Intelligent beef cattle breeding monitoring management system based on whole industry chain data

Abstract

The invention discloses an intelligent beef cattle breeding monitoring management system based on whole industry chain data. The system comprises a data acquisition module which is used for deploying sensors and data acquisition terminals in each link of a beef cattle whole industry chain for data acquisition; the data transmission and integration module is used for transmitting data acquired in each link to a cloud data center for processing by using a wireless communication technology; the intelligent analysis module is used for mining the integrated data and adjusting breeding processes such as seed selection and matching, a nutrition formula and a management strategy; the monitoring and early warning module is used for setting a key index threshold value, and when the key index exceeds the threshold value, the system automatically sends out early warning information to a mobile phone terminal or a computer client of a culture worker; and the breeding decision support module is used for providing guidance for the breeding direction according to long-term accumulated data and analysis results and screening out good beef cattle individuals and families thereof. The system has the advantages of high breeding efficiency, high beef cattle variety improvement speed and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of beef cattle breeding monitoring systems, and in particular to an intelligent beef cattle breeding monitoring and management system based on the whole industrial chain data. Background Art

[0002] In the beef cattle breeding industry, traditional breeding management methods often rely on the experience of farmers and simple recording means, lacking systematicness and accuracy. In the whole industrial chain links from the reproduction, growth, feeding of beef cattle to the final slaughter and processing, the data of each link exist in isolation and have not been effectively integrated and utilized. For example, the reproductive performance data of breeding cattle, the growth and development data of calves, the feed conversion data of fattening cattle, and the beef quality detection data are scattered in different farms, veterinary stations, slaughterhouses, etc., making it difficult to form a coherent information flow and unable to provide comprehensive and in-depth support for breeding decisions. This leads to low breeding efficiency, slow improvement of beef cattle breeds, and difficulty in meeting the growing market demand. There is an urgent need for an innovative intelligent management method to break the current situation. Summary of the Invention

[0003] The technical problem to be solved by the present invention is how to provide an intelligent beef cattle breeding monitoring and management system based on the whole industrial chain data with comprehensive data integration and utilization, accurate analysis and decision-making, and fast improvement of beef cattle breeds.

[0004] To solve the above technical problems, the technical solution adopted by the present invention is: an intelligent beef cattle breeding monitoring and management system based on the whole industrial chain data, including:

[0005] Data acquisition module: used to deploy sensors and data acquisition terminals at each link of the whole industrial chain of beef cattle for data acquisition;

[0006] Data transmission and integration module: used to utilize wireless communication technology to transmit the data collected at each link to the cloud data center for processing;

[0007] Intelligent analysis module: used to mine the integrated data through big data analysis and artificial intelligence algorithms, and adjust the breeding selection, nutrition formula and management and feeding strategies;

[0008] Monitoring and early warning module: used to set key index thresholds. When the key index exceeds the threshold, the system automatically sends a warning message to the mobile phone or computer client of the breeding personnel, and takes intervention measures according to the alarm information;

[0009] Breeding decision support module: used to provide guidance for the breeding direction according to the long-term accumulated data and analysis results, and screen out excellent beef cattle individuals and their families.

[0010] Furthermore, the data collected by the data acquisition module includes:

[0011] In the breeding stage, record the time nodes, body temperature and behavioral characteristics of the breeding cattle's estrus, breeding and conception; in the growth stage, monitor the weight, body size and daily weight gain data of beef cattle; in the feeding stage, collect the types of feed, feeding amount, remaining amount and feed energy efficiency parameters, and combine the environmental temperature and humidity, oxygen concentration and ammonia concentration information; in the slaughter and processing stage, measure the marbling score, tenderness and pH value of beef, and at the same time, enter the veterinary diagnosis records and vaccination status.

[0012] A further technical solution lies in that in the data transmission and integration module: in the cloud data center, through a preset data cleaning algorithm, outlier data and duplicate data are removed, data in different formats and sources are standardized, and multi-dimensional data such as breeding, growth, feeding, slaughter and processing are associated with the beef cattle individual number as the primary key to construct a complete beef cattle whole industry chain database.

[0013] A further technical solution lies in that the intelligent analysis module: based on a machine learning model, predicts the conception probability and calving interval of breeding cattle according to historical breeding data, and plans breeding; according to the growth curve model, combines the feeding conditions, adjusts the feed formula in real time, and optimizes the growth rate; through correlation analysis of meat quality indicators and feeding process data, finds the key factors to improve beef quality, and provides a basis for adjusting feeding strategies.

[0014] A further technical solution lies in that in the monitoring and warning module: when the daily weight gain of calves is lower than the normal range, the disease symptom indicators of fattening cattle are abnormal, and the quality of slaughtered beef does not meet the standards, the system automatically sends a warning message to the mobile phone or computer client of the breeding personnel, and adjusts the feeding and / or treatment plan according to different indicators to ensure the healthy growth and high-quality output of beef cattle.

[0015] A further technical solution lies in that the breeding decision support module: is used to provide guidance for the breeding direction according to the long-term accumulated data and analysis results, screen out beef cattle individuals and their families with excellent reproductive performance, fast growth and good meat quality, formulate a breeding selection and mating plan, and gradually optimize the beef cattle breed gene pool.

[0016] The beneficial effects of adopting the above technical solutions are as follows: 1) In this application, sensors and acquisition terminals are deployed throughout the whole industry chain from breeding, growth, feeding to slaughter and processing, comprehensively collecting the details of breeding cattle's estrus and conception, real-time indicators of beef cattle growth, feeding environment and feeding information, as well as accurate evaluation results of beef quality, and wirelessly transmitting them to the cloud for integration. This makes the data no longer fragmented, but constructs a complete database with the beef cattle individual number as the core, providing unprecedented depth and breadth support for breeding decisions, and making the data integration and utilization more comprehensive.

[0017] 2) This application introduces big data analysis and artificial intelligence algorithms. Based on machine learning models, it deeply mines historical breeding data, and the accuracy of predicting the conception probability of breeding cattle and the calving interval far exceeds manual estimation, making breeding more planned; according to the growth curve model combined with real-time feeding conditions, it dynamically adjusts the feed formula to ensure that the growth rate of beef cattle is always optimized, avoiding growth retardation or feed waste; by correlating meat quality and feeding data, it accurately locates the key factors such as specific feed ingredients that improve tenderness, and the feeding strategy adjustment is targeted, greatly enhancing the beef quality control ability and pushing breeding decisions towards scientific and precise leadership.

[0018] 3) This application sets key indicator thresholds and the system automatically monitors. Once the daily weight gain of calves deviates from normal, the fattening cattle show disease or physiological abnormalities, or the quality of slaughtered beef does not meet the standards, it immediately pushes warnings to the mobile phones or computer terminals of the breeding personnel. The information is detailed and accurate, including key contents such as the problem individuals and abnormal indicator values, enabling the breeding personnel to know and take corresponding feeding or treatment plans in the first time, greatly reducing the breeding risk and ensuring the healthy growth and high-quality output of beef cattle.

[0019] 4) Based on the data accumulated in the long term, this application screens beef cattle individuals and their pedigrees with excellent breeding, growth, and meat quality characteristics, formulates a rigorous breeding and mating plan according to genetic principles, and at the same time uses gene marker-assisted selection to continuously optimize the breed gene pool, improving the quality and performance of beef cattle generation by generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention will be further described in detail below in conjunction with the drawings and specific embodiments.

[0021] Figure 1 is the principle block diagram of the system according to the embodiment of the present invention;

[0022] Figure 2 is the principle block diagram of the data acquisition module in the system according to the embodiment of the present invention;

[0023] Figure 3 is the principle block diagram of the data transmission and integration module in the system according to the embodiment of the present invention;

[0024] Figure 4 is the principle block diagram of the intelligent analysis module in the system according to the embodiment of the present invention;

[0025] Figure 5 is the principle block diagram of the monitoring and warning module in the system according to the embodiment of the present invention;

[0026] Figure 6 is the principle block diagram of the breeding decision support module in the system according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] Combined with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0028] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0029] Generally, as Figure 1 shown, the embodiments of the present invention disclose an intelligent beef cattle breeding monitoring and management system based on the whole industrial chain data, including:

[0030] Data acquisition module 101: used to deploy sensors and data acquisition terminals at each link of the whole industrial chain of beef cattle to collect data;

[0031] Data transmission and integration module 102: used to utilize wireless communication technology to transmit the data collected at each link to the cloud data center for processing;

[0032] Intelligent analysis module 103: used to mine the integrated data through big data analysis and artificial intelligence algorithms, and adjust the breeding selection, nutritional formula and management feeding strategies;

[0033] Monitoring and warning module 104: used to set the threshold of key indicators. When the key indicators exceed the threshold, the system automatically sends a warning message to the mobile phone or computer client of the breeding personnel, and takes intervention measures according to the alarm information;

[0034] Breeding decision support module 105: used to provide guidance for the breeding direction according to the long-term accumulated data and analysis results, and screen out excellent beef cattle individuals and their families.

[0035] Furthermore, the data collected by the data acquisition module includes:

[0036] In the breeding link, record the time nodes, body temperature and behavioral characteristics of the breeding cattle's estrus, mating and conception; in the growth link, monitor the weight, body size and daily weight gain data of the beef cattle; in the feeding link, collect the feed type, feeding amount and remaining amount, as well as the feed energy efficiency parameters, and combine the environmental temperature, humidity, oxygen concentration and ammonia concentration information; in the slaughter and processing link, measure the marble score, tenderness and pH value of the beef, and at the same time, record the veterinary diagnosis records and vaccine inoculation conditions.

[0037] Further, as Figure 2 shown, the data acquisition module 101 includes:

[0038] 1) Reproduction link data acquisition unit 1011:

[0039] Install high-definition cameras in the cattle shed, combine intelligent image recognition algorithms, and monitor the behavior of breeding cattle all-weather. Through in-depth learning and analysis of behavioral characteristics such as the activity frequency of breeding cattle and the interaction patterns with other cattle, detect signs of estrus. Once the onset of estrus is determined, the system immediately records this time node; at the same time, wear a biosensor collar on the breeding cattle. The collar collects body temperature data every 5 - 15 minutes through an integrated high-precision body temperature sensor and wirelessly transmits it to the data acquisition terminal. The collar also has a minimally invasive hormone level monitoring module that regularly collects changes in the levels of estrogen and progesterone in the blood through a microprobe in contact with the cattle's body surface and uploads it in real time; in the breeding link, the breeding personnel use a handheld terminal device to enter the breeding time and the breeding bull number information into the management system while completing the breeding operation; combined with the regular ultrasound examination results of veterinarians and the subsequent monitoring of the physiological characteristics of the breeding cattle, comprehensively determine the time node of successful conception and also enter it into the system for subsequent analysis. Through the above steps, the time nodes of estrus, breeding, and conception of breeding cattle, as well as the body temperature and hormone levels, can be accurately recorded.

[0040] 2) Growth link data acquisition unit 1012:

[0041] Install high-precision electronic scales at the entrances and exits of the growing and fattening cattle sheds. Each time the beef cattle enter and leave for feeding and drinking, the weight measurement is automatically triggered, and the data is immediately transmitted to the data acquisition terminal; multiple groups of laser measuring instruments are reasonably arranged at intervals in the cattle shed. When the beef cattle stand in the measurement area of the laser measuring instrument, measure their body length, body height, and chest circumference data. The measurement frequency is once a day to obtain the changes in the growth and development body shape of the beef cattle; the daily weight gain data is automatically generated by the system through the difference calculation of the daily weight data, and combined with the corresponding date information, the growth rate trend of the beef cattle is presented;

[0042] 3) Feeding link data acquisition unit 1013:

[0043] The feed storage bin and feeding equipment are equipped with intelligent metering sensors. Before feeding the cattle, the system automatically records the type of feed and the initial feeding amount for this feeding. After the feeding is completed, by comparing the remaining feed amount, the actual feeding situation of the beef cattle is calculated to obtain the type of feed, the feeding amount, and the remaining amount. There are several multi-functional environmental sensing nodes distributed in the cattle shed. The environmental sensing nodes integrate temperature and humidity sensors, oxygen sensors, and ammonia sensors, and collect environmental data every 15 - 20 minutes to obtain the environmental temperature and humidity, oxygen concentration, and ammonia concentration information in the cattle shed. Once the environmental indicators deviate from the appropriate range, the system immediately triggers an alarm to remind the breeding personnel to adjust the operating status of equipment such as ventilation, cooling, and dehumidification in a timely manner to ensure that the beef cattle are in a comfortable feeding environment.

[0044] 4) Data acquisition unit 1014 for the slaughter and processing link:

[0045] Professional meat quality detection equipment and standardized processes are used to measure the marbling score, tenderness, and pH value of beef. A meat quality detection workstation is set up beside the production line in the slaughter workshop. When the beef carcass passes through a specific station, a near-infrared spectroscopy analyzer is used to scan the surface of the beef, and based on the spectral feature analysis, the marbling score is quickly given. With the help of a professional tenderness tester, a standard-sized meat sample is cut from the beef sample, and the shear force of the meat sample is measured by simulating the chewing process of the instrument to determine the tenderness index. The sampling detection ratio of each batch of beef is not less than 10%. A high-precision pH meter is used to measure the beef at multiple points 45 minutes after slaughter and 24 hours after acidification at 0 - 4°C, and the average value is taken as the pH value data of this batch of beef. The staff in the slaughterhouse retrieves the diagnosis records and vaccine inoculation status of the veterinarian during the whole process of beef cattle breeding from the cloud database according to the individual number of the beef cattle to complete the beef quality traceability chain.

[0046] Furthermore, in the data transmission and integration module: In the cloud data center, through a preset data cleaning algorithm, outlier data and duplicate data are removed, and data in different formats and sources are standardized. Using the individual number of the beef cattle as the primary key, multi-dimensional data such as breeding, growth, feeding, and slaughter and processing are associated to construct a complete beef cattle full-industry chain database.

[0047] Furthermore, as Figure 3 shown, the data transmission and integration module 102 includes:

[0048] 1) Outlier data removal unit 1021: For continuous data such as beef cattle weight, daily weight gain, environmental temperature and humidity, etc., calculate its mean value μ and standard deviation σ, set the threshold range, and use μ ± 3σ as the judgment basis to determine the data outside this range as outliers; for example, if the average daily weight gain of a beef cattle group is 1.5 kg and the standard deviation is 0.2 kg within a certain period, then the data points with a daily weight gain less than 0.9 kg or greater than 2.1 kg will be marked as outliers. For discrete data such as estrus cycle days, number of vaccine inoculations, etc., determine the value range based on historical data and industry experience, and consider the data outside this range as outliers; for data that changes dynamically over time, such as the fluctuation of breeding cattle body temperature within a day, use the time series analysis method to construct an autoregressive moving average ARIMA model to predict the expected value range at each time point. If the actual collected value deviates from the predicted interval by more than the set threshold, it is considered abnormal data. For example, if the normal body temperature of breeding cattle in the early morning should be between 37.5 - 38.5 °C, and the body temperature measured at a certain early morning is 40 °C and the model predicts that the normal interval of body temperature at this time does not include this value, it can be determined as abnormal; conduct joint analysis on related variables. For example, in the feeding process, if the feed intake increases significantly during a certain period, but the weight of the beef cattle does not increase but decreases instead, combined with environmental data (such as too high ammonia concentration may affect the appetite and health of beef cattle), analyze the logical relationship between related variables, and judge the existing abnormal data according to the corresponding logical relationship.

[0049] 2) Duplicate data processing unit 1022: Based on the beef cattle individual number, compare all uploaded data records. When two or more data records have exactly the same beef cattle individual number, data collection time, and data type, they are determined as duplicate data. Keep one of the records and delete the remaining duplicates. For example, if the system receives two identical numerical data marked as weight measurement from the same beef cattle individual number within the same second, remove the redundant one; for the situation where partial information is not completely consistent due to data collection equipment failure or human input error but actually refers to the same data, use the fuzzy matching algorithm to identify and merge suspected duplicate data by setting a tolerance range (such as a time deviation within 5 minutes and a measurement value deviation within the allowable error range of the instrument accuracy) to ensure the uniqueness of the data.

[0050] 3) Data standardization processing unit 1023: It is used to formulate a unified specification for the format differences of data from different sources. For text data, a standard name library is established to make the data consistent in expression; for data involving physical quantities, dimensional conversion is performed to unify the dimensions of different types of physical quantities; for categorical data, such as beef cattle breeds, genders, etc., a coding system is adopted. Common beef cattle breeds such as Simmental cattle are coded as 01, Charolais cattle are coded as 02, etc.; male cattle are coded as 1, and female cattle are coded as 2. The text description is replaced by digital coding, which not only saves storage space but also facilitates rapid computer processing and data analysis.

[0051] 4) Multi-dimensional data association unit 1024: It is used to set the beef cattle individual number as the unique primary key during the database design stage and construct an association relationship around this primary key; when new data is uploaded, the system first identifies the beef cattle to which it belongs based on the individual number, and then associates the corresponding data records in the breeding, growth, feeding, and slaughter and processing links with it. When receiving the slaughter and processing data of a certain beef cattle, which includes the beef marbling score, the previous growth records, feeding details, and breeding information of this beef cattle are found through the individual number, realizing the seamless connection of data in the entire beef cattle industry chain; multiple association tables are created, such as the "breeding - growth association table", which records the corresponding relationship between the key time nodes and growth indicators of each beef cattle from conception to each growth stage; the "feeding - slaughter association table", which reflects the influence path of feed factors on the final beef quality. Through these association tables, using the JOIN operation of the database, multi-dimensional data can be quickly integrated during query analysis.

[0052] Furthermore, the intelligent analysis module: Based on the machine learning model, it predicts the conception probability and calving interval of breeding cattle according to historical breeding data and plans breeding; according to the growth curve model, combined with feeding conditions, it adjusts the feed formula in real time to optimize the growth rate; through correlation analysis of meat quality indicators and feeding process data, it finds the key factors to improve beef quality and provides a basis for adjusting feeding strategies.

[0053] Furthermore, as Figure 4 shown, the intelligent analysis module 103 includes:

[0054] 1) Breeding analysis and breeding planning unit 1031:

[0055] First, collect a large amount of pre - processed historical breeding data, including beef cattle breed, age, past estrus cycles, breeding times, pregnancy status, and calving history information. Use a logistic regression model for pregnancy probability prediction and a random forest regression model for calving interval prediction. Divide the pre - processed historical breeding data into a training set and a test set according to a certain ratio, and input them into the selected models for training. By continuously adjusting the hyperparameters of the models and using the cross - validation method, make the models achieve high accuracy and stability on the test set.

[0056] When predicting the pregnancy probability of a certain beef cattle, process the current characteristic data of this beef cattle according to the set coding and standardization method, and input it into the trained logistic regression model. The model outputs a probability value between 0 and 1, indicating the pregnancy possibility of this beef cattle for this breeding. For calving interval prediction, input the breeding history and current status - related data of the beef cattle into the random forest regression model, and the model outputs the expected number of days of the calving interval.

[0057] According to the pregnancy probability prediction result, set a threshold. For example, when the pregnancy probability is higher than 60%, give priority to arranging this beef cattle for breeding. If it is lower than the threshold, further analyze the reasons, such as whether there are health problems, poor feeding environment, etc., make corresponding adjustments first and then consider the breeding timing. Combine the calving interval prediction to reasonably plan the breeding cycle of beef cattle, avoid overly dense or sparse breeding arrangements, so as to ensure the breeding efficiency and health status of the beef cattle population and achieve scientific breeding planning.

[0058] 2) Feed formula adjustment unit 1032:

[0059] Growth curve model construction: Collect historical growth data of beef cattle of different breeds, genders, and age groups, including daily weight and body measurement change information. Use the non - linear regression method to construct the Gompertz growth curve model to fit the changing trend of the growth rate during the growth process of beef cattle. Through fitting and training a large amount of historical data, determine the parameters in the model.

[0060] Feeding condition data fusion: Real - time collect the current feeding condition data of beef cattle, including feed type, feeding amount, environmental temperature and humidity, and ammonia concentration. Combine these data with the Gompertz growth curve model, analyze the influence direction and degree of the current feeding conditions on the growth rate of beef cattle, and compare the growth indicators that beef cattle should reach according to the normal growth curve with the actual measured values under the current feeding conditions through the model.

[0061] Feed formula adjustment: When the actual growth indicators of beef cattle deviate from the normal growth curve, the feed formula adjustment mechanism is activated; using the linear programming algorithm, with the goal of meeting the nutritional needs of beef cattle, reducing feed costs, and increasing growth rate, dynamically adjust the proportions of protein, energy, vitamins, and minerals in the feed. For example, if the growth of beef cattle is slow and it is found through analysis that the protein intake is insufficient, appropriately increase the proportion of high-protein feed raw materials such as soybean meal, and at the same time reduce the usage of some low-nutrient, high-fiber feeds. By dynamically adjusting the feed formula in real time, promote the growth of beef cattle to return to the ideal growth curve and achieve the optimization of growth rate.

[0062] 3) Key factor mining unit 1033 of beef quality:

[0063] Data collection and collation: Integrate the beef quality data in the slaughter and processing links, including marbling score, tenderness, and pH value data, as well as the corresponding feeding process data, including feed feeding records, disease prevention and control situations, and environmental data throughout the growth process of beef cattle;

[0064] Association analysis application: Use the Pearson correlation coefficient to analyze the linear correlation between continuous variables, and through ANOVA (Analysis of Variance), analyze the differences in the pH values of beef among different breeds of beef cattle under the same feeding conditions; Use the Apriori algorithm in data mining to mine frequent item sets and find the combinations of feeding factors that frequently appear in cases of high-quality beef production;

[0065] Key factor determination and feeding strategy adjustment: According to the results of association analysis, determine the key factors that have a significant impact on beef quality; For example, it is found that adding a certain proportion of linseed meal to the feed can significantly improve the marbling score of beef, or maintaining the environmental temperature at 15 - 20°C and the ammonia concentration below 10 ppm during the late fattening period of beef cattle helps to improve tenderness. Incorporate these key factors into the feeding strategy, formulate targeted feeding plans, such as adjusting the feed formula at specific growth stages and optimizing the environmental control parameters of the cattle shed, provide a scientific basis for improving beef quality, and achieve high-quality production of beef from the feeding source.

[0066] Furthermore, in the monitoring and warning module 104: When the daily weight gain of calves is lower than the normal range, the disease symptom indicators of fattening cattle are abnormal, and the quality of slaughtered beef does not meet the standards, the system automatically sends a warning message to the mobile phone or computer client of the breeding personnel, and adjusts the feeding and / or treatment plan according to different indicators to ensure the healthy growth and high-quality production of beef cattle.

[0067] Furthermore, as Figure 5 shown, the monitoring and warning module 104 includes:

[0068] 1) Warning trigger unit 1041

[0069] Calf daily weight gain warning: Based on historical data and industry standards, combined with the growth characteristics of calves of different breeds, genders, and feeding stages, a dynamic normal range model of daily weight gain is constructed; the actual weight data of calves is obtained through the data collection module every day, the daily weight gain is automatically calculated, and it is compared with the normal range in the corresponding model in real time. When it is detected that the daily weight gain of a certain calf is lower than the set lower limit value for 3 consecutive days, a warning signal is triggered;

[0070] Fattening cattle disease symptom index warning: In the feeding process, the data collection terminal continuously collects the body temperature, respiratory rate, and feeding behavior data of fattening cattle, constructs a disease prediction model through machine learning algorithms, and identifies the abnormal change patterns of each index under different disease states by learning a large amount of historical data of healthy and diseased fattening cattle. When the body temperature of a certain fattening cattle is higher than 40°C for 2 consecutive hours, the respiratory rate exceeds the normal upper limit by 20%, and the feed intake is reduced by more than 50% compared with the average feed intake in the past 7 days, it is determined that the fattening cattle has abnormal disease symptom indicators and a warning is issued;

[0071] Slaughter beef quality warning: In the slaughter and processing process, when the marbling score, tenderness, and pH value of beef are measured by the detection equipment, the system scores according to the preset quality standards. Taking the marbling score as an example, if the score of a certain batch of beef is lower than the industry high-quality standard (such as 3 points, with a full score of 5 points), or the tenderness index exceeds the upper limit of the appropriate shear force (such as 5 kgf), or the pH value deviates from the normal range (the pH value within 45 minutes after slaughter is not between 5.8 - 6.2), corresponding quality non-compliance warnings will be triggered.

[0072] 2) Warning information push unit 1042:

[0073] Using the pre-integrated communication interface, the warning information is sent to the mobile phone or computer client of the breeding personnel by means of text messages, instant messaging software, or system pop-ups; the content of the warning information includes: for the calf daily weight gain warning, it shows the individual number of the calf with problems, the location of the cattle shed, the current daily weight gain value, and the number of consecutive days below the normal range, and at the same time attaches the reference range of the normal daily weight gain of this breed of calf; for the fattening cattle disease warning information, it shows the number of the diseased fattening cattle, the current abnormal physiological index values, and preliminarily indicates the possible disease types involved to assist the breeding personnel in initially judging the condition; for the slaughter beef quality warning, it lists the batch number of the non-compliant beef, the specific quality index values, and the difference from the standard, such as "the marbling score of a certain batch of beef is 2 points, 1 point lower than the high-quality standard", so that the breeding personnel can clearly know the problem.

[0074] 3) Feeding and / or treatment plan adjustment unit 1043

[0075] Treatment of calf daily weight gain: After the system gives an early warning, first analyze the automatically generated detailed feeding analysis report. If it is found through analysis that the problem is caused by insufficient feed nutrition, the system recommends adjusting the feed formula. If it is found through analysis that the problem is caused by environmental factors, the system prompts to adjust the environmental control parameters of the cattle shed, turn on the heating equipment, strengthen ventilation and dehumidification, control the temperature at 15-20 °C, and maintain the relative humidity at 60%-70%. Continuously monitor the subsequent growth of the calves until the daily weight gain returns to normal;

[0076] Treatment of fattening cattle diseases: The system, based on the disease prompts in the early warning and in combination with the remote guidance of veterinarians, conducts preliminary isolation and observation on the diseased fattening cattle; if it is determined to be a common disease, the system, according to the preset treatment plan library, gives specific medication suggestions, including the name of the drug, dosage, and medication frequency; at the same time, adjust the feeding plan for the diseased cattle, provide easily digestible and high-nutrition feed, such as silage combined with a small amount of concentrate, to ensure the physical recovery of the diseased cattle. After the condition improves, resume the regular feeding, and continuously monitor various physiological indicators until they return to normal; if it is determined to be a rare disease, give a treatment plan through the remote guidance of veterinarians;

[0077] Improvement of slaughter beef quality: In response to the situation where the beef quality does not meet the standards, the system traces back the data of the entire feeding process of this batch of beef cattle, analyzes the root cause of the problem. If the feed formulation is unreasonable in the later stage of feeding, such as excessive energy and lack of minerals affecting the formation of marbling, when feeding the next batch of beef cattle, adjust the feed formula, optimize the addition amount of minerals such as calcium and phosphorus, and add them in a specific ratio (such as calcium: phosphorus = 1.5:1) during the fattening stage; if the pH value is abnormal due to stress before slaughter, improve the management of transportation and waiting-for-slaughter links, shorten the transportation time, optimize the waiting-for-slaughter environment, reduce the stress response of beef cattle, and ensure the subsequent beef quality meets the standards through continuous improvement of the feeding and slaughter processes.

[0078] Furthermore, the breeding decision support module: is used to provide guidance for the breeding direction based on the long-term accumulated data and analysis results, screen out beef cattle individuals and their families with excellent reproductive performance, fast growth, and good meat quality, formulate a breeding and mating plan, and gradually optimize the gene pool of beef cattle breeds.

[0079] Furthermore, as Figure 6 shown, the breeding decision support module 105 includes:

[0080] 1) Data accumulation and analysis integration unit 1051:

[0081] Construct a comprehensive database: Collect and store data from all links of the beef cattle full industry chain, including pedigrees of breeding cattle, conception success rates, calving intervals, and birth weights of calves in the breeding link; weight, body size change curves, and daily weight gain rates at different growth stages in the growth link; feed conversion rates and growth performance under different feed formulations in the feeding link; scores of various beef quality indicators in the slaughter and processing link; taking individual beef cattle as the basic unit, deeply associate and integrate these multi-dimensional data to form a full life cycle database of individual beef cattle;

[0082] Genetic evaluation and analysis: Through animal genetic evaluation algorithms, estimate genetic parameters for the accumulated data. By combining pedigree information with reproductive performance data, calculate the heritability and repeatability parameters of breeding cattle to measure the degree to which traits are affected by genetic factors; for growth rate, analyze the genetic variation of beef cattle in different families under similar feeding conditions to identify families with significant genetic advantages; for meat quality traits, according to genetic correlations, find gene markers or genetic background information closely associated with high-quality meat.

[0083] 2) Excellent individual and family screening unit 1052:

[0084] Formulate a screening index system: Based on market demand, breeding benefits, and industry development trends, establish a screening index system; in terms of reproductive performance, set a conception rate threshold (such as above 80%), a calving interval standard (average not exceeding 380 days), and an ideal range of calf birth weights (depending on the breed, such as 40 - 45 kg for Simmental calves); in terms of growth performance, set daily weight gain targets and body size development targets during the feeding period; in terms of meat quality, set grade standards based on market grading requirements for high-quality beef, such as the shear force of beef tenderness being lower than the threshold (such as 4 kgf), and the marbling score reaching the premium level (4 points and above).

[0085] Comprehensive multi-trait evaluation: Based on the screening index system, comprehensively evaluate individual beef cattle and their families through the multi-trait selection index method, and assign weights to each trait. For example, for a breeding direction focusing on meat production performance, the weights of growth rate and meat quality traits can be relatively high, set at 0.4 and 0.35 respectively, and the weight of reproductive performance is set at 0.25. By calculating the comprehensive selection index of each individual and family, sort them according to the index level, and screen out beef cattle individuals and their affiliated families with excellent comprehensive performance and high rankings;

[0086] 3) Breeding and mating plan formulation unit 1053

[0087] Selection strategy: According to the screening results, priority is given to selecting breeding cattle with a high comprehensive selection index as the core breeding source; for young breeding cattle with great genetic potential, even if some of the current trait data are not yet fully mature, but if their pedigree genetic advantages are prominent, combined with the prediction of growth and development trends, they are also included in the key breeding scope; at the same time, the breeding cattle population is updated regularly, and individuals with declining reproductive performance, slowing growth rate, and deteriorating meat quality as they age are eliminated.

[0088] Mating plan: Based on the theory of heterosis and the principle of avoiding inbreeding depression, a mating plan is formulated for the selected breeding cattle; through pedigree analysis, the coefficient of relationship is controlled within 0.25, and for the improvement requirements of different traits, if it is desired to improve the marbling score of beef, male and female cattle with complementary advantages in this trait are selected for mating, that is, the bull has a high-scoring genetic background and the cow performs well in other aspects such as growth rate. Through targeted mating, the optimization and improvement of offspring in multiple key traits are achieved.

[0089] 4) Gene bank optimization unit 1054

[0090] Gene marker-assisted selection: Using gene chips, gene markers closely associated with excellent traits of beef cattle are detected. During the selection and mating process, the gene marker information is included in the reference scope, and individuals carrying favorable gene marker combinations are preferentially selected to accelerate the diffusion frequency of excellent genes in the population and optimize the gene frequency distribution of the beef cattle breed gene bank.

[0091] Continuous monitoring and feedback adjustment: Establish a breeding effect monitoring mechanism to track the growth and development, reproductive performance, and meat quality performance of the offspring of selection and mating, and compare and analyze the actual data with the expected goals; according to the feedback results, adjust the screening index system, selection and mating strategies, and the focus of gene marker selection in a timely manner to make the breeding work meet the needs of industrial development.

[0092] In summary, the system has the advantages of comprehensive data integration and utilization, accurate analysis and decision-making, and fast improvement speed of beef cattle breeds.

Claims

1. An intelligent beef cattle breeding monitoring and management system based on data from the entire industry chain, characterized in that include: Data collection module: used to deploy sensors and data collection terminals in all links of the beef cattle industry chain for data collection; Data transmission and integration module: used to use wireless communication technology to transmit the data collected in each link to the cloud data center for processing; Intelligent analysis module: used to mine the integrated data through big data analysis and artificial intelligence algorithms, and adjust seed selection, nutritional formula and management strategies; Monitoring and early warning module: used to set the threshold of key indicators. When the key indicators exceed the threshold, the system automatically sends an early warning message to the mobile phone or computer client of the breeding personnel, and takes intervention measures according to the alarm information; Breeding decision support module: used to provide guidance for breeding direction based on long-term accumulated data and analysis results, and to screen out excellent beef cattle individuals and their families.

2. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 1, characterized in that: The data collected by the data collection module includes: In the breeding stage, the time points of estrus, mating and conception of the breeding cattle, as well as the body temperature and behavioral characteristics are recorded; in the growth stage, the weight, body size and daily weight gain data of the beef cattle are monitored; in the feeding stage, the feed type, feeding amount and remaining amount as well as the feed energy efficacy parameters are collected, and combined with the environmental temperature and humidity, oxygen concentration and ammonia concentration information; in the slaughtering and processing stage, the marble pattern score, tenderness and pH value of the beef are determined, and at the same time, the veterinary diagnosis records and vaccination status are entered.

3. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 2, characterized in that: The data acquisition module includes 1) Reproduction data acquisition unit: High-definition cameras are installed in the cowshed, combined with intelligent image recognition algorithms, to monitor the behavior of breeding cattle around the clock. Through deep learning analysis of the activity frequency of breeding cattle and the interaction mode with other cattle, signs of estrus are captured. When estrus is determined to have begun, the system immediately records the time node; at the same time, biosensor collars are worn for breeding cattle. The collars collect body temperature data every 5-15 minutes through the integrated high-precision body temperature sensor and transmit it wirelessly to the data acquisition terminal. The collars also use minimally invasive hormone level monitoring modules. Through micro-probes in contact with the skin of the cattle, the levels of estrogen and progesterone in the blood are regularly collected and uploaded in real time; in the breeding process, breeding personnel use handheld terminal devices to enter the breeding time and breeding bull number information into the management system while completing the breeding operation; combined with the results of regular veterinary ultrasound examinations and subsequent monitoring of changes in the physiological characteristics of breeding cattle, the time node of successful conception is comprehensively determined and also entered into the system for subsequent analysis; 2) Growth stage data acquisition unit: High-precision electronic scales are installed at the entrances and exits of growing and fattening cattle sheds. Every time the beef cattle enter or leave to eat or drink water, the weight measurement is automatically triggered, and the data is instantly transmitted to the data acquisition terminal; multiple groups of laser measuring instruments are arranged at intervals in the cattle shed. When the beef cattle stand in the measurement area of ​​the laser measuring instrument, their body length, height and chest circumference data are measured once a day to obtain the growth and development changes of the beef cattle; the daily weight gain data is automatically generated by the system through the difference calculation of the daily weight data, combined with the corresponding date information, to show the growth rate trend of the beef cattle; 3) Data acquisition unit for feeding process: The feed storage bin and feeding equipment are equipped with intelligent metering sensors. Before feeding, the system automatically records the type of feed and the initial feeding amount. After feeding, the system calculates the actual feeding situation of the beef cattle by comparing the remaining feed amount, and obtains the type of feed, feeding amount and remaining amount. Several multifunctional environmental sensor nodes are distributed in the cowshed. The environmental sensor nodes integrate temperature and humidity sensors, oxygen sensors and ammonia sensors. Environmental data is collected every 15-20 minutes to obtain the environmental temperature and humidity, oxygen concentration and ammonia concentration information in the cowshed. 4) Slaughtering and processing data acquisition unit: Determine the marble score, tenderness and pH value of beef. Use professional meat quality testing equipment and standardized processes. Set up a meat quality testing workstation next to the slaughterhouse assembly line. When the beef carcass passes through a specific station, use a near-infrared spectrometer to scan the beef surface. Based on the spectral feature analysis, quickly give a marble score. With the help of a tenderizer, cut a standard-sized meat sample from the beef sample, and use the instrument to simulate the chewing process to measure the shear force of the meat sample to determine the tenderness index. The sampling test ratio for each batch of beef is not less than 10%; Using a high-precision pH meter, multi-point measurements are taken on the beef 45 minutes after slaughter and 24 hours after acid excretion in a 0-4°C environment, and the average value is taken as the pH value data for the batch of beef. Slaughterhouse staff retrieve the veterinarian's diagnostic records and vaccination status of the entire beef cattle breeding process from the cloud database based on the individual cattle numbers, so as to complete the beef quality traceability chain.

4. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 1, characterized in that: In the data transmission and integration module: In the cloud data center, outlier data and duplicate data are removed through preset data cleaning algorithms, and data in different formats and sources are standardized. The individual beef cattle number is used as the primary key, and the breeding, growth, feeding, slaughtering and processing data are associated to build a complete beef cattle industry chain database.

5. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 1, characterized in that: The data transmission and integration module includes: 1) Outlier data removal unit: For continuous data, calculate its mean μ and standard deviation σ, set the threshold range, use μ±3σ as the basis for judgment, and judge the data beyond this range as outliers; for discrete data, determine the value range based on historical data and industry experience, and the data outside this range is considered abnormal; for data that changes dynamically over time, construct an autoregressive moving average ARIMA model to predict the expected value range at each time point. If the actual collected value deviates from the prediction interval by more than the set threshold, it is considered abnormal data; conduct joint analysis on correlated variables, analyze the logical relationship between correlated variables, and judge the existence of abnormal data based on the corresponding logical relationship; 2) Duplicate data processing unit: Based on the individual number of beef cattle, all uploaded data records are compared. When the individual number of beef cattle, data collection time and data type of more than two data records are completely consistent, it is determined to be duplicate data, one of the records is retained, and the remaining duplicates are deleted; in the case where some information is not completely consistent due to data collection equipment failure or human input errors, but actually points to the same data, a fuzzy matching algorithm is used to identify and merge suspected duplicate data by setting a tolerance range; 3) Data standardization processing unit: used to formulate unified specifications for format differences of data from different sources. For text data, a standard name library is established to make the data consistent in expression; the dimensions of data involving physical quantities are converted to unify the dimensions of different types of physical quantities; for classified data, a coding system is used to encode the data, replacing text descriptions with digital codes; 4) Multi-dimensional data association unit: used to set the individual number of beef cattle as the unique primary key during the database design phase, and to build an association relationship around the primary key; when new data is uploaded, the system first identifies the beef cattle based on the individual number, and then associates the data records corresponding to the reproduction, growth, feeding and slaughtering and processing links with it. When the slaughtering and processing data of a beef cattle is received, the slaughtering and processing data includes the beef marbling score. The early growth records, feeding details and breeding information of the beef cattle are found through the individual number, so as to realize the seamless connection of the data of the whole beef cattle industry chain; create multiple association tables to record the corresponding relationship between the key time nodes and growth indicators of each beef cattle from conception to growth. Through these association tables, the JOIN operation of the database is used to integrate multi-dimensional data.

6. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 1, characterized in that: The intelligent analysis module: Based on machine learning models, the conception probability and calving interval of breeding cattle are predicted according to historical breeding data, and breeding is planned; based on the growth curve model and combined with feeding conditions, the feed formula is adjusted in real time to optimize the growth rate; through correlation analysis of meat quality indicators and feeding process data, the key factors for improving beef quality are found, providing a basis for adjusting feeding strategies.

7. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 6, characterized in that: The intelligent analysis module includes: 1) Reproduction analysis and breeding planning unit: First, a large amount of pre-processed historical reproduction data is collected, including the breed, age, past estrus cycle, number of matings, conception status, and calving history information of the breeding cattle; a logistic regression model is used to predict the probability of conception, and a random forest regression model is used to predict the calving interval. The pre-processed historical reproduction data is divided into training sets and test sets according to a certain ratio, and input into the selected model for training. By continuously adjusting the hyperparameters of the model and using the cross-validation method, the model can achieve a high accuracy and stability on the test set. When predicting the probability of pregnancy of a certain breeder, the current characteristic data of the breeder is processed according to the set coding and standardization method, and input into the trained logistic regression model. The model outputs a probability value between 0 and 1, indicating the possibility of pregnancy of the breeder in this mating. For calving interval prediction, the breeding history and current status data of the breeder are input into the random forest regression model, and the model outputs the expected number of days of calving interval. A threshold is set based on the prediction results of the pregnancy probability. When the pregnancy probability is higher than the set threshold, the breeder cattle are given priority for breeding. If it is lower than the threshold, the cause is further analyzed and the breeding timing is considered after corresponding adjustments are made. The breeding cycle of the breeder cattle is planned in combination with the calving interval prediction. 2) Feed formula adjustment unit: Growth curve model construction: Collect historical growth data of beef cattle of different breeds, genders, and age groups, including daily weight and body size change information, and use nonlinear regression method to construct the growth curve Gompertz model to fit the growth rate change trend of beef cattle during growth. Determine the parameters in the model through fitting training of a large amount of historical data; Feeding condition data fusion: Real-time collection of current feeding condition data of beef cattle, including feed type, feeding amount, ambient temperature and humidity, and ammonia concentration, and combining these data with the Gompertz model of the growth curve to analyze the direction and degree of influence of current feeding conditions on the growth rate of beef cattle, and using the model to predict the growth indicators that beef cattle should achieve according to the normal growth curve under the current feeding conditions and compare them with the actual measured values; Feed formula adjustment: When the actual growth indicators of beef cattle deviate from the normal growth curve, the feed formula adjustment mechanism is activated; using linear programming algorithms, the proportion of protein, energy, vitamins and minerals in the feed is dynamically adjusted to meet the nutritional needs of beef cattle, reduce feed costs and increase growth rate. By adjusting the feed formula in real time and dynamically, the growth of beef cattle can return to the ideal growth curve; 3) Key factors of beef quality mining unit: Data collection and organization: Integrate beef quality data from the slaughtering and processing stages, including marbling scores, tenderness and pH data, as well as corresponding feeding process data, including feed feeding records, disease prevention and control conditions, and environmental data throughout the growth process of beef cattle; Application of association analysis: Pearson correlation coefficient is used to analyze the linear correlation between continuous variables. ANOVA is used to analyze the differences in pH values ​​of beef from different breeds of beef cattle under the same feeding conditions. Apriori algorithm in data mining is used to mine frequent item sets to find the feeding factor combinations that frequently appear in high-quality beef production cases. Identification of key factors and adjustment of feeding strategies: Based on the results of association analysis, key factors that have a significant impact on beef quality are determined, these key factors are incorporated into feeding strategies, and targeted feeding plans are developed.

8. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 1, characterized in that: In the monitoring and early warning module: When the daily weight gain of calves is lower than the normal range, fattening cattle show abnormal disease symptoms, and the quality of slaughtered beef does not meet the standards, the system automatically sends an early warning message to the breeder's mobile phone or computer client, and adjusts the feeding and / or treatment plan according to different indicators to ensure the healthy growth and high-quality output of beef cattle.

9. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data as claimed in claim 8, characterized in that: The monitoring and early warning module includes: 1) Early warning trigger unit: Calf daily weight gain warning: Based on historical data and industry standards, combined with the growth characteristics of calves of different breeds, genders, and feeding stages, a dynamic daily weight gain normal range model is constructed; the actual weight data of the calf is obtained through the data acquisition module every day, the daily weight gain is automatically calculated, and compared with the normal range in the corresponding model in real time. When the daily weight gain of a calf is detected to be lower than the set lower limit for three consecutive days, an early warning signal is triggered; Early warning of disease symptom indicators of fattening cattle: In the feeding process, the data collection terminal continuously collects the body temperature, respiratory rate and feeding behavior data of fattening cattle, builds a disease prediction model through machine learning algorithms, and identifies the abnormal change patterns of various indicators under different disease states by learning a large amount of historical data of healthy and diseased fattening cattle. When the body temperature of a fattening cattle is higher than 40°C for 2 consecutive hours, the respiratory rate exceeds the normal upper limit by 20%, and the feeding amount is reduced by more than 50% compared with the average feeding amount in the past 7 days, it is determined that the fattening cattle have abnormal disease symptom indicators and an early warning is issued; Slaughter beef quality warning: In the slaughtering and processing stage, after the marble pattern score, tenderness and pH value of the beef are measured by the testing equipment, the system will score it against the preset quality standards. When the marble pattern score is lower than the industry quality standard, or the tenderness index exceeds the upper limit of the appropriate shear force, or the pH value deviates from the normal range, the corresponding quality non-standard warning will be triggered; 2) Early warning information push unit: Using the pre-integrated communication interface, the warning information is sent to the mobile phone or computer client of the breeder through SMS, instant messaging software or system pop-up window; the warning information includes: for the daily weight gain warning of calves, the individual number of the calf with problems, the location of the cowshed, the current daily weight gain value and the number of consecutive days below the normal range are displayed, and the normal daily weight gain reference range of calves of this breed is attached; for the disease warning information of fattening cattle, the number of fattening cattle with the disease, the current abnormal physiological index value, and the preliminary prompt of the possible disease type are displayed to assist the breeder in making a preliminary judgment on the disease; for the quality warning of slaughtered beef, the batch number of the substandard beef, the specific quality index value and the difference with the standard are listed; 3) Feeding and / or treatment regimen adjustment unit: Calf daily weight gain problem handling: After the system gives an early warning, it first analyzes the report based on the automatically generated detailed feeding analysis report. If the analysis finds that the problem is caused by insufficient feed nutrition, the system recommends adjusting the feed formula. If the analysis finds that the problem is caused by environmental factors, it prompts you to adjust the cowshed environmental control parameters, turn on the heating equipment, strengthen ventilation and dehumidification, control the temperature at 15-20℃, maintain the relative humidity at 60%-70%, and continuously monitor the subsequent growth of the calves until the daily weight gain returns to normal; Disease treatment for fattening cattle: The system conducts preliminary isolation and observation of sick fattening cattle based on the disease indications in the early warning and remote guidance from veterinarians. If it is determined to be a common disease, the system will give specific medication recommendations based on the preset treatment plan library, including drug name, dosage and frequency of medication. At the same time, the system will adjust the feeding plan for sick cattle and provide easily digestible and highly nutritious feed, such as silage with a small amount of concentrated feed, to ensure that the sick cattle recover physically. After the condition improves, regular feeding will be resumed, and various physiological indicators will be continuously monitored until they return to normal. If it is determined to be a rare disease, a treatment plan will be given through remote guidance from veterinarians. Improvement of slaughter beef quality: In case of substandard beef quality, the system will trace back the data of the entire process of raising beef cattle for this batch and analyze the root cause of the problem. If the feed combination in the later stage of raising is unreasonable, the feed formula will be adjusted when raising the next batch of beef cattle, and the amount of minerals added will be optimized. They will be added in a specific proportion during the fattening stage. If the pH value is abnormal due to stress before slaughter, the management of transportation and slaughtering links will be improved to shorten the transportation time, optimize the slaughtering environment, and reduce the stress response of beef cattle.

10. The intelligent beef cattle breeding monitoring and management system based on the whole industry chain data according to claim 1, characterized in that: The breeding decision support module: It is used to provide guidance for breeding direction based on long-term accumulated data and analysis results, screen out beef cattle individuals and their families with excellent reproductive performance, fast growth and good meat quality, formulate breeding and mating plans, and gradually optimize the gene pool of beef cattle breeds.

Images