A method and system for monitoring the production safety of composite probiotics

Abstract

The present invention discloses a composite probiotic production safety monitoring method and system, which relates to the field of production monitoring technology, including deploying sensors to collect environmental data, analyzing environmental data to determine the stable state of the current environment and formulating corresponding measures, collecting raw material samples in strains and culture media for microbial safety detection after the environment is restored and stabilized, determining the qualified status of the raw materials and conducting pollution tracing analysis. The present invention collects environmental data in real time, and intelligently determines the initialization parameters of fermentation conditions according to the environmental state and the raw material state, thereby realizing full-process monitoring of the environment, fermentation conditions and strain morphological characteristics, and can timely capture the subtle characteristics of environmental fluctuations, raw material pollution and strain morphological changes, and generate corresponding control strategies based on abnormal classification results, effectively improving the safety and stability of production, realizing the precise positioning and classification of pollution sources in the production process, and being able to take targeted control strategies when pollution incidents occur to avoid production losses.

Description

Technical Field

[0001] The present invention relates to the technical field of production monitoring, and in particular to a method and system for monitoring the production safety of composite probiotics. Background Art

[0002] In recent years, with the rapid development of biotechnology, probiotics have been increasingly used in food, medicine, animal husbandry and other fields, especially in fermented products. Composite probiotics have become the focus of scientific research and industry because they can improve intestinal microecology, enhance immunity and prevent diseases through the synergistic effect of multiple strains. However, the production process of probiotics, especially the fermentation process, has extremely stringent environmental requirements. Factors such as temperature, humidity, and oxygen concentration directly affect the activity of the strain and the fermentation effect. With the advancement of industrial production, safety issues in probiotic production have received more and more attention, especially microbial contamination, abnormal fermentation conditions and other problems, which not only affect product quality, but also may bring potential safety hazards. Therefore, how to ensure the safety of the probiotic production process through effective monitoring methods has become an urgent problem to be solved. Summary of the invention

[0003] In view of the problems existing in the above-mentioned existing composite probiotic production safety monitoring method and system, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is how to ensure the safety of the probiotic production process through effective monitoring means, which has become an urgent problem to be solved.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for monitoring the production safety of a composite probiotic, comprising: deploying sensors to collect environmental data, analyzing the environmental data to determine the stability of the current environment and formulating corresponding measures, collecting raw material samples in the strains and culture medium for microbial safety testing after the environment is restored to stability, determining the eligibility of the raw materials and conducting pollution tracing analysis; initializing the fermentation conditions of the composite probiotics according to the environmental state and the raw material state, and monitoring the changes in condition parameters during the fermentation process in real time to determine the fluctuation of the conditions, collecting morphological images of the strains and extracting morphological features after the conditions fluctuate, detecting abnormal morphology of the strains and classifying them; formulating corresponding measures according to the abnormal classification results, storing the monitoring data during the production process and backing them up.

[0006] As a preferred solution of the composite probiotic production safety monitoring method of the present invention, wherein: the deployment of sensors to collect environmental data, analyzing the environmental data to determine the stability of the current environment and formulating corresponding measures refers to deploying various environmental sensors in the production workshop and fermentation tank, setting the sensor data collection frequency and collecting environmental data, performing format conversion and standardization processing on the collected real-time environmental data, and using a normalized format to map each type of environmental data to a standardized value between 0 and 1;

[0007] Set the safety thresholds corresponding to each parameter in the environmental data, and calculate the stability of each parameter based on the currently collected environmental data :

[0008] ;

[0009] In the formula, is the current parameter value, is the ideal parameter value, is the maximum fluctuation range of the parameter in the current time window, is the timeout penalty factor, is the cumulative time that the parameter exceeds the safety threshold;

[0010] The weighted sum of the stability indexes of all environmental parameters generates a comprehensive environmental stability index. , set the stability threshold ,like , it means that the overall environment is in a stable state, and the "environmental stability" mark is generated. , it means that the environment is unstable, and the "unstable environment" logo is generated, and the corresponding control equipment is started. When all environmental parameters remain stable for 10 consecutive minutes, the "qualified environment logo" is generated.

[0011] As a preferred solution of the composite probiotic production safety monitoring method of the present invention, wherein: the collecting of raw material samples in the strain and culture medium for microbial safety detection after the environment is restored to stability, judging the eligibility of the raw materials and conducting pollution tracing analysis refers to using a sterile sampler to collect strain samples and culture medium samples from the fermentation tank respectively, placing the collected samples into a sample pool of a PCR detection device for pre-treatment, and the PCR detection device automatically loads the required primers, probes and samples, and the detection is divided into two main tasks: pathogen concentration detection and strain concentration detection;

[0012] Set PCR amplification parameters according to the target microbial species to amplify the target gene sequence;

[0013] In each RCP amplification cycle, the fluorescence signal intensity of the target gene sequence in each sample is read by the fluorescent probe, and the amplified fluorescence signal intensity is converted into the target microorganism concentration Z:

[0014] ,

[0015] In the formula, is the number of target genes, is the number of PCR amplification cycles, V is the volume of the target sample involved in PCR amplification;

[0016] Compare the calculated bacterial strain concentration and pathogen concentration with the raw material safety standards:

[0017] If the concentration of pathogens exceeds 100 cfu / mL, it will be marked as "pathogen concentration exceeds the standard", otherwise it will be marked as "pathogen concentration is qualified". If the concentration of the target bacteria is less than 10,000 cfu / mL, it will be marked as "insufficient bacteria concentration", otherwise it will be marked as "qualified bacteria concentration". If the characteristic gene sequences of Salmonella, Listeria, and Escherichia coli are detected, it will be marked as "contaminants exceed the standard", otherwise it will be marked as "no contaminants";

[0018] Generate a raw material status mark based on the results. If all test items meet the standards, the raw material is qualified. If one or more test items do not meet the standards, the raw material is unqualified.

[0019] When the raw material test fails, traceability analysis is performed based on the environmental microbial concentration data to calculate the pollution traceability coefficient:

[0020] ,

[0021] Where H is the target microbial concentration detected in the raw material, PI is the microbial concentration in the ambient air, and j is the conversion coefficient;

[0022] Set the pollution concentration threshold to L. If the traceability coefficient is greater than the threshold L, it means that the pollutant concentration is higher than the environmental pollution concentration, and the pollution source is the raw material itself. Isolate the unqualified raw materials and stop the current fermentation operation. Otherwise, it indicates that the pollutant is related to the environmental pollution source. Suspend the current production batch and start the environmental re-cleaning operation.

[0023] As a preferred embodiment of the composite probiotic production safety monitoring method of the present invention, the method of initializing the fermentation conditions of the composite probiotics according to the environmental state and the raw material state and monitoring the changes of the condition parameters in the fermentation process in real time to determine the condition fluctuation refers to, after completing the safety confirmation of the environmental and raw material states, initially setting the fermentation parameters for the growth of the composite probiotics according to different raw materials and environmental characteristics, and automatically recording the real-time parameter values ​​in the fermentation tank as the fermentation benchmark data set after the initialization setting;

[0024] After fermentation, the fermentation parameters in the fermentation tank are collected every 10 seconds and the real-time data is compared with the benchmark data, and the parameter deviation value at each time point is recorded:

[0025] If the deviation value exceeds the ±5% range for more than 60 seconds, it is marked as a “conditional fluctuation event” and the specific time point and fluctuation parameters of the fluctuation are recorded.

[0026] As a preferred embodiment of the composite probiotic production safety monitoring method of the present invention, the method of collecting morphological images of bacterial strains and extracting morphological features after the conditions fluctuate, detecting abnormal morphological conditions of bacterial strains and classifying them means that after receiving the "condition fluctuation event" mark, 1-5 ml of fermentation liquid is automatically sampled from the fermentation tank every 2 hours, and after the sample liquid is stable, bacterial strain images are collected by laser confocal microscopy and the focus of the microscope is automatically adjusted, 5-10 continuous images are collected each time, and the images obtained under different lighting conditions are synthesized and image noise is removed using multispectral image fusion technology;

[0027] The denoised image is segmented using the U-Net image segmentation model to identify the morphological feature areas of bacterial cells, automatically annotate the features of each bacterial cell, and calculate the specific value of each morphological feature;

[0028] The specific value of calculating each morphological feature includes obtaining the hyphae extension length B of the strain by calculating the main axis length of each cell morphological region:

[0029] The specific value of calculating each morphological feature includes obtaining the hyphae extension length B of the strain by calculating the main axis length of each cell morphological region:

[0030] ;

[0031] Where n is the number of mycelium, and are the coordinates of the two ends of the ith mycelium, is the flow rate of the fermentation liquid, is the density of the fermentation liquid, is the liquid viscosity, is the spatial distribution factor of mycelium growth in the fermenter;

[0032] The growth density Y of the bacteria is determined by calculating the number of cells per unit area:

[0033] ;

[0034] In the formula, is the total number of cells per unit area, is the total sample area, The acceleration due to gravity, is the density of the cells, is the depth of the fermentation liquid, is the flow rate of the fermentation liquid;

[0035] The morphological consistency coefficient is calculated based on the area and shape characteristics of each cell to determine the consistency of the overall morphology of the strain C:

[0036] ;

[0037] In the formula, is the standard deviation of the cell area, is the mean cell area, is the aspect ratio deviation, is the ideal cell aspect ratio, is the average distance between cells, is the cell density, is the adjustment index parameter;

[0038] The current morphological feature data are compared with the historical benchmark data to obtain the comprehensive deviation value of all morphological features. The random forest model is used to perform abnormal classification according to the degree of deviation of the morphological features, and the abnormal conditions of the strain morphology are classified into mild abnormal state, moderate abnormal state and severe abnormal state.

[0039] As a preferred embodiment of the composite probiotic production safety monitoring method of the present invention, the formulating corresponding measures according to the abnormal classification results means that if the bacterial species morphology is slightly abnormal, the temperature control system parameters of the fermentation tank are adjusted to reduce the temperature to the target value;

[0040] If the bacterial morphology is moderately abnormal, the fermentation process is suspended, the stirring system of the fermentation tank is stopped, a fresh air source is introduced to increase the dissolved oxygen concentration of the fermentation liquid, and a fermentation liquid sample is collected for pollutant detection;

[0041] If the bacterial morphology is severely abnormal, stop the fermentation equipment, shut down all external air sources and nutrient supply systems, sterilize the fermentation tank with high-temperature steam, and eliminate all potential sources of contamination.

[0042] As a preferred scheme of the composite probiotic production safety monitoring method of the present invention, the storage and backup of monitoring data in the production process refers to classifying all production process data according to data type and archiving them according to timestamp and batch number, using local database and cloud storage system for data storage, and performing regular and incremental backup of data.

[0043] Another object of the present invention is to provide a composite probiotic production safety monitoring system, which comprises:

[0044] Environmental monitoring module, used to collect environmental data, determine the stability of environmental status and implement corresponding control measures;

[0045] The raw material collection module is used to collect bacterial strains and culture medium samples, analyze the eligibility of raw materials and conduct pollution traceability analysis under the premise of stable environmental conditions;

[0046] The morphology monitoring module is used to set the initial fermentation conditions according to the environmental state and the raw material state and compare the deviation values ​​of the real-time fermentation condition parameters and the benchmark data to generate a condition fluctuation event mark, and automatically collect fermentation liquid samples and perform morphological feature analysis when the fermentation conditions fluctuate;

[0047] The comprehensive evaluation module is used to formulate corresponding fermentation control strategies according to the classification results of morphological abnormalities and store the monitoring data and analysis results.

[0048] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of a composite probiotic production safety monitoring method when executing the computer program.

[0049] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for monitoring the production safety of composite probiotics.

[0050] The beneficial effects of the present invention are as follows: the present invention realizes full-process monitoring of the environment, fermentation conditions and strain morphological characteristics by collecting environmental data in real time and intelligently judging the initialization parameters of the fermentation conditions according to the environmental status and the raw material status, and can timely capture the subtle characteristics of environmental fluctuations, raw material pollution and strain morphological changes, and generate corresponding control strategies based on abnormal classification results, thereby effectively improving the safety and stability of production. Through multi-dimensional data fusion technology and pollution tracing, the present invention realizes the precise positioning and classification of pollution sources in the production process, and can adopt targeted control strategies when pollution incidents occur to avoid production losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0052] Figure 1 Schematic diagram of the process for production safety monitoring of compound probiotics.

[0053] Figure 2 This is a schematic diagram of the structure of the compound probiotic production safety monitoring system. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0056] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0057] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a method for monitoring the production safety of a composite probiotic. The method for monitoring the production safety of a composite probiotic includes:

[0058] S1. Deploy sensors to collect environmental data, analyze environmental data to determine the current environmental stability and formulate corresponding measures. After the environment is restored to stability, collect samples of raw materials in bacterial strains and culture media for microbial safety testing, determine the eligibility of raw materials and conduct pollution traceability analysis;

[0059] Specifically, deploying sensors to collect environmental data, analyzing environmental data to determine the stability of the current environment and formulating corresponding measures refers to deploying various environmental sensors in production workshops and fermentation tanks;

[0060] The various environmental sensors mentioned above refer to temperature sensors deployed at the top and bottom of the production workshop and fermentation tanks to ensure that the temperature gradient changes caused by fermentation during the production process can be detected to prevent local overheating;

[0061] Install humidity sensors on the inner wall of the fermentation tank and the air circulation port of the workshop to ensure that the humidity conditions in the workshop and the fermentation tank can be monitored in real time to avoid abnormal growth of bacteria caused by too low or too high humidity;

[0062] Deploy oxygen concentration sensors in fermentation tanks and near ventilation equipment in production workshops to detect oxygen concentration and ensure sufficient oxygen supply during fermentation;

[0063] Deploy carbon dioxide concentration sensors at the fermentation tank outlet and workshop exhaust port to monitor the carbon dioxide content generated during the fermentation process in real time to prevent carbon dioxide accumulation from inhibiting bacterial growth;

[0064] Set up multiple microbial sampling points in the production workshop, including the outer wall of the fermentation tank, equipment connection ports, operating tables, pipeline connection points, personnel hands and air, etc., to collect surface microbial concentration data and air microbial concentration data respectively;

[0065] Ensure that the production area is fully covered to avoid blind spots. Set the sensor data collection frequency to collect temperature, humidity, oxygen concentration, carbon dioxide concentration, and environmental microbial concentration. Convert and standardize the collected real-time environmental data. Use the normalized format to map each environmental data to a standardized value between 0 and 1.

[0066] Set the safety thresholds corresponding to each parameter in the environmental data through actual application scenarios and experience, and calculate the stability of each parameter based on the currently collected environmental data :

[0067] ;

[0068] In the formula, is the current parameter value, is the ideal parameter value, which is calculated by the median of the safety threshold. For example, if the temperature safety threshold is 25 to 35 degrees, the ideal parameter value is 27.5 degrees. is the maximum fluctuation amplitude of the parameter in the current time window (the difference between the maximum and minimum values), It is the timeout penalty factor, which is used to control the impact of exceeding the safety time on the stability index. It is usually set according to the sensitivity of the actual environment to fluctuations. is the cumulative time that the parameter exceeds the safety threshold;

[0069] Introducing the accumulation mechanism of fluctuation amplitude and timeout penalty factor on the basis of the existing deviation analysis formula can more comprehensively reflect the real-time stability of environmental parameters, rather than relying solely on the deviation value at the current moment. By adding the fluctuation amplitude, the dramatic fluctuation of parameters over a period of time can be effectively reflected, which is very important for measuring the short-term instability of parameters. In practical applications, some parameters may quickly return to the ideal value after a short-term fluctuation, but if a parameter is in an unstable state for a long time (for example, the temperature is higher than the set value for a long time), then even if the fluctuation amplitude is small, it should be regarded as a potential risk source. Therefore, the introduction of the timeout penalty factor can enhance the comprehensiveness of the formula;

[0070] The stability index of all environmental parameters is weighted and summed to generate a comprehensive environmental stability index S2. A stability threshold Q is set. If S2≤Q, it means that the overall environment is in a stable state, and the "environmental temperature" mark is generated. If S2>Q, it means that the environment is unstable, and the "environmental instability" mark is generated, and corresponding control measures are taken;

[0071] The control measures refer to starting the corresponding control equipment according to the specific exceeding parameters marked by the "unstable environment" logo:

[0072] When the temperature exceeds the standard, the heating or cooling device is started; when the humidity exceeds the standard, the humidification or dehumidification device is started; when the oxygen concentration is too low, the oxygen supply system is started; when the carbon dioxide concentration exceeds the standard, the workshop exhaust system is started; when the air microorganism concentration exceeds the standard, the air purifier is started; and the data of the parameters exceeding the standard are collected every 10 seconds until the parameters return to a safe range;

[0073] When all environmental parameters remain stable for 10 consecutive minutes, an "Environmental Qualification Mark" is generated. The 10-minute status stability confirmation can effectively avoid the generation of erroneous marks caused by short-term fluctuations, ensuring that subsequent raw material testing is carried out in a stable and safe environment.

[0074] The stability threshold Q analyzes the system response characteristics under different Q values ​​by collecting historical data of environmental parameters, and gradually adjusts the specific value of Q based on this. By deploying a variety of sensors in the production workshop and fermentation tank, all-round real-time monitoring of the production environment can be achieved. After format conversion and standardization, the real-time collected environmental data can unify the data formats of different sources, eliminate the format inconsistency problem in data fusion, and improve data processing efficiency. By setting safety thresholds for various environmental parameters (such as temperature, humidity, oxygen concentration, etc.) and calculating the stability of each parameter in real time (including the deviation between the current value and the ideal value, and the accumulation of exceeding the standard time), the potential unstable factors in the system can be effectively identified. In particular, the introduction of the timeout penalty factor can increase the system's sensitivity to long-term fluctuations and prevent the system from being in a sub-healthy state for a long time without being discovered. By weighted summing the stability indexes of all environmental parameters, the comprehensive environmental stability index S2 is generated, which can achieve effective fusion and comprehensive analysis of multi-dimensional data. Setting the stability threshold Q can intuitively determine whether the overall environmental state is stable and generate an "environmentally stable" or "environmentally unstable" logo. The comprehensive environmental stability index can quantify the degree of fluctuation of the overall environmental state, making it easier for the system to adopt further control strategies. When the environmental state is judged to be unstable, the system can automatically start the corresponding control equipment (such as temperature control system, air purification system, etc.) to accurately control the fluctuation parameters and ensure that the production environment quickly returns to a stable state.

[0075] Furthermore, after the environment is restored and stabilized, raw material samples of the strains and culture media are collected for microbial safety testing, and the qualification of the raw materials is judged and pollution traceability analysis is performed, which means using a sterile sampler to collect strain samples and culture media samples from the fermentation tank respectively, and immediately sealing the samples after each sampling and storing them in a sterile refrigerator to prevent changes in the samples before testing;

[0076] The collected samples are placed in the sample pool of the PCR detection device for pre-treatment (including sample homogenization and DNA / RNA extraction). The PCR detection device automatically loads the required primers, probes and samples, and divides the detection into two main tasks: pathogen concentration detection and bacterial species concentration detection;

[0077] Set PCR amplification parameters, including temperature, time, and number of cycles, according to the target microbial species (pathogens, target bacteria) to amplify the target gene sequence;

[0078] In each RCP amplification cycle, the fluorescence signal intensity of the target gene sequence in each sample is read by the fluorescent probe, and the amplified fluorescence signal intensity is converted into the target microorganism concentration Z:

[0079] ,

[0080] In the formula, is the number of target genes. The fluorescence signal intensity is converted into the number of target genes by the standard curve method. Cycle is the number of PCR amplification cycles. V is the volume of the target sample involved in PCR amplification. Before the PCR experiment, it is set according to the actual sampling volume and the requirements of the PCR reaction system. It is measured during sample collection and the volume is corrected according to the sample dilution.

[0081] The original concentration calculation formula is calculated based on the amount of target substance and sample volume. During PCR amplification, the amount of target gene is not measured directly, but the signal intensity is detected by fluorescent probe. Indirectly reflected, the conversion factor in the standard curve method is introduced , that is, to establish the fluorescence signal intensity through the standard sample The number of genes, The conversion relationship between them can more accurately obtain the number of target genes from the fluorescence signal, which can effectively solve the problem that the original formula cannot handle the complex relationship between the fluorescence signal and the number of genes, so that the number of target genes can be accurately estimated in each PCR amplification cycle. During the PCR amplification process, the number of genes increases exponentially, so the number of target genes changes dynamically in each cycle, and the multiplication factor is introduced. ,in is the number of PCR amplification cycles, which can describe the changing pattern of the number of genes in each cycle, that is, the number of genes increases exponentially with the number of amplification cycles;

[0082] Compare the calculated bacterial strain concentration and pathogen concentration with the raw material safety standards:

[0083] If the concentration of pathogens exceeds 100 cfu / mL, it will be marked as "pathogen concentration exceeds the standard", otherwise it will be marked as "pathogen concentration meets the standard". The 100 cfu / mL threshold is set according to the safety standards of food or biological products and relevant laws and regulations;

[0084] If the target bacterial concentration is less than 10,000 cfu / mL, it is marked as "insufficient bacterial concentration", otherwise it is marked as "qualified bacterial concentration". The 10,000 cfu / mL threshold is set according to the efficacy requirements and production standards of probiotic products;

[0085] If the characteristic gene sequences of Salmonella, Listeria, and Escherichia coli are detected, it will be marked as "contaminants exceed the standard", otherwise it will be marked as "no contaminants";

[0086] Generate raw material status identification based on the results:

[0087] If all test items meet the standards, it means that the raw materials are qualified. If one or more test items do not meet the standards, it means that the raw materials are unqualified.

[0088] When the raw material test fails, traceability analysis is performed based on the environmental microbial concentration data to calculate the pollution traceability coefficient:

[0089] ;

[0090] Wherein, H is the target microbial concentration detected in the raw material, which is directly measured by the microbial concentration detection technology PCR detection, PI is the microbial concentration in the ambient air, which is measured by the air sampling equipment, j is the conversion coefficient, which is used to convert the microbial concentration in the air (cfu / m³) into a unit consistent with the microbial concentration in the liquid, and j is calibrated by experiment and optimized and adjusted according to the experimental data;

[0091] The pollution concentration threshold L is set according to the environmental pollution level and the allowable limit of raw material pollution. If the traceability coefficient is greater than the threshold L, it means that the pollutant concentration is higher than the environmental pollution concentration, and the pollution source is the raw material itself. Isolate the unqualified raw materials and stop the current fermentation operation. Otherwise, it indicates that the pollutant is related to the environmental pollution source. Suspend the current production batch and start the environmental re-cleaning operation.

[0092] Aseptic sampling can effectively avoid the introduction of external sources of contamination during the sample collection process, ensuring the representativeness and purity of the sample. Sample pretreatment (including DNA / RNA extraction and sample homogenization) can improve detection sensitivity and ensure that the target gene is effectively amplified during the amplification process. In addition, by setting different amplification parameters, high-throughput detection of a variety of pathogens and probiotics can be performed, providing accurate data support for subsequent concentration calculations and pollution traceability. Through PCR amplification combined with fluorescent probe technology, pathogens and target probiotic species can be accurately detected, and the target gene concentration in each amplification cycle can be analyzed in real time. After the sample is tested, generating the corresponding raw material status identification based on the concentration of the pathogen and target species is a key step in determining whether the raw material is qualified. The generation of this logo effectively distinguishes whether the raw materials are qualified or not by comparing the test results with the raw material safety standards, and provides detailed information on pollutant exceeding the standard based on the types of pollutants detected (such as Salmonella, Listeria, and E. coli). When the source of pollution comes from the environment, environmental re-cleaning operations can be taken in time to avoid further spread of pollution; when the source of pollution is the raw materials themselves, unqualified raw materials can be quickly isolated and fermentation operations can be stopped, thereby effectively controlling the risk of pollution and improving the safety of the production process.

[0093] S2. Initialize the fermentation conditions of the composite probiotics according to the environmental state and the raw material state, and monitor the changes in the condition parameters during the fermentation process in real time to determine the fluctuation of the conditions. After the conditions fluctuate, collect the morphological images of the strains and extract the morphological features, detect the abnormal morphology of the strains and classify them;

[0094] Specifically, the fermentation conditions of the composite probiotics are initialized according to the environmental state and the raw material state, and the changes of the condition parameters during the fermentation process are monitored in real time to determine the condition fluctuations. This means that after completing the safety confirmation of the environment and the raw material state, the fermentation parameters for the growth of the composite probiotics are initially set according to the different raw materials and environmental characteristics, including temperature, pH value, dissolved oxygen concentration and nutrient concentration. After the initialization setting, the real-time parameter values ​​in the fermentation tank are automatically recorded and used as the fermentation benchmark data set;

[0095] After fermentation, the fermentation parameters in the fermentation tank are collected every 10 seconds and the real-time data is compared with the benchmark data, and the parameter deviation value at each time point is recorded:

[0096] If the deviation value exceeds the ±5% range for more than 60 seconds, it is marked as a “conditional fluctuation event” and the specific time point and fluctuation parameters of the fluctuation are recorded.

[0097] The ±5% range is set based on the biological needs of bacterial growth, the control accuracy of the equipment and the comprehensive consideration of existing technical specifications. During the fermentation process of compound probiotics, the response time of the temperature control system, pH control system or oxygen concentration control system is usually between 30 and 90 seconds. Therefore, when the parameter fluctuation lasts for more than 60 seconds, it usually means that the existing control system has failed to restore the parameter to the target value within a reasonable time. At this time, it is necessary to restore the parameter stability through system alarm or further intervention operation. By testing the environmental status (including temperature, humidity, air quality, microbial contamination, etc.) and raw material status (including microbial safety, nutritional components, pollutant concentration, etc. of the raw materials) before fermentation, it can effectively prevent fermentation failure or product contamination caused by environmental pollution or unqualified raw materials, and can significantly improve the safety and stability of the fermentation process, ensuring that the fermentation can be started under ideal conditions. Initializing the fermentation parameters for the growth of compound probiotics according to different raw materials and environmental characteristics can ensure that each strain starts the growth process under suitable conditions. Setting the initial fermentation conditions according to the type of raw materials (such as the ratio of carbon source and nitrogen source) and the characteristics of the environment (such as temperature, pH, oxygen concentration, etc.) can maximize the activity of the strains and enable them to quickly enter the logarithmic growth phase, thereby improving production efficiency. During the fermentation process, the fermentation parameters in the fermentation tank are collected every 10 seconds, and the real-time data is compared with the benchmark data, which can realize high-frequency dynamic monitoring of the fermentation process. It can timely detect small fluctuations in fermentation conditions (such as temperature deviation, pH change, oxygen concentration fluctuation, etc.), so as to take corresponding control measures before the problem expands, and avoid fermentation failure due to condition fluctuations. If the deviation value of a parameter exceeds the range of ±5% for more than 60 seconds, it is marked as a "condition fluctuation event", which can accurately capture major fluctuations in fermentation conditions and record the specific time point and fluctuation parameters of the fluctuation, thereby providing reliable data support for subsequent abnormal analysis and control strategies. This design can realize accurate monitoring of the production process and reduce the risk of abnormal growth of strains caused by condition fluctuations.

[0098] Furthermore, after the condition fluctuates, the morphological image of the strain is collected and the morphological features are extracted. The abnormal morphology of the strain is detected and classified. After receiving the "condition fluctuation event" mark, 1-5 ml of fermentation liquid is automatically sampled from the fermentation tank every 2 hours. After the sample liquid is stable, the strain image is collected by laser confocal microscopy and the focus of the microscope is automatically adjusted. 5-10 continuous images are collected each time, and the images obtained under different lighting conditions are synthesized using multispectral image fusion technology to eliminate reflection and shadow noise under single lighting conditions;

[0099] Use denoising autoencoder to perform deep denoising on the image to further remove background impurities;

[0100] The denoised image is segmented using the U-Net image segmentation model to identify the morphological feature areas of bacterial cell boundaries, cytoplasm distribution, and hyphae length, automatically annotate the features of each bacterial cell (such as cell number, hyphae length, cell diameter, etc.) and calculate the specific value of each morphological feature:

[0101] The specific value of calculating each morphological feature includes obtaining the hyphae extension length B of the strain by calculating the main axis length of each cell morphological region:

[0102] The existing formula is expressed in terms of the length of the geometric principal axis:

[0103] ;

[0104] In the actual fermentation environment, the growth of mycelium is affected by the flow rate of the fermentation liquid. When the flow rate of the fermentation liquid is too fast, the growth length of the mycelium will be inhibited, while when the flow rate is too slow, the mycelium may be over-extended. , which can better describe the actual effect of fermentation liquid flow on mycelial morphology. The physical properties of the fermentation liquid (such as density and viscosity) have a greater impact on mycelial growth. Liquids with higher density will inhibit mycelial growth, while liquids with higher viscosity will increase mycelial growth resistance. By introducing and , which can more comprehensively reflect the actual impact of the liquid environment on the growth of mycelium. In a high-density fermentation environment, mycelium will produce a squeezing effect on each other, thus affecting its growth length. , which can better describe the actual growth state of mycelium in a high-density environment:

[0105] ;

[0106] Where n is the number of mycelium, and are the coordinates of the two ends of the ith mycelium. The starting and ending coordinates of the mycelium in the microscopic image are determined by the morphological segmentation algorithm. It is the flow rate of the fermentation liquid. The average flow rate of the fermentation liquid in the fermentation tank is collected in real time through an online flow rate sensor (such as an ultrasonic flow meter or a vortex flow meter). is the density of the fermentation liquid, measured by a density sensor, is the liquid viscosity, obtained by a rotational viscometer, It is the spatial distribution factor of mycelium growth in the fermenter, indicating the length inhibition effect caused by mycelium squeezing each other in the fermenter. ,in, is the number of cell units occupied by mycelium, is the total number of available cell units in the fermenter;

[0107] The growth density Y of the bacteria is determined by calculating the number of cells per unit area:

[0108] The existing technology uses the geometric proportional method to calculate:

[0109] ;

[0110] In the actual fermentation process, cell density depends not only on the geometric area, but also on gravity and liquid depth. Gravity can cause uneven cell density distribution, especially when the fermentation liquid is deep. Therefore, it is necessary to introduce a gravity correction factor. , which can more accurately describe the density distribution of cells during the actual fermentation process:

[0111] ;

[0112] In the formula, is the total number of cells per unit area. The number of each cell in the microscopic image is identified and counted through OpenCV contour extraction. is the total area of ​​the sample, converted to actual area using the pixels of the microscopic image (microscopic images usually have a fixed pixel-to-physical size ratio), is the acceleration due to gravity, is the density of the cells, measured by the mass and volume of the cell suspension, is the depth of the fermentation liquid, measured by the actual height of the fermentation tank or by a level sensor. is the flow rate of the fermentation liquid;

[0113] The morphological consistency coefficient is calculated based on the area and shape characteristics of each cell to determine the consistency of the overall morphology of the strain C:

[0114] The existing consistency formula is usually expressed as the ratio of the standard deviation of the cell area to the mean:

[0115] ,

[0116] In the actual fermentation environment, the morphology of cells depends not only on the area distribution, but also on the aspect ratio and the interaction between cells. The change of aspect ratio can reflect the morphological characteristics of cells in different growth stages (such as stretching, contraction, degeneration, etc.). By introducing the aspect ratio correction factor, the growth state of cells in different morphologies can be better described. The morphological consistency is also affected by the spatial distribution characteristics of cells in the fermentation environment. When the cell density is high, the interaction force between cells will cause the variation of cell morphology, thereby affecting the overall consistency. By introducing the aspect ratio correction factor, the cell morphology can be better described. The ratio can more accurately describe the consistency of cell populations under different density and distribution conditions. The traditional morphological consistency formula is a linear expression, while the consistency of actual cell morphology usually has nonlinear characteristics. The exponential function is used for nonlinear correction. The exponential function can better reflect the nonlinear relationship under complex morphology, thereby improving the description accuracy of the model:

[0117] ;

[0118] In the formula, is the standard deviation of the cell area, is the mean cell area, is the aspect ratio deviation, which indicates the difference between the current cell aspect ratio and the ideal aspect ratio. is the ideal cell aspect ratio, which usually depends on the cell morphology (e.g. the ideal aspect ratio of a round cell is 1), is the average distance between cells, calculated by the nearest neighbor distance of each cell in the microscopic image, is the cell density, It is a regulation index parameter, which is used to describe the influence of the relationship between cell spacing (average distance) and cell density on morphological consistency. It is adjusted according to the growth state of cells in the actual fermentation process.

[0119] The current morphological feature data are compared with the historical benchmark data to obtain the comprehensive deviation value of all morphological features. The random forest model is used to perform abnormal classification according to the degree of deviation of the morphological features, and the abnormal conditions of the strain morphology are classified into mild abnormal state, moderate abnormal state and severe abnormal state.

[0120] When fermentation conditions fluctuate, by automatically collecting fermentation liquid samples, using laser confocal microscopy to collect morphological images of strains, and combining multispectral image fusion technology to eliminate image noise, it is possible to ensure that high-quality microscopic images are obtained under different lighting conditions, thereby providing reliable data support for subsequent morphological feature extraction, significantly improving the accuracy of strain morphological monitoring, ensuring the integrity of morphological features, and avoiding misclassification due to poor image quality. The denoised microscopic images are segmented by the U-Net image segmentation model, which can identify the specific morphological features of strain cells (such as edge contours, cell density, hyphae extension length, etc.), and automatically mark the characteristic area of ​​each bacterial cell. Compared with traditional manual analysis methods, this model can efficiently and accurately extract complex morphological feature data in a short time, and provide high-quality input data for subsequent feature quantification. This automated feature extraction method can greatly improve the efficiency of morphological monitoring, especially in large-scale fermentation production. The morphological features are quantitatively described by their respective biological and physical indicators (such as the coordinates of the two ends of the mycelium, the density of the cells, the flow rate of the liquid, etc.). These parameters can fully reflect the growth state and morphological changes of the strains during the fermentation process. By quantitatively calculating the morphological changes of strains at different time points, the impact of fluctuations in fermentation conditions on strain growth can be quickly identified. Based on the quantification of morphological features, the random forest model is used to compare the morphological feature data with historical benchmark data, calculate the comprehensive deviation value of the morphological features, and classify the morphological abnormalities according to the degree of deviation (mild abnormality, moderate abnormality, severe abnormality). This classification process can accurately distinguish different types of morphological abnormalities, which helps to take appropriate corrective measures in the early stages of the occurrence of abnormalities to avoid further deterioration of the fermentation process. Based on the classification results of morphological abnormalities, the system can automatically formulate corresponding fermentation control strategies to ensure that fermentation conditions can quickly return to a safe state after the abnormality occurs.

[0121] S3. Formulate corresponding measures based on the abnormal classification results, store the monitoring data in the production process and back it up;

[0122] Specifically, according to the abnormal classification results, corresponding measures are formulated, that is, if the bacterial morphology is slightly abnormal, the temperature control system parameters of the fermentation tank are adjusted to reduce the temperature to the target value;

[0123] If the bacterial morphology is moderately abnormal, the fermentation process is suspended, the stirring system of the fermentation tank is stopped, fresh air or pure oxygen is introduced to increase the dissolved oxygen concentration of the fermentation liquid, and a sample of the fermentation liquid is collected for contaminant detection;

[0124] If the bacterial morphology is severely abnormal, stop the fermentation equipment, shut down all external air sources and nutrient supply systems, sterilize the fermentation tank with high-temperature steam, and eliminate all potential sources of contamination.

[0125] When the morphology of the strain shows mild abnormality (such as slight swelling or uneven morphology on the surface of the bacteria), it is usually due to the imbalance of the metabolic rate of the strain caused by temperature fluctuations during the fermentation process. By fine-tuning the temperature setting of the temperature control system, the metabolic rate of the strain can be effectively reduced to restore it to its normal morphological state. Moderate abnormalities are usually manifested as significant changes in the morphology of the strain cells (such as mycelium breakage, cell shape deformation, etc.), and may be accompanied by a decrease in bacterial activity. At this time, simple temperature adjustment is not enough to restore the strain state. It is necessary to suspend the fermentation process and stop stirring to reduce the mechanical shear force that may be generated in the fermentation liquid. Further damage to the strain. At the same time, by introducing fresh air to increase the dissolved oxygen concentration, the growth environment of the strain under high oxygen consumption conditions can be improved to prevent the occurrence of anaerobic metabolism. When the strain morphology is seriously abnormal (such as bacterial disintegration, mass cell death, discoloration of the fermentation liquid, etc.), it indicates that the fermentation environment can no longer maintain the normal metabolism of the strain. The fermentation process must be stopped immediately and all external nutrient supplies must be isolated to prevent further spread of the abnormality. High-temperature steam sterilization can effectively kill possible pathogens and contaminants, and thoroughly clean the fermentation equipment to prevent contaminants from remaining in subsequent production batches, thereby avoiding cross-contamination.

[0126] Furthermore, storing and backing up the monitoring data during the production process means classifying all production process data according to the data type and archiving them according to timestamps and batch numbers, using local databases (such as SQL databases) and cloud storage systems (such as AWS S3 or Azure Blob Storage) for data storage, and performing regular and incremental backups of the data.

[0127] By classifying the monitoring data in the production process according to the data type, the efficiency of data management can be effectively improved, and the search time and complexity of data processing can be reduced. Archiving the monitoring data by timestamp and batch number can effectively realize the hierarchical management and traceability of production data. The advantage of the local database is that it can achieve low-latency access and local processing of data, which is suitable for storing data that needs frequent access and real-time processing (such as real-time monitoring data). The strategy of combining regular backup and incremental backup can greatly reduce the storage space and time required for backup while ensuring data integrity.

[0128] Example 2, reference Figure 2 , which is the second embodiment of the present invention, is different from the previous embodiment and provides a composite probiotic production safety monitoring system, which includes:

[0129] Environmental monitoring module, used to collect environmental data, determine the stability of environmental status and implement corresponding control measures;

[0130] The raw material collection module is used to collect bacterial strains and culture medium samples, analyze the eligibility of raw materials and conduct pollution traceability analysis under the premise of stable environmental conditions;

[0131] The morphology monitoring module is used to set the initial fermentation conditions according to the environmental state and the raw material state and compare the deviation values ​​of the real-time fermentation condition parameters and the benchmark data to generate a condition fluctuation event mark, and automatically collect fermentation liquid samples and perform morphological feature analysis when the fermentation conditions fluctuate;

[0132] The comprehensive evaluation module is used to formulate corresponding fermentation control strategies according to the classification results of morphological abnormalities and store the monitoring data and analysis results.

[0133] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0135] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0136] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. A method for monitoring the production safety of composite probiotics, characterized in that: include, Deploy sensors to collect environmental data, analyze environmental data to determine the current environmental stability and formulate corresponding measures, collect strains and raw material samples in culture medium for microbial safety testing after the environment is restored to stability, determine the eligibility of raw materials and conduct pollution traceability analysis; Initialize the fermentation conditions of the composite probiotics according to the environmental status and the raw material status, and monitor the changes in the condition parameters during the fermentation process in real time to determine the condition fluctuations. After the conditions fluctuate, collect the morphological images of the strains and extract the morphological features, detect the abnormal morphology of the strains and classify them; Formulate corresponding measures according to the abnormal classification results, store the monitoring data in the production process and back them up; deploy sensors to collect environmental data, analyze the environmental data to determine the stability of the current environment and formulate corresponding measures, which means deploying various environmental sensors in the production workshop and fermentation tank, collecting environmental data after setting the sensor data collection frequency, converting the format and standardizing the collected real-time environmental data, and mapping each environmental data to a standardized value between 0 and 1 using a normalized format; Set the safety thresholds corresponding to each parameter in the environmental data, and calculate the stability of each parameter based on the currently collected environmental data : ； In the formula, is the current parameter value, is the ideal parameter value, is the maximum fluctuation range of the parameter in the current time window, is the timeout penalty factor, is the cumulative time that the parameter exceeds the safety threshold; The weighted sum of the stability indexes of all environmental parameters generates a comprehensive environmental stability index. , set the stability threshold ,like , it means that the overall environment is in a stable state, and the "environmental stability" mark is generated. , it means that the environment is unstable, an "unstable environment" mark is generated, and the corresponding control equipment is started. When all environmental parameters remain stable for 10 consecutive minutes, an "environmental qualification mark" is generated; after the environment is restored to stability, the raw material samples in the strains and culture medium are collected for microbial safety testing, and the qualified status of the raw materials is judged and the pollution source traceability analysis is performed. It means that the strain samples and culture medium samples are collected from the fermentation tank respectively using a sterile sampler, and the collected samples are placed in the sample pool of the PCR detection device for pre-treatment. The PCR detection device automatically loads the required primers, probes and samples, and the detection is divided into two main tasks: pathogen concentration detection and strain concentration detection; Set PCR amplification parameters according to the target microbial species to amplify the target gene sequence; In each RCP amplification cycle, the fluorescence signal intensity of the target gene sequence in each sample is read by the fluorescent probe, and the amplified fluorescence signal intensity is converted into the target microorganism concentration Z: ， In the formula, is the number of target genes, is the number of PCR amplification cycles, V is the volume of the target sample involved in PCR amplification; Compare the calculated bacterial strain concentration and pathogen concentration with the raw material safety standards: If the concentration of pathogens exceeds 100 cfu / mL, it will be marked as "pathogen concentration exceeds the standard", otherwise it will be marked as "pathogen concentration meets the standard". If the concentration of the target bacteria is less than 10,000 cfu / mL, it will be marked as "insufficient bacteria concentration", otherwise it will be marked as "qualified bacteria concentration". If the characteristic gene sequences of Salmonella, Listeria, and Escherichia coli are detected, it will be marked as "contaminant exceeds the standard", otherwise it will be marked as "no contaminant". Generate a raw material status mark based on the results. If all test items meet the standards, the raw material is qualified. If one or more test items do not meet the standards, the raw material is unqualified. When the raw material test fails, traceability analysis is performed based on the environmental microbial concentration data to calculate the pollution traceability coefficient: ， Where H is the target microbial concentration detected in the raw material, PI is the microbial concentration in the ambient air, and j is the conversion coefficient; The pollution concentration threshold is set to L. If the traceability coefficient is greater than the threshold L, it means that the pollutant concentration is higher than the environmental pollution concentration, and the pollution source is the raw material itself. The unqualified raw material is isolated and the current fermentation operation is stopped. Otherwise, it indicates that the pollutant is related to the environmental pollution source, the current production batch is suspended and the environmental re-cleaning operation is started; the fermentation conditions of the composite probiotics are initialized according to the environmental state and the raw material state, and the changes in the condition parameters during the fermentation process are monitored in real time to determine the condition fluctuations. After completing the safety confirmation of the environment and raw material state, the fermentation parameters for the growth of the composite probiotics are initially set according to different raw materials and environmental characteristics. After the initialization setting, the real-time parameter values ​​in the fermentation tank are automatically recorded and used as the fermentation benchmark data set; After fermentation, the fermentation parameters in the fermentation tank are collected every 10 seconds and the real-time data is compared with the benchmark data, and the parameter deviation value at each time point is recorded: If the deviation value exceeds the range of ±5% for more than 60 seconds, it is marked as a "condition fluctuation event", and the specific time point and fluctuation parameters of the fluctuation are recorded; the said collecting morphological images of strains and extracting morphological features after the condition fluctuates, detecting abnormal morphology of strains and classifying them means that after receiving the "condition fluctuation event" mark, 1-5 ml of fermentation liquid is automatically sampled from the fermentation tank every 2 hours, and after the sample liquid is stable, the strain image is collected by laser confocal microscopy and the focus of the microscope is automatically adjusted, 5-10 continuous images are collected each time, and the images obtained under different lighting conditions are synthesized and image noise is removed using multispectral image fusion technology; The denoised image is segmented using the U-Net image segmentation model to identify the morphological feature areas of bacterial cells, automatically annotate the features of each bacterial cell, and calculate the specific value of each morphological feature; The specific value of calculating each morphological feature includes obtaining the hyphae extension length B of the strain by calculating the main axis length of each cell morphological region: ； Where n is the number of mycelium, and are the coordinates of the two ends of the ith mycelium, is the flow rate of the fermentation liquid, is the density of the fermentation liquid, is the liquid viscosity, is the spatial distribution factor of mycelium growth in the fermenter; The growth density Y of the bacteria is determined by calculating the number of cells per unit area: ； In the formula, is the total number of cells per unit area, is the total sample area, Gravity acceleration, is the density of the cells, is the depth of the fermentation liquid, is the flow rate of the fermentation liquid; The morphological consistency coefficient is calculated based on the area and shape characteristics of each cell to determine the consistency of the overall morphology of the strain C: ； In the formula, is the standard deviation of the cell area, is the mean cell area, is the aspect ratio deviation, is the ideal cell aspect ratio, is the average distance between cells, is the cell density, is the adjustment index parameter; The current morphological feature data is compared with the historical benchmark data to obtain the comprehensive deviation value of all morphological features, and the abnormality classification is performed according to the degree of deviation of the morphological features through the random forest model, and the abnormal conditions of the strain morphology are classified into a mild abnormal state, a moderate abnormal state and a severe abnormal state; the formulation of corresponding measures according to the abnormal classification results means that if the strain morphology is mildly abnormal, the temperature control system parameters of the fermentation tank are adjusted to reduce the temperature to the target value; If the bacterial morphology is moderately abnormal, the fermentation process is suspended, the stirring system of the fermentation tank is stopped, a fresh air source is introduced to increase the dissolved oxygen concentration of the fermentation liquid, and a sample of the fermentation liquid is collected for contaminant detection; If the morphology of the strain is severely abnormal, stop the fermentation equipment, shut down all external gas sources and nutrient supply systems, sterilize the fermentation tank with high-temperature steam, and eliminate all potential sources of contamination; storing and backing up the monitoring data during the production process means classifying all production process data according to the data type and archiving them according to timestamps and batch numbers, using local databases and cloud storage systems for data storage, and performing regular and incremental backups of the data.

2. A composite probiotic production safety monitoring system based on the composite probiotic production safety monitoring method according to claim 1, characterized in that: include, Environmental monitoring module, used to collect environmental data, determine the stability of environmental status and implement corresponding control measures; The raw material collection module is used to collect bacterial strains and culture medium samples, analyze the eligibility of raw materials and conduct pollution traceability analysis under the premise of stable environmental conditions; The morphology monitoring module is used to set the initial fermentation conditions according to the environmental state and the raw material state and compare the deviation values ​​of the real-time fermentation condition parameters and the benchmark data to generate a condition fluctuation event mark, and automatically collect fermentation liquid samples and perform morphological feature analysis when the fermentation conditions fluctuate; The comprehensive evaluation module is used to formulate corresponding fermentation control strategies according to the classification results of morphological abnormalities and store the monitoring data and analysis results.

3. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the composite probiotic production safety monitoring method of claim 1 are implemented.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the composite probiotic production safety monitoring method of claim 1 are implemented.

Images