding to each exceeding standard cause; By using a pre-built allergen component control knowledge base, according to the over-standard assessment data of the target allergen quality inspection node, a corresponding production control strategy is determined, so as to optimize production quality through the production control strategy, wherein the production control strategy includes a preventive node control strategy and a remedial node control strategy; Based on the production line parameters, an allergen cross-contact risk analysis is performed on the current beverage production line to determine at least one target allergen quality inspection node in the current beverage production line, specifically including: Obtaining the previous batch production information corresponding to each production node in the current beverage production line through the production log in the production line parameters, wherein the previous batch production information includes the previous batch beverage raw material information and the previous batch production time; Obtaining node location information of each of the production nodes in the production line parameters within the production park, so as to determine an allergen flow assessment area based on the node location information, so as to determine a node distance between an adjacent production node of an adjacent production line located within the allergen flow assessment area and a production node in the current beverage production line; Determine the risky production node in the current beverage production line according to the previous batch production information corresponding to each production node and the node distance, and determine the target allergen quality inspection node by the quality inspection node located after the risky production node; According to the real-time beverage quality index parameters and the pre-acquired real-time production line parameters, the causes of allergen exceeding the standard are traced and analyzed to determine the exceeding standard assessment data of the target allergen quality inspection node, specifically including: Acquiring the real-time production line parameters, wherein the real-time production line parameters include real-time processing technology data of each production node; According to the real-time production line parameters of the current beverage production line, real-time production risk identification is performed on the previous production node of the target allergen quality inspection node, and a production risk indicator set corresponding to the previous production node is determined, wherein the production risk indicator set includes a production line cross-contact risk indicator, a node cross-contact risk indicator, and a production process introduction risk indicator; A weight is assigned to each cause of allergen exceeding the standard through the production risk indicator set and the real-time beverage quality indicator parameters to determine the predicted amount of allergen exceeding the standard corresponding to each cause of allergen exceeding the standard, wherein the causes of allergen exceeding the standard include production line cross-contact, node cross-contact and introduction into the production process.

2. The production quality control method of a plant protein beverage according to claim 1, characterized in that: Determining risky production nodes in the current beverage production line according to the previous batch production information corresponding to each production node and the node distance specifically includes: The cross-contact risk of adjacent production lines of the production node is evaluated by the node distance and the area radius corresponding to the allergen flow assessment area, and the corresponding production line cross-contact risk index of the production node is determined; Determining a node cross-contact risk coefficient of each production node according to the raw material attributes of the previous batch of beverage raw material information corresponding to each production node, wherein the raw material attributes include allergen raw materials and non-allergen raw materials; Using the previous batch production time corresponding to each production node, determine the current node processing time corresponding to the production node, and determine the node cross-contact risk quantification parameter corresponding to the production node based on the current node processing time and a preset allergen component processing time threshold corresponding to the production node; Determining a node cross-contact risk index corresponding to the production node based on a node cross-contact risk coefficient of the production node and a node cross-contact risk quantification parameter corresponding to the production node; A cross-contact risk index is generated according to the production line cross-contact risk index and the node cross-contact risk index, and a risk production node whose cross-contact risk index meets a preset condition is determined in the current beverage production line.

3. The production quality control method of a plant protein beverage according to claim 1, characterized in that: Determine the real-time beverage quality index parameters corresponding to each target allergen quality inspection node through the allergen quality inspection data, specifically including: Acquire the allergen quality inspection data, wherein the allergen quality inspection data is beverage spectrum data; Performing data cleaning and preprocessing operations on the beverage spectrum data to determine spectrum processing data; The corresponding real-time beverage quality index parameters are determined based on the spectral processing data through a pre-constructed regression prediction model, wherein the real-time beverage quality index parameters include the allergen quality inspection content.

4. The production quality control method of a plant protein beverage according to claim 1, characterized in that: According to the real-time production line parameters of the current beverage production line, real-time production risk identification is performed on the previous production node of the target allergen quality inspection node, and a set of production risk indicators corresponding to the previous production node is determined, specifically including: According to the beverage raw material information in the current beverage production line, the process parameter standard value corresponding to each production node is matched, so as to determine the overflow ratio of the process parameter corresponding to the production node through the real-time processing technology data of the previous production node of the target allergen quality inspection node and the corresponding process parameter standard value, so as to determine the corresponding production process introduction risk index; Determine the first production timestamp of the previous allergen raw material corresponding to the previous production node of the target allergen quality inspection node through the real-time production log of each production node in the real-time production line parameters, so as to determine the allergen processing time of the production node, and determine the corresponding real-time node cross-contact risk index according to the allergen processing time and the allergen component processing time threshold corresponding to the previous allergen raw material; Determine the adjacent production line production nodes of each production node in the real-time production line parameters within the preset allergen flow assessment area, determine the current production time interval corresponding to the production node through the real-time production log of the production node, and determine the startup state of the adjacent production line production node within the current production time interval; In the startup state, determining a node cross-contact risk coefficient corresponding to the production node, and determining a production line cross-contact risk quantification parameter corresponding to the production node through a pre-acquired node distance between an adjacent production line production node and the production node and an area radius corresponding to the allergen flow assessment area, and determining a real-time production line cross-contact risk index according to the production line cross-contact risk quantification parameter and the node cross-contact risk coefficient; Based on the production process introduction risk index, the real-time node cross-contact risk index and the real-time production line cross-contact risk index, a production risk index set corresponding to the previous production node is determined.

5. The production quality control method of a plant protein beverage according to claim 1, characterized in that: By using the production risk index set and the real-time beverage quality index parameter, a weight is assigned to each allergen exceeding the limit reason to determine the allergen exceeding limit prediction amount corresponding to each allergen exceeding the limit reason, specifically including: Establish a mapping relationship between the cause of each allergen exceeding the limit and the corresponding risk indicator; According to the production risk indicator set, multiple risk indicators in the production risk indicator set are processed to determine the risk proportion of each risk indicator in the production risk indicator set; Determining the predicted excess amount corresponding to each risk indicator through the risk proportion of each risk indicator and the real-time beverage quality indicator parameter; Based on the mapping relationship and the predicted excess amount corresponding to each risk indicator, the predicted excess amount of allergen corresponding to the cause of excess of each allergen is determined.

6. The production quality control method of a plant protein beverage according to claim 1, characterized in that: Through the pre-built allergen component control knowledge base, according to the over-standard assessment data of the target allergen quality inspection node, the corresponding production control strategy is determined, specifically including: Determine the cause of allergen excess in the excess assessment data where the predicted amount of allergen excess is higher than a preset excess threshold as the cause of target control excess; Determine a corresponding target control object, wherein the target control object includes a production node corresponding to the target allergen quality inspection node and a next production node corresponding to the target allergen quality inspection node; Based on the cause of the target control exceeding the limit, the preventive node control strategy corresponding to the next production process corresponding to the production node in the target control object is determined by using the allergen component control knowledge base, wherein the preventive node control strategy includes any one or more of equipment cleaning operation, equipment isolation operation and process parameter adjustment operation; According to the predicted amount of allergen excess corresponding to the cause of the target control excess and the real-time beverage quality index parameter, the allergen component control knowledge base is used to determine the remedial node control strategy corresponding to the current production process corresponding to the next production node in the target control object, wherein the remedial node control strategy includes any one or more of physical treatment and chemical treatment.

7. A production quality control device for plant protein beverages, characterized in that: The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any one of claims 1-6.

8. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to execute the method according to any one of claims 1 to 6.

Images