Ecological pasture cow breeding and health management method and system based on meteorological data

Abstract

The application provides an ecological pasture cow breeding and health management method and system based on meteorological data, and relates to the technical field of cow breeding and health management. The method comprises the following steps: collecting meteorological data and cow breeding basic data in real time; extracting key meteorological factors affecting cow breeding; dividing meteorological suitable threshold corresponding to each breeding stage; realizing weather adaptation of cow herd screening; selecting temperature suitable period for mating operation; formulating meteorological guarantee scheme according to pregnancy stage; quantifying the influence of meteorological conditions on breeding core indicators, and optimizing breeding plan and resource allocation based on the evaluation results. Through the improved cow breeding management process, the application improves the utilization efficiency of breeding resources, provides a traceable and iterative analysis framework for breeding plan optimization, promotes the development of ecological pasture breeding management towards precision and high efficiency, and helps the pasture to reduce costs while realizing the double improvement of population quality and economic benefits.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of dairy cow reproduction and health management, and particularly relates to an ecological ranch dairy cow reproduction and health management method based on meteorological data and an ecological ranch dairy cow reproduction and health management system based on meteorological data. BACKGROUND

[0002] In the ecological ranch dairy cow breeding industry, dairy cow reproduction and health management is the core link that determines the quality of the dairy cow population, production performance and economic benefits of the ranch. Traditional dairy cow reproduction and health management relies on experience and focuses on the basic dimensions of dairy cow breed selection, pedigree analysis and nutrition supply, but it generally ignores the key influence of meteorological conditions on the physiological state, reproductive performance and stress resistance of dairy cows, resulting in problems such as limited breeding efficiency and unreasonable resource allocation.

[0003] From the industry status, with the intensification of global warming trend, extreme weather events (such as high temperature heat wave, low temperature cold wave, persistent rain) occur frequently, and the impact on the reproduction process of dairy cows is more and more significant. In high temperature environment, dairy cows are prone to heat stress, showing accelerated breathing rate and elevated rectal temperature, which leads to decreased estrus rate and reduced conception rate, and the survival rate of calves is obviously affected by environmental temperature fluctuations; low temperature cold wave increases the energy consumption of dairy cows, causes cold stress, reduces feed conversion rate and immunity, and indirectly affects the production performance in the breeding cycle. In addition, weather factors such as rain, wind, etc. will change the feeding amount and activity of dairy cows, and interfere with the accuracy of estrus behavior recognition, further exacerbating the incompleteness of breeding management. SUMMARY

[0004] The present application provides an ecological ranch dairy cow reproduction and health management method and system based on meteorological data to solve the defect of incomplete breeding management in the prior art.

[0005] In one aspect, the present application provides an ecological ranch dairy cow reproduction and health management method based on meteorological data, comprising:

[0006] Real-time collection of meteorological data and dairy cow breeding basic data and standardized processing.

[0007] Key meteorological factors affecting dairy cow breeding are extracted by principal component analysis, including temperature factor, weather factor and rainfall factor.

[0008] The quantitative correlation between the key meteorological factors and the breeding indexes is established by correlation analysis and multiple regression method, and the meteorological suitable threshold corresponding to each breeding stage is determined, and the breeding indexes include the estrus rate, conception rate, morbidity and production performance of dairy cows.

[0009] Based on key meteorological factors and their meteorological suitable thresholds, breeding cow selection criteria are formulated, the stress resistance of replacement heifers and lactating cows under different meteorological conditions is dynamically evaluated, and the selection weight is adjusted to realize meteorological adaptation of cow selection.

[0010] According to meteorological data, the estrus period of dairy cows is predicted, and artificial intelligence technology is used to identify estrus behavior, and breeding operation is performed in a temperature suitable period.

[0011] Meteorological guarantee schemes are formulated according to different pregnancy stages, and environmental regulation and nutrition adjustment measures are started according to real-time meteorological data to avoid stress reaction.

[0012] A breeding benefit evaluation model is constructed to quantify the influence of meteorological conditions on breeding core indicators, and breeding plans and resource allocation are optimized based on the evaluation results.

[0013] According to the ecological ranching dairy cow breeding and health management method based on meteorological data provided by the application, the meteorological data includes daily average temperature, daily minimum temperature, daily maximum temperature, daily average relative humidity, daily average wind speed, daily cumulative precipitation, sunshine duration and air pressure. The dairy cow breeding basic data includes dairy cow breed, physiological stage, reproduction index, health index and production performance.

[0014] According to the ecological ranching dairy cow breeding and health management method based on meteorological data provided by the application, the process of extracting key meteorological factors affecting dairy cow breeding by principal component analysis method includes:

[0015] Collect meteorological data and corresponding dairy cow breeding basic data in multiple complete production cycles to construct a multidimensional data matrix.

[0016] For each feature in the multidimensional data matrix, calculate the mean and standard deviation, and obtain the standardized data matrix.

[0017] Based on the standardized matrix, a covariance matrix is calculated, and the covariance matrix is subjected to eigenvalue decomposition to obtain eigenvalues and corresponding eigenvectors.

[0018] The number of principal components is determined according to the eigenvalue contribution rate, and the screening is stopped when the cumulative contribution rate reaches the preset proportion.

[0019] According to the ecological ranching dairy cow breeding and health management method based on meteorological data provided by the application, the process of delimiting the meteorological suitable threshold corresponding to each breeding stage includes:

[0020] The correlation coefficient of the key meteorological factor and each breeding index is calculated by Pearson correlation analysis.

[0021] The key meteorological factors whose correlation coefficients with breeding indexes reach the preset correlation threshold are selected by correlation test.

[0022] Taking key meteorological factors as independent variables and breeding indexes as dependent variables, a multiple linear regression model is established.

[0023] The regression coefficients are solved by using the least square method to minimize the residual sum of squares, and the meteorological suitable threshold values of each breeding stage are determined based on the regression model.

[0024] According to the ecological pasture cow breeding and health management method based on meteorological data provided by the application, the process of formulating the breeding cow group selection and breeding standard includes:

[0025] Combined with the meteorological suitable threshold values of each breeding stage, the basic selection and breeding indexes of the breeding cow group are determined, including age, body weight and pedigree parameters.

[0026] For the temperature factor, the heat tolerance score standard is formulated, and the respiratory rate and rectal temperature of the cow under specific temperature conditions are detected for scoring.

[0027] For the weather factor, the forage intake change of the cow under specific rainy conditions is evaluated to evaluate the stress resistance.

[0028] For the rainfall factor, the activity of the cow under specific windy weather is observed to evaluate the adaptability.

[0029] The adaptability of key meteorological factors is included in the selection and breeding weight calculation comprehensive score, and individuals with comprehensive score reaching a set score are selected into the breeding cow group.

[0030] According to the ecological pasture cow breeding and health management method based on meteorological data provided by the application, the process of realizing the meteorological adaptation of the cow group screening includes:

[0031] Selecting the heifer and lactating cow as the evaluation sample.

[0032] An index system for evaluating stress resistance performance is constructed, and the index system for evaluating stress resistance performance includes physiological indexes, production indexes and health indexes. The physiological indexes include respiratory rate and rectal temperature. The production indexes include milk yield and milk protein rate. The health indexes include morbidity and somatic cell count.

[0033] Tracking monitoring is carried out for a set duration under different meteorological conditions.

[0034] When the temperature factor is in a specific heat stress interval, the increase of the respiratory rate of the heifer compared with the suitable temperature interval and the decrease of the milk yield of the lactating cow are recorded.

[0035] When the rainfall factor is in a high influence interval, the morbidity change of the two types of cows is recorded.

[0036] The analytic hierarchy process is used to determine the weight of each evaluation index of the stress resistance performance evaluation index system.

[0037] Calculate the comprehensive index of the anti-adverse performance, and adjust the breeding weight according to the comprehensive index.

[0038] Through dynamic weight adjustment, screening once every set period to realize weather adaptation of the cow herd optimization.

[0039] The process of selecting a temperature suitable period for mating operation according to the ecological pasture cow breeding and health management method based on meteorological data comprises:

[0040] Collect historical meteorological data, and label the estrus rate of the cows corresponding to the historical meteorological data.

[0041] According to the historical meteorological data, an ARIMA model in time series analysis is used to predict the change trend of the key meteorological factors in the future set time length.

[0042] The optimal model parameters are determined by AIC criterion, and the daily average temperature change curve of the temperature factor is predicted.

[0043] According to the correlation between the estrus rate of the cows and the temperature factor in the historical data, when the predicted daily average temperature is in a specific interval, it is determined as an estrus period.

[0044] Install high-definition cameras and infrared sensors on the pasture to collect cow behavior data and physiological data. The behavior data includes activity frequency, standing time and climbing behavior. The physiological data includes body temperature and activity amount.

[0045] An artificial intelligence recognition model based on CNN-LSTM is constructed, and the cow behavior data and physiological data are taken as inputs to recognize the estrus behavior of the cows.

[0046] According to the temperature prediction result and the estrus behavior recognition result, the mating operation is selected to be performed within a set time length after the cows are in estrus when the daily average temperature is in a specific interval.

[0047] The process of formulating a meteorological guarantee scheme according to the ecological pasture cow breeding and health management method based on meteorological data comprises:

[0048] The cow pregnancy process is divided into early, middle and late stages, and meteorological guarantee schemes are formulated for different stages.

[0049] In the early stage, the temperature factor is monitored, and when the real-time daily average temperature is lower than the set value, the cowshed heating system is started, the temperature in the shed is controlled to a specific interval, and the feed formula is adjusted to increase the proportion of energy feed and improve the cold resistance of the cows. When the temperature is higher than the set value, the spray and fan combination system is started, the spray interval and the length of each spray are set, and the fan speed is adjusted.

[0050] The middle stage pays attention to weather factors, when the sunshine duration is lower than a set value, the artificial light is supplemented, the light time of the set duration is increased, when the average relative humidity of the day is higher than a set value, the dehumidification equipment is started, the humidity in the shed is controlled below the set value, and the antifungal agent is added in the feed, and the adding amount is set.

[0051] The late stage monitors the rainfall factor, when the daily cumulative precipitation reaches a set value, the cowshed roof is reinforced, the drainage channel is cleaned, and the accumulated water is prevented, when the average wind speed of the day reaches a set value, the cowshed side window is closed, and the windproof roller blind is installed to reduce the entry of cold air.

[0052] Real-time acquisition of meteorological data and physiological data of dairy cows, when the abnormal respiratory rate of dairy cows or the decrease of feed intake reaches a set proportion or more, the emergency control measures are automatically triggered.

[0053] The ecological ranch dairy cow breeding and health management method based on meteorological data provided by the application, the breeding benefit evaluation model is constructed, and the process of quantifying the influence of meteorological conditions on the breeding core index includes:

[0054] The estrus rate, conception rate, calf survival rate and milk yield of dairy cows are selected as the breeding core index, the key meteorological factors are selected as the input variable, the core index is selected as the output variable, and the BP neural network is used to construct the breeding benefit evaluation model.

[0055] The network structure of the breeding benefit evaluation model is set as a specific node combination, including an input layer, a hidden layer and an output layer. The input layer corresponds to the nodes of the number of key meteorological factors, the hidden layer is set to a certain number of nodes, and the output layer corresponds to the nodes of the number of core indexes.

[0056] The Sigmoid function is selected as the activation function, and the learning rate and the number of iterations are set.

[0057] The breeding benefit evaluation model is trained through historical data, so that the prediction error of the model is controlled within a set proportion.

[0058] Based on the trained model, the influence of different meteorological conditions on the core index is quantified by calculating the meteorological influence benefit value.

[0059] On the other hand, the application also provides an ecological ranch dairy cow breeding and health management system based on meteorological data, comprising:

[0060] The data acquisition and feature extraction module is used for real-time acquisition of meteorological data and dairy cow breeding basic data, and the principal component analysis method is used to extract the key meteorological factors affecting dairy cow breeding.

[0061] The meteorological threshold determination module is used to establish the quantitative correlation between the key meteorological factors and the breeding indexes by correlation analysis and multiple regression method, and to determine the meteorological suitable threshold corresponding to each breeding stage.

[0062] The adaptive cow herd screening module is used to formulate breeding cow herd selection criteria based on key meteorological factors and their meteorological suitability thresholds, dynamically evaluate the stress resistance performance of replacement cows and lactating cows under different meteorological conditions, and adjust the selection weight, so as to realize meteorological adaptation of cow herd screening.

[0063] The suitable period selection module is used to predict the estrus period of dairy cows according to meteorological data, and identify estrus behavior by using artificial intelligence technology, and select a temperature suitable period for mating operation.

[0064] The meteorological guarantee formulation module is used to formulate meteorological guarantee schemes for different pregnancy stages, and start environmental regulation and nutrition adjustment measures according to real-time meteorological data to avoid stress reaction.

[0065] The breeding effect evaluation module is used to construct a breeding benefit evaluation model, quantify the influence of meteorological conditions on breeding core indicators, and optimize breeding plans and resource allocation based on the evaluation results.

[0066] The ecological ranch dairy cow reproduction and health management method and system based on meteorological data provided by the application accurately extract key meteorological factors by principal component analysis method, and solve the meteorological influence factors. With the help of correlation analysis and multiple regression method, the quantitative correlation between key meteorological factors and breeding indicators such as estrus rate and conception rate is established, and the meteorological suitable threshold of each breeding stage is determined, so that the breeding management is changed from experience judgment to data support.

[0067] By formulating meteorologically adapted breeding cow selection criteria, the temperature, weather and rainfall factors are adapted into the selection weight, an index system for evaluating stress resistance performance is constructed, and dynamic screening of replacement cows and lactating cows is realized. Under high temperature conditions, by monitoring the respiratory frequency increase of replacement cows and the lactation yield decrease of lactating cows, and combining with the analytic hierarchy process to determine the index weight, the comprehensive index of stress resistance performance can be calculated, so that individuals with strong heat tolerance can be screened out, and the production performance loss of the population under extreme high temperature weather can be reduced. At the same time, the selection weight is dynamically adjusted according to the set period, so as to ensure that the cow herd is continuously adapted to the regional meteorological characteristics, and lay a foundation for creating a core population with strong stress resistance and stable production for ecological ranch.

[0068] Fusion of historical meteorological data and ARIMA time series model can predict the temperature change trend in the future set time, combined with the correlation between historical estrus rate and temperature, the estrus period can be accurately determined; and through the CNN-LSTM artificial intelligence model, the estrus characteristics such as cow climbing behavior and body temperature change are identified, so as to realize double confirmation of meteorological prediction and behavior identification.

[0069] The cow pregnancy is divided into early, middle and late stages to develop differentiated meteorological guarantee schemes, realizing closed-loop management of prediction, prevention and control and adjustment. In the early stage of pregnancy, the heating or spraying system is started for temperature factor, in the middle stage, artificial light is supplemented for insufficient sunlight, humidity is controlled and mildew inhibitor is added, and in the late stage, the cowshed is reinforced for heavy rain and strong wind, and wind roller shutter is prevented, which can avoid stress reaction caused by meteorological stress from the source.

[0070] The breeding benefit evaluation model constructed based on the BP neural network can quantify the influence of different meteorological conditions on the core indexes such as estrus rate and calf survival rate, and provide accurate basis for resource allocation. The pasture can adjust the breeding plan in a targeted manner through the meteorological influence benefit value output by the model, so as to avoid resource waste. BRIEF DESCRIPTION OF DRAWINGS

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

[0072] Figure 1 is a flowchart of the ecological pasture cow breeding and health management method based on meteorological data provided by the embodiments of the present application;

[0073] Figure 2 is a schematic diagram of the dynamic change of temperature inside and outside the cowshed in different seasons;

[0074] Figure 3 is a structural schematic diagram of the ecological pasture cow breeding and health management system based on meteorological data provided by the embodiments of the present application. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme in the present application will be described clearly and completely in the following combined with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the protection scope of the present application.

[0076] The ecological pasture cow breeding and health management method and system based on meteorological data of the present application will be described in the following combined with Figures 1-3

[0077] Figure 1 is a flowchart of the ecological pasture cow breeding and health management method based on meteorological data provided by the embodiments of the present application; is a flowchart of the ecological pasture cow breeding and health management method based on meteorological data provided by the embodiments of the present application;

[0078] As Figure 1 indicated, the method and system for ecological pasture cow breeding and health management based on meteorological data provided by the embodiment of the application, the execution subject can be a method for ecological pasture cow breeding and health management based on meteorological data, the method comprises:

[0079] Real-time collection of meteorological data and cow breeding basic data. The meteorological data includes daily average temperature, daily minimum temperature, daily maximum temperature, daily average relative humidity, daily average wind speed, daily cumulative precipitation, sunshine duration and air pressure. The cow breeding basic data includes cow breed, physiological stage, reproduction index, health index and production performance.

[0080] Figure 2 is a schematic diagram of daily dynamic changes of temperatures inside and outside the cowshed in different seasons in the embodiment of the application.

[0081] In the fixed monitoring aspect, in different types of demonstration pastures such as tunnel-type pastures, open pastures, and sunshine constant-temperature pastures, meteorological environment monitoring integrated equipment is arranged according to the “Meteorological Environment Automatic Monitoring Specification for Large-scale Cow Farms”. Each set of equipment integrates 13 types of monitoring components such as temperature and humidity sensors, illuminance sensors, wind speed and direction sensors, carbon dioxide transmitters, ammonia temperature and humidity transmitters, malodorous gas sensors, and noise sensors. The equipment accesses the cloud platform through the NB-IOT / 4G network, realizes the minute-level collection and real-time uploading of meteorological elements such as daily average temperature, daily minimum temperature, daily maximum temperature (measurement range -50~+50℃, accuracy ±0.2℃), daily average relative humidity (measurement range 5~100%, accuracy ±3%), daily average wind speed (measurement range 0~60m / s, accuracy ±0.5m / s), daily cumulative precipitation (measurement range 0~4mm / min, accuracy ±0.4mm), sunshine duration (measurement range 0~2000W / m², accuracy ±5%FS), and air pressure (measurement range 450~1100hPa, accuracy ±0.3hPa), and synchronously collects gas concentration data such as ammonia (0-50PPM, accuracy ≤1%FS) and hydrogen sulfide (0-100PPM, accuracy ≤1%FS) that affect the health of cows.

[0082] In the mobile patrol link, a cowshed meteorological environment automatic patrol device is developed and matched. The device is equipped with an automatic navigation vehicle module, a high-precision camera module, and a multi-element sensor module (including an infrared thermal imaging sensor), and can move autonomously in the cowshed along a preset route, focusing on collecting local meteorological data (such as calf breathing zone temperature and humidity) and cow physiological state data (such as body surface temperature and activity frequency) in key areas such as calf activity areas and lactating cow bed areas, making up for the monitoring blind area of fixed equipment, and realizing all-round data coverage of macro environment and micro area.

[0083] The collection of dairy cow breeding basic data adopts the collaborative mode of intelligent equipment and manual recording: through the existing cow herd management system of the farm (such as dairy cow estrus monitoring collar, milking machine data terminal), the data of dairy cow breed (such as Holstein cow), physiological stage (heifer, lactating cow, pregnant cow), reproductive index (estrus record, breeding time, pregnancy result), production performance (milk yield, milk protein rate, milk fat rate) and the like are automatically obtained; in terms of health index, the data of cow body temperature collected by infrared sensor are combined with the veterinary inspection records (morbidity, somatic cell count, disease type) to form a complete breeding basic database, and after standardization processing (such as unified time format, elimination of abnormal values), the data are stored in the cloud database in association with the meteorological data.

[0084] Through analysis of the monitoring data of a certain farm from July 2021 to February 2022, the influence of the spatio-temporal distribution law of the temperature inside and outside the cowshed on dairy cow breeding is determined: in summer (July-August), the daily average temperature outside the cowshed is as high as 33℃, and due to the regulation measures such as spraying and fan, the temperature inside the cowshed is 0.32℃ lower than that outside, and the temperature difference at different positions in the cowshed is small (the maximum temperature difference is 0.14℃), so it is necessary to focus on controlling the temperature factor to avoid the decrease of estrus rate caused by heat stress of dairy cows; in winter (December-February), the daily average temperature inside the cowshed is 2.97℃ higher than that outside, and the temperature at the monitoring points No. 1 and No. 2 (close to the lying bed area) is significantly higher than that at the monitoring points No. 3 and No. 4 (close to the air vent), with a temperature difference of 1.89℃, which indicates that when formulating the meteorological guarantee scheme for pregnant cows in winter, the heating strategy for the lying bed area needs to be optimized to prevent cold stress of calves; in spring and autumn (March-June and September-November), the temperature difference between inside and outside the cowshed is between 0.5-1.13℃, and the temperature fluctuates greatly during the day (for example, the temperature in the afternoon is 8-10℃ higher than that in the morning in spring), so it is necessary to dynamically adjust the frequency of environmental regulation to ensure the stability of key breeding indicators. Based on the above temperature variation law, the meteorological suitable threshold of each breeding stage can be further refined, for example, the suitable temperature range for lactating cows in summer is 22-26℃, and in winter it is 12-16℃.

[0085] The meteorological data and breeding basic data are standardized, and the principal component analysis method is used to extract the key meteorological factors affecting dairy cow breeding, including temperature factor, weather factor and rainfall factor. The process of extracting the key meteorological factors affecting dairy cow breeding includes:

[0086] Collect the meteorological data and corresponding dairy cow breeding basic data in multiple complete production cycles to construct a multi-dimensional data matrix X, where X=[x ij ] (n×m) , n represents the number of samples, and m represents the number of features.

[0087] For each feature in the data matrix, the mean μ j and standard deviation σ j, the formulas are respectively:

[0088]

[0089] And by Get the standardized data matrix Z.

[0090] Based on the standardized matrix Z, the covariance matrix S is calculated, and the formula is represented as:

[0091]

[0092] The eigenvalue decomposition of the covariance matrix is carried out, and the eigenvalues λ1≥λ2≥…≥λ m And the corresponding eigenvectors.

[0093] According to the eigenvalue contribution rate, the principal component number is determined, when the cumulative contribution rate reaches the preset proportion, the screening is stopped, and finally the first three principal components are selected as the key meteorological factors, wherein the first principal component is the temperature factor, which is composed of daily average temperature, daily minimum temperature, daily maximum temperature and air pressure. The second principal component is the weather factor, which is composed of daily average relative humidity, sunshine duration, daily minimum visibility. The third principal component is the rainfall factor, which is composed of daily average wind speed and daily cumulative precipitation.

[0094] Through correlation analysis and multiple regression method, the quantitative correlation between key meteorological factors and breeding indexes is established, and the meteorological suitable threshold corresponding to each breeding stage is determined. Breeding indexes include cow estrus rate, conception rate, morbidity and production performance.

[0095] The process of determining the meteorological suitable threshold corresponding to each breeding stage includes:

[0096] Pearson correlation analysis is used to calculate the correlation coefficient between the key meteorological factors and each breeding index, and the formula is represented as:

[0097]

[0098] In the formula, is the i th key meteorological factor data, is the i th breeding index data, And The mean value of the corresponding data.

[0099] Through correlation test, the key meteorological factors whose correlation coefficient with breeding index reaches the preset correlation threshold are screened out.

[0100] Taking the key meteorological factor as the independent variable x k , and the breeding index as the dependent variable y, a multiple linear regression model is established, and the formula is represented as:

[0101]

[0102] wherein, is the intercept term, is the regression coefficient, is the random error.

[0103] The regression coefficient is solved by the least square method, so that the residual sum of squares is minimum, wherein, is the predicted value.

[0104] Based on the regression model and the optimal interval of breeding indicators in historical data, the meteorological suitable threshold of each breeding stage is determined. For example, in the reserve cattle breeding stage, the suitable range of daily average temperature corresponding to the temperature factor is 10-16℃, and when it exceeds this range, the estrus rate decreases by more than 15%. In the gestation stage, the suitable range of sunshine duration corresponding to the weather factor is 6-8 hours / day, and when it is less than 4 hours or more than 10 hours, the incidence rate increases by more than 8%.

[0105] Based on the key meteorological factors and their meteorological suitable thresholds, the breeding cattle selection criteria are developed, the stress resistance performance of reserve cattle and lactating cattle under different meteorological conditions is dynamically evaluated, and the selection weight is adjusted to realize the meteorological adaptation of cattle selection.

[0106] The process of developing breeding cattle selection criteria includes:

[0107] Combined with the meteorological suitable threshold of each breeding stage, the basic selection indicators of breeding cattle are determined, including age, body weight, pedigree, etc. Basic parameters, among which the age of reserve cattle needs to be 14-16 months old, and the body weight is not less than 380 kg.

[0108] For the temperature factor, the heat tolerance score standard is developed, and the respiratory rate and rectal temperature of dairy cows at a daily average temperature of 30℃ are detected. The respiratory rate is 50-70 times / minute, and the rectal temperature is 39.0-39.5℃, which is evaluated as excellent heat tolerance.

[0109] For the weather factor, the change of feed intake of dairy cows under continuous rainy weather (daily cumulative precipitation≥20mm, sunshine duration≤3 hours) is evaluated, and the feed intake is decreased by not more than 10%, which is evaluated as good stress resistance.

[0110] For the rainfall factor, the activity ability of dairy cows under windy weather (daily average wind speed≥6m / s) is observed, and the activity range is reduced by not more than 20%, which is evaluated as strong adaptability.

[0111] The key meteorological factor adaptability is included in the selection weight, the basic parameter accounts for 50%, the temperature factor adaptability accounts for 25%, the weather factor adaptability accounts for 15%, and the rainfall factor adaptability accounts for 10%. Through comprehensive score = basic parameter score × 0.5 + temperature adaptation score × 0.25 + weather adaptation score × 0.15 + rainfall adaptation score × 0.1, the individuals with comprehensive score of 80 or more are selected into the breeding cattle population.

[0112] The process of selecting the cattle population with meteorological adaptation includes:

[0113] 100 reserve cows and 80 lactating cows are selected as evaluation samples, and an anti-adversity performance evaluation index system is constructed. The anti-adversity performance evaluation index system includes physiological indexes, production indexes and health indexes. Physiological indexes include respiratory rate and rectal temperature. Production indexes include milk yield and milk protein rate. Health indexes include morbidity and somatic cell count.

[0114] Under different meteorological conditions, 12 months of tracking monitoring are carried out. When the temperature factor is in the mild heat stress interval (daily average temperature 25-28℃), the increase of respiratory rate of reserve cows and the decrease of milk yield of lactating cows in the mild heat stress interval are recorded.

[0115] When the rainfall factor is in the high impact interval (daily average wind speed ≥8m / s), the morbidity changes of the two types of cattle are recorded.

[0116] The analytic hierarchy process is used to determine the weight of each evaluation index. The weight of physiological index is 0.35, the weight of production index is 0.4, and the weight of health index is 0.25. The anti-adversity performance comprehensive index is calculated, which is expressed as:

[0117]

[0118] In the formula, is the weight of the evaluation index, is the standardized score of the evaluation index.

[0119] According to the comprehensive index, the selection weight is adjusted. The reserve cows with the top 30% of temperature factor anti-adversity performance index have their selection weight increased by 15%.

[0120] The lactating cows with the last 20% of rainfall factor anti-adversity performance index have their selection weight reduced by 10%. Through dynamic weight adjustment, the cattle population is optimized every quarter to realize meteorological adaptation, so that the proportion of adaptive cattle population increases from 65% to more than 85%.

[0121] According to the meteorological data, the estrus high incidence period of dairy cows is predicted, and the estrus behavior is identified by artificial intelligence technology. The breeding operation is carried out in the temperature suitable period, and the process includes:

[0122] Based on the meteorological data in the past 5 years, the ARIMA model in time series analysis is used to predict the trend of key meteorological factors in the next 3 months. The model formula is:

[0123]

[0124] In the formula, φ(L) is the autoregressive operator, θ(L) is the moving average operator, d is the difference order, is the observation value of the time series at time t, is the random error term at time t, the optimal model parameters are determined by AIC criterion, and the daily average temperature variation curve of the temperature factor is predicted.

[0125] Combined with the correlation between the estrus rate of dairy cows and the temperature factor in historical data, when the predicted daily average temperature is in the range of 18-22℃, it is determined as the high incidence period of estrus, and the estrus rate of dairy cows in this period is 25% higher than that in other temperature intervals.

[0126] High-definition cameras and infrared sensors are installed on the pasture to collect dairy cow behavior data (such as activity frequency, standing time, and mounting behavior) and physiological data (such as body temperature and activity level), and an artificial intelligence recognition model based on CNN-LSTM is constructed to recognize dairy cow estrus behavior. The training set of the model uses 5000 labeled estrus behavior data, and the test set accuracy reaches 92%.

[0127] According to the temperature prediction results and the estrus behavior recognition results, the mating operation is carried out within 6-18 hours after the estrus of dairy cows is recognized when the daily average temperature is 18-22℃, and the conception rate in this period is 30% higher than that in other periods. Real-time meteorological data at the time of mating is recorded for subsequent effect analysis.

[0128] Meteorological guarantee schemes are developed for different stages of pregnancy, and environmental control and nutrition adjustment measures are started according to real-time meteorological data to avoid stress reactions. The process includes:

[0129] The pregnancy process of dairy cows is divided into early stage (1-3 months), middle stage (4-6 months), and late stage (7-9 months), and meteorological guarantee schemes are developed for different stages.

[0130] In the early stage, the temperature factor is monitored, and when the real-time daily average temperature is lower than 5℃, the cowshed heating system is started, the indoor temperature is controlled to 10-12℃, and the proportion of energy feed (such as corn) in the feed formula is increased to 50% to improve the cold resistance of dairy cows.

[0131] When the temperature is higher than 28℃, the combined system of spraying and fan is started, the spraying interval is set to 10 minutes / time, and the fan speed is adjusted to 3m / s.

[0132] In the middle stage, focus on weather factors, when the daily sunshine duration is less than 5 hours, supplement artificial light, increase 3 hours of light per day.

[0133] When the average daily relative humidity is higher than 80%, turn on the dehumidification equipment to control the humidity in the shed below 65%, and add a mildew inhibitor to the feed at a dosage of 0.2%.

[0134] In the late stage, focus on rainfall factors, when the daily cumulative precipitation is ≥30mm, reinforce the cowshed roof and clean the drainage channel to prevent water accumulation.

[0135] When the average daily wind speed is ≥7m / s, close the side windows of the cowshed and install windproof roller shutters to reduce the entry of cold air.

[0136] Real-time collection of meteorological data and physiological data of dairy cows (such as body temperature, feeding conditions), when abnormal respiratory rate (higher than 60 times / minute) or feeding amount decrease by more than 15% is detected, automatically trigger emergency control measures such as increasing the frequency of spraying or adjusting the nutritional ratio of feed.

[0137] Table 1: Evaluation criteria for the degree of heat stress in dairy cows.

[0138]

[0139] Construct a breeding benefit evaluation model to quantify the impact of weather conditions on key breeding indicators, and optimize breeding plans and resource allocation based on the evaluation results. The process includes:

[0140] Select the estrus rate, conception rate, calf survival rate, and milk yield as key breeding indicators to construct a breeding benefit evaluation model.

[0141] Take key meteorological factors as input variables and key indicators as output variables, use BP neural network to construct the model, set the network structure to 3-19-4 (3 nodes in the input layer corresponding to key meteorological factors, 19 nodes in the hidden layer, and 4 nodes in the output layer corresponding to key indicators), select Sigmoid function as the activation function, set the learning rate to 0.01, and set the number of iterations to 1000 times.

[0142] Train the model through historical data (1000 sets of meteorological-breeding data) to control the model prediction error within 8%.

[0143] Based on the trained model, calculate the meteorological impact benefit value to quantify the impact of different weather conditions on key indicators, such as a 1°C increase in temperature factor, a 2.1% decrease in estrus rate, and a 1.8% decrease in conception rate.

[0144] In the rainfall factor, a 1m / s increase in average daily wind speed reduces the calf survival rate by 0.5%.

[0145] The formula for calculating the meteorological impact benefit value is expressed as:

[0146]

[0147] In the formula, is the core index value after optimization of meteorological conditions, is the core index value under original meteorological conditions, is the core index weight (heat rate 0.25, pregnancy rate 0.3, calf survival rate 0.2, and milk yield 0.25).

[0148] In summary, the ecological pasture cow breeding and health management method based on meteorological data provided in this embodiment accurately extracts key meteorological factors through principal component analysis, solving the problem of meteorological impact factors. By establishing a quantitative correlation between key meteorological factors and breeding indicators such as heat rate and pregnancy rate using correlation analysis and multiple regression methods, the meteorological suitable threshold for each breeding stage is determined, enabling the transition of breeding management from experience-based judgment to data support.

[0149] By developing meteorologically adapted breeding cow selection and breeding standards, temperature, weather, and rainfall factors are incorporated into the selection and breeding weight, an index system for evaluating stress resistance is constructed, and dynamic screening of replacement cows and lactating cows is achieved. Under high temperature conditions, by monitoring the respiratory frequency increase of replacement cows and the milk yield decrease of lactating cows, and combining the analytic hierarchy process to determine the index weight, a comprehensive index of stress resistance can be calculated to screen out individuals with strong heat tolerance, reducing the loss of production performance of the population under extreme high temperature weather. At the same time, the selection and breeding weight is dynamically adjusted according to the set period to ensure that the cow population continuously adapts to the regional meteorological characteristics, laying the foundation for creating a core population with strong stress resistance and stable production for ecological pastures.

[0150] By integrating historical meteorological data and ARIMA time series models, the temperature trend for a future set period can be predicted, and the correlation between historical heat rate and temperature can be accurately determined to determine the heat period. Then, through the CNN-LSTM artificial intelligence model, the heat characteristics such as the climbing behavior of dairy cows and body temperature changes can be identified, achieving double confirmation of meteorological prediction and behavior recognition.

[0151] By dividing the pregnancy of dairy cows into early, middle, and late stages and developing differentiated meteorological protection programs, a closed-loop management of prediction, prevention, and adjustment is achieved. In the early stage of pregnancy, the heating or spraying system is started for temperature factors, in the middle stage, artificial light is supplemented for insufficient sunlight, humidity is controlled, and antifungal agents are added, and in the late stage, the cowshed is reinforced for heavy rain and strong wind, and wind roller shutters are used to avoid stress reactions caused by meteorological stress from the source.

[0152] The breeding benefit evaluation model constructed based on the BP neural network can quantize the influence of different meteorological conditions on core indexes such as estrus rate and calf survival rate, and provide accurate basis for resource allocation. Through the meteorological influence benefit value output by the model, the pasture can adjust the breeding plan in a targeted manner, and avoid resource waste.

[0153] Based on the same overall inventive concept, the present application also protects an ecological pasture cow breeding and health management system based on meteorological data. The ecological pasture cow breeding and health management system based on meteorological data provided by the present application is described below, and the ecological pasture cow breeding and health management system based on meteorological data described below can be mutually corresponding and referred to the ecological pasture cow breeding and health management method described above.

[0154] Figure 3 is a structural schematic diagram of the ecological pasture cow breeding and health management system based on meteorological data provided by the embodiment of the present application.

[0155] As Figure 3 shown, the ecological pasture cow breeding and health management system based on meteorological data includes a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor includes a data acquisition and feature extraction module, a meteorological threshold determination module, an adapted cow herd screening module, a suitable time period selection module, a meteorological guarantee development module, and a breeding effect evaluation module.

[0156] The data acquisition and feature extraction module is used to acquire meteorological data and cow breeding basic data in real time, and extract key meteorological factors affecting cow breeding by using principal component analysis method.

[0157] The meteorological threshold determination module is used to establish a quantitative correlation between the key meteorological factors and the breeding indexes by correlation analysis and multiple regression method, and to determine the meteorological suitable threshold corresponding to each breeding stage.

[0158] The adapted cow herd screening module is used to develop breeding cow herd selection criteria based on the key meteorological factors and their meteorological suitable thresholds, dynamically evaluate the stress resistance performance of replacement heifers and lactating cows under different meteorological conditions, and adjust the selection weight to realize meteorological adapted cow herd screening.

[0159] The suitable time period selection module is used to predict the estrus period of cows according to meteorological data, and to select a temperature suitable period for mating operation by using artificial intelligence technology to identify estrus behavior.

[0160] The meteorological guarantee development module is used to develop meteorological guarantee schemes for different pregnancy stages, to start environmental control and nutrition adjustment measures according to real-time meteorological data, and to avoid stress reaction.

[0161] The breeding effect evaluation module is used for constructing a breeding benefit evaluation model, quantifying the influence of weather conditions on a breeding core index, and optimizing a breeding plan and resource allocation based on an evaluation result.

[0162] Those skilled in the art can clearly understand the implementation of the embodiments by means of software and necessary general hardware platforms from the above description of the embodiments, and of course, the embodiments can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in the sense of contribution to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, or an optical disc, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0163] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for some technical features thereof; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for managing the reproduction and health of dairy cows in an ecological pasture based on meteorological data, characterized by, The application relates to a method for optimizing dairy cow breeding based on weather data. Real-time weather data and dairy cow breeding basic data are collected and standardized; Key weather factors affecting dairy cow breeding are extracted by principal component analysis, including temperature factors, weather factors and rainfall factors; Quantitative correlation between the key weather factors and breeding indexes is established by correlation analysis and multiple regression method, and weather suitable thresholds corresponding to each breeding stage are determined, including dairy cow estrus rate, conception rate, morbidity and production performance; Based on the key weather factors and their weather suitable thresholds, breeding cow selection standards are formulated, the stress resistance of reserve cows and lactating cows under different weather conditions is dynamically evaluated, and the selection weight is adjusted to realize weather-adapted cow selection; According to weather data, the estrus period of dairy cows is predicted, and artificial intelligence technology is used to identify estrus behavior, and breeding operation is performed in a temperature suitable period; Weather guarantee schemes are formulated according to different pregnancy stages, and environmental regulation and nutrition adjustment measures are started according to real-time weather data to avoid stress reaction; A breeding benefit evaluation model is constructed to quantify the influence of weather conditions on breeding core indexes, and breeding plans and resource allocation are optimized based on the evaluation results.

2. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The weather data includes daily average temperature, daily minimum temperature, daily maximum temperature, daily average relative humidity, daily average wind speed, daily cumulative precipitation, daily sunshine duration and air pressure; the dairy cow breeding basic data includes dairy cow breed, physiological stage, reproduction index, health index and production performance.

3. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The process of extracting key weather factors affecting dairy cow breeding by principal component analysis includes: Collecting weather data and corresponding dairy cow breeding basic data in multiple complete production cycles to construct a multidimensional data matrix; Calculating the mean and standard deviation of each feature in the multidimensional data matrix to obtain a standardized data matrix; Calculating the covariance matrix based on the standardized data matrix, and performing eigenvalue decomposition on the covariance matrix to obtain eigenvalues and corresponding eigenvectors; Determine the number of principal components according to the eigenvalue contribution rate, and stop screening when the cumulative contribution rate reaches the preset proportion.

4. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The process of determining the weather suitable threshold corresponding to each breeding stage includes: Calculating the correlation coefficient of the key weather factors and each breeding index by Pearson correlation analysis; Screening the key weather factors whose correlation coefficients with the breeding indexes reach the preset correlation threshold through correlation test; Establishing a multiple linear regression model with the key weather factors as independent variables and the breeding indexes as dependent variables; Solving the regression coefficients by the least square method to minimize the residual sum of squares, and determining the weather suitable threshold of each breeding stage based on the regression model.

5. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The process of formulating breeding cow selection standards includes: Determine the breeding cow basic selection indexes including age, weight and pedigree parameters according to the weather suitable threshold of each breeding stage; For the temperature factor, develop a heat tolerance scoring standard by detecting the respiratory rate and rectal temperature of dairy cows under specific temperature conditions; For the weather factor, evaluate the stress resistance of dairy cows by observing the change of their feed intake under specific rainy conditions; For the rainfall factor, observe the activity of dairy cows under specific windy weather to evaluate their adaptability. The key meteorological factors are adapted into the calculation of the breeding weight comprehensive score, and individuals with a comprehensive score higher than a set score are selected into the breeding herd.

6. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The process of selecting a herd adapted to the weather includes: Selecting heifers and lactating cows as evaluation samples; An index system for evaluating stress resistance performance is constructed, which includes physiological indicators, production indicators, and health indicators; the physiological indicators include respiratory rate and rectal temperature; the production indicators include milk yield and milk protein rate; the health indicators include morbidity and somatic cell count; Tracking and monitoring are carried out for a set period of time under different weather conditions; When the temperature factor is in a specific heat stress interval, record the increase in respiratory rate of heifers compared to the appropriate temperature interval, and the decrease in milk yield of lactating cows; When the rainfall factor is in the high impact interval, record the changes in morbidity of both types of cattle; The weights of each evaluation index in the stress resistance performance evaluation index system are determined using the analytic hierarchy process; The comprehensive index of stress resistance performance is calculated, and the breeding weight is adjusted according to the comprehensive index; Through dynamic weight adjustment, select once every set period to achieve optimization of the herd adapted to the weather.

7. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The process of selecting a temperature suitable period for breeding operation includes: Collect historical weather data and mark the corresponding cow estrus rate; According to the historical weather data, use the ARIMA model in time series analysis to predict the trend of key meteorological factors in the future for a set period of time; Determine the optimal model parameters through AIC criterion to predict the daily average temperature change curve of the temperature factor; Combine the correlation between cow estrus rate and temperature factor in historical data, and when the predicted daily average temperature is in a specific interval, determine it as an estrus period; Install high-definition cameras and infrared sensors on the farm to collect cow behavior data and physiological data; the behavior data includes activity frequency, standing time, and mounting behavior; the physiological data includes body temperature and activity amount; Build a CNN-LSTM-based artificial intelligence recognition model, input the cow behavior data and physiological data, and identify the cow estrus behavior; According to the temperature prediction results and the estrus behavior identification results, select a set period of time after the cow estrus when the daily average temperature is in a specific interval for breeding operation.

8. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The process of developing weather protection schemes for different pregnancy stages and adjusting environmental control and nutrition measures according to real-time weather data to avoid stress reactions includes: Divide the cow pregnancy process into early, middle, and late stages, and develop weather protection schemes for different stages; In the early stage, monitor the temperature factor; when the real-time daily average temperature is lower than a set value, start the cowshed heating system to control the indoor temperature to a specific interval, and adjust the feed formula to increase the proportion of energy feed to improve the cold resistance of cows; when the temperature is higher than the set value, start the combined system of spraying and fan, set the spraying interval and spraying time, and adjust the fan speed; In the middle stage, the weather factors are concerned. When the sunshine duration is less than a set value, the artificial light is supplemented to increase the light time by a set duration. When the average relative humidity is higher than a set value, the dehumidification equipment is started to control the humidity in the shed below the set value, and the antifungal agent is added in the feed with a set amount; In the late stage, the rainfall factor is monitored. When the daily cumulative precipitation reaches a set value, the shed roof is reinforced, and the drainage channel is cleaned to prevent water accumulation. When the average wind speed reaches a set value, the shed side window is closed, and the windproof roller blind is installed to reduce the cold air entering; The real-time meteorological data and dairy cow physiological data are collected. When the abnormal respiratory rate or the feed intake reduction of the dairy cow reaches a set proportion, the emergency control measures are automatically triggered.

9. The method of claim 1, wherein the weather data-based eco-pasture dairy cow reproduction and health management method is characterized by, The breeding benefit evaluation model is constructed to quantify the influence of meteorological conditions on the breeding core indicators. The process includes: The estrus rate, conception rate, calf survival rate, and milk yield are selected as the breeding core indicators. The key meteorological factors are used as input variables, and the core indicators are used as output variables. The breeding benefit evaluation model is constructed by using the BP neural network. The network structure of the breeding benefit evaluation model is set as a specific node combination, including an input layer, a hidden layer, and an output layer. The input layer corresponds to the nodes of the number of key meteorological factors. The hidden layer is set with a number of nodes. The output layer corresponds to the nodes of the number of core indicators. The Sigmoid function is selected as the activation function, and the learning rate and iteration number are set. The breeding benefit evaluation model is trained by using the historical data, and the model prediction error is controlled within a set proportion. Based on the trained model, the influence of different meteorological conditions on the core indicators is quantified by calculating the meteorological influence benefit value.

10. An eco-pasture dairy cow breeding and health management system based on meteorological data, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to realize the ecological ranch dairy cow reproduction and health management method based on meteorological data according to any one of claims 1-9.

Images