Intelligent construction system for rapid breeding of mutton sheep

Abstract

The invention discloses an intelligent construction system for rapid breeding of mutton sheep, and relates to the technical field of mutton sheep breeding, and the system comprises a data acquisition module, a gene analysis module, an intelligent decision module, a breeding execution module and a block chain transaction module. The data acquisition module is used for acquiring physiological data, gene data and behavior data of mutton sheep in real time through a sensor and a'Anhui sheep No.1 breeding chip '; the gene analysis module is used for carrying out accurate genotype analysis and gene scoring by utilizing the collected gene data; the intelligent decision-making module generates an optimal breeding scheme through a multi-objective optimization algorithm; the breeding execution module performs automatic hybridization, breeding management and offspring monitoring according to the intelligent decision result; and the block chain transaction module realizes automatic matching and decentralized transaction of breeding sheep resources through an intelligent contract. According to the invention, the health state of the sheep flock can be monitored in real time, the breeding decision is optimized, the manual intervention is reduced, the production efficiency is improved, and the transparency and automation of sheep transaction are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of meat sheep breeding, and particularly to an intelligent construction system for rapid meat sheep breeding. Background Art

[0002] In recent years, with the rapid development of information technology, artificial intelligence, and biotechnology, meat sheep breeding has gradually developed towards the direction of intelligence and precision. In the process of meat sheep breeding, traditional breeding methods have problems such as low efficiency and insufficient accuracy. With the continuous improvement of the requirements for selecting excellent traits of meat sheep, there is an urgent need for innovative breeding technologies and systems. The "Wanyang 1 Breeding Chip" successfully developed by the team led by Professor Ling Yinghui of Anhui Agricultural University has brought new opportunities for meat sheep breeding. The present invention is a new breeding system constructed based on this and related intelligent technologies.

[0003] Currently, there are some obvious technical deficiencies in meat sheep breeding, which are mainly reflected in the following aspects: the problem of data islands, where data between different technologies, devices, and systems cannot be effectively integrated, resulting in the inability to perform cross-module data analysis and decision-making, affecting the overall breeding effect; high cost and complex operation. Although genomics and gene editing technologies provide possible technical support for meat sheep breeding, the application of these technologies is still limited by high costs and operational complexity, making it difficult to popularize in large-scale farms; lack of real-time performance. Although existing intelligent systems can collect data in real time, due to the lack of powerful analysis and decision-making support functions, timely feedback and adjustment cannot be achieved, resulting in the inability to make a quick response in the face of emergencies or health problems; insufficient accuracy of traditional breeding methods. Although traditional experience-based breeding methods have optimized the reproductive performance of meat sheep to a certain extent, they often lack scientific basis and precise decision-making support and cannot meet the increasing demand for refined management.

[0004] In view of the above problems, the present invention proposes an intelligent meat sheep breeding system that comprehensively applies modern Internet of Things, big data, artificial intelligence, genomics and other technologies to solve the problems of data islands, high costs, poor real-time performance, and insufficient accuracy. Summary of the Invention

[0005] The present invention aims at the above problems and provides an intelligent construction system for rapid meat sheep breeding to solve the problems of data islands, insufficient accuracy, low transaction efficiency, and lack of real-time decision-making support in the prior art.

[0006] To solve the above technical problems, the present invention provides the following technical solution: An intelligent construction system for rapid meat sheep breeding, comprising: Data acquisition module, which collects physiological data, behavioral data and genetic data of meat sheep, and provides input for subsequent gene analysis, intelligent decision-making and breeding execution modules; Gene analysis module, which conducts a detailed analysis of the genetic data collected from the data acquisition module, and matches the genotypes related to meat performance and lactation performance; Intelligent decision-making module, which generates the optimal breeding decision based on the gene scores and genotype data provided by the gene analysis module, as well as the physiological data, behavioral data and environmental data provided by the data acquisition module; Breeding execution module, which executes specific mating operations, reproductive management and health tracking of offspring according to the breeding plan generated by the intelligent decision-making module; Blockchain transaction module, which realizes the automation and transparency of breeding sheep transactions through smart contracts and decentralized transactions; Data processing and decision-making layer module, which preprocesses, analyzes and conducts health prediction on the data obtained from the data acquisition module, gene analysis module and intelligent decision-making module, and generates optimized decisions for the meat sheep population.

[0007] The data acquisition module includes a physical sign data acquisition unit and a genetic data acquisition unit; The physical sign data acquisition unit uses a body temperature sensor, a heart rate sensor and an acceleration sensor to collect the physical sign data of meat sheep in real time. The physical sign data includes body temperature, heart rate and motion state; the behavior state of meat sheep is identified through the acceleration sensor, and the acquisition frequency of physical sign data is adjusted according to the behavior state. The sampling period of meat sheep in the stationary state is longer than that in the motion state. The sampling period in the stationary state of meat sheep is set as and the sampling period in the motion state is , specifically as shown in the formula:

[0008] where is the sampling frequency of physical sign data, and are the sampling periods in the stationary state and the motion state respectively; The genetic data acquisition unit collects the genetic data of meat sheep through the "Wanyang No. 1 Breeding Chip", and extracts the gene information related to meat performance and lactation performance. The gene information related to meat performance includes fast growth genes and disease resistance genes, and the gene information related to lactation performance includes milk yield genes and milk fat genes.

[0009] The data analysis module includes a gene data analysis unit and a gene scoring unit; The gene data analysis unit analyzes and identifies gene information related to the growth and milk production of meat sheep by receiving gene data collected by the "Wanyang No. 1 breeding chip". Compare the collected genotype data with the known excellent trait genotypes, evaluate the genetic potential of each meat sheep for specific traits, where the specific traits include meat performance and lactation performance; through genome-wide association analysis, determine the association between genotypes and target traits, and genome-wide association analysis is used to identify gene information significantly related to specific traits of meat sheep. The genotype refers to the genetic composition of an individual at specific gene loci, and the gene information refers to these gene loci and their effects on specific traits. The model is specifically as shown in the formula:

[0010] Where, is the phenotypic data of the is the genotype data of the is the effect value of the is the constant term, representing the basic relationship between genotype and phenotype; is the error term, representing the influence of other unconsidered factors on the phenotype; is the total number of gene information; The gene scoring unit evaluates the genotype of each meat sheep to generate a gene score, representing its genetic potential for specific traits. The gene score will provide key genetic information for the intelligent decision-making module to support the optimization of the breeding plan. The gene score is generated by analyzing the association between the genotype data of each meat sheep and excellent traits, specifically as shown in the formula:

[0011] Where, is the gene score of the is the genotype data of the is the degree of influence of the is the total number of gene information.

[0012] The intelligent decision-making module includes a multi-objective decision-making unit and a genetic algorithm decision-making unit; The multi-objective decision-making unit uses the GECO model to generate a breeding plan by comprehensively considering gene data, environmental data, and economic data. The objective function is designed as a multi-objective optimization problem, and multiple objectives will jointly affect the breeding decision. The multiple objectives include growth traits, reproductive traits, environmental adaptability, and economic benefits. The optimization formula of the GECO model is specifically as shown in the formula:

[0013] Among them, represents the genetic objective, represents the environmental adaptability objective, represents the economic benefit objective, is the decision variable, representing the genotype, mating combination, and feeding strategy of meat sheep; The genetic algorithm decision-making unit is based on the genetic algorithm and is used to generate the optimal solution in the multi-objective optimization process. The genetic algorithm searches for the optimal solution in a large-scale solution space through selection, crossover, and mutation operations, and finally selects those solutions that can achieve the best balance among multiple objectives. According to the gene scores and genotype data provided by the gene analysis module, combined with the environmental data and market data provided by the data collection module, comprehensively evaluate the biological benefits and economic benefits of each mating plan, and generate personalized breeding suggestions for each meat sheep according to the evaluation results, including mating combinations, feeding strategies, and adjustment strategies during the breeding period; The genetic algorithm generates the optimal breeding plan through the following steps: Initialize the population, create the initial population, and each individual in the population represents a possible breeding decision; Fitness evaluation, calculate the fitness of each individual, and the fitness function is usually the weighted sum of each objective function, representing the comprehensive performance under different objectives; Selection operation, select individuals with high fitness to enter the next generation; Crossover and mutation, generate new candidate solutions through crossover and mutation operations. The crossover operation combines the information of the parent solutions to generate new offspring solutions, and the mutation operation introduces random changes to ensure the diversity of solutions; Iterative update, through multiple generations of evolution, the system gradually approaches the optimal solution through selection, crossover, and mutation operations; The specific form of the fitness function is as shown in the formula:

[0014] Among them, is the multi-objective function, is the weight of each objective function.

[0015] The breeding execution module includes a breeding management unit and a cycle monitoring unit; After the breeding management unit follows the optimal breeding plan of the intelligent decision-making module, it automatically selects suitable breeding sheep for mating. The decision-making plan is usually based on gene scoring, breeding history, and market demand. The mating task is completed through automated equipment, and suitable combinations of breeding sheep are selected for mating and reproduction, as specifically shown in the formula:

[0016] Among them, is the gene score of the breeding sheep, is the breeding history of the breeding sheep, is the score of market demand, are the weights of each factor; The cycle monitoring unit monitors the pregnancy status of ewes, the fetal development status, and the health status of meat sheep in real time through sensors. By monitoring the growth and health status of the offspring, and through sensor data and gene data, it tracks the healthy development of the offspring in real time, predicts possible health problems, and automatically adjusts the subsequent breeding plan according to the health status and growth performance of the offspring, optimizing the selected breeding sheep and mating strategies.

[0017] The blockchain transaction module includes an intelligent contract execution unit and a decentralized breeding sheep trading unit; The intelligent contract execution unit sets trading conditions according to the breeding plan generated by the breeding execution module and the intelligent decision-making module, and automatically completes the trading of breeding sheep when the conditions are met. According to gene scoring and breeding history factors, it determines which breeding sheep can be traded for mating. When the intelligent contract detects breeding sheep that meet the trading conditions, it automatically executes the trading process; The trading process includes determining the buyer and seller, generating trading terms, and confirming the trading price. The setting of the intelligent contract trading conditions is specifically shown in the formula:

[0018]

[0019] Among them, is the gene score of the i-th meat sheep, is the threshold of the gene score, is the breeding success rate of the i-th meat sheep, is the threshold of the breeding success rate; The decentralized breeding sheep trading unit, based on blockchain technology, allows multiple participants to directly conduct breeding sheep trading without an intermediary. All trading information is recorded on the blockchain, and the real data of each meat sheep can be verified. The trading execution is specifically shown in the formula:

[0020] Among them, Represents the genetic score of breeding sheep, Represents the breeding history score of breeding sheep, Are the thresholds of the genetic score and the breeding history score respectively.

[0021] The data processing and decision-making layer module includes a data processing unit and a health prediction unit; The data processing unit includes two functions: anomaly rejection and time compression; For the anomaly rejection, an anomaly detection algorithm based on statistical methods is used to reject the abnormal data that significantly deviates from the normal range from the sensor and health monitoring data. The abnormal data includes equipment failures and error data; For the time compression, based on the huge amount of data generated by each meat sheep in large-scale farms, the time window method is used to compress the time series data. By dividing the time series data into several time windows and calculating the statistical features within the window to represent the data of this time period, the purpose of time compression is achieved. Specifically, as shown in the formula:

[0022]

[0023] Where, Is the compressed data, And Are the start and end times of the time window; The health prediction unit predicts the health status of meat sheep through the LSTM long short-term memory network, identifies potential health problems, and predicts the health status of meat sheep in the future period based on the historical data collected by sensors, and early warns of potential health problems. The prediction formula is specifically shown in the formula:

[0024] Where, Is the hidden state at the current moment, that is, the prediction result; And Are weight matrices, connecting the hidden state of the previous moment and the current input respectively; Is the bias term, Is the input data at the current moment; Is the activation function.

[0025] Compared with the prior art, the beneficial effects of the present invention are: The target optimization algorithm is used for breeding decision-making, which improves the accuracy of meat sheep breeding. In addition, an optimal breeding plan is generated based on scientific analysis methods, avoiding the problem of poor breeding effects caused by insufficient experience or misjudgment.

[0026] Through the Internet of Things and big data analysis technologies, the present invention realizes the real-time collection and integration of multi-data sources, monitors the health status of meat sheep in real time through the LSTM health prediction model, and makes decisions quickly based on real-time data, which can significantly improve the timeliness and response speed of decisions, ensure the timely identification and solution of health problems during the breeding process, and thus effectively reduce disease transmission and breeding problems.

[0027] The present invention reduces manual intervention by adopting intelligent devices and algorithms. The application of smart contracts and blockchain technologies decentralizes and automates the process of breeding sheep trading and resource matching, improves trading efficiency, and significantly reduces intermediary fees and labor management costs.

[0028] By adopting advanced algorithms to optimize the breeding plan, the present invention can find the best balance among multiple goals, not only maximizing the growth and reproductive ability of meat sheep, but also minimizing feeding costs and resource waste, significantly improving the production efficiency of meat sheep, and ensuring higher economic returns for the farm in the short term. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0030] Figure 1 It is the system architecture diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but is merely for the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0032] Please refer to Figure 1 It is a schematic diagram of an intelligent construction system for rapid breeding of meat sheep provided by an embodiment of the present invention, including: The data acquisition module collects physiological data, behavioral data, and genetic data of meat sheep, providing input for subsequent gene analysis, intelligent decision-making, and breeding execution modules. The gene analysis module conducts a detailed analysis of the genetic data collected from the data acquisition module, matching genotypes related to meat production performance and lactation performance. The intelligent decision-making module generates optimal breeding decisions based on the gene scores and genotype data provided by the gene analysis module, as well as the physiological data, behavioral data, and environmental data provided by the data acquisition module. The breeding execution module executes specific mating operations, reproductive management, and health tracking of offspring according to the breeding plan generated by the intelligent decision-making module. The blockchain transaction module realizes the automation and transparency of breeding sheep transactions through smart contracts and decentralized transactions. The data processing and decision-making layer module mainly aims to preprocess, analyze, and perform health prediction on the data obtained from the data acquisition module, gene analysis module, and intelligent decision-making module, and generate optimized decisions for the meat sheep population.

[0033] The data acquisition module includes a physical sign data acquisition unit and a genetic data acquisition unit. The physical sign data acquisition unit uses a body temperature sensor, a heart rate sensor, and an acceleration sensor to collect the physiological data of meat sheep in real time, including body temperature, heart rate, and motion state; it identifies the behavioral state of the sheep through the acceleration sensor and dynamically adjusts the frequency of physiological data collection according to the state. Specifically, the sampling period in the stationary state of the sheep is longer than that in the motion state. The sampling period in the stationary state of the sheep is and the sampling period in the motion state is to reduce power consumption and ensure more accurate data collection during motion. Specifically, as shown in the formula:

[0034] where is the sampling frequency of physical sign data, and are the sampling periods in the stationary state and motion state respectively.

[0035] The genetic data acquisition unit collects the genetic data of meat sheep through the "Wanyang No. 1 breeding chip", extracts gene information related to meat production performance and lactation performance. Among them, the gene information related to meat production performance includes fast growth genes and disease-resistant genes, and the gene information related to lactation performance includes milk yield genes and milk fat genes, etc.; based on the genetic data, a gene score for each meat sheep is generated, indicating its genetic potential for specific traits.

[0036] ​It should be noted that the purpose of dynamically adjusting the sampling frequency is to adjust the data acquisition frequency according to the activity state of the sheep, providing a higher frequency when more data is needed, and reducing the sampling frequency when the sheep is stationary or resting, saving energy and extending the service time of the sensor.

[0037] The behavior state of the meat sheep is identified through a three-axis acceleration sensor, and the acceleration data of the sheep is analyzed in real time to determine whether the sheep is moving, eating or resting, and the sampling frequency of the physical signs is adjusted according to the behavior state, where the physical signs include body temperature and heart rate; the sampling period is dynamically updated according to the change of the behavior state to ensure that the sampling frequency is adjusted according to the activity state of the sheep.

[0038] The data analysis module includes a gene data analysis unit and a gene scoring unit; The gene data analysis unit receives the gene data collected by the "Wan Sheep No. 1 breeding chip" from the gene data collection unit, analyzes these gene data, and identifies the gene information related to excellent traits such as the growth and milk production of the meat sheep; compares the collected genotype data with the known excellent trait genotypes to evaluate the genetic potential of each meat sheep in specific traits, where the specific traits include meat performance and lactation performance; determines the association between the genotype and the target trait through genome-wide association analysis, where genome-wide association analysis is used to identify the gene information significantly related to specific traits of the meat sheep, and the model is specifically as shown in the formula:

[0039] Among them, is the phenotypic data of the th individual, is the genotype data of the th i individual at the th j gene information, is the effect value of the th j gene information, indicating the influence of this gene on the phenotype; is the constant term, indicating the basic relationship between the genotype and the phenotype; is the error term, indicating the influence of other unconsidered factors on the phenotype; is the total number of gene information.

[0040] The purpose of genome-wide association analysis is to find the gene information related to the target trait by analyzing the relationship between the genotype and the phenotype, so as to provide a basis for breeding decisions.

[0041] The gene scoring unit evaluates the genotype of each meat sheep and generates a gene score, indicating its genetic potential in specific traits. The gene score will provide key genetic information for the intelligent decision-making module to support the optimization of the breeding plan, where the gene score is generated by analyzing the genotype data of each meat sheep associated with excellent traits, and is specifically as shown in the formula:

[0042] Among them, is the genetic score of the ith meat sheep, representing its genetic potential for target traits, is the genotype data of the ith meat sheep for the jth gene information, is the influence degree of the jth gene information on the target trait, is the total number of gene information.

[0043] The intelligent decision-making module includes a multi-objective decision-making unit and a genetic algorithm decision-making unit; The multi-objective decision-making unit uses the GECO model. By comprehensively considering gene data, environmental data, and economic data, it generates an optimal breeding plan to maximize the genetic benefit of meat sheep while minimizing feeding costs and other economic objectives. In this module, the objective function is designed as a multi-objective optimization problem, and multiple objectives will jointly affect the breeding decision. The multiple objectives include growth traits, reproductive traits, environmental adaptability, and economic benefits. The optimization formula of the GECO model is specifically as shown in the formula:

[0044] Among them, represents the genetic objective, represents the environmental adaptability objective, represents the economic benefit objective, is the decision variable, representing the genotype, mating combination, and feeding strategy of meat sheep.

[0045] The genetic algorithm decision-making unit, based on the genetic algorithm, is used to generate the optimal solution in the multi-objective optimization process. The genetic algorithm searches for the optimal solution in a large-scale solution space through selection, crossover, and mutation operations, and finally selects those solutions that can achieve the best balance among multiple objectives. The system comprehensively evaluates the biological and economic benefits of each mating plan according to the gene scores and genotype data provided by the gene analysis module, combined with the environmental data provided by the data acquisition module and market data. Based on these evaluation results, the system generates personalized breeding suggestions for each meat sheep, including mating combinations, feeding strategies, and adjustment strategies during the breeding period.

[0046] The genetic algorithm generates the optimal breeding plan through the following steps: Initialize the population, create the initial population, and each individual in the population represents a possible breeding decision; Fitness evaluation, calculate the fitness of each individual. The fitness function is usually the weighted sum of each objective function, representing the comprehensive performance under different objectives; Selection operation, select individuals with high fitness to enter the next generation; Crossing and mutation generate new candidate solutions through crossing and mutation operations. The crossing operation combines the information of the parent solutions to generate new offspring solutions, and the mutation operation introduces random changes to ensure the diversity of solutions; Iterative update. After multiple generations of evolution, the system gradually approaches the optimal solution through operations such as selection, crossing, and mutation.

[0047] Among them, the fitness function is specifically as shown in the formula:

[0048] Among them, is a multi-objective function, is the weight of each objective function.

[0049] The breeding execution module includes a breeding management unit and a cycle monitoring unit; The breeding management unit, after according to the optimal breeding plan of the intelligent decision-making module, automatically selects suitable breeding sheep for breeding. The decision-making plan is usually based on gene scoring, breeding history, and market demand, and completes the breeding task through automated equipment. The breeding operation will ensure the reproduction of the most suitable combination of breeding sheep. The selection of the breeding combination is specifically as shown in the formula:

[0050] Among them, is the gene score of the breeding sheep, is the breeding history of the breeding sheep, is the score of market demand, is the weight of each factor.

[0051] The cycle monitoring unit monitors the pregnancy status of ewes, the fetal development status, and the health status of meat sheep in real time through sensors to ensure the smooth progress of the entire breeding process; by monitoring the growth and health status of offspring. Through sensor data and gene data, it tracks the healthy development of offspring in real time and predicts possible health problems. According to the health status and growth performance of offspring, it automatically adjusts the subsequent breeding plan and optimizes the selected breeding sheep and breeding strategy.

[0052] In addition, the breeding execution module collects and feeds back various data in real time, including breeding results, pregnancy status, offspring health status, etc., and adjusts the breeding strategy according to the real-time feedback data.

[0053] For example, if the offspring of a certain breeding combination perform poorly, the system will adjust the breeding plan or feeding strategy for the next round.

[0054] The blockchain transaction module includes an intelligent contract execution unit Smart contract execution unit. According to the breeding plan generated by the breeding execution module and the intelligent decision-making module, the smart contract will set trading conditions and automatically complete the transaction of breeding sheep when the conditions are met. The smart contract will determine which breeding sheep can be traded for breeding based on factors such as gene scores and breeding history. When the smart contract detects breeding sheep that meet the trading conditions, it will automatically execute the trading process, which includes determining the buyer and seller, generating trading terms, confirming the trading price, etc. The smart contract ensures that the trading conditions, price, and information of both trading parties are clearly recorded on the blockchain, ensuring that all transactions can be traced and cannot be tampered with. Among them, the setting of smart contract trading conditions is specifically as shown in the formula:

[0055]

[0056] Among them, is the gene score of the i-th meat sheep, is the threshold of the gene score, is the breeding success rate of the i-th meat sheep, is the threshold of the breeding success rate.

[0057] Decentralized breeding sheep trading unit. This unit is based on blockchain technology and allows multiple participants to directly conduct breeding sheep transactions without intermediaries. All transaction information (such as the gene score, breeding history, health status, etc. of breeding sheep) will be recorded on the blockchain, ensuring the transparency of breeding sheep transactions and allowing the verification of the true data of each meat sheep, avoiding false information and fraud. Among them, the transaction execution is specifically as shown in the formula:

[0058] Among them, represents the gene score of the breeding sheep, represents the breeding history score of the breeding sheep, are the thresholds of the gene score and the breeding history score respectively.

[0059] In addition, the detailed information of all breeding sheep transactions will be recorded on the blockchain, including the gene score, breeding history, health data, transaction amount, transaction time, etc. of breeding sheep. These data are public, transparent, and cannot be tampered with, providing participants with the ability to trace data in all aspects. Although all transaction information can be publicly traced, privacy data such as personal information (such as the identity of the farm) will be encrypted to ensure data security.

[0060] The data processing and decision-making layer module includes a data processing unit and a health prediction unit; The data processing unit includes two functions: anomaly elimination and time compression; Anomaly rejection: An anomaly detection algorithm based on statistical methods is used to reject anomaly data that significantly deviates from the normal range from sensor and health monitoring data. Anomaly data includes equipment failures and error data.

[0061] Time compression: Since the amount of data generated by each meat sheep in large-scale farms is extremely large, a time window method is used to compress time series data, reducing the complexity of data processing. By dividing the time series data into several time windows and calculating the statistical features within the window to represent the data of that time period, the purpose of time compression is achieved. Specifically, as shown in the formula:

[0062]

[0063] Where, is the compressed data, and are the start and end times of the time window.

[0064] Health prediction unit: Based on the LSTM long short-term memory network, the health status of meat sheep is predicted to identify potential health problems. LSTM is suitable for processing time series data and can capture long-term dependencies in the data. Based on the historical data collected by sensors, the LSTM model can predict the health status of meat sheep in the future for a period of time and give early warnings of potential health problems. The prediction formula is specifically shown as follows:

[0065] Where, is the hidden state at the current moment, that is, the prediction result; and are weight matrices, connecting the hidden state of the previous moment and the current input respectively; is the bias term, is the input data at the current moment; is the activation function.

[0066] The above is only the preferred embodiment of the present invention and is not used to limit the present invention. For those skilled in the art, there are various changes and modifications to the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent construction system for rapid breeding of meat sheep, characterized in that, Including: A data acquisition module, which collects physiological data, behavioral data, and gene data of meat sheep, providing inputs for subsequent gene analysis, intelligent decision-making, and breeding execution modules; A gene analysis module, which conducts a detailed analysis of the gene data collected from the data acquisition module, and matches the genotypes related to meat production performance and lactation performance; An intelligent decision-making module, which generates an optimal breeding decision based on the gene scores and genotype data provided by the gene analysis module, as well as the physiological data, behavioral data, and environmental data provided by the data acquisition module; A breeding execution module, which executes specific mating operations, reproductive management, and health tracking of offspring according to the breeding plan generated by the intelligent decision-making module; A blockchain transaction module, which realizes the automation and transparency of breeding sheep transactions through smart contracts and decentralized transactions; A data processing and decision-making layer module, which preprocesses, analyzes, and conducts health prediction on the data obtained from the data acquisition module, gene analysis module, and intelligent decision-making module, and generates optimized decisions for the meat sheep population.

2. The intelligent construction system for rapid breeding of meat sheep according to claim 1, characterized in that: The data acquisition module includes a physical sign data acquisition unit and a gene data acquisition unit; The physical sign data acquisition unit uses a body temperature sensor, a heart rate sensor, and an acceleration sensor to collect the physical sign data of the mutton sheep in real time. The physical sign data includes body temperature, heart rate, and motion state. The behavior state of the mutton sheep is identified through the acceleration sensor, and the acquisition frequency of the physical sign data is adjusted according to the behavior state. The sampling period of the mutton sheep in the stationary state is longer than that in the motion state. It is set that the sampling period of the mutton sheep in the stationary state is , and the sampling period in the motion state is , specifically as shown in the formula: ； Among them, is the sampling frequency of the physical sign data, and are the sampling periods in the static state and the moving state respectively; The gene data acquisition unit collects the gene data of meat sheep through the "Wan Sheep No. 1 breeding chip", and extracts the gene information related to meat production performance and lactation performance. The gene information related to meat production performance includes fast growth genes and disease-resistant genes, and the gene information related to lactation performance includes milk yield genes and milk fat genes.

3. The intelligent construction system for rapid breeding of meat sheep according to claim 1, characterized in that: The data analysis module includes a gene data analysis unit and a gene scoring unit; The gene data analysis unit analyzes and identifies the gene information related to the growth and milk yield of meat sheep by receiving the gene data collected by the "Wan Sheep No. 1 breeding chip"; Compare the collected genotype data with the known excellent trait genotypes, evaluate the genetic potential of each meat sheep in specific traits, and the specific traits include meat production performance and lactation performance; determine the association between genotypes and target traits through genome-wide association analysis, where genome-wide association analysis is used to identify the gene information significantly related to specific traits of meat sheep; The genotype refers to the genetic composition of an individual at specific gene loci, and the gene information refers to these gene loci and their effects on specific traits; The model is specifically as shown in the formula: ； Among them, is the phenotypic data of the \(i\)th individual, is the genotype data of the \(i\)th individual for the \(j\)th gene information, is the effect value of the \(j\)th gene information, indicating the impact of this gene on the phenotype; is the constant term, representing the basic relationship between genotype and phenotype; is the error term, representing the impact of other unconsidered factors on the phenotype; is the total number of gene information; The gene scoring unit evaluates the genotype of each meat sheep, generates a gene score, indicating its genetic potential in specific traits. The gene score will provide key genetic information for the intelligent decision-making module to support the optimization of the breeding plan. The gene score is generated by analyzing the association between the genotype data of each meat sheep and excellent traits, and is specifically as shown in the formula: ； Among them, is the gene score of the i-th meat sheep, indicating its genetic potential for the target trait, is the genotype data of the i-th meat sheep for the j-th gene information, is the degree of influence of the j-th gene information on the target trait, is the total number of gene information.

4. The intelligent construction system for rapid breeding of meat sheep according to claim 1, characterized in that: The intelligent decision-making module includes a multi-objective decision-making unit and a genetic algorithm decision-making unit; The multi-objective decision-making unit uses the GECO model to generate a breeding plan by comprehensively considering gene data, environmental data, and economic data. The objective function is designed as a multi-objective optimization problem, and multiple objectives will jointly affect the breeding decision. The multiple objectives include growth traits, reproductive traits, environmental adaptability, and economic benefits. The optimization formula of the GECO model is specifically as shown in the formula: ； Among them, represents the genetic goal, represents the environmental adaptability goal, represents the economic benefit goal, is a decision variable, representing the genotype, breeding combination and feeding strategy of meat sheep; The genetic algorithm decision-making unit is based on the genetic algorithm and is used to generate the optimal solution in the multi-objective optimization process. The genetic algorithm searches for the optimal solution in a large-scale solution space through selection, crossover, and mutation operations, and finally selects those solutions that can achieve the best balance among multiple objectives. According to the gene scores and genotype data provided by the gene analysis module, combined with the environmental data and market data provided by the data collection module, comprehensively evaluate the biological benefits and economic benefits of each breeding plan, and generate personalized breeding suggestions for each mutton sheep according to the evaluation results, including breeding combinations, feeding strategies, and adjustment strategies during the breeding period; The genetic algorithm generates the optimal breeding plan through the following steps: Initialize the population, create the initial population, and each individual in the population represents a possible breeding decision; Fitness evaluation, calculate the fitness of each individual. The fitness function is usually the weighted sum of each objective function, representing the comprehensive performance under different objectives; Selection operation, select individuals with high fitness to enter the next generation; Crossover and mutation, generate new candidate solutions through crossover and mutation operations. The crossover operation combines the information of the parent solutions to generate new offspring solutions, and the mutation operation introduces random changes to ensure the diversity of solutions; Iterative update, through multiple generations of evolution, the system gradually approaches the optimal solution through selection, crossover, and mutation operations; The specific form of the fitness function is as shown in the formula: ； Among them, is a multi-objective function, is the weight of each objective function.

5. The intelligent construction system for rapid breeding of mutton sheep according to claim 1, characterized in that: The breeding execution module includes a breeding management unit and a cycle monitoring unit; The breeding management unit, after the optimal breeding plan of the intelligent decision-making module, automatically selects suitable breeding sheep for breeding. The decision-making plan is usually based on gene scores, breeding history, and market demand, and completes the breeding task through automated equipment, selects suitable breeding sheep combinations for breeding and reproduction, specifically as shown in the formula: ； Among them, is the gene score of breeding sheep, is the breeding history of breeding sheep, is the score of market demand, are the weights of various factors; The cycle monitoring unit monitors the pregnancy status of ewes, the fetal development status, and the health status of mutton sheep in real time through sensors. By monitoring the growth and health status of the offspring, through sensor data and gene data, it tracks the healthy development of the offspring in real time and predicts possible health problems. According to the health status and growth performance of the offspring, it automatically adjusts the subsequent breeding plan and optimizes the selected breeding sheep and breeding strategies.

6. The intelligent construction system for rapid breeding of mutton sheep according to claim 1, characterized in that: The blockchain trading module includes an intelligent contract execution unit and a decentralized breeding sheep trading unit; The intelligent contract execution unit sets trading conditions according to the breeding plan generated by the breeding execution module and the intelligent decision-making module, and automatically completes the trading of breeding sheep when the conditions are met. It determines which breeding sheep can be traded for breeding based on gene scoring and breeding history factors. When the intelligent contract detects breeding sheep that meet the trading conditions, it automatically executes the trading process; The trading process includes determining the buyer and seller, generating trading terms, and confirming the trading price. The setting of the intelligent contract trading conditions is specifically shown as follows: ； ； Among them, is the gene score of the i-th meat sheep, is the threshold of the gene score, is the reproductive success rate of the i-th meat sheep, is the threshold of the reproductive success rate; The decentralized breeding sheep trading unit, based on blockchain technology, allows multiple participants to directly conduct breeding sheep trading without intermediaries. All trading information is recorded on the blockchain, and the true data of each meat sheep can be verified. The trading execution is specifically shown as follows: ； Among them, represents the gene score of the breeding sheep, represents the breeding history score of the breeding sheep, are the thresholds of the gene score and the breeding history score respectively.

7. The intelligent construction system for rapid breeding of meat sheep according to claim 1, characterized in that: The data processing and decision-making layer module includes a data processing unit and a health prediction unit; The data processing unit includes two functions: anomaly rejection and time compression; For anomaly rejection, an anomaly detection algorithm based on statistical methods is used to remove abnormal data that significantly deviates from the normal range from sensor and health monitoring data. Abnormal data includes equipment failures and error data; For time compression, based on the huge amount of data generated by each meat sheep in large-scale farms, a time window method is used to compress time series data. By dividing the time series data into several time windows and calculating the statistical features within the window to represent the data of that time period, the purpose of time compression is achieved, specifically shown as follows: ； ； Among them, is the compressed data, and are the start and end times of the time window; The health prediction unit predicts the health status of meat sheep through an LSTM long short-term memory network, identifies potential health problems, and predicts the health status of meat sheep in the future for a period of time based on historical data collected by sensors, and gives early warnings of potential health problems. The prediction formula is specifically shown as follows: ； Among them, is the hidden state at the current moment, that is, the prediction result; and are weight matrices, connecting the hidden state at the previous moment and the current input respectively; is the bias term, is the input data at the current moment; is the activation function.

Images