Intelligent filling laboratory management method and system, electronic equipment and medium

Abstract

The invention discloses an intelligent filling laboratory management method and system, electronic equipment and a medium, and aims to quantitatively evaluate the pollution risk between filling tasks through a preset filling equipment pollution risk evaluation model, so as to provide a reliable reference for the sorting of the filling tasks and the selection of a cleaning strategy during the switching of the filling tasks. For a multi-task situation, dynamic prediction of a long-term pollution risk and influence of chain transmission of pollution are realized, improvement of the operation efficiency of a filling laboratory is facilitated, a scheduling target is expanded from a single time dimension to a double dimension of time efficiency and pollution risk control, and the scheduling efficiency of the filling laboratory is improved. And a comprehensive technical solution is provided for intelligent management of the liquid food filling laboratory.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart laboratory, in particular to a smart filling laboratory management method, system, electronic device and medium. BACKGROUND

[0002] In modern scientific research and industrial development, the laboratory as the core place for carrying out scientific experiments, technical verification and product development, its management efficiency is directly related to the research and development progress and the quality of the results. Laboratory management not only involves the overall deployment of resources such as equipment, materials and personnel, but also includes the optimization and intelligent scheduling of experimental processes. Especially in the smart filling laboratory of liquid food, the experimental task often involves the filling, testing and analysis of multiple flavors of liquid, and the process has the characteristics of strong continuity, high process requirements and significant cross-contamination risk. The traditional way of relying on manual arrangement of experimental tasks and equipment usage time is not only inefficient, but also difficult to deal with scheduling conflicts caused by various uncertain factors in the experiment.

[0003] Therefore, various laboratory intelligent management systems have been introduced, for example, the system and method based on laboratory process intelligent management disclosed in Chinese patent CN119624056A. This kind of system can improve equipment utilization efficiency and task continuity by classifying experimental tasks, predicting execution time, identifying task time overlap and dynamically adjusting reservation period. However, in the specific experimental scenario of liquid food filling, the existing system still has obvious limitations. Especially in experiments involving continuous filling of different flavors of liquid, the filling equipment usually needs to be cleaned after each task, and the cleaning time needs to be considered. If it is heavy cleaning, the experimental equipment can be used for experiments for a very short time. If it is light cleaning, there is a concern about flavor contamination. Such residues may cause subsequent experimental sample contamination and affect the effectiveness of experimental results. Although the existing management system can predict the duration of the "experimental task" itself and schedule it based on historical data, it generally does not consider the impact of residual risk on task sequence arrangement. SUMMARY

[0004] Technical purpose: In view of the deficiencies of the existing filling laboratory, the present application discloses a smart filling laboratory management method, system, electronic device and medium which can effectively evaluate the contamination risk between filling tasks, sequence and schedule continuous multi-task operation, and improve the efficiency of laboratory operation.

[0005] Technical scheme: In order to achieve the above technical purpose, the present application adopts the following technical scheme: A smart filling laboratory management method, comprising the steps of: S01, a filling equipment pollution risk assessment model is established in advance, which is used to assess the flavor pollution risk of residual in the filling equipment to the product of the next filling task, and determine the cleaning type of the filling equipment when the filling task is switched based on the pollution risk; When the pollution risk is assessed, the initial residual content in the filling equipment that affects the next filling task is determined, and the pollution risk between the filling tasks is assessed according to the residual content that can be tolerated by the product of the next filling task; S02, the filling tasks of the same filling line are clustered, the filling tasks with similar flavors are divided into the same cluster based on the liquid formula data of the filling tasks, or the pollution risks between the filling tasks are calculated, and the filling tasks with smaller pollution risks are divided into the same cluster; S03, the filling tasks in the same cluster are sorted, the pollution risks between the filling tasks are assessed by the established filling equipment pollution risk assessment model, and the task with the smallest pollution risk to the current filling equipment state is selected for filling sorting; S04, for the sorting of different clusters, the filling tasks of the cluster are abstracted as virtual tasks as a whole, the pollution risk between the virtual task and the first executed filling task of the next cluster is calculated by the established filling equipment pollution risk assessment model, the cluster is sorted, and finally the sorting of the filling tasks is completed.

[0006] Preferably, in step S01, when the initial residual content in the filling equipment is determined, the residual state is divided into liquid phase residual and gas phase residual; When calculating the liquid phase residual, the surface area of the filling equipment directly contacted with the liquid phase is determined according to the three-dimensional structure model of the filling equipment, and the theoretical liquid phase residual of various substances attached to the contact surface of the filling equipment is determined according to the adsorption amount of the corresponding filling material components on the surface of the filling equipment. When calculating the gas phase residual, the headspace area of the filling equipment containing volatile gas phase substances is determined according to the three-dimensional structure model of the filling equipment, and the theoretical gas phase residual of various substances in the filling equipment is determined according to the filling operation temperature and the saturated vapor pressure of each component of the filling product.

[0007] Preferably, in S01, when the cleaning type of the filling equipment when the filling task is switched is determined based on the pollution risk, a cleaning model of the cleaning type to the residual is established, the cleaning effect of various cleaning types on the initial residual is calculated according to the determined initial residual content in the filling equipment, the residual content in the filling equipment after cleaning is determined, and the pollution risk corresponding to each cleaning type is assessed according to the residual content in the filling equipment after cleaning and the residual content that can be tolerated by the product of the next filling task, and the cleaning type with the shortest time and the pollution risk within the range is selected.

[0008] Preferably, the cleaning model of the present invention is an exponential decay model, where the cleaning residue Ma = M * exp(-kc * tc), where M represents the initial residue content, kc represents the cleaning efficiency coefficient, which is obtained through experimental calibration of the cleaning type, and tc is the cleaning time.

[0009] Preferably, when assessing the pollution risk, this invention uses the theoretical concentration of residue mixed into the product of the next filling task for evaluation, comparing the residue content per unit volume of the filling task product with the residue content that the filling product can tolerate; the theoretical concentration C1 = (Mal + Ma2 × η) / V, where Ma1 represents liquid phase residue, Ma2 represents gas phase residue, η represents the release probability coefficient of gas phase residue, with a value range of 0.5-1.0, and V represents the volume of the filling task product; the dose ratio of residue mixed into the filling task product to the residue content that the filling product can tolerate is R = C1 / C2, where C2 is the residue concentration that the filling product can tolerate, and the dose ratio R is mapped to a risk value of 0-1 using a nonlinear function for pollution risk assessment.

[0010] Preferably, when conducting contamination risk assessment between filling tasks, the present invention uses the residual amount of key flavor compounds in the filled product to conduct contamination risk assessment between filling tasks. When the filling equipment is operating continuously, a residue state vector of the filling equipment is established. When performing a filling task, the dimension of the residue state vector is increased accordingly for each additional key flavor compound. When conducting contamination risk assessment between filling tasks, the cumulative residual amount of key flavor compounds that affect the flavor of the product in the next filling task is used to conduct contamination risk assessment.

[0011] Preferably, when calculating the residual amount of the corresponding key flavor compounds, the present invention calibrates the key flavor compounds one-to-one with the residual retention rate of the corresponding cleaning type. The residual retention rate = residual amount after cleaning / residual amount before cleaning. A diagonal matrix is ​​constructed using the residual retention rate, and the diagonal elements of the matrix are the residual retention rates corresponding to the key flavor compounds in the residue state vector. The residual amount of key flavor compounds in the cleaning and filling equipment is calculated in batches through the diagonal matrix to conduct a contamination risk assessment.

[0012] This invention discloses a smart liquid food filling laboratory management system. Using the aforementioned smart liquid food filling laboratory management method, it includes a filling data processing module, a contamination risk assessment module, and a filling equipment control module for controlling the operation of filling equipment to perform filling and cleaning operations. The filling data processing module receives filling tasks and performs clustering and sorting. During the clustering and sorting process, the contamination risk assessment module assesses the contamination risk between filling tasks and determines the cleaning type between filling tasks, formulating a cleaning strategy for the filling task execution process. The filling equipment control module then performs filling according to the determined filling tasks and cleaning strategies.

[0013] This invention discloses an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to enable the device to perform the above-described intelligent filling laboratory management method for liquid food.

[0014] This invention discloses a computer-readable storage medium storing a computer program. When the computer program is run, the device running the computer program implements the above-mentioned intelligent filling laboratory management method for liquid food.

[0015] Beneficial Effects: The intelligent filling laboratory management method, system, electronic equipment, and medium disclosed in this invention have the following beneficial effects: 1. This invention quantifies the contamination risk between filling tasks by using a pre-set contamination risk assessment model for filling equipment. This provides a reliable benchmark for the sequencing of filling tasks and the selection of cleaning strategies when switching filling tasks, which is conducive to improving the operating efficiency of the filling laboratory. It expands the scheduling objective from a single time dimension to a dual dimension of time efficiency and contamination risk control, and provides a comprehensive technical solution for the intelligent management of liquid food filling laboratories.

[0016] 2. This invention clusters filling tasks into the same cluster based on the similarity of the flavors of the filling products, and performs contamination risk assessments both within and between clusters to sort the filling tasks. This can reduce the contamination risk between products with different flavors, reduce computational complexity, and facilitate task scheduling in the laboratory.

[0017] 3. This invention can make intelligent decisions on the intensity and timing of cleaning based on pollution risk assessment, thereby improving the utilization rate of filling equipment, avoiding the pollution risk caused by insufficient cleaning, and also avoiding the waste of time and resources caused by excessive cleaning.

[0018] 4. In cases where filling equipment performs multiple filling tasks, this invention accumulates and evaluates the key flavor compounds for each filling task, constructs a residue state vector for the filling equipment, and, in conjunction with a diagonal matrix, can quickly monitor the residues of multiple key flavor compounds. Based on the next filling task, the corresponding key flavor compounds are selected for contamination risk assessment, which can track and predict the chain-cumulative effect of contamination and is suitable for long-term, continuous experimental planning. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0020] Figure 1 This is a schematic diagram of the operation process of the intelligent filling laboratory management system for liquid food of the present invention. Detailed Implementation

[0021] Reference will now be made in detail to embodiments of the present disclosure, one or more of which are set forth herein. Each embodiment and example is provided by way of explanation of the apparatus, composition, and materials of the present disclosure, and not by way of limitation. Rather, the following description provides convenient illustrations for implementing exemplary embodiments of the present disclosure. Indeed, it will be apparent to those skilled in the art that various modifications and variations can be made to the teachings of the present disclosure without departing from the scope or spirit of the present disclosure.

[0022] This invention discloses a smart filling laboratory management method, comprising the following steps: S01. Establish a contamination risk assessment model for filling equipment in advance to assess the risk of flavor contamination caused by residues in the filling equipment to the product in the next filling task, and determine the cleaning type of the filling equipment when switching filling tasks based on the contamination risk. When conducting a contamination risk assessment, first determine the initial residue content in the filling equipment that will affect the next filling task, and then assess the contamination risk between filling tasks based on the residue content that the product in the next filling task can tolerate.

[0023] First, the initial residue content is assessed. In step S01, when determining the corresponding initial residue content in the filling equipment, the present invention classifies the residue into liquid phase residue and gas phase residue according to the state of the residue. When calculating liquid phase residue, first determine the surface area of ​​the filling equipment that is in direct contact with the liquid phase based on the three-dimensional structural model of the filling equipment, and then determine the theoretical liquid phase residue of various substances that are theoretically attached to the contact surface of the filling equipment based on the adsorption amount of the corresponding filling material components on the surface of the filling equipment. The adsorption amount of the corresponding filling material components on the surface of the filling equipment is determined based on the composition of the filling product and the material of the contact surface of the filling equipment. The saturated adsorption amount per unit area is then multiplied by the surface area of ​​the filling equipment in direct contact with the liquid phase. This data can be obtained by testing the adsorption intensity of different substances on the filling equipment under corresponding filling conditions in advance, and constructing a substance-material adsorption database. When the filling equipment is used for filling operations, the theoretical liquid phase residue of various substances theoretically attached to the contact surface of the filling equipment can be directly calculated by calling the corresponding data based on the filling product.

[0024] Similarly, when calculating gaseous residues, the headspace area of ​​the equipment that accommodates volatile gaseous substances is determined based on the three-dimensional structural model of the filling equipment. The headspace area is the effective surface area of ​​the top space inside the equipment during the filling process, which is confirmed by the three-dimensional structural model. Conventional filling equipment has a uniform filling cross-sectional area to ensure accurate control of the filling volume. Therefore, the amount of gaseous evaporation that the equipment can accommodate can be determined by the headspace area and the specific dimensions of the filling equipment. Based on the filling operation temperature and the saturated vapor pressure of each component of the filling product, the maximum gaseous concentration that can be achieved in the headspace of the equipment is calculated using the ideal gas law or equation of state. This determines the theoretical gaseous residues of various substances in the filling equipment. Gaseous residues mainly target volatile flavor substances, such as limonene.

[0025] When calculating the residue content, this invention considers the easily overlooked gaseous contamination sources, enabling accurate assessment of contamination risks between filling tasks. This reduces the impact of contaminants on the filled products, improves the accuracy of experiments, and allows for the selection of appropriate cleaning types based on different contaminants, achieving effective cleaning of residues. This avoids situations where water washing is used for gaseous residues but the cleaning effect fails to meet requirements.

[0026] In S01, when determining the cleaning type of the filling equipment based on the pollution risk during the switching of filling tasks, this invention first establishes a cleaning model for the cleaning type on the residue. Based on the determined initial residue content in the filling equipment, it calculates the removal effect of various cleaning types on the initial residue, determines the residue content in the filling equipment after cleaning, and assesses the pollution risk corresponding to each cleaning type based on the residue content in the filling equipment after cleaning and the residue content that the product of the next filling task can tolerate. Finally, it selects the cleaning type with the lowest pollution risk within the range and the shortest time consumption.

[0027] The cleaning type is a pre-set cleaning mode for the filling equipment. For example, it can be divided into several cleaning types based on the cleaning fluid flow rate, temperature, whether a cleaning agent is used, and the cleaning procedure (such as the turbulence effect of CIP), corresponding to different cleaning intensities. Experiments and evaluations can be carried out based on the cleaning type settings of the filling equipment, so that the management method of this invention can be used to manage the filling tasks of the filling equipment for different filling equipment.

[0028] In the embodiments of this invention, the cleaning model is an exponential decay model, where the cleaning residue Ma = M * exp(-kc * tc), and M represents the initial residue content, kc represents the cleaning efficiency coefficient. The higher the cleaning intensity of the cleaning type, the larger the corresponding cleaning efficiency coefficient. This is determined through experimental calibration of different cleaning types. The cleaning efficiency coefficient kc varies depending on the residue. For the same cleaning procedure (such as standard water washing), the kc values ​​differ for liquid-phase residues and gas-phase residues. For different key flavor compounds (such as limonene and vanillin), the kc values ​​also differ due to differences in adsorption characteristics and chemical properties. The parameters corresponding to each substance can be calibrated experimentally. The cleaning efficiency coefficient is not universally applicable but is bound to specific residues, forming a one-to-one correspondence. It can be pre-calibrated experimentally, and a corresponding database can be established based on experiments with commonly used filling products, so that data can be directly retrieved for calculation and evaluation during contamination risk assessment.

[0029] To improve the accuracy of pollution risk assessment and effectively quantify pollution risk, this invention assesses pollution risk by using the theoretical concentration of residue mixed into the next filling task product. It compares the residue content per unit volume of the filling task product with the tolerable residue content of the filling product. The theoretical concentration C1 = (Mal + Ma2 × η) / V, where Ma1 represents liquid phase residue, Ma2 represents gas phase residue, η represents the release probability coefficient of gas phase residue (range 0.5-1.0), and V represents the volume of the filling task product. The dilution effect of the filling task product volume on residue is considered in the pollution risk assessment process, rather than solely relying on residue content. The dose ratio of residue mixed into the filling task product to the tolerable residue content of the filling product is R = C1 / C2, where C2 is the tolerable residue concentration of the filling product. The dose ratio R is mapped to a risk value of 0-1 using a nonlinear function for pollution risk assessment. The nonlinear function can be an sigmoid function Hill equation, and the pollution risk value TTCI = R^n / (1 + R^n). Where n is the Hill coefficient, used to adjust the steepness of the curve, typically n≥2; this mapping perfectly expresses the abrupt change characteristics of risk. When R<1, the pollution risk value approaches 0; when R>1, the pollution risk value rises sharply and approaches 1. This makes the pollution risk value TTCI a sensitive and reliable indicator.

[0030] Once a reliable and effective pollution risk assessment model is obtained, the operating strategy of the filling equipment can be adjusted accordingly. When the filling task switching is infrequent, pollution risk assessment is only required during task switching. In this case, the initial residue content can be directly calculated according to the scheme of this invention to conduct pollution risk assessment and select cleaning strategies. Furthermore, in this situation, since cleaning is infrequent, even if heavy cleaning is used, the improvement in the efficiency of the filling equipment in performing filling tasks by the management method of this invention cannot be significantly reflected. The heavy cleaning refers to the definition of the equipment's cleaning intensity for contaminants. Here, heavy cleaning specifically refers to a cleaning method with good cleaning effect and long time consumption. In practical applications, the cleaning intensity can be classified and defined according to the cleaning type of the specific filling equipment, and is not limited by the definition of this invention.

[0031] In multi-tasking scenarios, frequent switching of filling products is required. Each time a switch is made, a contamination risk assessment and cleaning of the filling equipment are necessary. While heavy cleaning can effectively remove residues when switching filling tasks, it also takes a lot of time and is not conducive to improving the operational efficiency of the laboratory.

[0032] For scenarios involving multitasking, this invention discloses the following steps: S02. Perform task clustering on the same filling line. Based on the liquid formulation data of the filling tasks, divide the filling tasks with similar flavors into the same cluster. For example, based on the liquid formulation data of the tasks, use clustering algorithms (such as K-means, hierarchical clustering) for analysis. The feature vector may include: the concentration vector of the main flavor substances; or directly calculate the contamination risk between filling tasks, divide the filling tasks with lower contamination risk into the same cluster, and obtain several task sets. S03. Sort the filling tasks within the same cluster. Assess the contamination risk between filling tasks using the established filling equipment contamination risk assessment model. Select the task with the lowest contamination risk relative to the current filling equipment state for filling sorting. This will determine the execution order of filling tasks within the cluster. Within each cluster, perform detailed task sorting to further reduce the risk of internal contamination chains. Since the tasks within the cluster have similar characteristics, the contamination risk is relatively controllable. Heuristic algorithms (such as greedy algorithms: always select the task with the lowest contamination risk relative to the current equipment state) can also be used for rapid production scheduling.

[0033] S04. For the sorting of different clusters, the filling tasks of the cluster are abstracted as virtual tasks. The pollution risk between the virtual task and the first filling task executed in the next cluster is calculated by using the established filling equipment pollution risk assessment model. The clusters are then sorted, and finally the sorting of filling tasks is completed.

[0034] By clustering and grouping filling tasks, it is possible to effectively sort and plan according to the different products being filled, thereby reducing the risk of contamination of key flavor compounds in the filled products, shortening cleaning time, and improving the operating efficiency of filling equipment. For cleaning types between clusters, heavy cleaning can be forcibly selected to block the transmission of contamination, thereby reducing the computational load of subsequent filling tasks. At the same time, a contamination risk assessment model can be used to assess the contamination risk and select a suitable cleaning type, further improving the operating efficiency of filling equipment and shortening the total completion time.

[0035] Meanwhile, to more accurately describe the long-term contamination status of the equipment when continuously performing multiple tasks, the filling equipment is abstracted as a dynamic system with memory. When assessing the contamination risk between filling tasks, this invention uses the residual amount of key flavor compounds in the filled product. During continuous operation, a residue state vector is established for the filling equipment. Each time a key flavor compound is added during a filling task, the dimension of the residue state vector is increased accordingly, updating the key flavor compounds. Furthermore, when assessing the contamination risk between filling tasks, the cumulative residual amount of key flavor compounds affecting the flavor of the product in the next filling task is used for contamination risk assessment. This ensures that when performing multiple filling tasks, the contamination risk of different types of accumulated key flavor compounds on the next filled product can be comprehensively considered.

[0036] In calculating the residual amount of corresponding key flavor compounds, this invention establishes a one-to-one correspondence between the key flavor compounds and the residual retention rates of the corresponding cleaning types. The residual retention rate is calculated as: residual amount after cleaning / initial residual amount before cleaning = Ma / M = exp(-kc_j * tc). A cleaning effect matrix is ​​constructed using the residual retention rate. This cleaning effect matrix is ​​a diagonal matrix, and the diagonal elements of the matrix represent the residual retention rates corresponding to the key flavor compounds in the residue state vector. By using the diagonal matrix, the residual amount of key flavor compounds collected during cleaning in the filling equipment is calculated in batches to conduct a contamination risk assessment and improve calculation efficiency.

[0037] The cleaning effect matrix is ​​essentially a state transition matrix, establishing a dynamic relationship between operation and the operated state vector. When a preset cleaning program is initiated, mathematically, this is represented by multiplying this matrix by the residual state vector at the end of the previous task. This multiplication is not a simple overall cleaning, but rather an independent, proportional attenuation of the residual amount of each contaminant in the state vector, resulting in a new residual state vector after cleaning. This accurately simulates the different responses of substances with different chemical properties to the same cleaning program in the real world. For example, water washing is highly efficient at removing sugar residues, but may have a high retention rate for oily substances. This invention integrates multiple parallel exponential decay processes into a single mathematical structure, achieving a crucial leap from treating single contaminants to simultaneously batch-processing all monitored contaminants. This allows the system to predict the overall effect of complex cleaning actions on complex contamination with extremely high efficiency. Furthermore, through repeated application of the cleaning effect matrix, the system achieves dynamic tracking and prediction of the chain-like transmission of contamination.

[0038] The continuous process of the equipment performing tasks can be mathematically viewed as an iterative sequence of state vectors: upon completion of each task, a new contamination vector is added to the current state vector; and each cleaning process during task switching involves applying a cleaning effect matrix transformation to the current accumulated state vector. This cycle of superposition-transformation-re-superposition accurately characterizes the cumulative effects of contaminants across different tasks, enabling the system not only to simulate the risks of current task switching but also to predict potential contamination trends over long-term operation, thus possessing accurate predictive capabilities.

[0039] This invention also discloses a smart liquid food filling laboratory management system. Using the aforementioned smart liquid food filling laboratory management method, it includes a filling data processing module, a contamination risk assessment module, and a filling equipment control module for controlling the operation of filling equipment for filling and cleaning operations. The filling data processing module receives filling tasks and performs clustering and sorting. During the clustering and sorting process, the contamination risk assessment module assesses the contamination risk between filling tasks and determines the cleaning type between them, formulating a cleaning strategy for the filling task execution process. The filling equipment control module then performs filling according to the determined filling tasks and cleaning strategies. Furthermore, those skilled in the art can add corresponding functional modules based on the operational management and monitoring needs of the filling equipment, without departing from the scope of protection of this invention.

[0040] like Figure 1As shown, taking the execution of a liquid food filling task with multiple flavors as an example, the task input is first performed. The filling data processing module receives filling tasks of liquid foods with different flavors, conducts a risk assessment of the geophysical risks between filling tasks, calculates the contamination risk index between tasks, and for filling tasks with multiple similar flavors, tasks can be clustered according to flavor similarity and assigned to the same cluster. Then, the execution order of tasks within the same cluster and between clusters is sorted through contamination risk assessment, and the cleaning type between adjacent tasks is determined. The complete execution plan for the filling task is output, and the filling equipment control module controls the operation of the filling equipment according to the execution plan to complete the execution of the filling task.

[0041] This invention discloses an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to enable the device to perform the above-described intelligent filling laboratory management method for liquid food.

[0042] This invention discloses a computer-readable storage medium storing a computer program. When the computer program is run, the device running the computer program implements the above-mentioned intelligent filling laboratory management method for liquid food.

Claims

1. A smart filling laboratory management method, characterized in that, Including the following steps: S01. Establish a contamination risk assessment model for filling equipment in advance to assess the risk of flavor contamination caused by residues in the filling equipment to the product in the next filling task, and determine the cleaning type of the filling equipment when switching filling tasks based on the contamination risk. When conducting a contamination risk assessment, first determine the initial residue content in the filling equipment that will affect the next filling task, and then assess the contamination risk between filling tasks based on the residue content that the product in the next filling task can tolerate. S02. Cluster the filling tasks on the same filling line and group filling tasks with similar flavors into the same cluster based on the liquid formulation data of the filling tasks; or calculate the contamination risk between filling tasks and group filling tasks with lower contamination risk into the same cluster. S03. Sort the filling tasks within the same cluster, assess the contamination risk between filling tasks using the established filling equipment contamination risk assessment model, and select the task with the lowest contamination risk relative to the current state of the filling equipment for filling sorting. S04. For the sorting of different clusters, the filling tasks of the cluster are abstracted as virtual tasks. The pollution risk between the virtual task and the first filling task executed in the next cluster is calculated by using the established filling equipment pollution risk assessment model. The clusters are then sorted, and finally the sorting of filling tasks is completed.

2. The intelligent filling laboratory management method for liquid food according to claim 1, characterized in that, In step S01, when determining the corresponding initial residue content in the filling equipment, the residue is classified into liquid phase residue and gas phase residue according to its state. When calculating liquid phase residue, first determine the surface area of ​​the filling equipment that is in direct contact with the liquid phase based on the three-dimensional structural model of the filling equipment, and then determine the theoretical liquid phase residue of various substances that are theoretically attached to the contact surface of the filling equipment based on the adsorption amount of the corresponding filling material components on the surface of the filling equipment. When calculating the gas phase residue, the headspace area of ​​the equipment that contains volatile gas phase substances is determined based on the three-dimensional structural model of the filling equipment. The theoretical gas phase residue of various substances in the filling equipment is determined based on the filling operation temperature and the saturated vapor pressure of each component of the filling product.

3. A method for intelligent filling laboratory management of liquid food according to claim 1 or 2, characterized in that, In S01, when determining the cleaning type of the filling equipment during the switching of filling tasks based on contamination risk, a cleaning model for the cleaning type on the residue is established. Based on the determined initial residue content in the filling equipment, the removal effect of various cleaning types on the initial residue is calculated, the residue content in the filling equipment after cleaning is determined, and the contamination risk corresponding to each cleaning type is evaluated based on the residue content in the filling equipment after cleaning and the residue content that the product of the next filling task can tolerate. The cleaning type with the lowest contamination risk and the shortest time is selected.

4. The intelligent filling laboratory management method for liquid food according to claim 3, characterized in that, The cleaning model is an exponential decay model, where the cleaning residue Ma = M * exp(-kc * tc), where M represents the initial residue content, kc represents the cleaning efficiency coefficient (obtained through experimental calibration of the cleaning type), and tc represents the cleaning time.

5. The intelligent filling laboratory management method for liquid food according to claim 4, characterized in that, When assessing the pollution risk, the theoretical concentration of residue mixed into the product of the next filling task is used for evaluation. The residue content per unit volume of the filling task product is compared with the residue content that the filling product can tolerate. The theoretical concentration C1 = (Mal + Ma2 × η) / V, where Ma1 represents liquid phase residue, Ma2 represents gas phase residue, η represents the release probability coefficient of gas phase residue, with a value range of 0.5-1.0, and V represents the volume of the filling task product. The dose ratio of the residue mixed into the filling task product to the residue content that the filling product can tolerate is R = C1 / C2, where C2 is the residue concentration that the filling product can tolerate. The dose ratio R is mapped to a risk value of 0-1 using a nonlinear function for pollution risk assessment.

6. The intelligent filling laboratory management method for liquid food according to claim 1, characterized in that, When conducting contamination risk assessment between filling tasks, the residual amount of key flavor compounds in the filled product is used to assess the contamination risk between filling tasks. When the filling equipment is operating continuously, a residue state vector of the filling equipment is established. When performing a filling task, the dimension of the residue state vector is increased accordingly for each additional key flavor compound. When conducting contamination risk assessment between filling tasks, the cumulative residual amount of key flavor compounds that affect the flavor of the product in the next filling task is used to assess the contamination risk.

7. The intelligent filling laboratory management method for liquid food according to claim 6, characterized in that, When calculating the residual amount of corresponding key flavor compounds, the key flavor compounds are calibrated one-to-one with the residual retention rate of the corresponding cleaning type. The residual retention rate = residual amount after cleaning / residual amount before cleaning. A diagonal matrix is ​​constructed using the residual retention rate, and the diagonal elements of the matrix are the residual retention rates corresponding to the key flavor compounds in the residue state vector. The residual amount of key flavor compounds in the cleaning of the filling equipment is calculated in batches through the diagonal matrix to conduct a contamination risk assessment.

8. A smart liquid food filling laboratory management system, using the smart liquid food filling laboratory management method according to any one of claims 1-7, characterized in that, It includes a filling data processing module, a contamination risk assessment module, and a filling equipment control module for controlling the operation of filling equipment to carry out filling and cleaning operations. The filling data processing module receives filling tasks and performs clustering and sorting. During the clustering and sorting process, the contamination risk assessment module assesses the contamination risk between filling tasks and determines the cleaning type between filling tasks, formulates cleaning strategies during the execution of filling tasks, and performs filling according to the determined filling tasks and cleaning strategies through the filling equipment control module.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute the computer program to cause the device to perform the intelligent filling laboratory management method for liquid food as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is run, the device running the computer program implements the intelligent filling laboratory management method for liquid food as described in any one of claims 1-7.

Images