Livestock disease prevention and control monitoring and early warning system based on Internet of Things

Abstract

The invention relates to the technical field of Internet of Things and artificial intelligence, in particular to a livestock disease prevention, control, monitoring and early warning system based on Internet of Things, which comprises a data acquisition module, a data preprocessing module, a self-adaptive data analysis module, an early warning generation module and a user interface module, wherein the data acquisition module is used for acquiring physiological data, behavior data and environment data of livestock; the data preprocessing module is used for cleaning and processing the data to eliminate noise and interference; the self-adaptive data analysis module is used for dynamically adjusting monitoring parameters and early warning threshold values and identifying potential disease risks; and the early warning generation module is used for generating early warning information of different levels and sending early warning. According to the invention, by combining the Internet of Things, machine learning and deep learning technologies, real-time monitoring of livestock health, accurate identification of disease risks and dynamic adjustment of monitoring parameters and early warning thresholds are realized, and the accuracy and timeliness of disease prevention and control are significantly improved.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things and artificial intelligence technology, and in particular to a livestock disease prevention and control monitoring and early warning system based on the Internet of Things. Background Art

[0002] With the rapid development of the global livestock industry, livestock health management has become an important part of ensuring the production efficiency and animal welfare of the livestock industry. Traditional livestock disease prevention and control usually relies on manual inspections and regular physical examinations, which is not only inefficient but also easily affected by human factors, resulting in the inability to detect and effectively control diseases in a timely manner. With the development of Internet of Things (IoT) technology, many modern farms have begun to use sensor equipment to monitor the physiological and behavioral data of livestock in real time. These data include information such as body temperature, heart rate, exercise, appetite, etc., which can provide managers with real-time health status reference. However, although existing technologies can achieve data collection and monitoring, there is still a great technical challenge for the huge amount of data and complex health assessments due to the lack of efficient analysis and processing methods.

[0003] Most of the current livestock disease prevention and control systems rely on fixed thresholds for monitoring, lack adaptive analysis capabilities, and are unable to adjust monitoring parameters and warning thresholds in real time to adapt to different environmental conditions and changes in livestock health conditions; in addition, the existing warning mechanisms usually simply set warning thresholds based on static data, fail to effectively use big data analysis and machine learning technology for deep data mining, and are unable to accurately identify potential disease risks and predict the possibility of disease occurrence. Therefore, how to improve the accuracy and timeliness of disease warnings through intelligent, data-driven technical methods has become an urgent problem to be solved. Summary of the invention

[0004] Based on the above objectives, the present invention provides a livestock disease prevention and control monitoring and early warning system based on the Internet of Things.

[0005] A livestock disease prevention and control monitoring and early warning system based on the Internet of Things, including a data acquisition module, a data preprocessing module, an adaptive data analysis module, an early warning generation module and a user interface module; wherein:

[0006] Data acquisition module: used to collect the physiological data, behavioral data and environmental data of livestock in real time through various sensors installed on the surface of livestock or its surrounding environment, and transmit the collected raw data to the data preprocessing module;

[0007] Data preprocessing module: used to receive the raw data from the data acquisition module, and use filtering, normalization and feature extraction technology to clean and process the data to eliminate noise and interference, and pass the processed data to the adaptive data analysis module;

[0008] Adaptive data analysis module: receives processed data from the data preprocessing module, analyzes the preprocessed data in real time based on machine learning and deep learning algorithms, dynamically adjusts monitoring parameters and warning thresholds, and evaluates the health status of livestock and identifies potential disease risks;

[0009] Warning generation module: used to generate warning information of different levels according to the disease risks identified by the adaptive data analysis module, and send warnings to managers through various communication methods;

[0010] User interface module: provides a visual interface to display livestock health status, system operation status and early warning information, and presents data analysis results in the form of charts and reports.

[0011] Optionally, the data acquisition module includes a physiological data acquisition unit, a behavioral data acquisition unit, an environmental data acquisition unit and a data transmission unit; wherein:

[0012] Physiological data acquisition unit: used to collect physiological data of livestock in real time, including body temperature and heart rate data, through temperature sensors and heart rate sensors installed on the surface of livestock. The temperature sensor is installed by attaching it to the surface of livestock skin or by means of wearable devices, and the heart rate sensor is installed on the chest of livestock.

[0013] Behavior data collection unit: used to collect the behavior data of livestock through activity monitoring sensors installed on the surface of livestock. The activity monitoring sensors are acceleration sensors or gyroscopes installed on the back or legs of livestock to monitor the gait, activity level and range of motion of livestock;

[0014] Environmental data acquisition unit: used to collect environmental data through temperature and humidity sensors and air quality sensors installed in the environment around the livestock. The temperature and humidity sensors are installed in the activity area where the livestock are located, and are used to collect environmental temperature and humidity in real time; the air quality sensors are installed in the air circulation area where the livestock are located, and are used to collect concentration data of harmful gases in the environment;

[0015] Data transmission unit: used to transmit the collected physiological data, behavioral data and environmental data to the data preprocessing module in real time through wireless communication technology, and the wireless communication technology includes low-power wide area network, Bluetooth and Wi-Fi.

[0016] Optionally, the data preprocessing module includes a filtering unit, a normalization unit, a feature extraction unit and a data integration unit; wherein:

[0017] The filtering unit is used to perform signal filtering processing on the raw data from the data acquisition module, using a low-pass filter or a median filter to remove high-frequency noise and random interference in the sensor data;

[0018] Normalization unit: used to normalize the filtered data according to a predetermined range, using linear normalization or Z-score normalization method to standardize the data to a uniform scale;

[0019] Feature extraction unit: used to extract relevant features from the normalized data, using the principal component analysis algorithm to identify and extract characteristic parameters that affect the health status of livestock, including body temperature change trends, heart rate variability and activity data.

[0020] Optionally, the feature extraction unit includes:

[0021] Covariance matrix calculation subunit: used to calculate the covariance matrix of normalized data, the formula is: Among them, ∑ is the covariance matrix, which represents the correlation between the features of the data set; X i is the feature vector of the i-th sample; μ is the mean vector of the data set; n is the total number of samples; T is the matrix transpose operation;

[0022] Solving subunit: used to solve the eigenvalues ​​and corresponding eigenvectors of the covariance matrix, sort them according to the size of the eigenvalues, and select the first k largest eigenvalues ​​and their corresponding eigenvectors as the principal components;

[0023] Principal component projection subunit: used to project the original data matrix onto the selected principal component matrix to generate a new feature space representation;

[0024] Feature parameter extraction subunit: Based on the results of principal component analysis, identify and extract feature parameters that affect the health status of livestock, including body temperature change trends, heart rate variability and activity data.

[0025] Optionally, the adaptive data analysis module includes a machine learning analysis unit, a deep learning analysis unit, a parameter adjustment unit and a health assessment unit; wherein:

[0026] Machine learning analysis unit: Based on the machine learning algorithm of support vector machine, it analyzes the extracted feature parameters in real time to evaluate the health status of livestock;

[0027] Deep learning analysis unit: Based on the convolutional neural network algorithm, deep feature learning and pattern recognition are performed on the extracted feature parameters to identify potential disease risks;

[0028] Parameter adjustment unit: used to dynamically adjust monitoring parameters and warning thresholds based on real-time analysis results to ensure that the system can adapt to different environmental conditions and changes in livestock numbers;

[0029] Health Assessment Unit: Used to comprehensively analyze the outputs of the machine learning analysis unit and the deep learning analysis unit to generate a health assessment report for livestock, including health indicators, risk levels, and recommended prevention and control measures.

[0030] Optionally, the machine learning analysis unit includes:

[0031] Data receiving subunit: used to receive the extracted characteristic parameters from the data preprocessing module, including body temperature change trend, heart rate variability and activity data;

[0032] Model training subunit: used to train the support vector machine model using training data with pre-labeled health status labels, and to build a classification model by analyzing historical health data and corresponding health status labels;

[0033] Decision function subunit: Use the trained support vector machine model to perform real-time classification analysis on the newly received processed data, and determine whether the current health status of the livestock is within the normal range based on the classification results of the model.

[0034] Optionally, the deep learning analysis unit includes:

[0035] Data input subunit: used for receiving processed data from the data preprocessing module, the data including various physiological, behavioral and environmental characteristics of livestock;

[0036] Convolution operation subunit: used to perform convolution calculation on input data through convolutional neural network to generate new feature maps;

[0037] Activation operation subunit: used to apply a nonlinear activation function to the convolution result after the convolution operation, truncating negative values ​​to zero and retaining only positive values;

[0038] Pooling operation subunit: used to reduce the dimension of the feature map after the convolution operation, specifically by sampling the maximum value or average value of the local area to reduce the dimension of the feature map;

[0039] Fully connected layer unit: used to input feature information after convolution, activation and pooling processing into the fully connected layer for final disease risk classification.

[0040] Optionally, the parameter adjustment unit specifically includes:

[0041] A real-time data receiving subunit, used to receive real-time analysis results from the machine learning analysis unit and the deep learning analysis unit as a basis for adjusting monitoring parameters and warning thresholds;

[0042] Monitoring parameter adjustment subunit: used to dynamically adjust the monitoring parameters in the system according to the real-time analysis results; adjust the threshold range of each monitoring parameter according to the changing trend of livestock health status. When the analysis results show that a certain characteristic parameter has abnormal fluctuations, the monitoring range of the corresponding parameter will be automatically adjusted to ensure that the system can detect early signs of the disease;

[0043] Warning threshold adjustment subunit: used to dynamically adjust the warning threshold according to the real-time analysis results. When a potential disease risk is detected, the warning threshold will be automatically adjusted according to the risk assessment results to improve the accuracy and timeliness of the warning. The specific adjustment rules include:

[0044] Rule 1: When an increase in disease risk is detected, the warning threshold is automatically lowered;

[0045] Rule 2: When an improvement in health status is detected, the warning threshold is increased.

[0046] Optionally, the warning generation module includes a risk assessment unit, a warning information generation unit and a communication interface unit; wherein:

[0047] Risk assessment unit: used to classify disease risks into three levels: low risk, medium risk and high risk;

[0048] Warning information generation unit: used to generate corresponding warning information text according to different levels of risk generated by the risk assessment unit, including risk level description, recommended prevention and control measures and emergency response instructions;

[0049] Communication interface unit: used to send warning information to management personnel through various communication methods, including mobile application interface, SMS sending and email sending.

[0050] Optionally, the user interface module includes

[0051] Data display unit: used to display the health data of livestock in a graphical way and indicate whether each indicator exceeds the normal range;

[0052] Chart generation unit: used to display data analysis results in the form of charts, and automatically generate charts in the form of line charts, bar charts or pie charts according to the analysis results output by the adaptive data analysis module;

[0053] Report generation unit: used to organize livestock health status, monitoring parameters, risk assessment results and early warning information into a report format and provide it to management personnel for detailed review.

[0054] Beneficial effects of the present invention:

[0055] The present invention can accurately monitor the health status of livestock by integrating multiple sensors for real-time data collection and using data preprocessing, machine learning and deep learning algorithms for analysis; the system can also dynamically adjust monitoring parameters and warning thresholds, and optimize the detection range in real time according to the physiological changes of livestock and environmental conditions, avoiding the insensitivity and false alarm problems caused by traditional fixed threshold settings; through this adaptive data analysis and real-time adjustment mechanism, the system can identify potential disease risks at the earliest stage and provide managers with more timely and effective warning information.

[0056] The present invention simplifies the data analysis process through visual display via the user interface module, so that managers can intuitively understand the health status of livestock, system operation status and early warning information; the form of charts and reports makes the data analysis results easier to understand and communicate, greatly improving management efficiency and decision-making accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 This is a schematic diagram of a livestock disease prevention and control monitoring and early warning system according to an embodiment of the present invention;

[0059] Figure 2 Schematic diagram of an adaptive data analysis module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0061] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0062] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0063] like Figure 1-Figure 2 As shown, a livestock disease prevention and control monitoring and early warning system based on the Internet of Things includes a data acquisition module, a data preprocessing module, an adaptive data analysis module, an early warning generation module and a user interface module; wherein:

[0064] Data acquisition module: used to collect the physiological data, behavioral data and environmental data of livestock in real time through various sensors installed on the surface of livestock or its surrounding environment, and transmit the collected raw data to the data preprocessing module;

[0065] Data preprocessing module: used to receive the raw data from the data acquisition module, and use filtering, normalization and feature extraction technology to clean and process the data to eliminate noise and interference, and pass the processed data to the adaptive data analysis module;

[0066] Adaptive data analysis module: receives processed data from the data preprocessing module, analyzes the preprocessed data in real time based on machine learning and deep learning algorithms, dynamically adjusts monitoring parameters and warning thresholds, and evaluates the health status of livestock and identifies potential disease risks;

[0067] Warning generation module: used to generate warning information of different levels according to the disease risks identified by the adaptive data analysis module, and send warnings to managers through various communication methods;

[0068] User interface module: provides a visual interface to display livestock health status, system operation status and early warning information, and presents data analysis results in the form of charts and reports to facilitate management personnel to view and operate data.

[0069] The data acquisition module includes a physiological data acquisition unit, a behavioral data acquisition unit, an environmental data acquisition unit and a data transmission unit; wherein:

[0070] Physiological data acquisition unit: used to collect the physiological data of livestock in real time, including body temperature and heart rate data, through temperature sensors and heart rate sensors installed on the surface of livestock. The temperature sensor is installed by attaching it to the surface of livestock skin or through wearable devices, and the heart rate sensor is installed on the chest of livestock. The collected body temperature and heart rate data are transmitted to the data preprocessing module by wireless means.

[0071] Behavior data acquisition unit: used to collect the behavior data of livestock through activity monitoring sensors installed on the surface of livestock. The activity monitoring sensors are acceleration sensors or gyroscopes installed on the back or legs of livestock to monitor the gait, activity level and range of motion of livestock, and transmit the data to the data preprocessing module through wireless communication;

[0072] Environmental data acquisition unit: used to collect environmental data through temperature and humidity sensors and air quality sensors installed in the environment around the livestock. The temperature and humidity sensors are installed in the activity area where the livestock are located to collect environmental temperature and humidity in real time; the air quality sensors are installed in the air circulation area where the livestock are located to collect concentration data of harmful gases (such as ammonia, carbon dioxide, etc.) in the environment. All collected data are transmitted to the data preprocessing module by wireless means;

[0073] Data transmission unit: used to transmit the collected physiological data, behavioral data and environmental data to the data preprocessing module in real time through wireless communication technology to ensure the real-time and reliability of data transmission. Wireless communication technologies include low-power wide area network, Bluetooth and Wi-Fi. Through the setting of the above data acquisition module, the physiological, behavioral and environmental data of livestock can be accurately collected, and the real-time and accuracy of data transmission can be guaranteed. The combination of different types of sensors and installation methods can efficiently obtain key data without interfering with the normal life of livestock, thereby providing reliable data support for subsequent disease warning and health monitoring.

[0074] The data preprocessing module includes a filtering unit, a normalization unit, a feature extraction unit and a data integration unit; wherein:

[0075] The filtering unit is used to perform signal filtering on the raw data from the data acquisition module, using a low-pass filter or a median filter to remove high-frequency noise and random interference in the sensor data to ensure the smoothness and stability of the data;

[0076] Normalization unit: used to normalize the filtered data according to a predetermined range, using linear normalization or Z-score normalization method to standardize the data to a uniform scale for subsequent data analysis and model training;

[0077] Feature extraction unit: used to extract relevant features from the normalized data, using the principal component analysis (PCA) algorithm to identify and extract characteristic parameters that affect the health status of livestock, including temperature change trends, heart rate variability and activity data, so as to reduce data dimensions and improve analysis efficiency; through the collaborative work of the above-mentioned filtering, normalization and feature extraction units, the data preprocessing module can effectively clean and process the collected raw data, eliminate noise and interference, and improve the quality and consistency of the data.

[0078] The feature extraction unit includes:

[0079] Covariance matrix calculation subunit: used to calculate the covariance matrix of normalized data, the formula is: Among them, ∑ is the covariance matrix, which represents the correlation between the features of the data set; X i is the feature vector of the i-th sample, indicating the value of each feature in the i-th data sample; μ is the mean vector of the data set, indicating the average value of each feature in the entire data set; n is the total number of samples, indicating the number of samples in the data set; T is the matrix transpose operation;

[0080] Solving subunit: used to solve the eigenvalue λ of the covariance matrix j and the corresponding eigenvector v j , and sort them according to the size of the eigenvalues, and select the first k largest eigenvalues ​​λ j and its corresponding eigenvector v j , as the principal component;

[0081] Principal component projection subunit: used to project the original data matrix onto the selected principal component matrix to generate a new feature space representation Y, the calculation formula of which is: Y = XV k , where Y is the new feature space representation, representing the projected data; X is the normalized data matrix, containing multiple samples and the values ​​of their features; V k is a matrix composed of the first k eigenvectors, representing the selected principal components;

[0082] Feature parameter extraction subunit: Based on the results of principal component analysis, identify and extract feature parameters that affect the health status of livestock, including body temperature change trend, heart rate variability and activity data; among them, body temperature change trend includes extracting the amplitude and change rate of body temperature fluctuation; heart rate variability includes extracting the mean, maximum, minimum and frequency of heart rate changes; activity data includes extracting exercise frequency, activity duration and resting time to reflect the behavior and health status of livestock; the above subunits can effectively reduce the dimension of data, remove redundant information, extract the most representative feature parameters, and improve the accuracy and timeliness of disease warning through the application of principal component analysis (PCA) algorithm.

[0083] The adaptive data analysis module includes a machine learning analysis unit, a deep learning analysis unit, a parameter adjustment unit, and a health assessment unit; wherein:

[0084] Machine learning analysis unit: Based on the machine learning algorithm of support vector machine, it analyzes the extracted feature parameters in real time to evaluate the health status of livestock, including identifying health indicators such as abnormal body temperature and heart rate fluctuations;

[0085] Deep learning analysis unit: Based on the convolutional neural network algorithm, deep feature learning and pattern recognition are performed on the extracted feature parameters to identify potential disease risks, such as respiratory diseases, digestive system diseases, etc.

[0086] Parameter adjustment unit: used to dynamically adjust monitoring parameters and warning thresholds based on real-time analysis results, ensuring that the system can adapt to different environmental conditions and changes in livestock numbers, and maintain the accuracy and real-time nature of monitoring and warning;

[0087] Health Assessment Unit: Used to comprehensively analyze the outputs of the machine learning analysis unit and the deep learning analysis unit to generate a health assessment report for livestock, including health indicators, risk levels, and recommended prevention and control measures.

[0088] The machine learning analysis unit includes:

[0089] Data receiving subunit: used to receive the extracted characteristic parameters from the data preprocessing module, including body temperature change trend, heart rate variability and activity data;

[0090] Model training subunit: used to train the support vector machine (SVM) model using training data with pre-labeled health status labels, and build a classification model by analyzing historical health data and corresponding health status labels to distinguish between healthy and abnormal states;

[0091] Decision function subunit: Use the trained support vector machine model to perform real-time classification analysis on the newly received processed data, and determine whether the current health status of the livestock is within the normal range based on the classification results of the model.

[0092] The specific process of support vector machine algorithm evaluation is as follows:

[0093] First, receive the processed feature parameters D = [d1, d2, ..., d m ], where m represents the number of features;

[0094] Next, we use a labeled dataset (D, t), where t = [t1, t2, ..., t n ] is the label of each sample, and the SVM model is trained to solve the following optimization problem: Among them, v is the weight vector, c is the bias term, ζ i is a slack variable used to allow the degree of classification error, C is a penalty parameter, and n1 is the number of training samples;

[0095] Then, the new data d is classified and judged by the following decision function, and the decision function expression is: Among them, f(d) is the output of the decision function, indicating the classification result; is the mapping feature of the input data d, which may transform the input data into a high-dimensional space through nonlinear mapping;

[0096] Finally, according to the output result f(d) of the support vector machine decision function, the health status of the livestock is comprehensively analyzed and the health assessment value is output; if f(d)>0, it is judged to be healthy; if f(d)≤0, it is judged to have a health risk; by using the support vector machine algorithm, the machine learning analysis unit can perform accurate real-time analysis on the processed data from the data preprocessing module. Through model training and judgment of the decision function, the system can dynamically evaluate the health status of livestock and identify potential disease risks in a timely manner, thereby improving the intelligence level of livestock health monitoring.

[0097] The Deep Learning Analysis Unit includes:

[0098] Data input subunit: used to receive processed data from the data preprocessing module, which includes various physiological, behavioral and environmental characteristics of livestock, and pass these characteristic data to the convolution operation unit for further analysis;

[0099] Convolution operation subunit: used to perform convolution calculation on the input data through the convolutional neural network, and generate a new feature map by performing dot multiplication and summation between the convolution kernel and the local area of ​​the input data; the convolution operation can effectively identify potential local patterns from the input data, helping the system to identify key features related to disease risk;

[0100] Activation operation subunit: used to apply a nonlinear activation function (such as the ReLU function) to the convolution result after the convolution operation, truncating negative values ​​to zero and retaining only positive values; this operation helps improve the nonlinear learning ability of the model, enabling it to handle more complex patterns and features, and effectively improve the sensitivity and accuracy of disease risk identification;

[0101] Pooling operation subunit: used to reduce the dimension of the feature map after the convolution operation. Specifically, it reduces the dimension of the feature map by sampling the maximum value or average value of the local area. The pooling operation helps to reduce the amount of calculation, avoid overfitting, and improve the generalization ability of the model.

[0102] Fully connected layer unit: used to input feature information after convolution, activation and pooling processing into the fully connected layer for final disease risk classification.

[0103] The specific operation steps of the convolutional neural network algorithm are as follows:

[0104] Step 1, receiving processed feature data F from the data preprocessing module, where F = [f1, f2, ..., f m ], each f represents the feature vector of a sample, and m is the number of features;

[0105] Step 2: Perform convolution calculation on the received data to extract local features. The calculation formula for each layer of convolution is: Among them, h ij It is the element in the feature map output by the current convolutional layer; The element in the input feature map of the previous layer; w pq is the weight of the convolution kernel; b is the bias term; l represents the layer number of the convolution layer;

[0106] Step 3: Apply a nonlinear activation function to the output of the convolutional unit to enhance the model's ability to learn complex patterns. The activation function is calculated as: g(h ij )=max(0,h ij ), where g(h ij ) is the output value after activation, using the ReLU function to retain positive values ​​and truncate negative values;

[0107] Step 4: Perform dimensionality reduction on the activated feature map to reduce computational complexity and retain important features. The calculation formula for the pooling operation is:ij =max k,l∈ wg(h k,l ), where p ij is the output after pooling; W is the size of the pooling window; g(h k,l ) is the activation value within the pooling window;

[0108] Step 5 is used to expand the feature map after the pooling operation into a one-dimensional vector and perform matrix multiplication with the fully connected layer to generate the classification result; its calculation formula is: k =∑ i v ik ·p i +c k , where z k is the output of the kth classification node; v ik is the weight of the fully connected layer; p i is the input after the pooling operation; c k is the bias term;

[0109] Step 6: Based on the output of the fully connected operation unit, the Softmax function is applied to generate the probability value of each category, thereby identifying the health status and potential disease risk of livestock; the calculation formula of the Softmax function is: Among them, P(y=k|z) is the probability that the sample belongs to the kth class; z k is the output of the kth classification node; K is the total number of classification categories.

[0110] The parameter adjustment unit specifically includes:

[0111] A real-time data receiving subunit, used to receive real-time analysis results from the machine learning analysis unit and the deep learning analysis unit as a basis for adjusting monitoring parameters and warning thresholds;

[0112] Monitoring parameter adjustment subunit: used to dynamically adjust the monitoring parameters in the system according to the real-time analysis results; adjust the threshold range of various monitoring parameters (such as body temperature, heart rate, activity level, etc.) according to the changing trend of livestock health status. When the analysis results show that a certain characteristic parameter has abnormal fluctuations, the monitoring range of the corresponding parameter will be automatically adjusted to ensure that the system can detect early signs of disease;

[0113] Warning threshold adjustment subunit: used to dynamically adjust the warning threshold according to the real-time analysis results. When a potential disease risk is detected, the warning threshold will be automatically adjusted according to the risk assessment results to improve the accuracy and timeliness of the warning. The specific adjustment rules include:

[0114] Rule 1: When an increase in disease risk is detected, the warning threshold is automatically lowered to increase the sensitivity of the warning;

[0115] Rule 2: When an improvement in health status is detected, the warning threshold is raised to avoid false alarms.

[0116] The warning generation module includes a risk assessment unit, a warning information generation unit and a communication interface unit; wherein:

[0117] Risk assessment unit: used to classify disease risks into three levels: low risk, medium risk and high risk;

[0118] Warning information generation unit: used to generate corresponding warning information text according to the different levels of risk generated by the risk assessment unit. The warning information generation unit formulates detailed warning content according to the warning level, including risk level description, recommended prevention and control measures and emergency response instructions;

[0119] Communication interface unit: used to send early warning information to management personnel through a variety of communication methods, including mobile application interface, SMS sending and email sending. It can select appropriate communication channels according to the early warning level to ensure that the early warning information can be transmitted to relevant management personnel in a timely and accurate manner. Through the setting of the above-mentioned early warning generation module, the system can accurately classify and generate early warning information of different levels according to the disease risk level identified by the adaptive data analysis module, ensuring the pertinence and practicality of the early warning content.

[0120] The user interface module includes

[0121] Data display unit: used to display the health data of livestock in a graphical way and indicate whether each indicator exceeds the normal range;

[0122] Chart generation unit: used to display data analysis results in the form of charts. According to the analysis results output by the adaptive data analysis module, it automatically generates charts in the form of line charts, bar charts or pie charts to intuitively display the health trends of livestock, changes in monitoring parameters and changes in warning levels. Different types of charts help managers understand the dynamic changes of data and potential disease risks more clearly.

[0123] Report generation unit: used to organize livestock health status, monitoring parameters, risk assessment results and early warning information into a report format and provide it to management personnel for detailed review; through the design of the above user interface module, the system can clearly and intuitively display the health status of livestock, the operating status of the system and early warning information.

[0124] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0125] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A livestock disease prevention and control monitoring and early warning system based on the Internet of Things, characterized in that: It includes a data acquisition module, a data preprocessing module, an adaptive data analysis module, an early warning generation module and a user interface module; among which: Data acquisition module: used to collect the physiological data, behavioral data and environmental data of livestock in real time through various sensors installed on the surface of livestock or its surrounding environment, and transmit the collected raw data to the data preprocessing module; Data preprocessing module: used to receive the raw data from the data acquisition module, and use filtering, normalization and feature extraction technology to clean and process the data to eliminate noise and interference, and pass the processed data to the adaptive data analysis module; Adaptive data analysis module: receives processed data from the data preprocessing module, analyzes the preprocessed data in real time based on machine learning and deep learning algorithms, dynamically adjusts monitoring parameters and warning thresholds, and evaluates the health status of livestock and identifies potential disease risks; Warning generation module: used to generate warning information of different levels according to the disease risks identified by the adaptive data analysis module, and send warnings to managers through various communication methods; User interface module: provides a visual interface to display livestock health status, system operation status and early warning information, and presents data analysis results in the form of charts and reports.

2. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 1 is characterized in that: The data acquisition module includes a physiological data acquisition unit, a behavioral data acquisition unit, an environmental data acquisition unit and a data transmission unit; wherein: Physiological data acquisition unit: used to collect physiological data of livestock in real time, including body temperature and heart rate data, through temperature sensors and heart rate sensors installed on the surface of livestock. The temperature sensor is installed by attaching it to the surface of livestock skin or by means of wearable devices, and the heart rate sensor is installed on the chest of livestock. Behavior data collection unit: used to collect the behavior data of livestock through activity monitoring sensors installed on the surface of livestock. The activity monitoring sensors are acceleration sensors or gyroscopes installed on the back or legs of livestock to monitor the gait, activity level and range of motion of livestock; Environmental data acquisition unit: used to collect environmental data through temperature and humidity sensors and air quality sensors installed in the environment around the livestock. The temperature and humidity sensors are installed in the activity area where the livestock are located, and are used to collect environmental temperature and humidity in real time; the air quality sensors are installed in the air circulation area where the livestock are located, and are used to collect concentration data of harmful gases in the environment; Data transmission unit: used to transmit the collected physiological data, behavioral data and environmental data to the data preprocessing module in real time through wireless communication technology, and the wireless communication technology includes low-power wide area network, Bluetooth and Wi-Fi.

3. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 1 is characterized in that: The data preprocessing module includes a filtering unit, a normalization unit, a feature extraction unit and a data integration unit; wherein: The filtering unit is used to perform signal filtering processing on the raw data from the data acquisition module, using a low-pass filter or a median filter to remove high-frequency noise and random interference in the sensor data; Normalization unit: used to normalize the filtered data according to a predetermined range, using linear normalization or Z-score normalization method to standardize the data to a uniform scale; Feature extraction unit: used to extract relevant features from the normalized data, using the principal component analysis algorithm to identify and extract characteristic parameters that affect the health status of livestock, including body temperature change trends, heart rate variability and activity data.

4. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 3 is characterized in that: The feature extraction unit comprises: Covariance matrix calculation subunit: used to calculate the covariance matrix of normalized data, the formula is: Among them, ∑ is the covariance matrix, which represents the correlation between the features of the data set; X i is the feature vector of the i-th sample; μ is the mean vector of the data set; n is the total number of samples; T is the matrix transpose operation; Solving subunit: used to solve the eigenvalues ​​and corresponding eigenvectors of the covariance matrix, sort them according to the size of the eigenvalues, and select the first k largest eigenvalues ​​and their corresponding eigenvectors as the principal components; Principal component projection subunit: used to project the original data matrix onto the selected principal component matrix to generate a new feature space representation; Feature parameter extraction subunit: Based on the results of principal component analysis, identify and extract feature parameters that affect the health status of livestock, including body temperature change trends, heart rate variability and activity data.

5. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 1 is characterized in that: The adaptive data analysis module includes a machine learning analysis unit, a deep learning analysis unit, a parameter adjustment unit and a health assessment unit; wherein: Machine learning analysis unit: Based on the machine learning algorithm of support vector machine, it analyzes the extracted feature parameters in real time to evaluate the health status of livestock; Deep learning analysis unit: Based on the convolutional neural network algorithm, deep feature learning and pattern recognition are performed on the extracted feature parameters to identify potential disease risks; Parameter adjustment unit: used to dynamically adjust monitoring parameters and warning thresholds based on real-time analysis results to ensure that the system can adapt to different environmental conditions and changes in livestock numbers; Health Assessment Unit: Used to comprehensively analyze the outputs of the machine learning analysis unit and the deep learning analysis unit to generate a health assessment report for livestock, including health indicators, risk levels, and recommended prevention and control measures.

6. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 5 is characterized in that: The machine learning analysis unit comprises: Data receiving subunit: used to receive the extracted characteristic parameters from the data preprocessing module, including body temperature change trend, heart rate variability and activity data; Model training subunit: used to train the support vector machine model using training data with pre-labeled health status labels, and to build a classification model by analyzing historical health data and corresponding health status labels; Decision function subunit: Use the trained support vector machine model to perform real-time classification analysis on the newly received processed data, and determine whether the current health status of the livestock is within the normal range based on the classification results of the model.

7. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 6 is characterized in that: The deep learning analysis unit comprises: Data input subunit: used for receiving processed data from the data preprocessing module, the data including various physiological, behavioral and environmental characteristics of livestock; Convolution operation subunit: used to perform convolution calculation on input data through convolutional neural network to generate new feature maps; Activation operation subunit: used to apply a nonlinear activation function to the convolution result after the convolution operation, truncating negative values ​​to zero and retaining only positive values; Pooling operation subunit: used to reduce the dimension of the feature map after the convolution operation, specifically by sampling the maximum value or average value of the local area to reduce the dimension of the feature map; Fully connected layer unit: used to input feature information after convolution, activation and pooling processing into the fully connected layer for final disease risk classification.

8. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 7 is characterized in that: The parameter adjustment unit specifically includes: A real-time data receiving subunit, used to receive real-time analysis results from the machine learning analysis unit and the deep learning analysis unit as a basis for adjusting monitoring parameters and warning thresholds; Monitoring parameter adjustment subunit: used to dynamically adjust the monitoring parameters in the system according to the real-time analysis results; adjust the threshold range of each monitoring parameter according to the changing trend of livestock health status. When the analysis results show that a certain characteristic parameter has abnormal fluctuations, the monitoring range of the corresponding parameter will be automatically adjusted to ensure that the system can detect early signs of the disease; Warning threshold adjustment subunit: used to dynamically adjust the warning threshold according to the real-time analysis results. When a potential disease risk is detected, the warning threshold will be automatically adjusted according to the risk assessment results to improve the accuracy and timeliness of the warning. The specific adjustment rules include: Rule 1: When an increase in disease risk is detected, the warning threshold is automatically lowered; Rule 2: When an improvement in health status is detected, the warning threshold is increased.

9. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 1, characterized in that: The warning generation module includes a risk assessment unit, a warning information generation unit and a communication interface unit; wherein: Risk assessment unit: used to classify disease risks into three levels: low risk, medium risk and high risk; Warning information generation unit: used to generate corresponding warning information text according to different levels of risk generated by the risk assessment unit, including risk level description, recommended prevention and control measures and emergency response instructions; Communication interface unit: used to send warning information to management personnel through various communication methods, including mobile application interface, SMS sending and email sending.

10. The livestock disease prevention and control monitoring and early warning system based on the Internet of Things according to claim 1, characterized in that: The user interface module includes Data display unit: used to display the health data of livestock in a graphical way and indicate whether each indicator exceeds the normal range; Chart generation unit: used to display data analysis results in the form of charts, and automatically generate charts in the form of line charts, bar charts or pie charts according to the analysis results output by the adaptive data analysis module; Report generation unit: used to organize livestock health status, monitoring parameters, risk assessment results and early warning information into a report format and provide it to management personnel for detailed review.

Images