A microbial control analysis method and system

Abstract

The present invention discloses a microbial control analysis method and system, which relates to the field of microbial control analysis. The method comprises the following steps: identifying key microbial communities; constructing a microbial diffusion prediction model based on historical microbial concentration information, environmental parameters and identified key microbial community data, and using the trained microbial diffusion prediction model to predict the diffusion trend of microorganisms and potential contaminated areas in the future time period; synchronizing the prediction results and real-time monitoring data into a digital twin model, updating the simulation state of microbial diffusion; using the digital twin model to conduct simulation experiments to verify the effectiveness of pre-set microbial culture environment control measures; the system comprises an identification module, a prediction module, a diffusion simulation module and a measure verification module. The present invention evaluates the effect of control measures without intervening in the actual environment, reducing the cost of trial and error.

Description

Technical Field

[0001] The present invention relates to the field of microbial control analysis, and in particular to a microbial control analysis method and system. Background Art

[0002] Microorganisms play an important role in the fields of medicine, agriculture, and food industry, such as disease diagnosis and treatment in medicine and fermentation processes in the food industry. The control and analysis of microorganisms are crucial to the development of these fields. Traditional microbial experiments usually rely on manual operations and a single data source, and the monitoring methods are relatively limited, making it difficult to achieve comprehensive and real-time monitoring and control of the microbial growth environment. This leads to a certain degree of challenge in the reliability and real-time nature of the experimental results.

[0003] As an emerging virtual modeling and simulation technology, digital twin technology has been widely used in the fields of industry, medicine, etc. in recent years. Through digital twin technology, a digital model that is updated synchronously with the actual system can be established to achieve accurate monitoring and prediction of entities or systems. Microbial control analysis methods based on digital twin technology can effectively solve the problems existing in traditional methods, realize intelligent monitoring and control of microbial growth environments, and improve experimental efficiency and the credibility of results. The application of digital twin technology will bring new development opportunities and breakthroughs to the field of microbiology.

[0004] For example, Chinese patent 202311448693.6 discloses a soil microbial control analysis method and system based on digital twins, which uses digital twin technology to construct a plant simulation growth model to predict the growth and distribution of plant rhizosphere, which can improve soil health and agricultural production. However, the above method still has the following shortcomings: the method does not identify key microbial communities, which will weaken the simulation effect of subsequent digital twin models, and the failure to identify key microbial communities may lead to insufficient effectiveness of control measures.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a microbial control analysis method and system to overcome the above technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] According to one aspect of the present invention, a microbial control analysis method is provided, the microbial control analysis method comprising the following steps:

[0009] S1. Collect microbial distribution information and identify key microbial communities.

[0010] S2. Based on historical microbial concentration information, environmental parameters and identified key microbial community data, a microbial diffusion prediction model is constructed, and the trained microbial diffusion prediction model is used to predict the diffusion trend of microorganisms and potential contaminated areas in future time periods.

[0011] S3. Synchronize the prediction results and real-time monitoring data into the digital twin model and update the simulation status of microbial diffusion.

[0012] S4. Use the digital twin model to conduct simulation experiments to verify the effectiveness of pre-set microbial culture environment control measures.

[0013] Furthermore, collecting microbial distribution information and identifying key microbial communities include the following steps:

[0014] S11, assigning a weight based on ecological importance to each microorganism to obtain a microbial distribution data set containing the weight, performing data cleaning on the microbial distribution information, and encoding the microbial species;

[0015] S12, converting the collected microbial distribution information into a sample database, each sample representing a specific microbial sample;

[0016] S13, converting the sample database into a matrix form to obtain a sample matrix, wherein the elements of the sample matrix represent whether the microbial species appear in the sample;

[0017] S14, applying data compression storage technology, converting the sample matrix into a matrix containing only 0 and 1, obtaining a compressed matrix, and using two arrays to record the row position and column position of 1 in the matrix;

[0018] S15, construct support function by combining the weight and occurrence frequency of item sets;

[0019] S16. Use the improved association rule algorithm combined with the weight-adjusted support threshold to solve frequent item sets from the compression matrix;

[0020] S17. Preset a weight threshold, identify microbial communities whose weights are higher than the weight threshold and appear in the frequent item set, and obtain key microbial communities.

[0021] Furthermore, when the support function is constructed by combining the weight and frequency of the item set, the support function is:

[0022] ;

[0023] In the formula, x represents the type of microorganism;

[0024] S ( x) represents the support of microbial species x;

[0025] A ( x ) represents the number of occurrences of microbial species x;

[0026] λ ( x ) represents the weight of microbial species x;

[0027] B Indicates the total number of samples.

[0028] Furthermore, using the improved association rule algorithm, combined with the support threshold adjusted by weight, solving the frequent itemsets from the compression matrix includes the following steps:

[0029] S161, multiplying the weight of each microorganism by a preset minimum support threshold to calculate an adjusted support threshold for each microorganism;

[0030] S162, traversing the compression matrix, obtaining the number of times each microbial species appears in the sample, and calculating the support of each microbial species according to the support function;

[0031] S163, selecting microorganism species whose support is higher than the adjusted support threshold as candidate item sets, and obtaining a candidate item set list;

[0032] S164. Screen the candidate item set list, exclude item sets that are less than the minimum support threshold, and obtain frequent item sets.

[0033] Furthermore, based on the historical microbial concentration information, environmental parameters and identified key microbial community data, a microbial diffusion prediction model is constructed, and the trained microbial diffusion prediction model is used to predict the diffusion trend of microorganisms and potential contaminated areas in the future time period, including the following steps:

[0034] S21. Collect key microbial community data, historical microbial concentration information and environmental parameters for historical periods and integrate them into a unified data set, where each data point contains key microbial community data, historical microbial concentration information and environmental parameters within a certain period of time;

[0035] S22. Standardize the data in the unified data set;

[0036] S23, taking historical microbial concentration information, environmental parameters and data of key microbial communities as feature sets, and using the concentration at a certain point in the future as a label;

[0037] S24. Use sliding windows to construct feature sets for time series data.

[0038] S25, using the feature set of the label data and the time series data to train a model based on a gated recurrent unit to obtain a microbial diffusion prediction model;

[0039] S26. Evaluate the microbial diffusion prediction model using an independent test set, and use the microbial diffusion prediction model to output the microbial diffusion trend in the future time period based on the current microbial concentration information, environmental parameters and key microbial community data;

[0040] S27. Obtain potential contaminated areas based on microbial diffusion trends.

[0041] Furthermore, the following steps are included before the model based on gated recurrent units is trained using the feature set of label data and time series data to obtain the microbial diffusion prediction model:

[0042] Construct a loss function and measure the difference between the predicted value and the true value of the microbial diffusion prediction model. The loss function is:

[0043] ;

[0044] In the formula, a Represents the loss function value;

[0045] l Indicates the number of prediction points involved in calculating the loss value;

[0046] y ( t ) indicates the t The true value at a time point;

[0047] z ( t ) indicates the t Predicted value at a time point;

[0048] An optimization algorithm is used to update the parameters of the microbial diffusion prediction model according to the gradient of the loss function to minimize the loss function value.

[0049] Further, obtaining potential contaminated areas based on microbial diffusion trends includes the following steps:

[0050] S271. Combined with geographic information system technology, the microbial diffusion trend data predicted by the microbial diffusion prediction model is mapped to specific geographical locations, and through geographic information system analysis, the geographical areas with increased microbial concentrations are identified and potential contaminated areas are located;

[0051] S272. According to the potential impact of microbial concentration on the environment and public health, different levels of risk thresholds are set, and the microbial diffusion trend data predicted by the microbial diffusion prediction model are converted into specific pollution levels based on the use of risk thresholds;

[0052] S273. Visualize the polluted area on a map and display areas with different pollution levels.

[0053] Furthermore, the prediction results and real-time monitoring data are synchronized into the digital twin model, and the simulation state of microbial diffusion is updated, including the following steps:

[0054] S31. Integrate the prediction results and real-time monitoring data into a unified database and convert them into the format required by the digital twin model;

[0055] S32, importing the prediction results and real-time monitoring data into the digital twin model, and updating the state variables in the digital twin model;

[0056] S33. Run the simulation of the digital twin model to simulate the dynamic process of microbial diffusion.

[0057] Furthermore, the digital twin model is used to conduct simulation experiments to verify the effectiveness of the pre-set microbial culture environment control measures, including the following steps:

[0058] S41. Implement predetermined microbial culture environment control measures in the digital twin model;

[0059] S42. Run the digital twin model to obtain simulation results under different microbial culture environment control measures;

[0060] S43. Compare the effects of various microbial culture environment control measures to obtain microbial culture environment control measures that can achieve the predetermined goals. If the predetermined goals can be achieved, the corresponding microbial culture environment control measures are effective.

[0061] According to another aspect of the present invention, a microbial control analysis system is provided, which includes an identification module, a prediction module, a diffusion simulation module and a measure verification module.

[0062] Among them, the identification module is used to collect microbial distribution information and identify key microbial communities.

[0063] The prediction module is used to build a microbial diffusion prediction model based on historical microbial concentration information, environmental parameters and identified key microbial community data, and use the trained microbial diffusion prediction model to predict the diffusion trend of microorganisms and potential contaminated areas in future time periods.

[0064] The diffusion simulation module is used to synchronize the prediction results and real-time monitoring data into the digital twin model and update the simulation status of microbial diffusion.

[0065] The measure verification module is used to conduct simulation experiments using the digital twin model to verify the effectiveness of pre-set microbial culture environment control measures.

[0066] The beneficial effects of the present invention are:

[0067] (1) The present invention provides a microbial control analysis method and system that can achieve accurate prediction and effective control of microbial spread. Through detailed analysis and processing of microbial distribution data, the key microbial communities with the greatest impact on the environment are identified, providing important input for subsequent microbial spread prediction models. This helps narrow the scope of analysis and focus on the most critical factors, which is conducive to improving the simulation effect of the digital twin model, thereby making the control measures more effective.

[0068] (2) By using historical data and key microbial community information to build a microbial diffusion prediction model, the future microbial diffusion trend and potential contaminated areas can be accurately predicted. This is conducive to early identification of risk areas and the formulation of prevention and control measures. By synchronizing the prediction results and real-time monitoring data to the digital twin model, the digital twin model status can be updated in real time, so that the digital twin model reflects the current and future microbial diffusion situation, providing a dynamic and real-time updated data basis for simulation and prediction. Testing different control measures in the digital twin model can evaluate their effects without intervening in the actual environment, which helps to select the most effective measures, optimize resource allocation, and reduce trial and error costs.

[0069] (3) When identifying key microbial communities, the present invention assigns each microorganism a weight based on its ecological importance, which ensures that the analysis process pays more attention to those microbial species that have a greater impact on the environment. The microbial distribution information is systematized and stored in a structured form that is easy to process, namely, a sample database and a sample matrix, which provides a data basis for data mining and pattern recognition. Data compression technology reduces the resources required for data storage and processing while maintaining the integrity and availability of the data. By combining the weight and frequency of occurrence of the item set to construct a support function and using an improved association rule algorithm, frequently occurring item sets can be effectively mined from the data. By setting a weight threshold and identifying microbial communities with weights above the threshold and appearing in frequent item sets, the microbial communities that are most important to the ecosystem can be accurately determined.

[0070] (4) When constructing the microbial diffusion prediction model, the present invention measures the difference between the model prediction value and the true value through the loss function. By quantifying the difference between the prediction and the actual situation, the direction and degree of optimization can be clarified, that is, the model parameters are continuously adjusted to reduce the prediction error. The model parameters are continuously iterated using the optimization algorithm to improve the prediction accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0072] Figure 1 is a flow chart of a microbial control analysis method according to an embodiment of the present invention;

[0073] Figure 2 It is a principle block diagram of a microbial control and analysis system according to an embodiment of the present invention.

[0074] In the figure:

[0075] 1. Identification module; 2. Prediction module; 3. Diffusion simulation module; 4. Measures verification module. DETAILED DESCRIPTION

[0076] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0077] According to an embodiment of the present invention, a microbial control analysis method and system are provided.

[0078] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a microbial control analysis method is provided, and the microbial control analysis method comprises the following steps:

[0079] S1. Collect microbial distribution information, including species, location, and quantity, and identify key microbial communities.

[0080] In a further embodiment, collecting microbial distribution information and identifying key microbial communities comprises the following steps:

[0081] S11. Assign a weight based on ecological importance to each microorganism or community to obtain a microbial distribution data set containing weights, and the weights are based on the physiological characteristics, environmental impacts or importance of the microorganism in previous studies. The microbial distribution information is cleaned, missing values ​​and outliers are processed, and the microbial species are encoded. The purpose of encoding is to convert the name or identifier of the microorganism into a digital form that can be processed by a computer, such as one-hot encoding;

[0082] Assigning weights to each microorganism or microbial community based on its ecological importance, physiological properties, environmental impact, or importance in previous studies can help distinguish which microorganisms are more important for the functioning and stability of the ecosystem.

[0083] S12. Convert the collected microbial distribution information into a sample database, where each sample represents a specific microbial sample or observation point, which helps to organize and manage the microbial data during the analysis process.

[0084] S13. Convert the sample database into a matrix form to obtain a sample matrix, wherein the elements of the sample matrix represent whether the microbial species appear in the sample, thereby facilitating statistical analysis and pattern recognition of microorganisms.

[0085] S14, applying data compression storage technology to convert the sample matrix into a matrix containing only 0 and 1 to obtain a compressed matrix to reduce storage space, and using two arrays to record the row position and column position of 1 in the matrix to optimize data processing speed;

[0086] Data compression storage technology is a technology used to reduce data storage space. By compressing data, it can effectively reduce data storage and transmission costs, while improving data processing efficiency. Binarize the original data or convert it using a compression algorithm.

[0087] Binarization: keep the non-zero elements in the sample matrix as 1, and keep the zero elements as 0. Alternatively, define a threshold for the elements in the sample matrix, keep the elements greater than the threshold as 1, and set the elements less than the threshold as 0.

[0088] Utilize compression algorithm: Use sparse matrix compression algorithm to compress non-zero elements in the sample matrix into 1 positions, and use additional data structures to record the position information of non-zero elements. This can greatly reduce the storage space of the matrix and quickly restore the original matrix.

[0089] S15. Construct a support function by combining the weight and frequency of item sets to better mine infrequent but important microbial community patterns;

[0090] In a further embodiment, when constructing a support function in combination with the weight and occurrence frequency of the item set, the support function is:

[0091] ;

[0092] In the formula, x represents the type of microorganism;

[0093] S ( x ) represents the support of microbial species x;

[0094] A ( x ) represents the number of occurrences of microbial species x;

[0095] λ ( x ) represents the weight of microbial species x;

[0096] B Indicates the total number of samples. The support function helps identify microbial communities that are less frequent but have some importance in the entire dataset.

[0097] S16. Use the improved association rule algorithm combined with the weight-adjusted support threshold to solve frequent item sets from the compression matrix;

[0098] In a further embodiment, using an improved association rule algorithm in combination with a weight-adjusted support threshold, solving frequent item sets from a compression matrix includes the following steps:

[0099] S161. Multiply the weight of each microorganism by the preset minimum support threshold to calculate the adjusted support threshold of each microorganism; adjust the support threshold according to the importance or influence of the microorganism to mine frequent item sets more specifically. By considering the weight of the microorganism, the mining of microbial communities is more flexible and accurate, and the microbial communities with greater impact on the ecosystem can be better discovered.

[0100] S162, traverse the compression matrix, obtain the number of times each microbial species appears in the sample, and calculate the support of each microbial species according to the support function; combined with the support threshold after weight adjustment, effectively filter out low-frequency item sets, reduce unnecessary calculation and storage overhead, and improve the efficiency of the algorithm.

[0101] S163. Select microbial species with support higher than the adjusted support threshold as candidate item sets to obtain a candidate item set list. By dynamically adjusting the support threshold, frequent item sets can be mined more flexibly according to actual conditions, thereby improving the accuracy and interpretability of the mining results.

[0102] S164. Screen the candidate item set list, exclude the item sets with less than the minimum support threshold, and obtain frequent item sets. That is, filter out the frequent item sets with support higher than the threshold, extract the microbial communities with greater impact on the ecosystem, and provide valuable information for subsequent analysis and application.

[0103] It should be noted that the improved association rule algorithm adjusts the support threshold according to the weight of the microorganisms, flexibly mines frequent item sets, improves the efficiency of the algorithm and the accuracy of the results, is suitable for the mining and analysis of microbial community data, and can better discover key ecological information.

[0104] S17. Analyze frequent item sets and preset weight thresholds to identify microbial communities with weights higher than the weight thresholds and appearing in frequent item sets, obtain key microbial communities, and determine key microbial communities that have an important impact on ecosystem stability and function.

[0105] S2. Based on historical microbial concentration information, environmental parameters and identified key microbial community data, a microbial diffusion prediction model is constructed, and the trained microbial diffusion prediction model is used to predict the diffusion trend of microorganisms and potential contaminated areas in future time periods.

[0106] In a further embodiment, based on historical microbial concentration information, environmental parameters and identified key microbial community data, a microbial diffusion prediction model is constructed, and the trained microbial diffusion prediction model is used to predict the diffusion trend of microorganisms and potential contaminated areas in the future time period, including the following steps:

[0107] S21. Collect key microbial community data, historical microbial concentration information and environmental parameters for historical periods and integrate them into a unified data set, where each data point contains key microbial community data, historical microbial concentration information and environmental parameters within a certain time period.

[0108] S22. Standardize the data in the unified data set so that the numerical ranges of different features are relatively consistent, which is conducive to the stability and convergence speed of the model training process. Standardization avoids the impact of dimensional differences between data, making it easier for the model to learn the true relationship between data and improving the generalization ability of the model.

[0109] S23. Use historical microbial concentration information, environmental parameters, and key microbial community data as feature sets, and use the concentration at a certain point in the future as labels; establish a correspondence between features and labels to provide training data for the model. The feature set contains the key factors that affect microbial diffusion, while the label is the target of model prediction.

[0110] S24. Use sliding windows to construct feature sets for time series data to enhance the learning ability of the microbial diffusion prediction model for time series data. The process of microbial diffusion usually has a certain time correlation. The microbial concentration and environmental conditions at the previous moment will affect the subsequent diffusion. Therefore, time series data is considered to better capture the temporal relationship.

[0111] S25, using the feature set of the label data and the time series data to train a model based on a gated recurrent unit (GRU) to obtain a microbial diffusion prediction model;

[0112] In a further embodiment, the following steps are included before training a model based on a gated recurrent unit using a feature set of label data and time series data to obtain a microbial diffusion prediction model:

[0113] Construct a loss function and measure the difference between the predicted value and the true value of the microbial diffusion prediction model. The loss function is:

[0114] ;

[0115] In the formula, a Represents the loss function value, that is, the prediction error of the model. The smaller this value is, the closer the prediction result of the model is to the true value, and the better the model performance is.

[0116] l Indicates the number of prediction points involved in calculating the loss value, that is, the number of samples in the data set. In time series prediction, it is the total number of time points used for prediction.

[0117] y ( t ) indicates the t The true value at a time point refers to the true measured value of the actual microbial concentration at time point t.

[0118] z ( t ) indicates the t The predicted value at a time point is predicted by the model based on past data. t Predicted value of microbial concentration at a time point.

[0119] Use optimization algorithms, such as gradient descent method, Adam optimization algorithm, etc., and update the parameters of the microbial diffusion prediction model according to the gradient of the loss function to minimize the loss function value.

[0120] S26. Evaluate the microbial diffusion prediction model using an independent test set, and use the microbial diffusion prediction model to output the microbial diffusion trend in the future time period based on the current (or recent) microbial concentration information, environmental parameters and key microbial community data;

[0121] S27. Obtain potential pollution areas based on microbial diffusion trends. Predicting potential pollution areas can help decision makers take targeted measures to prevent or mitigate possible pollution incidents as early as possible and protect the stability of the environment and ecosystem.

[0122] In a further embodiment, obtaining a potential contaminated area based on a microbial diffusion trend comprises the following steps:

[0123] S271. Combined with geographic information system (GIS) technology, the microbial diffusion trend data predicted by the microbial diffusion prediction model is mapped to a specific geographical location, and through GIS analysis, the geographical areas with increased microbial concentrations are identified and potential contaminated areas are located. GIS technology can combine the data predicted by the model with geographical location information, converting abstract microbial diffusion trends into specific areas on the map, which is conducive to intuitively displaying the situation of microbial diffusion.

[0124] S272. According to the potential impact of microbial concentration on the environment and public health, different levels of risk thresholds are set, and based on the use of risk thresholds, the microbial diffusion trend data predicted by the microbial diffusion prediction model are converted into specific pollution levels, such as no pollution, light pollution, moderate pollution and heavy pollution.

[0125] S273. Visualize polluted areas on maps and show areas with different pollution levels to help managers and decision makers understand which areas may be affected.

[0126] S3. Synchronize the prediction results and real-time monitoring data into the digital twin model and update the simulation status of microbial diffusion.

[0127] In a further embodiment, synchronizing the prediction results and real-time monitoring data into the digital twin model and updating the simulation state of microbial diffusion includes the following steps:

[0128] S31. Integrate the prediction results and real-time monitoring data into a unified database and convert them into the format required by the digital twin model.

[0129] S32, import the prediction results and real-time monitoring data into the digital twin model, and update the state variables in the digital twin model, so that the digital twin model can timely reflect the changes and status of the real situation. The digital twin model is a virtual simulation of the real system. By combining real-time data with the model, the model can maintain the synchronization with the actual system, so that the model can more accurately simulate the dynamic process of microbial diffusion.

[0130] S33. Run the simulation of the digital twin model to simulate the dynamic process of microbial diffusion.

[0131] The digital twin model is a model based on the combination of physical models and data-driven models. It achieves virtual simulation and prediction of physical systems by matching the physical systems in the real world with digital simulation models. The digital twin model is used for simulation experiments, that is, by simulating the behavior and response of the digital twin model, the operation and changes of the system in the real world are simulated to better understand and predict the behavior of the real system.

[0132] S4. Use the digital twin model to conduct simulation experiments to verify the effectiveness of pre-set microbial culture environment control measures and make necessary adjustments.

[0133] In a further embodiment, a simulation experiment using a digital twin model to verify the effectiveness of a pre-set microbial culture environment control measure includes the following steps:

[0134] S41. Implement predetermined microbial culture environment control measures in the digital twin model, such as adjusting parameters, changing operating conditions, etc.; by simulating microbial culture environment control measures, explore the degree of influence of different control measures on the actual situation.

[0135] Among them, the microbial culture environment control measures are implemented through the automated control system, and the control measures include:

[0136] Environmental parameter adjustment: Adjust environmental parameters such as temperature, humidity, and pH value according to the needs of the microbial community.

[0137] Nutrient supply control: Control the supply of nutrients in the culture medium to optimize microbial growth conditions.

[0138] Bioreactor operation: Operate the bioreactor's mixing rate, oxygen supply, etc. to maintain optimal growth conditions.

[0139] Light management: For photosynthetic microorganisms, control the intensity and cycle of light.

[0140] Chemical addition: Inhibitors or promoters are added to control the growth of specific microorganisms.

[0141] S42. Run the digital twin model to obtain simulation results under different microbial culture environment control measures.

[0142] S43. Compare the effects of various microbial culture environment control measures to obtain microbial culture environment control measures that can achieve the predetermined goals. If the predetermined goals can be achieved, the corresponding microbial culture environment control measures are effective. The predetermined goals include reducing pollution, optimizing energy consumption, and protecting public health.

[0143] In addition, control measures are implemented in the actual microbial culture environment, such as adjusting temperature, humidity, pH value, etc. through an automated system, and data after control implementation is collected and fed back to the microbial diffusion prediction model and digital twin model for updating the microbial diffusion prediction model and digital twin model. The specific steps include:

[0144] Implement specific control measures through automated control systems to adjust environmental parameters to target values.

[0145] Use sensors and monitoring equipment to collect real-time data on environmental parameters (temperature, humidity, pH, etc.) and microbial growth data.

[0146] The collected data are recorded in a database.

[0147] The collected data are preprocessed to make them meet the input requirements of the microbial diffusion prediction model and the digital twin model.

[0148] The processed data are input into the microbial diffusion prediction model and digital twin model for model updating or calibration.

[0149] like Figure 2 As shown, according to another embodiment of the present invention, a microbial control analysis system is provided, which includes an identification module 1, a prediction module 2, a diffusion simulation module 3 and a measure verification module 4.

[0150] Among them, the identification module 1 is used to collect microbial distribution information and identify key microbial communities.

[0151] Prediction module 2 is used to build a microbial diffusion prediction model based on historical microbial concentration information, environmental parameters and identified key microbial community data, and use the trained microbial diffusion prediction model to predict the diffusion trend of microorganisms and potential contaminated areas in future time periods.

[0152] The diffusion simulation module 3 is used to synchronize the prediction results and real-time monitoring data into the digital twin model and update the simulation status of microbial diffusion.

[0153] The measure verification module 4 is used to conduct simulation experiments using the digital twin model to verify the effectiveness of pre-set microbial culture environment control measures.

[0154] In summary, the microbial control analysis method and system provided by the present invention can realize accurate prediction and effective control of microbial diffusion. Through detailed analysis and processing of microbial distribution data, the key microbial communities with the greatest impact on the environment are identified, which provides important input for the subsequent microbial diffusion prediction model. It helps to narrow the scope of analysis and focus on the most critical factors, which is conducive to improving the simulation effect of the digital twin model, thereby making the control measures more effective. By using historical data and key microbial community information to construct a microbial diffusion prediction model, the future microbial diffusion trend and potential contaminated areas can be accurately predicted. It is conducive to early identification of risk areas and formulation of prevention and control measures. And by synchronizing the prediction results and real-time monitoring data into the digital twin model, the digital twin model status can be updated in real time, so that the digital twin model reflects the current and future microbial diffusion situation, providing a dynamic and real-time updated data basis for simulation and prediction. Testing different control measures in the digital twin model can evaluate their effects without intervening in the actual environment, which helps to select the most effective measures, optimize resource allocation, and reduce trial and error costs. When identifying key microbial communities, the present invention can ensure that the analysis process pays more attention to those microbial species that have a greater impact on the environment by assigning each microorganism a weight based on its ecological importance. The microbial distribution information is systematized and stored in a structured form that is easy to process, namely, a sample database and a sample matrix, which provides a data basis for data mining and pattern recognition. The resources required for data storage and processing are reduced by data compression technology, while maintaining the integrity and availability of the data. By combining the weight and frequency of occurrence of the item set to construct a support function, and using an improved association rule algorithm, frequently occurring item sets can be effectively mined from the data. By setting a weight threshold, microbial communities with weights higher than the threshold and appearing in the frequent item set are identified, and those microbial communities that are most important to the ecosystem can be accurately determined. When constructing a microbial diffusion prediction model, the present invention measures the difference between the model prediction value and the true value through a loss function. By quantifying the difference between the prediction and the actual situation, the direction and degree of optimization can be clarified, that is, the model parameters are continuously adjusted to reduce the prediction error. The model parameters are continuously iterated using an optimization algorithm to improve the prediction accuracy of the model.

[0155] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A microbial control analysis method, characterized in that: The microbial control analysis method comprises the following steps: S1. Collect microbial distribution information and identify key microbial communities; S2. Based on historical microbial concentration information, environmental parameters and identified key microbial community data, a microbial diffusion prediction model is constructed, and the trained microbial diffusion prediction model is used to predict the diffusion trend of microorganisms and potential contaminated areas in the future time period; S3, synchronize the prediction results and real-time monitoring data into the digital twin model and update the simulation status of microbial diffusion; S4. Use the digital twin model to conduct simulation experiments to verify the effectiveness of the pre-set microbial culture environment control measures; The method of collecting microbial distribution information and identifying key microbial communities includes the following steps: S11, assigning a weight based on ecological importance to each microorganism to obtain a microbial distribution data set containing the weight, performing data cleaning on the microbial distribution information, and encoding the microbial species; S12, converting the collected microbial distribution information into a sample database, each sample representing a specific microbial sample; S13, converting the sample database into a matrix form to obtain a sample matrix, wherein the elements of the sample matrix represent whether the microbial species appear in the sample; S14, applying data compression storage technology, converting the sample matrix into a matrix containing only 0 and 1, obtaining a compressed matrix, and using two arrays to record the row position and column position of 1 in the matrix; S15, construct support function by combining the weight and occurrence frequency of item sets; S16. Use the improved association rule algorithm combined with the weight-adjusted support threshold to solve frequent item sets from the compression matrix; S17, presetting a weight threshold, identifying microbial communities whose weights are higher than the weight threshold and appear in the frequent item set, and obtaining key microbial communities; When the support function is constructed by combining the weight and occurrence frequency of the item set, the support function is: ； In the formula, x represents the type of microorganism; S ( x ) represents the support of microbial species x; A ( x ) represents the number of occurrences of microbial species x; λ ( x ) represents the weight of microbial species x; B represents the total number of samples; The method of using the improved association rule algorithm and the support threshold adjusted by weight to solve the frequent item sets from the compression matrix includes the following steps: S161, multiplying the weight of each microorganism by a preset minimum support threshold to calculate an adjusted support threshold for each microorganism; S162, traversing the compression matrix, obtaining the number of times each microbial species appears in the sample, and calculating the support of each microbial species according to the support function; S163, selecting microorganism species whose support is higher than the adjusted support threshold as candidate item sets, and obtaining a candidate item set list; S164. Screen the candidate item set list, exclude item sets that are less than the minimum support threshold, and obtain frequent item sets.

2. A microbial control analysis method according to claim 1, characterized in that: The method of constructing a microbial diffusion prediction model based on historical microbial concentration information, environmental parameters and identified key microbial community data, and using the trained microbial diffusion prediction model to predict the diffusion trend of microorganisms and potential contaminated areas in the future time period includes the following steps: S21. Collect key microbial community data, historical microbial concentration information and environmental parameters for historical periods and integrate them into a unified data set, where each data point contains key microbial community data, historical microbial concentration information and environmental parameters within a certain period of time; S22. Standardize the data in the unified data set; S23, taking historical microbial concentration information, environmental parameters and data of key microbial communities as feature sets, and using the concentration at a certain point in the future as a label; S24. Use sliding windows to construct feature sets for time series data. S25, using the feature set of the label data and the time series data to train a model based on a gated recurrent unit to obtain a microbial diffusion prediction model; S26. Evaluate the microbial diffusion prediction model using an independent test set, and use the microbial diffusion prediction model to output the microbial diffusion trend in the future time period based on the current microbial concentration information, environmental parameters and key microbial community data; S27. Obtain potential contaminated areas based on microbial diffusion trends.

3. A microbial control analysis method according to claim 2, characterized in that: The method of using the feature set of label data and time series data to train a model based on a gated recurrent unit to obtain a microbial diffusion prediction model includes the following steps: Construct a loss function and measure the difference between the predicted value and the true value of the microbial diffusion prediction model. The loss function is: ； In the formula, a Represents the loss function value; l Indicates the number of prediction points involved in calculating the loss value; y ( t ) indicates the t The true value at a time point; z ( t ) indicates the t Predicted value at a time point; An optimization algorithm is used to update the parameters of the microbial diffusion prediction model according to the gradient of the loss function to minimize the loss function value.

4. A microbial control analysis method according to claim 3, characterized in that: The method of obtaining the potential contaminated area based on the microbial diffusion trend comprises the following steps: S271. Combined with geographic information system technology, the microbial diffusion trend data predicted by the microbial diffusion prediction model is mapped to specific geographical locations, and through geographic information system analysis, the geographical areas with increased microbial concentrations are identified and potential contaminated areas are located; S272. According to the potential impact of microbial concentration on the environment and public health, different levels of risk thresholds are set, and the microbial diffusion trend data predicted by the microbial diffusion prediction model are converted into specific pollution levels based on the use of risk thresholds; S273. Visualize the polluted area on a map and display areas with different pollution levels.

5. A microbial control analysis method according to claim 4, characterized in that: The method of synchronizing the prediction results and real-time monitoring data into the digital twin model and updating the simulation state of microbial diffusion includes the following steps: S31. Integrate the prediction results and real-time monitoring data into a unified database and convert them into the format required by the digital twin model; S32, importing the prediction results and real-time monitoring data into the digital twin model, and updating the state variables in the digital twin model; S33. Run the simulation of the digital twin model to simulate the dynamic process of microbial diffusion.

6. A microbial control analysis method according to claim 5, characterized in that: The simulation experiment using the digital twin model to verify the effectiveness of the pre-set microbial culture environment control measures includes the following steps: S41. Implement predetermined microbial culture environment control measures in the digital twin model; S42. Run the digital twin model to obtain simulation results under different microbial culture environment control measures; S43. Compare the effects of various microbial culture environment control measures to obtain microbial culture environment control measures that can achieve the predetermined goals. If the predetermined goals can be achieved, the corresponding microbial culture environment control measures are effective.

7. A microbial control analysis system, used to implement the microbial control analysis method according to any one of claims 1 to 6, characterized in that: The microbial control analysis system includes an identification module, a prediction module, a diffusion simulation module and a measure verification module; Wherein, the identification module is used to collect microbial distribution information and identify key microbial communities; The prediction module is used to construct a microbial diffusion prediction model based on historical microbial concentration information, environmental parameters and identified key microbial community data, and use the trained microbial diffusion prediction model to predict the diffusion trend of microorganisms and potential contaminated areas in the future time period; The diffusion simulation module is used to synchronize the prediction results and real-time monitoring data into the digital twin model and update the simulation status of microbial diffusion; The measure verification module is used to conduct simulation experiments using a digital twin model to verify the effectiveness of pre-set microbial culture environment control measures.

Images