A cow estrus and mating monitoring system based on blockchain edge computing

Abstract

The application discloses a kind of based on blockchain edge computing's cow estrus mating monitoring system, it is related to the technical field of livestock breeding, the system includes: intelligent necklace terminal, for wearing in the neck of cow, the physical data of cow is monitored in real time by built-in sensor;Data management platform is used to receive the physical data obtained by monitoring, the estrus event information of each cow is centrally managed and stored by the database constructed;Event management platform is used to obtain the estrus event information of data management platform, and forwards the application operation from the display interaction application;Display interaction application is used to connect event management platform, inquires and shows the estrus state of cow;SMS notification platform is used to send SMS notification to registered user by wireless communication mode.The application monitors the physical data of cow in real time, utilizes built-in intelligent LED lamp to visually display estrus state, timely identification and take mating measures, improve the management efficiency of cow estrus monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of livestock breeding, in particular to a cow estrus and breeding monitoring system based on blockchain edge computing. BACKGROUND

[0002] Whether it is dairy cattle breeding, beef cattle breeding, cow breeding is directly related to the operation and income of the ranch. In order to ensure the quality of breeding, most of the domestic ranches currently use artificial breeding. Taking dairy cows as an example, the estrus cycle of dairy cows is 21 days, and the estrus duration is about 20 hours. Similar to dairy cows, there may be differences in some cattle breeds.

[0003] The premise of artificial breeding is to observe the estrus of the cow. In a large-scale ranch, there is an urgent need for a large-scale and accurate means to monitor the estrus of the cow. At the same time, it is also necessary to provide a quick means to find the cow that has estrus.

[0004] Although there are some pedometer collars on the market at present, the pedometer collar mainly reports the activity data of the cow, and the ranch manager needs to check the pace data of each cow many times a day, and judge whether the cow is in estrus in combination with the basic data of the cow. The ordinary pedometer collar faces two problems, first: the ranch manager needs to perform thousands of times of manual data comparison; second: it is necessary to quickly find the estrus cow in a large herd. These are a great challenge to the operation of the ranch. Simply relying on the personal experience of employees to record and predict, the accuracy of the data is not high, so that the production efficiency of the cattle farm has not been greatly improved.

[0005] In view of the problems in the related art, no effective solution has been proposed so far. SUMMARY

[0006] (I) Technical problems solved

[0007] In view of the deficiencies in the prior art, the present application provides a cow estrus and breeding monitoring system based on blockchain edge computing, which has the advantages of improving the accuracy of cow estrus monitoring and management efficiency, reducing breeding costs, and enhancing data security and system scalability, thereby solving the problem of low production efficiency caused by the use of artificial experience to determine whether the cow is in estrus in traditional ranch management.

[0008] (II) Technical solutions

[0009] In order to achieve the above-mentioned advantages of improving the accuracy of cow estrus monitoring and management efficiency, reducing breeding costs, and enhancing data security and system scalability, the specific technical solutions adopted by the present application are as follows:

[0010] A cow estrus and mating monitoring system based on blockchain edge computing, the system comprises:

[0011] An intelligent collar terminal is used for wearing on the neck of a cow, which monitors the physical data of the cow in real time through the built-in sensor, and dynamically displays the estrus state of the cow through the intelligent LED light;

[0012] A data management platform is used for receiving the monitored physical data, centrally managing and storing the estrus event information of each cow through the constructed database, and establishing information interaction with the event management platform;

[0013] An event management platform is used for obtaining the estrus event information of the data management platform, forwarding the application operation from the display interaction application, and providing an access interface to realize information interaction between platforms;

[0014] A display interaction application is used for connecting the event management platform, querying and displaying the estrus state of the cow;

[0015] A short message notification platform is used for sending short message notifications to registered users through wireless communication.

[0016] Further, the intelligent collar terminal comprises an internal task management module, an estrus monitoring module, an LED light module, and a communication module;

[0017] The internal task management module is used for monitoring the running state of each module, optimizing the resource use of each module, and managing data flow and processing, and executing the preset working logic;

[0018] The estrus monitoring module is used for collecting the physical data of the cow through the built-in sensor, and identifying the estrus state of the cow in the monitoring period by means of machine learning algorithm through edge computing, the physical data including activity data, rumination data and body temperature data;

[0019] The LED light module is used for prompting the estrus state of the cow through different flashing modes;

[0020] The communication module is used for uploading the collected and processed physical data to the data management platform, and providing terminal identification and positioning functions, binding the device number of the collar and the individual number of the cow.

[0021] Further, monitoring the running state of each module, optimizing the resource use of each module, and managing data flow and processing, and executing the preset working logic comprises:

[0022] Checking the hardware and software state of each module, when the function is normal, allocating initial computing storage resources, and continuously monitoring the running index of each module, when exceeding the standard threshold, triggering the early warning mechanism;

[0023] Based on the actual operation of the module, the resource allocation is adjusted in real time, and the liquidity processing procedure of data acquisition, processing, storage and transmission is managed;

[0024] The running state and resource usage of each module are recorded, node logs are generated, and when an exception occurs, a report is automatically generated and sent to the data management platform.

[0025] Further, the estrus monitoring module includes a motion sensor, a pressure sensor, a body temperature sensor, an activity recording unit, a rumination recording unit, and a state evaluation unit.

[0026] The motion sensor measures three-axis acceleration, angular velocity, and angle to collect motion parameters generated by the cow during feeding.

[0027] The pressure sensor measures the waveform transformation of the cow during chewing and swallowing to obtain pressure parameters when the cow is ruminating.

[0028] The body temperature sensor collects real-time body temperature data of the cow during feeding.

[0029] The activity recording unit records the motion parameters of the cow and analyzes the activity data of the cow in the current monitoring period according to the preset monitoring period.

[0030] The rumination recording unit records the pressure parameters of the cow and inputs them into a pre-built behavior classification model to identify the rumination data of the cow in the current monitoring period.

[0031] The state evaluation unit evaluates the estrus state of the cow based on the activity data, body temperature data, and rumination data of the cow in the current monitoring period.

[0032] Further, the pressure parameters of the cow are recorded and input into a pre-built behavior classification model to identify the rumination data of the cow in the current monitoring period.

[0033] After obtaining the historical pressure parameters of the cow, the data is standardized and normalized to construct a behavior data set, which is divided into a training set and a test set according to an 8:2 ratio.

[0034] The behavior data set is used as an input variable to build a long short-term memory network structure, and the firefly algorithm is used to optimize the parameters of the long short-term memory network to construct a behavior classification model.

[0035] The pressure parameters of the current monitoring period are input to the behavior classification model, the behavior classification results of the cattle in the current monitoring period are counted, and the duration of the ruminant behavior of the cattle is extracted as the ruminant data; wherein, the behavior classification results include ruminant behavior, eating behavior, drinking behavior and food stopping behavior.

[0036] Further, the network structure of the long short-term memory network is built, the parameter optimization of the long short-term memory network is performed by using the moth flame algorithm, and the behavior classification model is constructed, including:

[0037] The number and value range of the model parameters to be optimized of the long short-term memory network are set; wherein, the model parameters include the maximum training axis, the learning rate, the block size, the first layer hidden layer neuron data and the second layer hidden layer neuron number;

[0038] The accuracy of the evaluation index of the long short-term memory network is set as the fitness function output;

[0039] The moth matrix is initialized, and the moth fitness vector is calculated according to the moth matrix and the fitness function;

[0040] It is judged whether the maximum iteration number is reached, if the current iteration number exceeds the maximum iteration number, the iteration is ended, if the current iteration number does not reach the maximum iteration number, the population is updated by using the moth position update formula and the optimization is continued, and the moth position update formula is:

[0041] S(M i ,F j )=D i ·e bt ·cos(2πt)+F j

[0042] D i =|F j -M i |

[0043] In the formula, S represents a logarithmic spiral function; D i represents the distance from the i-th moth to the j-th flame; M i represents the i-th moth; F j represents the j-th flame; b represents the constant of S; t represents a random number; e represents a natural constant;

[0044] After all the iteration numbers are finished, the optimal model parameters are output, and the behavior classification model is obtained.

[0045] Further, based on the activity data, the body temperature data set and the ruminant data of the cattle in the current monitoring period, the estrus state of the cattle is evaluated, including:

[0046] Based on the monitoring result of the activity data, the activity type of the cow is divided into active posture and non-active posture, and the percentage of the active posture of the cow in the total time length of the current monitoring period is calculated as the activity amount feature of the cow in the current monitoring period;

[0047] Based on the monitoring result of the rumination data, the percentage of the duration of the rumination behavior of the cow in the total time length of the monitoring period and the percentage of the duration of the rumination behavior in the total feeding time are calculated as the rumination feature of the cow in the current monitoring period;

[0048] Based on the monitoring result of the body temperature data, the body temperature change amount of the cow in the current monitoring period is recorded, and the temperature difference between the cow and the standard body temperature is calculated as the body temperature feature of the cow in the current monitoring period;

[0049] The activity amount feature, the rumination feature and the body temperature feature are used to construct a feature matrix as the model input, and the recognition label of the estrus state of the cow is pre-set to construct a state recognition model based on a long short-term memory network, and the estrus state of the cow in the current monitoring period is recognized by using the state recognition model.

[0050] Further, the data management platform comprises an edge management module, a data receiving module, a data management module, a double-chain storage module, an access control module and an interface management module.

[0051] The edge management module is used for deploying an Internet of Things gateway, managing all intelligent necklace terminals in the breeding area in a partitioned manner, and coordinating distributed processing of the physical sign data and the estrus state recognition.

[0052] The data receiving module is used for receiving and integrating real-time data uploaded by the intelligent necklace terminals.

[0053] The data management module is used for associating the individual numbers of the cows with their respective physical sign data and estrus states, forming estrus event information corresponding to the cows, and performing partitioned management according to the breeding areas.

[0054] The double-chain storage module is used for constructing data blocks according to the real-time data uploaded by the intelligent necklace terminals, and storing the data blocks into server nodes in an asynchronous concurrent manner to form a double-chain storage structure.

[0055] The access control module is used for loading a ciphertext access mechanism, giving data access permissions to data owners and authorized persons, and regularly querying and verifying the integrity of the blockchain data.

[0056] The interface management module is used for providing data interfaces between the event management platform and the display interaction application, and supporting conversion of multiple data formats and protocols.

[0057] Further, the real-time data uploaded by the intelligent collar terminal is constructed into data blocks, and the data blocks are stored into the server nodes in an asynchronous and concurrent manner to form a double-chain storage structure, including:

[0058] The real-time data uploaded by the intelligent collar terminal of each cow is generated into data blocks according to different breeding areas, and the data blocks are stored into the server nodes in a time sequence;

[0059] Each server node stores data blocks of multiple breeding areas, and when the server node stores the blocks, the storage rules are verified first, and after the verification, the block storage is performed;

[0060] The storage rules of the first two blocks are preset, and starting from the third block, with the increase of each block, according to the storage rules of the first two blocks, when the server obtains the storage permission of other breeding areas, the storage information is published to the block chain, and the block is attached to the tail of the server node, and other server nodes give up the storage right competition of the block.

[0061] Further, the event management platform includes: an event processing module, a real-time monitoring module, a strategy management module, an event cache module and an extension interface module;

[0062] The event processing module is used for classifying, sorting and priority processing of estrus event information according to preset rules, and timely responding to cows with abnormal estrus state;

[0063] The real-time monitoring module is used for real-time monitoring of the estrus time processing process, and recording all processing logs, and then storing the processing logs through block chain encryption;

[0064] The strategy management module is used for supporting the administrator to define the processing strategy of the estrus event;

[0065] The event cache module is used for temporarily storing the estrus event information which is not processed and is being processed;

[0066] The extension interface module is used for providing data interfaces with display interaction applications, data management platforms and other modules, and is responsible for conversion and adaptation of data format and communication protocol.

[0067] (Three) beneficial effects

[0068] Compared with the prior art, the present application provides a cow estrus and breeding monitoring system based on block chain edge computing, which has the following beneficial effects:

[0069] (1) By combining the intelligent collar terminal, data management platform, event management platform, display interaction application and SMS notification platform, a complete cow estrus monitoring and management solution system is formed; by monitoring the physical data of the cow in real time, after analyzing the estrus state, the estrus state is displayed intuitively by using the built-in intelligent LED lamp, so that the manager or operator can identify and take mating measures in time, thereby effectively improving the accuracy of cow estrus monitoring and management efficiency, and reducing the breeding cost.

[0070] (2) By using edge computing and blockchain technology, the data is ensured to be processed and stored quickly and safely, and cross-platform information interaction and access control are supported, the efficiency and reliability of data management are improved, the security of data and the expansibility of the system are effectively enhanced; at the same time, the user is provided with a convenient query and monitoring interface, the visualization and tracking of the estrus state become more intuitive and easy to use, and the wireless communication technology is used to send SMS notification, so that the manager can receive the notification in time when the cow is in estrus, and the probability of mating failure caused by manual monitoring lag is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0071] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0072] Figure 1 is a system principle block diagram of a cow estrus mating monitoring system based on blockchain edge computing according to an embodiment of the present application;

[0073] Figure 2 is a main body structure schematic diagram of an intelligent collar terminal in a cow estrus mating monitoring system based on blockchain edge computing according to an embodiment of the present application;

[0074] Figure 3 is a structure block diagram of an intelligent collar terminal in a cow estrus mating monitoring system based on blockchain edge computing according to an embodiment of the present application;

[0075] Figure 4 is a communication schematic diagram of a cow estrus mating monitoring system based on blockchain edge computing according to an embodiment of the present application;

[0076] Figure 5 is a logic flow diagram of an intelligent collar terminal in a cow estrus mating monitoring system based on blockchain edge computing according to an embodiment of the present application;

[0077] Figure 6This is a schematic diagram of the system composition of a cattle estrus and mating monitoring system based on blockchain edge computing according to an embodiment of the present invention;

[0078] Figure 7 This is a logical flowchart of a cattle estrus and mating monitoring system based on blockchain edge computing according to an embodiment of the present invention.

[0079] In the picture:

[0080] 1. Smart collar terminal; 2. Data management platform; 3. Event management platform; 4. Display and interactive application; 5. SMS notification platform. Detailed Implementation

[0081] According to an embodiment of the present invention, a cattle estrus and mating monitoring system based on blockchain edge computing is provided.

[0082] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 and Figure 6 As shown, according to an embodiment of the present invention, a cattle estrus and mating monitoring system based on blockchain edge computing includes:

[0083] The smart collar terminal 1 is worn around the neck of cattle. It monitors the cattle's vital signs in real time through built-in sensors and dynamically displays the cattle's estrus status through smart LED lights.

[0084] In the description of this invention, as Figure 2 and Figure 3 As shown, the intelligent collar terminal 1 (in this invention, it is worn around the neck of a cow as an intelligent LED flashing collar) includes: an internal task management module, an estrus monitoring module, and an LED light module ( Figure 2 As shown in L), it includes a communication module (which can be divided into a wireless communication module and a serial communication module according to its function), as well as a battery and its battery management function.

[0085] The internal task management module is used to monitor the running status of each module, optimize the resource usage of each module, manage data flow and processing, and execute preset work logic.

[0086] In the description of this invention, monitoring the operating status of each module, optimizing the resource usage of each module, managing data flow and processing, and executing preset working logic include:

[0087] Step S101: Check the hardware and software status of each module. When the functions are normal, allocate initial computing and storage resources and continuously monitor the operating indicators of each module. When the standard threshold is exceeded, trigger the early warning mechanism.

[0088] Step S102, based on the actual operation of the module, real-time adjustment of resource allocation, and management of data collection, processing, storage and transmission flow processing procedure.

[0089] Step S103, record the running state and resource usage of each module, generate node log, and automatically generate report when abnormality occurs, and send to data management platform 2.

[0090] The estrus monitoring module is used for collecting the physical data of the cattle through the built-in sensor, and identifying the estrus state of the cattle in the monitoring period by means of edge computing and machine learning algorithm, and the physical data includes activity data, rumination data and body temperature data.

[0091] In the description of the present application, the estrus monitoring module comprises a motion sensor, a pressure sensor, a body temperature sensor, an activity recording unit, a rumination recording unit and a state evaluation unit.

[0092] The motion sensor is used for collecting the motion parameters generated by the cattle during feeding by measuring three-axis acceleration, angular velocity and angle.

[0093] Specifically, the motion sensor measures the acceleration of the cattle along X, Y and Z directions through the built-in accelerometer. The acceleration data can reflect the motion intensity and direction change of the cattle, thereby helping to analyze the activity behavior of the cattle, such as walking, running or standing, etc. The motion sensor also has a built-in gyroscope for measuring the rotation speed of the cattle around each axis in space. The angular velocity data can capture the action of the cattle turning its head, turning its body or other rapid movements. Through comprehensive analysis of the acceleration and angular velocity data, the motion sensor can calculate the posture angle of each part of the cattle's body, such as the head tilt angle and the body tilt angle, which helps to further understand the behavior state of the cattle, such as whether it is resting, eating grass or drinking water.

[0094] The pressure sensor is used for obtaining the pressure parameters of the cattle during rumination behavior by measuring the waveform transformation of the cattle during mastication and swallowing.

[0095] Specifically, the pressure sensor can detect the pressure change generated during mastication and swallowing in real time. Each rumination behavior will produce a specific pressure waveform, and the sensor can identify the rumination action by measuring these waveform changes. Through the collected pressure data, the sensor can analyze the mastication frequency, mastication intensity and swallowing frequency of the cattle at different time periods. These information can be used to judge whether the cattle is performing rumination behavior, and to distinguish rumination from other feeding behaviors.

[0096] The body temperature sensor is used for collecting the body temperature data of the cattle during feeding in real time.

[0097] An activity recording unit is configured to record the motion parameters of the cattle and analyze the activity data of the cattle in the current monitoring period according to a preset monitoring period.

[0098] A rumination recording unit is configured to record the stress parameters of the cattle and identify the rumination data of the cattle in the current monitoring period by inputting into a pre-constructed behavior classification model.

[0099] In the description of the present application, recording the stress parameters of the cattle and identifying the rumination data of the cattle in the current monitoring period by inputting into a pre-constructed behavior classification model comprises:

[0100] In step S111, the historical stress parameters monitored by the cattle are obtained, and after data standardization and normalization processing, a behavior data set is constructed, and the behavior data set is divided into a training set and a test set according to a ratio of 8:2.

[0101] In step S112, a network structure of a long short-term memory network is built with the behavior data set as an input variable, and a moth-flame optimization algorithm is used to optimize parameters of the long short-term memory network to construct a behavior classification model.

[0102] Specifically, the moth-flame optimization algorithm (MFO) is a novel natural heuristic algorithm, which is inspired by the mathematical model of the phototactic behavior of moths. MFO simulates the trajectory of moths flying around a flame to perform global optimization search and is widely used to solve various optimization problems. The long short-term memory network (LSTM) is a special recurrent neural network (RNN) that is specifically designed to process and predict time series data. LSTM solves the problem of gradient disappearance or explosion in long sequence training of traditional RNN by introducing a memory unit.

[0103] The combination of the moth-flame optimization algorithm (MFO) and the long short-term memory network (LSTM) is to optimize the hyperparameters of the LSTM network, which has the advantages of fast optimization speed and not easy to fall into local optimum, thereby improving the performance.

[0104] In the description of the present application, the network structure of the long short-term memory network is built, the moth-flame optimization algorithm is used to optimize the parameters of the long short-term memory network, and the behavior classification model is constructed.

[0105] In step S1121, the number and value range of the model parameters to be optimized of the long short-term memory network are set. The model parameters include the maximum training axis, the learning rate, the block size, the first layer hidden layer neuron data, and the second layer hidden layer neuron number.

[0106] In step S1122, the evaluation index accuracy of the long short-term memory network is set as an output of the fitness function.

[0107] In step S1123, a firefly matrix is initialized, and a firefly fitness vector is calculated according to the firefly matrix and the fitness function.

[0108] In step S1124, it is determined whether the maximum iteration number is reached, if the current iteration number exceeds the maximum iteration number, the iteration is ended, if the current iteration number does not reach the maximum iteration number, the population is updated by using a firefly position updating formula, and the optimization is continued, the firefly position updating formula is:

[0109] S(M i ,F j )=D i ·e bt ·cos(2πt)+F j

[0110] D i =|F j -M i |

[0111] In the formula, S represents a logarithmic spiral function, D i represents a distance from the i-th firefly to the j-th flame, M i represents the i-th firefly, F j represents the j-th flame, b represents a constant of S, t represents a random number, and e represents a natural constant.

[0112] In step S1125, after the iteration is ended, optimal model parameters are output, and a behavior classification model is obtained.

[0113] In step S113, the pressure parameters in the current monitoring period are input into the behavior classification model, the behavior classification results of the cattle in the current monitoring period are counted, and the duration of the ruminant behavior of the cattle is extracted as the ruminant data. The behavior classification results include the ruminant behavior, the eating behavior, the drinking behavior and the stop-eating behavior.

[0114] The state evaluation unit is configured to evaluate the estrus state of the cattle based on the activity data, the body temperature data set and the ruminant data of the cattle in the current monitoring period.

[0115] In the description of the present application, the estrus state of the cattle is evaluated based on the activity data, the body temperature data set and the ruminant data of the cattle in the current monitoring period.

[0116] In step S121, based on the monitoring result of the activity data, the activity type of the cattle is divided into an active posture and an inactive posture, the percentage of the active posture of the cattle in the total duration of the monitoring period in the current monitoring period is counted as the activity amount feature of the cattle in the current monitoring period.

[0117] Specifically, the activity of the cow is monitored by the motion sensor, and active postures (such as standing, walking) and inactive postures (such as resting, lying) are distinguished. Then, the duration of the active posture of the cow in the current monitoring period is counted, and the percentage of the total duration of the monitoring period is calculated. This percentage is the activity feature, which reflects the activity level of the cow.

[0118] Step S122, based on the monitoring result of the rumination data, the percentage of the duration of the rumination behavior of the cow in the current monitoring period to the total duration of the monitoring period, and the percentage of the duration of the rumination behavior to the total feeding time are counted as the rumination feature of the cow in the current monitoring period.

[0119] Step S123, based on the monitoring result of the body temperature data, the body temperature change of the cow in the current monitoring period is recorded, and the temperature difference between the cow and the standard body temperature is calculated as the body temperature feature of the cow in the current monitoring period.

[0120] Step S124, the activity feature, rumination feature and body temperature feature are used to construct a feature matrix as the model input, and the recognition label of the estrus state of the cow is pre-set to construct a state recognition model based on long short-term memory network. The estrus state of the cow in the current monitoring period is recognized by using the state recognition model.

[0121] Specifically, all the extracted feature values are combined into a matrix. Assuming that there are n samples, each sample has 3 main features (activity feature, rumination feature, body temperature feature), and the dimension of the feature matrix X is n x 3.

[0122] In order to make the model more easily convergent, the feature matrix is usually standardized to make the range of each feature value between 0 and 1, or standardized to a normal distribution with mean 0 and variance 1. Set the estrus state label: for each sample x i , assign the estrus state label y i , the label can be binary classification (estrus or non-estrus), or multi-classification (for example, 0 represents non-estrus, 1 represents impending estrus, and 2 represents estrus).

[0123] LSTM network structure design: LSTM is a neural network structure suitable for processing time series data, which can capture long-term dependencies. In order to apply the LSTM model to identify the estrus state of the cow, the following structure is usually designed:

[0124] Input layer: the feature matrix as the data of the input layer, a feature vector x i is input at each time step.

[0125] LSTM layer: includes one or more LSTM layers to learn patterns in time series. The output of the LSTM layer is passed to the next layer or directly connected to the output layer.

[0126] Fully connected layer: connects the output of the LSTM layer to one or more fully connected layers for further processing of the features extracted by LSTM.

[0127] Output layer: depending on the type of problem, the output layer can be a Sigmoid activation function for binary classification problems or a Softmax activation function for multi-classification problems. The labeled estrus status labels (y values) are used to train the LSTM model. Define the loss function (e.g., binary cross-entropy loss or multi-class cross-entropy loss) and optimize the model weights through the backpropagation algorithm.

[0128] LED light module for prompting the estrus status of the cow through different flashing modes.

[0129] Communication module for uploading the collected and processed physical data to the data management platform and providing terminal identification and positioning functions, binding the device number of the collar with the individual number of the cow.

[0130] In addition, as Figure 5 As shown in Figure 7 The present application uses an intelligent collar terminal to collect the physical data of the cow and its related processing logic flow, the purpose is to help the operator or manager quickly grasp the estrus status of the cow through continuous monitoring and LED feedback, so as to realize quick response.

[0131] Data management platform 2 for receiving the monitored physical data, centrally managing and storing the estrus event information of each cow through the constructed database, and establishing information interaction with the event management platform.

[0132] In the description of the present application, the data management platform 2 includes: edge management module (not labeled in the figure), data receiving module (not labeled in the figure), data management module (not labeled in the figure), double-chain storage module (not labeled in the figure), access control module (not labeled in the figure), interface management module (not labeled in the figure).

[0133] Among them, the edge management module is used to deploy the Internet of Things gateway, manage all intelligent collar terminals 1 in the breeding area in zones, and coordinate the distributed processing of physical data and estrus status recognition.

[0134] The data receiving module is used to receive and integrate the real-time data uploaded by the intelligent collar terminal 1.

[0135] A data management module is configured to associate individual numbers of the cattle with respective physical data and estrus states, form estrus event information corresponding to the cattle, and perform partition management according to breeding areas.

[0136] A double-chain storage module is configured to construct data blocks according to real-time data uploaded by the smart collar terminal 1, and store the data blocks into server nodes in an asynchronous and concurrent manner to form a double-chain storage structure.

[0137] In the description of the present application, the construction of the data blocks according to the real-time data uploaded by the smart collar terminal 1 and the storage of the data blocks into the server nodes in an asynchronous and concurrent manner to form the double-chain storage structure include:

[0138] In step S201, real-time data uploaded by the smart collar terminal 1 worn by each cattle is generated into data blocks according to different breeding areas, and the data blocks are stored into server nodes in a time sequence.

[0139] In step S202, each server node stores data blocks of multiple breeding areas, and when the server node stores the blocks, the storage rules are verified first, and the block storage is performed after the verification.

[0140] In step S203, the storage rules of the first two blocks are preset, and starting from the third block, with the increase of each block, the storage information is published to the block chain according to the storage rules of the first two blocks when the server obtains the storage permission of other breeding areas, and the block is attached to the tail of the server node, and other server nodes give up the storage right competition of the block.

[0141] Specifically, the advantages of the double-chain storage structure are as follows: a) high efficiency: since the consensus mechanism is storage power, computing power and time sequence, complex algorithms are not required, and data can be directly stored and confirmed with other nodes after being generated, and the storage efficiency is high; b) safety: in the private chain, each data storage needs to verify the previous two blocks (as a reference rule), that is, a chain can form a trust chain after three times of storage. The double-chain backup storage improves the security of the data; c) decentralization: the double-chain storage structure has the advantages of decentralization of the block chain, and all servers manage the data together without a centralized management mechanism; d) distributed storage: the data of each breeding area is managed by multiple servers, and multiple breeding areas can be recorded at the same time to realize distributed storage.

[0142] An access control module is configured to load a ciphertext access mechanism, grant data access rights to data owners and authorized persons, and regularly query and verify the integrity of the block chain data.

[0143] An interface management module is configured to provide data interfaces between the event management platform 3 and the display interaction application and support conversion between multiple data formats and protocols.

[0144] The event management platform 3 is configured to obtain estrus event information from the data management platform, forward application operations from the display interaction application, and provide access interfaces to realize information interaction between platforms.

[0145] In the description of the present application, the event management platform 3 includes an event processing module (not labeled in the figure), a real-time monitoring module (not labeled in the figure), a policy management module (not labeled in the figure), an event cache module (not labeled in the figure), and an extension interface module (not labeled in the figure).

[0146] The event processing module is configured to classify, sort, and prioritize estrus event information according to preset rules and respond to estrus state abnormal cattle in a timely manner.

[0147] The real-time monitoring module is configured to monitor the estrus time processing process in real time, record all processing logs, and store the processing logs through blockchain encryption.

[0148] The policy management module is configured to support administrators to define estrus event processing strategies.

[0149] The event cache module is configured to temporarily store estrus event information that is not processed or is being processed.

[0150] The extension interface module is configured to provide data interfaces with the display interaction application, the data management platform, and other modules, and is responsible for conversion and adaptation of data formats and communication protocols.

[0151] The display interaction application 4 is configured to connect the event management platform, query, and display the estrus state of the cattle.

[0152] Specifically, as shown in Figure 4 The wireless access and communication mode diagram of the present system is shown. The display interaction application 4 plays a key role in connecting users and system data in the entire cattle estrus and breeding monitoring system based on blockchain edge computing. Its main function is to query and display the estrus state of the cattle through interaction with the event management platform 3, so that users can intuitively and timely obtain key information.

[0153] The display interaction application 4 provides an intuitive and easy-to-use user interface (UI) to display the estrus state of the cattle. It supports multiple display modes such as charts, lists, timelines, etc., allowing users to clearly view the changes in the estrus state of the cattle. Real-time display of the estrus state changes of the cattle, through regular data synchronization with the event management platform, ensures the timeliness and accuracy of the display information, and provides a state update notification function to timely remind users when the estrus state of the cattle changes.

[0154] Users can query historical estrus state data of specific cattle through the application for longitudinal comparison analysis. It supports filtering data based on different conditions (such as time period, estrus state, etc.) to help users quickly locate the content of interest. Estrus state data obtained from the event management platform is visualized to generate charts or reports, helping users better understand and analyze the estrus trends of the cattle; provides data export function, users can export the analysis results in multiple formats, convenient for further processing or saving.

[0155] The SMS notification platform 5 is used to send SMS notifications to registered users through wireless communication.

[0156] In addition, the present application combines edge computing and blockchain technology to build an IoT edge computing blockchain cloud architecture, which can be divided into five layers according to the system architecture, namely device layer, container layer, middle layer, application layer and user layer. In the system framework of the present application, the corresponding relationship between each layer and the system module is as follows:

[0157] 1. Device layer (corresponding to smart collar terminal)

[0158] The device layer is the basic layer of the entire system, mainly including hardware devices for data collection and preliminary processing. In the estrus monitoring system, the smart collar terminal serves as the core node of the device layer, which monitors the physical data of the cattle (such as activity, body temperature, rumination behavior, etc.) in real time, and performs preliminary data processing and estrus state analysis locally on the collar.

[0159] Function: The device layer reduces the delay of data transmission through the edge computing capabilities of the smart collar terminal, improves the real-time performance of data processing, and provides high-quality data sources for the upper-layer platform.

[0160] 2. Container layer (corresponding to data management platform)

[0161] The container layer is the data storage and management center of the system. The data management platform is responsible for receiving data from the smart collar terminal, storing, classifying, managing, and preliminarily analyzing these data to support the upper-layer applications. The design of the container layer focuses on the scalability and security of data, especially when using blockchain technology, to ensure the integrity and tamper resistance of data.

[0162] Function: The container layer provides a stable and secure data processing and storage environment for the system, ensuring the reliability and consistency of data transmission and usage within the system.

[0163] 3. Middle layer (corresponding to the event management platform)

[0164] The middle layer is the coordination center of the system. The event management platform in this layer is responsible for managing and coordinating information exchange and operation requests between various functional modules. It receives estrus event information from the data management platform and handles requests and operations for the display interactive application and SMS notification platform, ensuring that all parts of the system work effectively in coordination.

[0165] Function: Through the role of the event management platform, the middle layer achieves orderly collaboration between different modules in the system, ensuring smooth information flow and efficient operation execution.

[0166] 4. Application layer (corresponding to the display interactive application)

[0167] The application layer is the part of the system that users directly interact with, mainly responsible for converting data into visual display content. The display interactive application in this layer displays the estrus state information of cattle through an intuitive interface and allows users to perform related operations such as query, analysis, and report generation.

[0168] Function: The application layer converts complex system data into easy-to-understand charts and information displays through a user-friendly interface, helping users make accurate decisions.

[0169] 5. User layer (corresponding to the SMS notification platform)

[0170] The user layer is the level at which the system directly interacts with users. The SMS notification platform delivers key information about the estrus state of cattle to users through SMS or other communication methods. This layer ensures that users can receive important notification information at any time and place without needing to constantly check the system.

[0171] Function: The user layer ensures that users can promptly learn about the estrus state of cattle through the real-time communication function of the SMS notification platform, thereby effectively managing breeding.

[0172] In summary, by means of the technical scheme of the present application, a complete cattle estrus monitoring and management solution system is formed by combining the intelligent collar terminal, the data management platform, the event management platform, the display interaction application and the SMS notification platform; by monitoring the physical data of the cattle in real time, the estrus state is analyzed and the estrus state is displayed intuitively by using the built-in intelligent LED light, so that the manager or the operator can identify and take mating measures in time, thereby effectively improving the accuracy of cattle estrus monitoring and the management efficiency and reducing the breeding cost. By using edge computing and blockchain technology, the data is processed and stored quickly and safely, and cross-platform information interaction and access control are supported, the efficiency and reliability of data management are improved, and the security of data and the expansibility of the system are effectively enhanced; at the same time, a convenient query and monitoring interface is provided for the user, the visualization and tracking of the estrus state become more intuitive and easy to use, and the wireless communication technology is used to send SMS notification, so that the manager can receive the notification in time when the cattle is in estrus, and the probability of mating failure caused by manual monitoring lag is reduced.

[0173] The above merely describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A cattle estrus and mating monitoring system based on blockchain edge computing, characterized in that, The system includes: The intelligent collar terminal is worn around the neck of cattle. It monitors the cattle's vital signs in real time through built-in sensors and dynamically displays the cattle's estrus status through intelligent LED lights. The smart collar terminal includes: an internal task management module, an estrus monitoring module, an LED light module, and a communication module. The internal task management module monitors the operational status of each module, optimizes resource usage, manages data flow and processing, and executes preset work logic. The estrus monitoring module collects vital sign data from cattle using built-in sensors and employs edge computing and machine learning algorithms to identify the estrus state of the cattle within the monitoring period. The vital sign data includes activity data, rumination data, and body temperature data. The LED light module uses different flashing patterns to indicate the estrus state of the cattle. The communication module uploads the collected and processed vital sign data to a data management platform and provides terminal identification and positioning functions, binding the collar's device number to the cattle's individual number. The estrus monitoring module includes: a motion sensor, a pressure sensor, a body temperature sensor, an activity recording unit, a rumination recording unit, and a state assessment unit. The motion sensor collects motion parameters generated by cattle during feeding by measuring triaxial acceleration, angular velocity, and angle. The pressure sensor obtains pressure parameters during rumination by measuring waveform changes during chewing and swallowing. The body temperature sensor collects real-time body temperature data of cattle during feeding. The activity recording unit records the cattle's motion parameters and analyzes the cattle's activity data within a preset monitoring period. The rumination recording unit records the cattle's pressure parameters and identifies rumination data within the current monitoring period by inputting them into a pre-built behavioral classification model. The state assessment unit assesses the estrus status of the cattle based on the activity data, body temperature data, and rumination data within the current monitoring period. The process of recording cattle stress parameters and inputting them into a pre-constructed behavioral classification model to identify cattle rumination data within the current monitoring period includes: acquiring historical stress parameters from cattle monitoring; constructing a behavioral dataset after data standardization and normalization; dividing the dataset into training and testing sets in an 8:2 ratio; using the behavioral dataset as input variables to build a Long Short-Term Memory (LSTM) network structure; optimizing the parameters of the LSM network using the Moth to a Flame algorithm to construct a behavioral classification model; inputting the stress parameters of the current monitoring period into the behavioral classification model; statistically analyzing the behavioral classification results of cattle within the current monitoring period; and extracting the duration of rumination behavior as rumination data; wherein the behavioral classification results include rumination behavior, feeding behavior, drinking behavior, and fasting behavior. The data management platform is used to receive vital sign data obtained from monitoring, centrally manage and store estrus event information of each cattle through a constructed database, and establish information interaction with the event management platform; the data management platform includes: an edge management module, a data receiving module, a data management module, a dual-chain storage module, an access control module, and an interface management module; The event management platform is used to obtain estrus event information from the data management platform, forward application operations from the display and interactive application, and provide access interfaces to realize information exchange between platforms. The interactive application is used to connect to the event management platform to query and display the estrus status of cattle; SMS notification platform, used to send SMS notifications to registered users via wireless communication.

2. The cattle estrus and mating monitoring system based on blockchain edge computing according to claim 1, characterized in that, The monitoring of the operational status of each module, optimization of resource utilization of each module, management of data flow and processing, and execution of preset working logic include: Check the hardware and software status of each module. If the functions are normal, allocate initial computing and storage resources and continuously monitor the operating indicators of each module. If the standard threshold is exceeded, trigger the early warning mechanism. Based on the actual operation of the modules, resource allocation is adjusted in real time, and the flow of data collection, processing, storage and transmission processes is managed. Record the running status and resource usage of each module, generate node logs, and automatically generate reports when an anomaly occurs, sending them to the data management platform.

3. The cattle estrus and mating monitoring system based on blockchain edge computing according to claim 1, characterized in that, The network structure for building a Long Short-Term Memory (LSTM) network, using the moth-to-a-flame algorithm to optimize the parameters of the LSM network, and constructing the behavior classification model includes: The number and range of model parameters to be optimized for the Long Short-Term Memory Network are set; wherein, the model parameters include the maximum training axis, learning rate, block size, number of neurons in the first hidden layer, and number of neurons in the second hidden layer; The accuracy metric for evaluating Long Short-Term Memory (LSTM) networks is set as the output of the fitness function. Initialize the moth matrix, and calculate the moth fitness vector based on the moth matrix and the fitness function; Determine if the maximum number of iterations has been reached. If the current iteration count exceeds the maximum, the iteration ends. If the current iteration count has not reached the maximum, the population is updated using the moth position update formula, and the search for optimization continues. The moth position update formula is: In the formula, S Represents the logarithmic spiral function; D i Indicates the first i Only one moth reached the first j The distance of a flame; M i Indicates the first i A moth; F j Indicates the first j A flame; b express S The constant; t Represents a random number; e Represents the natural constant; After all iterations are completed, the optimal model parameters are output, resulting in the behavior classification model.

4. A cattle estrus and mating monitoring system based on blockchain edge computing according to claim 1, characterized in that, The assessment of the estrus status of a cow based on its activity data, body temperature dataset, and rumination data during the current monitoring period includes: Based on the monitoring results of activity data, the activity types of cattle are divided into active postures and inactive postures. The percentage of active postures of cattle in the current monitoring period is counted as the activity level characteristics of cattle in the current monitoring period. Based on the monitoring results of rumination data, the percentage of the duration of rumination behavior in cattle during the current monitoring period and the percentage of the duration of rumination behavior in the total feeding time are statistically analyzed as the rumination characteristics of cattle during the current monitoring period. Based on the monitoring results of body temperature data, the change in body temperature of cattle during the current monitoring period is recorded, and the temperature difference between the cattle and the standard body temperature is calculated as the body temperature characteristics of cattle during the current monitoring period. A feature matrix was constructed using activity level, rumination, and body temperature characteristics as model input. A state recognition model based on a long short-term memory network was built by pre-setting identification labels for the estrus state of cattle. The state recognition model was then used to identify the estrus state of cattle in the current monitoring period.

5. A cattle estrus and mating monitoring system based on blockchain edge computing according to claim 1, characterized in that, The edge management module is used to deploy IoT gateways, manage all smart collar terminals in the breeding area by zone, and coordinate the distributed processing of vital sign data and estrus status identification. The data receiving module is used to receive and integrate real-time data uploaded by the smart collar terminal; The data management module is used to associate the individual numbers of cattle with their respective vital signs and estrus status, forming estrus event information corresponding to each cattle, and to manage them in zones according to the breeding area. The dual-chain storage module is used to construct data blocks based on real-time data uploaded by the smart collar terminal, and to store the data blocks to the server node in an asynchronous and concurrent manner, forming a dual-chain storage structure. The access control module is used to carry out the encrypted access mechanism, grant data access permissions to data owners and authorizers, and periodically query and verify the integrity of blockchain data; The interface management module is used to provide data interfaces between the event management platform and the display and interaction application, and supports the conversion of various data formats and protocols.

6. A cattle estrus and mating monitoring system based on blockchain edge computing according to claim 5, characterized in that, The step of constructing data blocks based on real-time data uploaded by the smart collar terminal and storing the data blocks to the server node in an asynchronous concurrent manner to form a dual-chain storage structure includes: The real-time data uploaded by each cattle wearing smart collar terminals is generated into data blocks according to different breeding areas, and then the data blocks are stored in the server node in chronological order. Each server node stores data blocks from multiple aquaculture areas. When storing blocks, the server node first verifies the storage rules and then stores the blocks after the verification is successful. The storage rules for the first two blocks are pre-defined. Starting from the third block, for each additional block, based on the storage rules of the first two blocks, when the server obtains storage rights for other breeding areas, it publishes the storage information to the blockchain and appends the block to the end of the chain of the server node. Other server nodes give up participating in the competition for storage rights of the block.

7. A cattle estrus and mating monitoring system based on blockchain edge computing according to claim 1, characterized in that, The event management platform includes: an event processing module, a real-time monitoring module, a policy management module, an event caching module, and an extension interface module; The event processing module is used to classify, sort, and prioritize estrus event information according to preset rules, and to respond promptly to cattle with abnormal estrus states. The real-time monitoring module is used to monitor the estrus time processing process in real time, record all processing logs, and then encrypt and store the processing logs through blockchain. The strategy management module is used to support administrators in defining strategies for handling estrus events; The event caching module is used to temporarily store information on unprocessed and currently being processed estrus events; The extended interface module is used to provide data interfaces with interactive display applications, data management platforms and other modules, and is responsible for the conversion and adaptation of data formats and communication protocols.

Images