Pig bacterial disease prevention and control decision support system and optimization method

Abstract

The invention discloses a pig bacterial disease prevention and control decision support system and an optimization method, and particularly relates to the technical field of pig bacterial disease prevention and control, and the system specifically comprises the following modules: a data acquisition module, a pathogen analysis module, a prediction and early warning module, a decision control module, an execution module and a self-adaptive optimization module. According to the invention, environment and swinery health micro-changes are accurately captured in real time through the data acquisition module, and dynamic matching and parameter correction of the pathogen analysis module are combined to construct a'dynamic fingerprint 'model of a bacteria growth environment; the prediction and early warning module realizes 48-hour risk level prediction based on LSTM and a random forest algorithm, and generates a'prevention and control strategy matrix 'through risk mapping; and the adaptive optimization module provides a low-cost and high-efficiency bacterial disease prevention and control solution for the pig farm through a reinforcement learning continuous iteration strategy.

Description

Technical Field

[0001] This invention relates to the field of swine bacterial disease prevention and control technology, and in particular to a decision support system and optimization method for swine bacterial disease prevention and control. Background Technology

[0002] In large-scale pig farming, the traditional prevention and control model for swine bacterial diseases relies on manual inspections and empirical medication. This model suffers from problems such as lagging monitoring, ambiguous pathogen identification, crude environmental control and medication strategies, and static prevention and control strategies. As a result, some growth problems occur during pig farming, such as the inability of humans to capture fluctuations in environmental parameters and abnormalities in the pig herd in real time, easy misdiagnosis of pathogens, lagging equipment control, high medication waste rate, difficulty in dynamically correcting the system model, high resource consumption, and isolated data with insufficient regional coordination. It is difficult to achieve accurate prediction, targeted intervention and dynamic optimization, so there is an urgent need for a new generation of intelligent, data-driven prevention and control systems. Summary of the Invention

[0003] The main objective of this invention is to provide a decision support system and optimization method for the prevention and control of swine bacterial diseases, which can effectively solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: Decision support system and optimization methods for the prevention and control of swine bacterial diseases, including: Data acquisition module: used to collect real-time data on piggery environmental parameters and pig health status; Pathogen analysis module: used to determine the types of bacteria infecting the target pig herd and their optimal growth environment parameter range; Prediction and early warning module: Based on machine learning models, it analyzes the correlation between historical environmental data and epidemics, predicts the risk of pathogen proliferation in the next 24-48 hours, and triggers audible and visual alarms and pushes emergency response plans to user terminals when environmental parameters deviate from the target control range or the abnormal proportion of pigs exceeds the threshold. Decision control module: used to perform correlation analysis between the pig herd health status data and the optimal growth environment parameter range of bacteria, and generate environmental regulation instructions and drug administration strategies; Execution module: used to adjust the environmental parameters of the pigsty according to the environmental control instructions, and to administer targeted medication to the target pig herd according to the drug administration strategy; Adaptive optimization module: It dynamically adjusts the control strategy through reinforcement learning to balance the control effect and resource consumption.

[0005] Preferably, the data acquisition module includes: Environmental sensor unit: installed inside the pigsty, used to collect data on temperature, humidity, ammonia concentration and ventilation volume; A health monitoring unit includes a body temperature sensor and an image recognition device. The body temperature sensor is attached to the surface of the pig's body, and the image recognition device is used to capture abnormal behavioral characteristics of the pig herd. The data transmission unit is connected to the environmental sensor group and the health monitoring module, and is used to transmit the collected data to the decision control module in real time.

[0006] Preferably, the pathogen analysis module specifically includes: Bacterial database unit: Stores optimal growth temperature, humidity, pH value, and drug resistance data for common swine bacterial pathogens; Matching unit: used to match the corresponding pathogenic bacteria species and their sensitive drug types from the bacterial database based on the pig herd health status data; Dynamic correction unit: used to update the environmental adaptation model of pathogenic bacteria by combining real-time environmental parameters with historical epidemic data.

[0007] Preferably, the decision control unit specifically includes: Parameter comparison unit: used to calculate the difference between the current pigsty environmental parameters and the range of parameters for the optimal growth environment of bacteria, and to determine the target control range for inhibiting bacterial growth; Strategy generation unit: used to generate composite instructions for ventilation, temperature and humidity adjustment and disinfection frequency based on the target control range using a fuzzy control algorithm; Priority allocation unit: used to allocate medication order and dosage according to the degree of infection in the pig herd.

[0008] Preferably, the execution module includes: Environmental control unit: includes variable frequency fan, spray cooling device and ultraviolet disinfection lamp; Intelligent drug delivery device: includes a drug storage bin, a metering pump, and a directional nozzle. The directional nozzle is installed above the feeding trough in the pig pen and is used to spray atomized drugs. Both the environmental control component and the intelligent drug delivery device are electrically connected to the decision control unit.

[0009] Preferably, the directional nozzle of the intelligent drug delivery device is a rotatable structure, the spray angle is controlled by a servo motor, and an anti-clogging filter is provided at the nozzle outlet.

[0010] Preferably, the prediction and early warning module includes: Data preprocessing unit: Cleans, standardizes, and extracts features from the input pathogen types, environmental parameters, and historical epidemic data to provide structured data for model training; Model training unit: Based on the random forest algorithm, the prediction model is trained to learn the correlation between environmental parameters and pathogen proliferation, and output the risk level for the next 12-48 hours; Risk mapping unit: Transforms risk levels into specific prevention and control recommendations, and labels them with confidence levels; Historical database: Stores long-term environmental monitoring data, pathogen detection results, and records of control effects, supporting iterative model training and validation; The adaptive optimization module includes: Effectiveness evaluation sub-unit: Real-time collection of actual data after implementation, calculation of prevention and control effectiveness and cost indicators; Strategy optimization subunit: Based on reinforcement learning algorithm, dynamically adjust parameters such as temperature and humidity thresholds and drug dosage; Parameter update subunit: Feeds back the optimized strategy parameters to the decision control unit in real time; Real-time feedback interface: connects the execution unit and the strategy optimization subunit.

[0011] This invention also provides a method for optimizing decision-making in the prevention and control of swine bacterial diseases, comprising the following steps: S1, Data Acquisition: Real-time acquisition of pig house environmental parameters, including but not limited to temperature, humidity, ammonia concentration and ventilation volume, and acquisition of pig health data, including but not limited to abnormal body temperature signals and abnormal behavioral characteristics. S2. Multi-source data fusion: Integrating environmental data, pig genome information, and feed composition data to construct a multi-dimensional decision-making model; S3. Pathogen analysis: Based on the pig herd health data, a preset bacterial database is matched to determine the target pathogen bacteria species and their optimal growth temperature and humidity range; S4. Decision generation: Dynamically compare the current environmental parameters with the optimal growth range to generate a composite decision instruction that includes temperature and humidity control thresholds, disinfection frequency, and drug administration priority. S5. Execution and control: According to the composite decision command, the ventilation and spraying equipment in the pig house are adjusted in a coordinated manner to control the temperature and humidity, and the targeted drug delivery device is activated to spray antibacterial agents onto the target pig herd according to priority. S6. Dynamic calibration: The number of pathogen colonies is collected every 6 hours, and the temperature and humidity thresholds and drug dosage are corrected in real time through feedback control algorithm.

[0012] Preferably, in step S4, determining the temperature and humidity control threshold includes: calculating the deviation between the current environmental parameters and the optimal growth range of bacteria using a fuzzy control algorithm.

[0013] Preferably, the allocation of drug administration priority includes: prioritizing the administration of high-concentration drugs to high-risk pigs based on the severity of individual infection in the pig herd and the drug resistance data of pathogenic bacteria, and providing real-time feedback on efficacy data.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention forms a full-chain intelligent prevention and control system through the closed-loop logic of data acquisition, analysis, decision-making, and execution optimization. The data acquisition module captures temperature, humidity, ammonia concentration, pig herd temperature, and behavioral data in real time through environmental sensors and health monitoring units. The data is then uploaded to the pathogen analysis module in seconds via the data transmission unit. The latter uses the KNN algorithm to match the bacterial database and combines the gradient descent method to dynamically correct the pathogen growth model, thereby identifying potential pathogens and their sensitive agents.

[0015] 2. This invention uses a prediction and early warning module to analyze historical data based on a random forest algorithm to predict pathogen proliferation risk 48 hours in advance. A decision control module compares the current environment with the optimal growth parameters of the pathogen, generates composite instructions using a fuzzy control algorithm, and executes them precisely through a variable frequency fan, spraying device, and intelligent drug delivery system. The directional nozzles, powered by servo motors, achieve "millimeter-level" drug coverage in the pig pen, saving 60% of the drug compared to administering medication to the entire herd. Simultaneously, an adaptive optimization module, based on a DQN reinforcement learning algorithm, dynamically adjusts temperature and humidity thresholds and drug dosage. These modules work together to construct a smart prevention and control closed loop of "real-time monitoring - precise source tracing - intelligent early warning - targeted intervention - dynamic optimization," providing an efficient, precise, and sustainable bacterial disease prevention and control solution for large-scale farming. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system flow of the present invention; Figure 2 This is a schematic diagram of the operation flow of the method of the present invention. Detailed Implementation

[0017] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0018] like Figure 1 As shown, this embodiment provides a decision support system for the prevention and control of swine bacterial diseases, including: Data acquisition module: used to collect real-time data on piggery environmental parameters and pig health status; The data acquisition module includes: Environmental sensor unit: installed inside the pigsty, used to collect data on temperature, humidity, ammonia concentration and ventilation volume; The health monitoring unit includes a body temperature sensor and an image recognition device. The body temperature sensor is attached to the pig's body surface, and the image recognition device is used to capture abnormal behavioral characteristics of the pig herd, such as capturing pig behavior (e.g., coughing, lameness) through a camera. Timely detection of abnormal body temperature (threshold: T) pig Temperatures >40℃ and abnormal behaviors (such as lying down for more than 80% of the time) The data transmission unit is connected to the environmental sensor group and the health monitoring module to transmit the collected data to the decision control module in real time.

[0019] By linking environmental sensor groups (temperature / humidity, ammonia, ventilation volume) with health monitoring units (body temperature sensor + image recognition) in real time, a three-in-one monitoring network of "environment, physiology and behavior" is formed. For example, when the image recognition device captures that the pigs spend more than 80% of their time lying down and their body temperature is >40℃, the system automatically triggers the pathogen analysis module and matches potential pathogens (such as Streptococcus suis) within 0.5 seconds using the KNN algorithm, which improves efficiency by more than 90% compared to traditional manual inspection.

[0020] Pathogen analysis module: used to determine the types of bacteria infecting the target pig herd and their optimal growth environment parameter range; The pathogen analysis module specifically includes: Bacterial database unit: Stores optimal growth temperature, humidity, pH value and drug resistance data of common swine bacterial pathogens, such as storing the optimal growth parameters of pathogenic bacteria; Matching unit: Used to match the corresponding pathogenic bacteria species and their sensitive drug types from the bacterial database based on the health status data of the pig herd; Specifically, based on pig herd health data (such as body temperature and behavioral abnormalities), the KNN algorithm is used to match pathogen types in the database. The formula for calculating the matching degree is as follows: Where: x i For current data characteristics (such as body temperature, humidity); y i For pathogen characteristics in the database; ω i The feature weights are (e.g., body temperature weight = 0.6).

[0021] Dynamic correction unit: used to update the environmental adaptability model of pathogenic bacteria by combining real-time environmental parameters with historical epidemic data; If gradient descent is used to adjust the optimal parameters, the specific formula involved is as follows: Where α is the learning rate (default 0.1); The original optimal temperature (known or initial setting value); The current ambient temperature; This is the updated optimal temperature, used to dynamically adapt to environmental changes.

[0022] Prediction and early warning module: Based on machine learning models (such as random forest), it analyzes the correlation between historical environmental data and epidemics, predicts the risk of pathogen proliferation in the next 24-48 hours, and triggers audible and visual alarms and pushes emergency response plans to user terminals when environmental parameters deviate from the target control range or the abnormal proportion of pigs exceeds the threshold. The prediction and early warning module includes: Data preprocessing unit: Normalizes the raw data based on the input pathogen type, environmental parameters (temperature, humidity, etc.), and historical epidemic data, and extracts time-series features (such as the 24-hour moving average). The specific normalization formula is as follows: in The data points are standardized, with a mean of 0 and a standard deviation of 1; σ is the standard deviation of the dataset, which measures the dispersion of the data; μ is the mean of the dataset; and x is the original data point, representing the specific value collected at a certain moment (such as temperature, humidity, etc.).

[0023] Model training unit: Based on the random forest algorithm, the prediction model is trained to learn the correlation between environmental parameters and pathogen proliferation, and outputs the risk level (low / medium / high) for the next 12-48 hours; the formulas involved in the above random forest algorithm are as follows: in, Input data (environmental parameters + historical epidemic data); The number of decision trees is set (default 100), thus enabling 48-hour risk level prediction; Risk mapping unit: Converts risk levels into specific prevention and control recommendations (e.g., "high risk" corresponds to "increase disinfection frequency to once per hour"), and marks the confidence level; Historical database: Stores long-term environmental monitoring data, pathogen detection results, and records of control effects, supporting iterative model training and validation; Decision control module: used to perform correlation analysis between pig herd health status data and the optimal growth environment parameter range of bacteria, and generate environmental regulation instructions and drug administration strategies; The decision control unit specifically includes: Parameter comparison unit: Used to calculate the difference between the current pigsty environmental parameters and the optimal bacterial growth environment parameter range, to determine the target control range for inhibiting bacterial growth; the specific formula for calculating the difference between the current environment and the optimal bacterial parameters is as follows: Strategy generation unit: Used to generate composite instructions for ventilation, temperature and humidity adjustment, and disinfection frequency based on the target control range using a fuzzy control algorithm. Specifically, the temperature control formula is as follows: Achieve temperature and humidity control error of less than ±1°; Priority allocation unit: used to allocate medication order and dosage according to the degree of infection in the pig herd.

[0024] Execution module: Used to adjust the environmental parameters of the pig house according to environmental control instructions, and to administer targeted drugs to the target pig herd according to the drug administration strategy; The execution module includes: Environmental control unit: includes variable frequency fan, spray cooling device and ultraviolet disinfection lamp. The variable frequency fan and spray cooling device are linked to achieve precise temperature and humidity regulation. Intelligent drug delivery device: includes a drug storage bin, a metering pump, and a directional nozzle. The directional nozzle is installed above the feeding trough in the pig pen for spraying atomized drugs. The intelligent drug delivery device has a rotatable nozzle, whose spray angle is controlled by a servo motor, and an anti-clogging filter is provided at the nozzle outlet. The directional nozzle is adjusted at an angle by a servo motor to ensure that the medicine covers the target pigs; Both the environmental control components and the intelligent drug delivery device are electrically connected to the decision control unit.

[0025] Adaptive optimization module: Dynamically adjusts control strategies through reinforcement learning to balance control effectiveness with resource consumption (such as energy saving and reducing drug use). The adaptive optimization module includes: Effectiveness evaluation sub-unit: Real-time collection of actual data after implementation (pathogen inhibition rate, pig herd health recovery rate, energy consumption and drug consumption), and calculation of prevention and control effectiveness and cost indicators; The strategy optimization subunit, based on reinforcement learning algorithms (such as DQN), dynamically adjusts parameters such as temperature and humidity thresholds and drug dosage with the goal of maximizing pathogen inhibition rate and minimizing resource consumption. The strategy is optimized using the DQN algorithm, with the reward function as follows: Where λ i These are the weighting coefficients, default. ; Further improve the efficiency of prevention and control.

[0026] Parameter update subunit: Feeds back the optimized strategy parameters (such as "lower the temperature threshold by 1℃") to the decision control unit in real time for the next round of decision generation; Real-time feedback interface: connects the execution unit and the strategy optimization subunit to ensure low-latency data transmission and support second-level response.

[0027] The adaptive optimization module dynamically balances the control effect and resource consumption through the DQN algorithm. For example, when the activity level of pigs decreases at night, the system automatically lowers the temperature threshold by 1°C and reduces ventilation by 20%, ensuring a pathogen inhibition rate of >90% while reducing energy consumption by 15% and drug usage by 12%.

[0028] In the operation of this embodiment, a closed-loop logic of data acquisition, analysis, decision-making, and execution optimization is coordinated to form a full-chain intelligent prevention and control system. The data acquisition module captures temperature and humidity, ammonia concentration, pig herd temperature and behavior data in real time through environmental sensors and health monitoring units. The data is then uploaded to the pathogen analysis module in seconds via the data transmission unit. The latter uses the KNN algorithm to match the bacterial database and combines the gradient descent method to dynamically correct the pathogen growth model, locking in potential pathogens and their sensitive agents within 0.5 seconds.

[0029] Meanwhile, the prediction and early warning module analyzes historical data based on the random forest algorithm, predicts the risk of pathogen proliferation 48 hours in advance, and normalizes the feature values ​​through the data preprocessing unit, mapping the risk level to an "environmental control + drug administration" strategy.

[0030] The same decision control module compares the current environment with the optimal growth parameters of the pathogen, uses a fuzzy control algorithm to generate compound instructions, and executes them precisely through the variable frequency fan, spray device and intelligent drug delivery system of the execution module. The directional nozzle achieves "millimeter-level" drug coverage in the pig pen through a servo motor, saving 60% of the drug dosage compared to administering the drug to the entire herd.

[0031] Furthermore, the adaptive optimization module, based on the DQN reinforcement learning algorithm, dynamically adjusts temperature and humidity thresholds and drug dosage. For example, it automatically lowers the temperature by 1°C and reduces ventilation by 20% at night, reducing energy consumption by 15% while maintaining an antibacterial rate of >90%. All modules ultimately construct a smart prevention and control closed loop of "real-time monitoring - precise traceability - intelligent early warning - targeted intervention - dynamic optimization," reducing drug usage by 50%-70% and breeding costs by 15%-20%, providing an efficient, precise, and sustainable bacterial disease prevention and control solution for large-scale farming.

[0032] This invention also provides a method for optimizing decision-making in the prevention and control of swine bacterial diseases, comprising the following steps: S1, Data Acquisition: Real-time acquisition of pig house environmental parameters, including but not limited to temperature, humidity, ammonia concentration and ventilation volume, and acquisition of pig health data, including but not limited to abnormal body temperature signals and abnormal behavioral characteristics. S2. Multi-source data fusion: Integrate environmental data, pig genome information (such as susceptibility gene markers), and feed composition data to construct a multi-dimensional decision-making model; S3. Pathogen Analysis: Based on pig herd health data, match a pre-set bacterial database to determine the target pathogen bacteria species and their optimal growth temperature and humidity range; S4. Decision generation: Dynamically compare the current environmental parameters with the optimal growth range to generate a composite decision instruction that includes temperature and humidity control thresholds, disinfection frequency, and drug administration priority. In S4, the determination of the temperature and humidity control threshold includes: calculating the deviation between the current environmental parameters and the optimal growth range of bacteria through a fuzzy control algorithm, dynamically adjusting the target temperature and humidity range to ensure that it simultaneously meets the requirements of inhibiting bacterial reproduction and the comfort of the pig herd. The allocation of drug administration priorities includes: prioritizing the administration of high-concentration drugs to high-risk pigs based on the severity of individual infections in the pig herd and data on the drug resistance of pathogenic bacteria, and providing real-time feedback on efficacy data to adjust subsequent drug administration strategies; S5. Execution and control: Based on the compound decision-making instructions, the ventilation and spraying equipment in the pig house are adjusted in a coordinated manner to control the temperature and humidity, and the targeted drug delivery device is activated to spray antibacterial agents onto the target pig herd according to priority. S6. Dynamic calibration: The number of pathogen colonies is collected every 6 hours, and the temperature and humidity thresholds and drug dosage are corrected in real time through feedback control algorithms (such as model predictive control).

[0033] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A decision support system for the prevention and control of swine bacterial diseases, characterized in that, include: Data acquisition module: used to collect real-time data on piggery environmental parameters and pig health status; Pathogen analysis module: used to determine the types of bacteria infecting the target pig herd and their optimal growth environment parameter range; Prediction and early warning module: Based on machine learning models, it analyzes the correlation between historical environmental data and epidemics, predicts the risk of pathogen proliferation in the next 24-48 hours, and triggers audible and visual alarms and pushes emergency response plans to user terminals when environmental parameters deviate from the target control range or the abnormal proportion of pigs exceeds the threshold. Decision control module: used to perform correlation analysis between the pig herd health status data and the optimal growth environment parameter range of bacteria, and generate environmental regulation instructions and drug administration strategies; Execution module: used to adjust the environmental parameters of the pigsty according to the environmental control instructions, and to administer targeted medication to the target pig herd according to the drug administration strategy; Adaptive optimization module: It dynamically adjusts the control strategy through reinforcement learning to balance the control effect and resource consumption.

2. The decision support system for the prevention and control of swine bacterial diseases according to claim 1, characterized in that: The data acquisition module includes: Environmental sensor unit: installed inside the pigsty, used to collect data on temperature, humidity, ammonia concentration and ventilation volume; A health monitoring unit includes a body temperature sensor and an image recognition device. The body temperature sensor is attached to the surface of the pig's body, and the image recognition device is used to capture abnormal behavioral characteristics of the pig herd. The data transmission unit is connected to the environmental sensor group and the health monitoring module, and is used to transmit the collected data to the decision control module in real time.

3. The decision support system for the prevention and control of swine bacterial diseases according to claim 1, characterized in that: The pathogen analysis module specifically includes: Bacterial database unit: Stores optimal growth temperature, humidity, pH value, and drug resistance data for common swine bacterial pathogens; Matching unit: used to match the corresponding pathogenic bacteria species and their sensitive drug types from the bacterial database based on the pig herd health status data; Dynamic correction unit: used to update the environmental adaptation model of pathogenic bacteria by combining real-time environmental parameters with historical epidemic data.

4. The decision support system for the prevention and control of swine bacterial diseases according to claim 1, characterized in that: The decision control unit specifically includes: Parameter comparison unit: used to calculate the difference between the current pigsty environmental parameters and the range of parameters for the optimal growth environment of bacteria, and to determine the target control range for inhibiting bacterial growth; Strategy generation unit: used to generate composite instructions for ventilation, temperature and humidity adjustment and disinfection frequency based on the target control range using a fuzzy control algorithm; Priority allocation unit: The order and dosage of medication are allocated according to the degree of infection in the pig herd.

5. The decision support system for the prevention and control of swine bacterial diseases according to claim 1, characterized in that: The execution module includes: Environmental control unit: includes variable frequency fan, spray cooling device and ultraviolet disinfection lamp; Intelligent drug delivery device: includes a drug storage bin, a metering pump, and a directional nozzle. The directional nozzle is installed above the feeding trough in the pig pen and is used to spray atomized drugs. Both the environmental control component and the intelligent drug delivery device are electrically connected to the decision control unit.

6. The decision support system for the prevention and control of swine bacterial diseases according to claim 5, characterized in that: The directional nozzle of the intelligent drug delivery device is a rotatable structure, and its spray angle is controlled by a servo motor. An anti-clogging filter is provided at the nozzle outlet.

7. The decision support system for the prevention and control of swine bacterial diseases according to claim 1, characterized in that: The prediction and early warning module includes: Data preprocessing unit: Cleans, standardizes, and extracts features from the input pathogen types, environmental parameters, and historical epidemic data to provide structured data for model training; Model training unit: Based on the random forest algorithm, the prediction model is trained to learn the correlation between environmental parameters and pathogen proliferation, and output the risk level for the next 12-48 hours; Risk mapping unit: Transforms risk levels into specific prevention and control recommendations, and labels them with confidence levels; Historical database: Stores long-term environmental monitoring data, pathogen detection results, and records of control effects, supporting iterative model training and validation; The adaptive optimization module includes: Effectiveness evaluation sub-unit: Real-time collection of actual data after implementation, calculation of prevention and control effectiveness and cost indicators; Strategy optimization subunit: Based on reinforcement learning algorithm, dynamically adjust parameters such as temperature and humidity thresholds and drug dosage; Parameter update subunit: Feeds back the optimized strategy parameters to the decision control unit in real time; Real-time feedback interface: connects the execution unit and the strategy optimization subunit.

8. The present invention also includes a method for optimizing decision-making in the prevention and control of swine bacterial diseases, characterized in that, Includes the following steps: S1, Data Acquisition: Real-time acquisition of pig house environmental parameters, including but not limited to temperature, humidity, ammonia concentration and ventilation volume, and acquisition of pig health data, including but not limited to abnormal body temperature signals and abnormal behavioral characteristics. S2. Multi-source data fusion: Integrating environmental data, pig genome information, and feed composition data to construct a multi-dimensional decision-making model; S3. Pathogen analysis: Based on the pig herd health data, a preset bacterial database is matched to determine the target pathogen bacteria species and their optimal growth temperature and humidity range; S4. Decision generation: Dynamically compare the current environmental parameters with the optimal growth range to generate a composite decision instruction that includes temperature and humidity control thresholds, disinfection frequency, and drug administration priority. S5. Execution and control: According to the composite decision command, the ventilation and spraying equipment in the pig house are adjusted in a coordinated manner to control the temperature and humidity, and the targeted drug delivery device is activated to spray antibacterial agents onto the target pig herd according to priority. S6. Dynamic calibration: The number of pathogen colonies is collected every 6 hours, and the temperature and humidity thresholds and drug dosage are corrected in real time through feedback control algorithm.

9. The method for optimizing decision-making in the prevention and control of swine bacterial diseases according to claim 8, characterized in that: In step S4, determining the temperature and humidity control threshold includes: calculating the deviation between the current environmental parameters and the optimal growth range of bacteria using a fuzzy control algorithm.

10. The method for optimizing decision-making in the prevention and control of swine bacterial diseases according to claim 8, characterized in that: The allocation of drug administration priority includes: prioritizing the administration of high-concentration drugs to high-risk pigs based on the severity of individual infections in the pig herd and data on the drug resistance of pathogenic bacteria, and providing real-time feedback on efficacy data.

Images