Pig epidemic prevention and control flow regulation system and method based on big data

Abstract

The invention discloses a live pig epidemic prevention and control flow regulation system and method based on big data, and relates to the technical field of animal health information, and the method comprises the steps: recognizing abnormal live pig signals and track intersection features of structured epidemic prevention feature data through an intelligent algorithm, comparing the virus gene difference degree, constructing a virus propagation network diagram, and carrying out the analysis of the virus propagation network diagram; outputting a high-risk propagation node map; constructing an aerosol dynamics three-dimensional space model by taking a super propagation hub node of the high-risk propagation node map as an original point and combining culture environment parameters, simulating a virus air diffusion path, and generating an environment propagation thermal distribution map; fusing the high-risk propagation node map and the environment propagation thermal distribution map, identifying a cross infection area through a multi-dimensional risk superposition model to perform risk dynamic grading, and outputting a grading early warning report; according to the invention, breakthrough improvement of the live pig epidemic prevention and control capability is realized, key nodes and potential risk areas of epidemic propagation can be accurately identified, and the pertinence of prevention and control measures is improved.

Description

Technical Field

[0001] The present invention relates to the field of animal health information technology, and in particular to a pig epidemic prevention and control epidemiological investigation system and method based on big data. Background Art

[0002] In the field of animal epidemic prevention and control, intelligent monitoring systems based on the Internet of Things (IoT) have gradually become the industry standard. The current mainstream technology system deploys a distributed sensor network to achieve real-time monitoring of the entire pig production process. It integrates multiple types of data collection terminals to build a comprehensive monitoring network. Optimized machine learning classification algorithms are used to process collected physiological characteristic data in real time. Using geographic information technology, the analysis results are correlated with spatial location information to generate risk heat maps with graded early warning capabilities. This type of technical solution has formed a complete technical chain from data perception, transmission and processing to decision-making output, providing a digital and intelligent means of epidemic monitoring for the modern livestock industry.

[0003] The current animal epidemic prevention and control system still has limitations in analyzing virus transmission pathways. Existing technical systems primarily rely on static environmental parameter collection and simplified transmission models, making it impossible to accurately track the dynamic diffusion of viral aerosols in complex environments. This technical shortcoming leads to significant deviations between risk assessment results and actual transmission conditions, resulting in a lack of targeted basis for the formulation of prevention and control measures. The inability to accurately quantify the spread and risk intensity of viral particles under different environmental conditions makes it difficult to precisely define isolation areas and set disinfection frequencies. Static analysis models struggle to reflect the spatiotemporal variations in virus transmission, resulting in a mismatch between the allocation of prevention and control resources and the actual distribution of risks. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a pig epidemic prevention and control epidemiological investigation method based on big data to solve the problem of insufficient quantitative analysis of the virus environmental transmission path, resulting in the lack of targeted prevention and control measures.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for epidemiological investigation of pig epidemic prevention based on big data, which includes collecting pig epidemic prevention characteristic data and preprocessing and outputting structured epidemic prevention characteristic data; the pig epidemic prevention characteristic data includes pig vital signs data, breeding environment parameters and logistics trajectory data; Intelligent algorithms are used to identify abnormal pig signals and trajectory intersection features in structured epidemic prevention feature data, compare viral genetic differences, construct a virus transmission network diagram, and output a map of high-risk transmission nodes; Taking the super-spreading hub node of the high-risk transmission node map as the origin, a three-dimensional aerosol dynamics model was constructed in combination with the breeding environment parameters to simulate the air diffusion path of the virus and generate an environmental transmission thermal distribution map; By integrating the high-risk transmission node map and the environmental transmission heat distribution map, a multi-dimensional risk superposition model is used to identify cross-infection areas, dynamically grade risks, and output graded warning reports. Based on the graded early warning report, the strategy knowledge base is called to generate prevention and control strategies, and the prevention and control strategies are optimized through machine learning to output the optimal prevention and control strategies.

[0007] As an optimal solution of the pig epidemic prevention and control epidemiological investigation method based on big data described in the present invention, the pig epidemic prevention feature data is subjected to outlier filtering, missing value filling, trace drift correction, and fusion of multi-source time labels and spatial location labels to output structured epidemic prevention feature data.

[0008] As a preferred solution of the pig epidemic prevention and control epidemiological investigation method based on big data of the present invention, wherein: the abnormal pig signal and trajectory intersection characteristics of the structured epidemic prevention feature data are identified by an intelligent algorithm, the virus gene difference is compared, and a virus transmission network diagram is constructed, and a high-risk transmission node map is output. The specific steps are as follows: Use historical structured epidemic prevention feature data to train a deep convolutional neural network to obtain a trained deep convolutional neural network; The trained deep convolutional neural network scans the pig vital signs data in the structured epidemic prevention feature data to identify abnormal pig body temperature and sudden drop in exercise volume, and obtain abnormal pig signals; Based on the logistics trajectory data of structured epidemic prevention feature data, the DBSCAN algorithm is used to calculate the overlap of transport vehicles at transport nodes and output trajectory intersection features; Extract viral gene sequences from pig vital signs data from structured epidemic prevention feature data, perform multiple sequence alignment with the historical gene feature library, calculate the viral evolutionary distance, and output the viral gene difference; By integrating abnormal pig signals, trajectory intersection characteristics and virus gene differences, we use graph theory algorithms to construct a virus transmission network diagram, identify super-transmission hub nodes, and output a high-risk transmission node map.

[0009] As a preferred solution of the pig epidemic prevention and control epidemiological investigation method based on big data described in the present invention, wherein: taking the super transmission hub node of the high-risk transmission node map as the origin, combining the breeding environment parameters to construct an aerosol dynamics three-dimensional space model, simulate the virus air diffusion path, and generate an environmental transmission thermal distribution map, the specific steps are as follows: Analyze the longitude and latitude information of super-spreading hub nodes in the high-risk transmission node map to generate the location coordinates of super-spreading hub nodes; The coordinates of the super-spreading hub nodes are combined with the real-time aquaculture environment parameters, and a three-dimensional spatial model of aerosol dynamics is constructed through the fluid mechanics control equations. Based on the three-dimensional space model of aerosol dynamics, the finite volume method is used to perform three-dimensional numerical analysis of the virus motion trajectory to obtain the virus aerosol diffusion trajectory; The viral aerosol diffusion trajectory is subjected to spatial kernel density analysis and risk concentration annotation to obtain an environmental transmission thermal distribution map.

[0010] As a preferred solution of the pig epidemic prevention and control epidemiological investigation method based on big data of the present invention, wherein: the high-risk transmission node map and the environmental transmission heat distribution map are integrated, the cross-infection area is identified through a multi-dimensional risk superposition model to perform risk dynamic classification, and a graded early warning report is output. The specific steps are as follows: Extract biological transmission factors from high-risk transmission node maps and analyze environmental transmission factors from environmental transmission heat distribution maps; The biological transmission factors were geographically weighted using the spatial overlay analysis method to obtain a geographically weighted risk grid. The environmental transmission factors were dynamically weighted and optimized for consistency using the improved analytic hierarchy process to generate an environmental risk dynamic weight matrix. Perform spatial weighted fusion and dynamic coupling analysis on the geographically weighted risk grid and the environmental risk dynamic weight matrix to obtain a multi-dimensional risk superposition model; Align the topological data of the high-risk transmission node map with the spatial grid of the environmental transmission heat distribution map to generate composite risk data; Using a multi-dimensional risk superposition model, spatial probability density analysis and risk threshold grading assessment are performed on the composite risk data, the risk concentration matrix is ​​output, and risk level visualization rendering is performed to obtain a graded early warning report.

[0011] As a preferred solution of the pig epidemic prevention and control epidemiological investigation method based on big data of the present invention, wherein: the strategy knowledge base is called based on the hierarchical early warning report to generate a prevention and control strategy, the specific steps are as follows: Perform feature extraction and policy association mapping on historical structured epidemic prevention feature data to obtain a policy knowledge base; Utilize graded early warning reports to retrieve historical epidemic response cases in the strategy knowledge base and output prevention and control strategies.

[0012] As a preferred solution of the pig epidemic prevention and control epidemiological investigation method based on big data of the present invention, wherein: the prevention and control strategy is optimized by machine learning and the optimal prevention and control strategy is output. The specific steps are as follows: Using historical prevention and control strategies, we trained the domain knowledge-enhanced XGBoost framework to obtain a prevention and control strategy effect prediction model; The prevention and control strategy is input into the prevention and control strategy effect prediction model for multi-dimensional effect evaluation to obtain strategy execution effect prediction data, and then optimized and iterated through a multi-objective genetic algorithm to output the optimal prevention and control strategy.

[0013] In the second aspect, the present invention provides a pig epidemic prevention and control epidemiological investigation system based on big data, comprising: The data collection module is used to collect pig epidemic prevention characteristic data and pre-process and output structured epidemic prevention characteristic data; the pig epidemic prevention characteristic data includes pig vital signs data, breeding environment parameters and logistics trajectory data; The transmission analysis module is used to identify abnormal pig signals and trajectory intersection characteristics in structured epidemic prevention feature data through intelligent algorithms, compare virus genetic differences, and construct a virus transmission network diagram to output a map of high-risk transmission nodes; The diffusion simulation module is used to construct a three-dimensional aerosol dynamics model based on the super-spreading hub node of the high-risk transmission node map and the aquaculture environment parameters, simulate the air diffusion path of the virus, and generate an environmental transmission thermal distribution map; The decision-making module is used to integrate the high-risk transmission node map and the environmental transmission heat distribution map, identify cross-infection areas through a multi-dimensional risk superposition model, dynamically grade risks, and output a graded early warning report; The optimization module is used to call the strategy knowledge base based on the hierarchical early warning report to generate prevention and control strategies, optimize the prevention and control strategies through machine learning, and output the optimal prevention and control strategies.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the pig epidemic prevention and control epidemiological investigation method based on big data as described in the first aspect of the present invention.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the pig epidemic prevention and control epidemiological investigation method based on big data as described in the first aspect of the present invention.

[0016] The beneficial effects of the present invention are as follows: by integrating virus transmission network analysis with aerosol diffusion simulation, a breakthrough improvement in the pig epidemic prevention and control capabilities has been achieved, and the key nodes and potential risk areas of epidemic transmission can be accurately identified, thereby improving the pertinence of prevention and control measures; at the same time, by dynamically simulating the virus transmission path, the lag problem existing in the static evaluation method is effectively solved. This multi-dimensional analysis technology not only improves the accuracy of epidemic tracing and early warning, but also optimizes the allocation efficiency of prevention and control resources, making the prevention and control response more timely and effective. The formed hierarchical early warning and optimization strategy output mechanism provides decision-making support for breeding enterprises, realizes the transition from passive response to active prevention and control, and improves the prevention and control level of major animal epidemics. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 any creative work.

[0018] Figure 1 This is a flow chart of the pig epidemic prevention and control epidemiological investigation method based on big data.

[0019] Figure 2 This is a schematic diagram of the pig epidemic prevention and control epidemiological investigation system based on big data.

[0020] Figure 3 Flowchart for outputting a high-risk propagation node map.

[0021] Figure 4 Flowchart for outputting the optimal prevention and control strategy. DETAILED DESCRIPTION

[0022] 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 with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. 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.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides a pig epidemic prevention and control epidemiological investigation method based on big data, comprising the following steps: S1. Collect pig epidemic prevention characteristic data, and pre-process and output structured epidemic prevention characteristic data; the pig epidemic prevention characteristic data includes pig vital signs data, breeding environment parameters and logistics trajectory data.

[0026] S1.1. Perform outlier filtering, missing value filling, trace drift correction, and fusion of multi-source time labels and spatial location labels on the pig epidemic prevention feature data to output structured epidemic prevention feature data.

[0027] Specifically, the system performs outlier filtering on the temperature monitoring records in the pig vital signs data, marking records with temperature values ​​exceeding the exemplary upper limit of 42 degrees Celsius and the exemplary lower limit of 37 degrees Celsius as invalid data; performs missing value filling on the exercise monitoring records; performs error correction on the temperature and humidity sensor readings in the breeding environment parameters; and performs trajectory drift correction on the GPS coordinate points in the logistics trajectory data. The timestamps of pig epidemic prevention characteristic data, including pig vital signs data, breeding environment parameters and logistics trajectory data, are aligned in a fixed time window; the body temperature values, exercise values, temperature and humidity values, ammonia concentration values ​​and GPS coordinate points are spatially associated and mapped according to the aligned timestamps; and a structured epidemic prevention characteristic data table containing timestamps, spatial coordinates and values ​​of various monitoring indicators is output.

[0028] S2. Use intelligent algorithms to identify abnormal pig signals and trajectory intersection features in structured epidemic prevention feature data, compare viral gene differences, construct a virus transmission network diagram, and output a high-risk transmission node map.

[0029] S2.1. Use historical structured epidemic prevention feature data to train the deep convolutional neural network to obtain the trained deep convolutional neural network.

[0030] Specifically, we extracted pig vital signs data, breeding environment parameters, and logistics trajectory data from historical structured epidemic prevention feature data. We then performed feature engineering on the temperature monitoring records and exercise volume monitoring records in the pig vital signs data, as well as the temperature and humidity readings and ammonia concentrations in the breeding environment parameters, to generate a training sample set. We then standardized the training sample set. The normalized training sample set was input into a deep convolutional neural network (DCNN) consisting of alternating convolutional and pooling layers. The convolutional layers used an exemplary 3×3 kernel size for local feature extraction, and each convolution operation was immediately followed by a ReLU activation function combined with a nonlinear transformation. The pooling layers employed a max pooling strategy, with an exemplary 2×2 sampling window and a downsampling parameter of stride 2. A fully connected layer was placed at the end of the DCNN, with the output dimension matching the classification task requirements. Training was performed using a mini-batch gradient descent algorithm with an exemplary batch size of 32 samples. Prediction error was calculated using the cross-entropy loss function, and the DCNN weight parameters were adjusted layer by layer using the backpropagation algorithm. The learning rate was initialized to an exemplary 0.001 with an exponential decay strategy, and the training epoch was set to an exemplary 100 full epochs. After each training epoch, the DCNN performance was evaluated on the validation set. Training was terminated early if the validation loss did not decrease for an exemplary five consecutive epochs. Finally, the weight parameters of the best-performing DCNN were saved to obtain the trained DCNN. It should be noted that the temperature monitoring records, exercise volume monitoring records, temperature and humidity readings and ammonia concentrations in the vital signs data of live pigs are subjected to feature engineering processing. Specifically, the temperature monitoring records are discretized in segments, the sliding window statistics are calculated for the exercise volume monitoring records, the time series features are extracted for the temperature and humidity sensor readings, the ammonia concentration detection values ​​are normalized, and the processed features are spliced ​​to form a training sample set.

[0031] S2.2. Use the trained deep convolutional neural network to scan the pig vital signs data in the structured epidemic prevention feature data, identify abnormal pig body temperature and sudden drop in exercise volume, and obtain abnormal pig signals.

[0032] Specifically, the pig vital signs data in the structured epidemic prevention feature data are input into the trained deep convolutional neural network, and the pig body temperature monitoring records and pig exercise volume monitoring records are respectively extracted with spatiotemporal features by the convolution layer, and are processed by the ReLU activation function and passed to the pooling layer for feature compression; the fully connected layer outputs the abnormal body temperature probability value and the abnormal exercise volume probability value, and the exemplary abnormal probability threshold is set to 0.85. When the abnormal body temperature probability value of the pig exceeds the abnormal probability threshold, it is determined to be abnormal body temperature. When the abnormal exercise volume probability value exceeds the abnormal probability threshold and the decrease rate exceeds the exemplary 40% compared with the previous period, it is determined to be an abnormal sudden drop in exercise volume; the pig individuals with abnormal body temperature and abnormal sudden drop in exercise volume are marked as abnormal pig signals, and the abnormal pig signal containing the abnormal type, occurrence time and individual identification is output; It should be noted that the process of setting the abnormality probability threshold is to balance the recognition accuracy and false alarm rate by analyzing the distribution of prediction results of the trained deep convolutional neural network on the validation set; for example, the abnormality probability threshold of 0.85 maintains a high level of recognition rate for real abnormal cases, while keeping the proportion of healthy pigs misjudged as abnormal within a low range.

[0033] S2.3. Based on the logistics trajectory data of structured epidemic prevention feature data, the DBSCAN algorithm is used to calculate the overlap of transport vehicles at transport nodes and output the trajectory intersection characteristics; the transport nodes include farms and trading markets.

[0034] Specifically, the logistics trajectory data is extracted from the structured epidemic prevention feature data to obtain the GPS coordinate points and timestamps of the transport vehicles; spatial clustering analysis is performed on the GPS coordinate points, and the neighborhood radius of the DBSCAN algorithm is set to an exemplary 100 meters, and the minimum number of neighborhood points is set to an exemplary 3, to identify the stay areas of transport vehicles around the transport nodes of farms and trading markets; the overlap of the stay time windows of different transport vehicles in the same farms and trading markets is obtained, and the exemplary time overlap threshold is set to 30 minutes. When the stay time periods of two or more transport vehicles in the same farms and trading markets overlap and exceed the time overlap threshold, it is determined to be a trajectory intersection event; the trajectory intersection features including the location of the transport node, the identification of the intersecting vehicle, and the overlapping time period are output; It should be noted that the time overlap threshold was set by collecting loading and unloading operation records from typical farms and trading markets, and statistically analyzing the actual dwell time distribution of transport vehicles at transport nodes. The data sample covers farms of varying sizes (exemplarily ranging from 500 to 5,000 head) and three types of trading markets (production, sales, and transit) to ensure the universality of the time overlap threshold. Approximately 85% of routine loading and unloading operations take place within the exemplary 30-45 minute range, so the exemplary time overlap threshold was set at 30 minutes.

[0035] S2.4. Extract the viral gene sequence of the pig vital signs data from the structured epidemic prevention feature data, perform multi-sequence alignment with the historical gene feature library, calculate the viral evolutionary distance, and output the viral gene difference.

[0036] Specifically, the viral gene sequence is extracted from the pig vital signs data of the structured epidemic prevention feature data to obtain the nucleotide sequence of the specific segment of the viral genome; the extracted viral gene sequence is compared with the reference sequence in the historical gene feature library, and the insertion, deletion and substitution mutation sites are identified using a bit-by-bit comparison algorithm; the number of differential nucleotides in the viral gene sequence is statistically compared, and the ratio of the number of differential nucleotides to the total length of the viral gene sequence is obtained as the original difference rate. When the original difference rate exceeds the difference rate example of 0.05, it is determined to be a significant viral gene difference; the viral gene difference degree including the number of differential sites, mutation type and original difference rate is output.

[0037] S2.5. Integrate abnormal pig signals, trajectory intersection characteristics and virus gene differences, use graph theory algorithms to construct a virus transmission network diagram, identify super-transmission hub nodes, and output a high-risk transmission node map.

[0038] Specifically, the individual pig identifiers in the abnormal pig signal are associated and matched with the transport vehicle identifiers in the trajectory intersection features to establish a contact relationship network between individual pigs and transport nodes; the initial transmission network diagram is constructed with individual pigs as vertices and transport vehicle trajectory intersection events as edges; the viral gene difference is used as the edge weight, and the greater the viral gene difference, the higher the edge weight; the betweenness centrality algorithm in graph theory is used to calculate the centrality score of the node in the initial transmission network diagram, and the exemplary centrality threshold is set to 0.75. When the node centrality score exceeds the threshold, it is determined to be a super transmission hub node; the high-risk transmission node map containing node position, centrality score and associated transmission path is output; It should be noted that the betweenness centrality algorithm in graph theory is used to calculate the centrality scores of nodes in the initial propagation network graph: ; in, is the node in the initial propagation network graph The betweenness centrality score of is the node in the initial propagation network graph To the node in the initial propagation network graph The shortest path length, is the distance function, It must pass through the nodes in the initial propagation network graph Nodes in the initial propagation network graph To the node in the initial propagation network graph The shortest path length, is the node in the initial propagation network graph To the node in the initial propagation network graph The path, is the node in the initial propagation network graph To the node in the initial propagation network graph The edge weight of .

[0039] S3. Taking the super-transmission hub node of the high-risk transmission node map as the origin, a three-dimensional spatial model of aerosol dynamics is constructed in combination with the breeding environment parameters to simulate the virus air diffusion path and generate an environmental transmission thermal distribution map.

[0040] S3.1. Analyze the longitude and latitude information of the super-spreading hub nodes in the high-risk transmission node map to generate the location coordinates of the super-spreading hub nodes.

[0041] Specifically, the transportation node information marked as super-spreading hub nodes is extracted from the high-risk transmission node map, and the GPS positioning coordinates of the transportation node information of the super-spreading hub nodes recorded in the logistics trajectory data are obtained; the GPS coordinates are verified for data validity, and the signal drift points are eliminated. Example drift judgment criteria: speed > 120km / h or horizontal accuracy > 10 meters; the valid GPS coordinates are converted into longitude and latitude values ​​under the WGS84 coordinate system, retaining 6 decimal places of accuracy; and the super-spreading hub node location coordinates containing the node identifier, longitude, and latitude fields are output.

[0042] S3.2. Combine the location coordinates of the super-spreading hub node with the real-time breeding environment parameters, and construct a three-dimensional spatial model of aerosol dynamics through a group of fluid mechanics control equations.

[0043] Specifically, the longitude and latitude information of the super-spreading hub node position coordinates are extracted from the super-spreading hub node position coordinates as the spatial origin of the aerosol diffusion simulation; the real-time breeding environment parameters are synchronously obtained, including temperature and humidity readings, wind speed and direction monitoring data, and ammonia concentration detection values; the super-spreading hub node position coordinates and environmental parameters are aligned according to timestamps to establish a three-dimensional data cube containing spatial coordinates, environmental parameters, and time stamps; the fluid mechanics control equations are constructed based on the Navier-Stokes control equations, and the exemplary initial conditions are set as a wind speed of 2m / s, a relative humidity of 60%, and a temperature of 25°C, and the finite volume method is used to discretize the solution domain; the super-spreading hub node is defined as a point pollution source in the computational grid, and the super-spreading hub node position coordinates and real-time breeding environment parameters are brought into the fluid mechanics control equations and iteratively solved to obtain the aerosol concentration distribution in the three-dimensional data cube, and the aerosol dynamics three-dimensional spatial model is output; S3.3. Based on the three-dimensional spatial model of aerosol dynamics, the virus motion trajectory is numerically analyzed in three dimensions using the finite volume method to obtain the virus aerosol diffusion trajectory.

[0044] Specifically, the discretized numerical solution of the fluid mechanics governing equations is extracted from the three-dimensional spatial model of aerosol dynamics to obtain aerosol concentration distribution data on a three-dimensional data cube; the Lagrangian particle tracking algorithm is used to release an exemplary 10,000 virtual virus particles in the computational domain, each carrying a viral load proportional to the local aerosol concentration, and the particle motion trajectory is calculated by integrating the fourth-order Runge-Kutta method, with the time step set to an exemplary 0.1 second; tracking is terminated when the particle motion exceeds the computational boundary or the simulation time reaches an exemplary 4 hours; the motion paths of all particles in the three-dimensional data cube are counted, and the viral aerosol diffusion trajectory containing the trajectory starting point coordinates, the motion path point sequence, and the end point position is output; It should be noted that the expression for calculating the particle motion trajectory by integrating the fourth-order Runge-Kutta method is: ; in, is the phase 1 slope, is the velocity field function, Particles in the The spatial coordinate vector of the time step, It is The moment value of a time step; ; in, is the phase 2 slope, is the time step; ; in, is the stage 3 slope; ; in, is the stage 4 slope; ; in, is the number of viral aerosol particles at the completion of the current time step Calculated moving particle trajectories.

[0045] S3.4. Perform spatial kernel density analysis and risk concentration annotation on the viral aerosol diffusion trajectory to obtain an environmental transmission thermal distribution map.

[0046] Specifically, the motion path point sequence of all particles was extracted from the viral aerosol diffusion trajectory, and the three-dimensional spatial coordinates and timestamp of each path point were obtained; the three-dimensional data cube was divided into an exemplary 0.5m×0.5m×0.2m aerosol concentration grid unit, the number of particles passing through each aerosol concentration grid unit was counted, and the product of the particle residence time and the local aerosol concentration was obtained as the risk density value; the Gaussian kernel density estimation algorithm was used to spatially smooth the risk density value, and the kernel function bandwidth was set to an exemplary 1.0m; according to the distribution range of the risk density value, the viral aerosol concentration was divided into 5 exemplary levels (low, high, and medium). Medium-low, medium, medium-high, and high), respectively, are marked with different colors. For example, the color markings for the virus aerosol concentration levels are: low (green RGB (0,128,0)), medium-low (yellow RGB (255,255,0)), medium (orange RGB (255,165,0)), medium-high (red RGB (255,0,0)), and high (dark red RGB (139,0,0)), corresponding to 0-20%, 20-40%, 40-60%, 60-80%, and 80-100% percentile concentration ranges, respectively; the output is an environmental transmission thermal distribution map containing grid coordinates, risk levels, and aerosol concentration values.

[0047] S4. Integrate the high-risk transmission node map and the environmental transmission heat distribution map, identify cross-infection areas through a multi-dimensional risk superposition model, perform dynamic risk classification, and output a graded warning report.

[0048] S4.1. Extract biological transmission factors from the high-risk transmission node map and analyze environmental transmission factors from the environmental transmission heat distribution map.

[0049] Specifically, the betweenness centrality score, the number of associated transmission paths and the weight data of super-transmission hub nodes are extracted from the high-risk transmission node map as biological transmission factors; the risk level and aerosol concentration value of each aerosol concentration grid unit are read from the environmental transmission thermal distribution map, and the mean aerosol concentration, spatial distribution range and diffusion direction of high-risk areas (marked in red and dark red) are extracted as environmental transmission factors; the biological transmission factors and environmental transmission factors are aligned according to spatial coordinates.

[0050] S4.2. Use the spatial overlay analysis method to perform geographically weighted calculations on biological transmission factors to obtain a geographically weighted risk grid. Use the improved hierarchical analysis method to perform dynamic weight calculations and consistency optimization on environmental transmission factors to generate an environmental risk dynamic weight matrix.

[0051] Specifically, the betweenness centrality score, the number of associated transmission paths and the weight data of the super-transmission hub nodes are extracted from the biological transmission factors and converted into a spatial grid format, and the grid resolution is set to an exemplary 0.5m×0.5m; the biological transmission factors are spatially interpolated using the inverse distance weighted method, and the weight attenuation coefficient is set to an exemplary 2.0 to generate a geographically weighted risk grid; the aerosol concentration mean, spatial distribution range and diffusion direction of the high-risk areas are extracted from the environmental transmission factors to construct a judgment matrix, and the judgment matrix element values ​​are exemplary scales of 1-9; the judgment matrix is ​​corrected for consistency by the improved hierarchical analysis method, and the eigenvectors of the judgment matrix are extracted in combination with the eigenroot optimization as the initial relative weights of each environmental transmission factor; the initial relative weights of each environmental transmission factor are tested for consistency, and the environmental transmission factor weight results are output when the consistency ratio is less than an exemplary 0.1, otherwise the judgment matrix is ​​adjusted until the conditions are met; the optimized environmental transmission factor weights are multiplied by the corresponding environmental transmission factors to generate an environmental risk dynamic weight matrix; It should be noted that the improved AHP improves upon the traditional AHP by adding a dynamic weight adjustment mechanism and real-time consistency optimization. Specific improvements include: replacing static expert scoring with automatic judgment matrix generation based on real-time monitoring data, such as updating matrix elements every hour based on environmental sensor readings; Use machine learning algorithms to automatically correct the consistency ratio. For example, when the CR (consistency ratio) is greater than 0.1, the judgment matrix is ​​adjusted by gradient descent. We added a timeliness constraint to the weights and set an exemplary weight decay coefficient of 0.95 / hour to ensure that the contribution of old data decreases over time. These improvements enable the weight allocation to dynamically respond to changes in aquaculture environment parameters, addressing the lack of adaptability of traditional AHP in epidemic transmission scenarios.

[0052] S4.3. Perform spatial weighted fusion and dynamic coupling analysis on the geographically weighted risk grid and the environmental risk dynamic weight matrix to obtain a multi-dimensional risk superposition model.

[0053] Specifically, the geographically weighted risk grid and the environmental risk dynamic weight matrix are aligned according to spatial coordinates to ensure that each geographically weighted risk grid cell corresponds to a one-to-one correspondence with an element of the environmental risk dynamic weight matrix; the biological transmission factor value at each spatial location is multiplied by the environmental transmission factor weight to obtain a comprehensive risk value; The dynamic coupling coefficient was used to adjust the contribution ratio of the biological transmission factor value to the environmental transmission factor, and the exemplary initial weights were set as 0.6 for the biological transmission factor and 0.4 for the environmental transmission factor. Local data noise was eliminated through the sliding window analysis method (exemplary window size 3×3 grid). The fused comprehensive risk value was normalized and converted into a relative risk index in the range of 0-1. The multidimensional risk superposition model containing the coordinates of each geographically weighted risk grid cell, the comprehensive risk value and the risk level was output.

[0054] S4.4. Align the topological data of the high-risk transmission node map with the spatial grid of the environmental transmission heat distribution map to generate composite risk data.

[0055] Specifically, the super-spreading hub node topological data, including super-spreading hub node identifiers, connection edge weights and spatial coordinates, are extracted from the high-risk transmission node map; the center coordinates and risk values ​​of the geographically weighted risk grid units are read from the environmental transmission heat distribution map; the super-spreading hub node topological data and the center coordinates and risk values ​​of the geographically weighted risk grid units are uniformly converted to the WGS84 geographic coordinate system, and the plane projection adopts UTM partitioning; the node coordinates are matched with the center coordinates of the geographically weighted risk grid by the nearest neighbor, and the exemplary matching tolerance is set to 0.5 meters; when the super-spreading hub node falls into a certain geographically weighted risk grid unit, the super-spreading hub node attributes are associated with the geographically weighted risk grid risk value; and the composite risk data including the super-spreading hub node identifier, topological features, matching geographically weighted risk grid coordinates and corresponding risk values ​​are output.

[0056] S4.5. Use a multi-dimensional risk superposition model to conduct spatial probability density analysis and risk threshold grading assessment on composite risk data, output a risk concentration matrix, and perform risk level visualization to obtain a graded warning report.

[0057] Specifically, the super-spreading hub node attributes and geographically weighted risk grid risk values ​​are extracted from the composite risk data, and input into the multidimensional risk superposition model for spatial kernel density analysis, with the kernel function bandwidth set to an exemplary 1.0 meter. The multidimensional risk superposition model integrates temporal evolution characteristics, spatial adjacency, and transmission factor weight information to achieve joint modeling and response intensity assessment of risk factors from different sources. The kernel density function uses a Gaussian kernel function to smoothly weight risk events in the local space, highlighting high-frequency clustering areas. The risk probability density value of each spatial location is obtained to generate a risk concentration matrix. The probability density values ​​in the risk concentration matrix are percentile graded, with the 20%, 40%, 60%, and 80% percentiles set as examples, to divide the risk into five levels: low (0-20%), medium-low (20-40%), medium (40-60%), medium-high (60-80%), and high (80-100%). The risk probability density of different regions is standardized and sorted to ensure that the distribution of each risk level is reasonable and spatially distinguishable. The classification results are mapped to a spatial coordinate system and rendered with color bands in combination with regional boundaries and environmental layers to form a risk level visualization layer. The color scale scheme can adopt a standard five-segment continuous color spectrum, from cold to warm tones, namely blue (low risk), light green (medium-low risk), yellow (medium risk), orange (medium-high risk) and red (high risk), to achieve intuitive risk identification; output graded warning reports S5. Based on the hierarchical early warning report, the strategy knowledge base is called to generate a prevention and control strategy.

[0058] S5.1. Perform feature extraction and strategy association mapping on historical structured epidemic prevention feature data to obtain a strategy knowledge base.

[0059] Specifically, pig vital signs data (body temperature, exercise volume), breeding environment parameters and logistics trajectory data are extracted from historical structured epidemic prevention feature data; anomaly detection is performed on the pig vital signs data, and body temperature exceeding the exemplary 39.5°C or exercise volume below the exemplary 200 units / hour is marked as abnormal; breeding environment parameters and logistics trajectory data are aligned by timestamp to generate a spatiotemporal correlation data set; the FP-Growth algorithm is used to mine high-frequency prevention and control strategy combinations, and the minimum support is set to the exemplary 0.3; the effectiveness score of the prevention and control strategy is obtained, and when the accuracy rate exceeds the exemplary 90%, it is stored in the strategy knowledge base; and the strategy knowledge base containing strategy rules, implementation conditions and effect scores is output.

[0060] S5.2. Use the graded early warning report to search the historical epidemic response cases in the strategy knowledge base and output the prevention and control strategy.

[0061] Specifically, risk level, spatial location and time range information are extracted from the graded warning report, and historical epidemic response cases that match the conditions are retrieved in the strategy knowledge base. When the risk level, spatial distribution pattern and virus genotype similarity all exceed the exemplary 0.8, it is determined to be a valid matching case; the prevention and control strategy records in the matching cases are extracted and arranged in descending order according to the implementation effect score; and the prevention and control strategy containing the strategy name, applicable conditions, implementation steps and expected effects is output.

[0062] S6. Optimize prevention and control strategies through machine learning and output the optimal prevention and control strategy.

[0063] S6.1. Using historical prevention and control strategies, the domain knowledge-enhanced XGBoost framework is trained to obtain a prevention and control strategy effectiveness prediction model.

[0064] Specifically, the strategy feature vectors are extracted from historical prevention and control strategies, including implementation conditions (risk level, environmental parameters), measure content (disinfection intensity, isolation scope) and effect indicators (control period, loss rate); the feature vectors are converted into numerical matrices, and missing values ​​are filled with the mean of the same strategy category; the hyperparameters of the domain knowledge enhanced XGBoost framework are configured, the maximum depth of the tree is set to an exemplary 6 layers, and the learning rate is set to an exemplary 0.01; the training set and validation set are divided into a 7:3 ratio, and the domain knowledge enhanced XGBoost framework is trained using 5-fold cross-validation; training is stopped when the mean absolute error on the validation set is lower than an exemplary 0.15, and the domain knowledge enhanced XGBoost framework structure and parameters are saved; and the prevention and control strategy effect prediction model including feature importance ranking and prediction error indicators is output.

[0065] S6.2. Input the prevention and control strategy into the prevention and control strategy effect prediction model for multi-dimensional effect evaluation to obtain strategy execution effect prediction data, and optimize and iterate through a multi-objective genetic algorithm to output the optimal prevention and control strategy.

[0066] Specifically, risk level data and environmental parameter data are extracted from the prevention and control strategy, measure content data is obtained, and the isolation fence radius and quarantine frequency are recorded; non-numeric data are converted into standardized numerical features, missing fields are filled with historical means of similar strategies, and outliers are eliminated according to the box plot rule; the processed prevention and control strategy is input into the prevention and control strategy effect prediction model, and multi-dimensional effect evaluation calculations are performed, including forward propagation calculation of control cycle prediction value, backpropagation optimization loss function gradient and cross-validation evaluation of the stability of the prevention and control strategy effect prediction model, and the estimated number of days of the epidemic control cycle is output, with an exemplary range of 3 to 15 days; the estimated economic loss in 10,000 yuan is output, with an exemplary accuracy of plus or minus 50,000 yuan; the cost of human resources and materials consumed by resources is output, with an exemplary unit of 10,000 yuan; time indicators are logarithmically converted to the range of 0 to 1, economic indicators are standardized to the range of 0 to 1 using Min-Max, and resource indicators are converted to budget percentages of 0% to 100%; and strategy execution effect prediction data is generated; A three-objective optimization function is established to minimize the control cycle and economic losses, and to constrain resource consumption not to exceed the hard budget constraint. The NSGA-II framework is used to configure the genetic algorithm parameters, with the population size set to an exemplary 100 individuals, the crossover operator using a simulated binary crossover probability of exemplary 0.9, the mutation operator using a polynomial mutation probability of 0.1, and the selection mechanism using a tournament selection scale of 3. Initialization randomly generates an initial population that meets the constraints, and calculates the three-objective function values ​​for each individual. Non-dominated sorting is performed to divide the frontier level, and the congestion distance is obtained to maintain diversity. The convergence of the hypervolume indicator is detected to terminate the iteration. The feasibility of the Pareto front solution set is verified, and the solutions with an isolation radius exceeding the exemplary 5 kilometers are eliminated. A three-dimensional target space scatter plot visualization output is generated. Finally, a structured optimal strategy report is output.

[0067] This embodiment also provides a pig epidemic prevention and control epidemiological investigation system based on big data, including: The data collection module is used to collect pig epidemic prevention characteristic data and pre-process and output structured epidemic prevention characteristic data; the pig epidemic prevention characteristic data includes pig vital signs data, breeding environment parameters and logistics trajectory data; The transmission analysis module is used to identify abnormal pig signals and trajectory intersection characteristics in structured epidemic prevention feature data through intelligent algorithms, compare virus genetic differences, and construct a virus transmission network diagram to output a map of high-risk transmission nodes; The diffusion simulation module is used to construct a three-dimensional aerosol dynamics model based on the super-spreading hub node of the high-risk transmission node map and the aquaculture environment parameters, simulate the air diffusion path of the virus, and generate an environmental transmission thermal distribution map; The decision-making module is used to integrate the high-risk transmission node map and the environmental transmission heat distribution map, identify cross-infection areas through a multi-dimensional risk superposition model, dynamically grade risks, and output a graded early warning report; The optimization module is used to call the strategy knowledge base based on the hierarchical early warning report to generate prevention and control strategies, optimize the prevention and control strategies through machine learning, and output the optimal prevention and control strategies.

[0068] This embodiment also provides a computer device, which is suitable for the pig epidemic prevention and control epidemiological investigation method based on big data, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the pig epidemic prevention and control epidemiological investigation method based on big data proposed in the above embodiment.

[0069] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0070] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the pig epidemic prevention and control flow investigation method based on big data as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0071] In summary, the present invention achieves a breakthrough improvement in the pig epidemic prevention and control capabilities by: integrating virus transmission network analysis and aerosol diffusion simulation, and can accurately identify key nodes and potential risk areas of epidemic spread, thereby improving the pertinence of prevention and control measures; at the same time, by dynamically simulating the virus transmission path, it effectively solves the lag problem existing in the static evaluation method. This multi-dimensional analysis technology not only improves the accuracy of epidemic tracing and early warning, but also optimizes the allocation efficiency of prevention and control resources, making the prevention and control response more timely and effective. The formed hierarchical early warning and optimization strategy output mechanism provides decision-making support for breeding enterprises, realizes the transition from passive response to active prevention and control, and improves the prevention and control level of major animal epidemics.

[0072] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for epidemiological investigation of pig epidemic prevention and control based on big data, characterized by: include, Collect pig epidemic prevention characteristic data, and pre-process and output structured epidemic prevention characteristic data; the pig epidemic prevention characteristic data includes pig vital signs data, breeding environment parameters and logistics trajectory data; Intelligent algorithms are used to identify abnormal pig signals and trajectory intersection features in structured epidemic prevention feature data, compare viral genetic differences, construct a virus transmission network diagram, and output a map of high-risk transmission nodes; Taking the super-spreading hub node of the high-risk transmission node map as the origin, a three-dimensional aerosol dynamics model was constructed in combination with the breeding environment parameters to simulate the air diffusion path of the virus and generate an environmental transmission thermal distribution map; By integrating the high-risk transmission node map and the environmental transmission heat distribution map, a multi-dimensional risk superposition model is used to identify cross-infection areas, dynamically grade risks, and output graded warning reports. Based on the graded early warning report, the strategy knowledge base is called to generate prevention and control strategies, and the prevention and control strategies are optimized through machine learning to output the optimal prevention and control strategies.

2. The method for epidemiological investigation of pig epidemic prevention and control based on big data according to claim 1, characterized in that: The pig epidemic prevention feature data is filtered for outliers, filled with missing values, corrected for trace drift, and fused with multi-source time labels and spatial location labels to output structured epidemic prevention feature data.

3. The method for epidemiological investigation of pig epidemic prevention and control based on big data according to claim 2, characterized in that: The method uses intelligent algorithms to identify abnormal pig signals and trajectory intersection features of structured epidemic prevention feature data, compares virus gene differences, constructs a virus transmission network diagram, and outputs a high-risk transmission node map. The specific steps are as follows: Use historical structured epidemic prevention feature data to train a deep convolutional neural network to obtain a trained deep convolutional neural network; The trained deep convolutional neural network scans the pig vital signs data in the structured epidemic prevention feature data to identify abnormal pig body temperature and sudden drop in exercise volume, and obtain abnormal pig signals; Based on the logistics trajectory data of structured epidemic prevention feature data, the DBSCAN algorithm is used to calculate the overlap of transport vehicles at transport nodes and output trajectory intersection features; Extract viral gene sequences from pig vital signs data from structured epidemic prevention feature data, perform multiple sequence alignment with the historical gene feature library, calculate the viral evolutionary distance, and output the viral gene difference; By integrating abnormal pig signals, trajectory intersection characteristics and virus gene differences, we use graph theory algorithms to construct a virus transmission network diagram, identify super-transmission hub nodes, and output a high-risk transmission node map.

4. The method for epidemiological investigation of pig epidemic prevention and control based on big data according to claim 3, characterized in that: The method uses the super-spreading hub node of the high-risk transmission node map as the origin, combines the breeding environment parameters to construct an aerosol dynamics three-dimensional spatial model, simulates the virus air diffusion path, and generates an environmental transmission heat distribution map. The specific steps are as follows: Analyze the longitude and latitude information of super-spreading hub nodes in the high-risk transmission node map to generate the location coordinates of super-spreading hub nodes; The coordinates of the super-spreading hub nodes are combined with the real-time aquaculture environment parameters, and a three-dimensional spatial model of aerosol dynamics is constructed through the fluid mechanics control equations. Based on the three-dimensional space model of aerosol dynamics, the finite volume method is used to perform three-dimensional numerical analysis of the virus motion trajectory to obtain the virus aerosol diffusion trajectory; The viral aerosol diffusion trajectory is subjected to spatial kernel density analysis and risk concentration annotation to obtain an environmental transmission thermal distribution map.

5. The method for epidemiological investigation of pig epidemic prevention and control based on big data according to claim 4, characterized in that: The method integrates the high-risk transmission node map and the environmental transmission heat distribution map, identifies the cross-infection area through a multi-dimensional risk superposition model, performs risk dynamic classification, and outputs a graded warning report. The specific steps are as follows: Extract biological transmission factors from high-risk transmission node maps and analyze environmental transmission factors from environmental transmission heat distribution maps; The biological transmission factors were geographically weighted using the spatial overlay analysis method to obtain a geographically weighted risk grid. The environmental transmission factors were dynamically weighted and optimized for consistency using the improved analytic hierarchy process to generate an environmental risk dynamic weight matrix. Perform spatial weighted fusion and dynamic coupling analysis on the geographically weighted risk grid and the environmental risk dynamic weight matrix to obtain a multi-dimensional risk superposition model; Align the topological data of the high-risk transmission node map with the spatial grid of the environmental transmission heat distribution map to generate composite risk data; Using a multi-dimensional risk superposition model, spatial probability density analysis and risk threshold grading assessment are performed on the composite risk data, the risk concentration matrix is ​​output, and risk level visualization rendering is performed to obtain a graded early warning report.

6. The method for epidemiological investigation of pig epidemic prevention and control based on big data according to claim 5, characterized in that: The specific steps of calling the strategy knowledge base to generate the prevention and control strategy based on the hierarchical early warning report are as follows: Perform feature extraction and policy association mapping on historical structured epidemic prevention feature data to obtain a policy knowledge base; Utilize graded early warning reports to retrieve historical epidemic response cases in the strategy knowledge base and output prevention and control strategies.

7. The method for epidemiological investigation of pig epidemic prevention and control based on big data according to claim 6, characterized in that: The specific steps for optimizing the prevention and control strategy through machine learning and outputting the optimal prevention and control strategy are as follows: Using historical prevention and control strategies, we trained the domain knowledge-enhanced XGBoost framework to obtain a prevention and control strategy effect prediction model; The prevention and control strategy is input into the prevention and control strategy effect prediction model for multi-dimensional effect evaluation to obtain strategy execution effect prediction data, and then optimized and iterated through a multi-objective genetic algorithm to output the optimal prevention and control strategy.

8. A pig epidemic prevention and control epidemiological investigation system based on big data, based on the pig epidemic prevention and control epidemiological investigation method based on big data according to any one of claims 1 to 7, characterized in that: include, The data collection module is used to collect pig epidemic prevention characteristic data and pre-process and output structured epidemic prevention characteristic data; the pig epidemic prevention characteristic data includes pig vital signs data, breeding environment parameters and logistics trajectory data; The transmission analysis module is used to identify abnormal pig signals and trajectory intersection characteristics in structured epidemic prevention feature data through intelligent algorithms, compare virus genetic differences, and construct a virus transmission network diagram to output a map of high-risk transmission nodes; The diffusion simulation module is used to construct a three-dimensional aerosol dynamics model based on the super-spreading hub node of the high-risk transmission node map and the aquaculture environment parameters, simulate the air diffusion path of the virus, and generate an environmental transmission thermal distribution map; The decision-making module is used to integrate the high-risk transmission node map and the environmental transmission heat distribution map, identify cross-infection areas through a multi-dimensional risk superposition model, dynamically grade risks, and output a graded early warning report; The optimization module is used to call the strategy knowledge base based on the hierarchical early warning report to generate prevention and control strategies, optimize the prevention and control strategies through machine learning, and output the optimal prevention and control strategies.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the pig epidemic prevention and control epidemiological investigation method based on big data are implemented as described in any one of claims 1 to 7.

10. 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 pig epidemic prevention and control epidemiological investigation method based on big data as described in any one of claims 1 to 7 are implemented.

Images