Feed regulation and control method and system for beef cattle breeding

Abstract

The invention discloses a feed regulation and control method and system for beef cattle breeding, and relates to the technical field of feed regulation and control, and the method comprises the steps: carrying out grouping management according to individual differences of beef cattle, and collecting feed component data, growth stage data and climate environment data of beef cattle breeding in each group in real time; on the basis of the feeding and defecation amount of the beef cattle, the nutrient substance digestion and metabolism capacity of the beef cattle is evaluated, a dynamic feeding model driven by multi-modal data is constructed, and a feeding mode and an optimal feed ratio are matched; the optimal feed proportion is converted into a control instruction, and equipment is driven to execute feed blending and feeding; beef cattle growth condition data are monitored in real time, and the feed ratio is optimized according to actual deviation feedback. According to the method, the cattle in different physiological states are finely distinguished, multi-dimensional data such as environmental parameters and metabolic capability evaluation are combined, a dynamic feeding model is constructed, the feed conversion efficiency and the stability of growth performance are remarkably improved, and healthy growth of the beef cattle is guaranteed through effective feeding management.

Description

Technical Field

[0001] The present invention relates to the technical field of feed regulation, and particularly to a feed regulation method and system for beef cattle breeding. Background Art

[0002] Beef cattle breeding has significant economic benefits. High-quality beef has high demand and price in the market. With the improvement of people's living standards, the consumption of beef is increasing continuously, making the sales revenue of beef cattle considerable. Beef cattle bred from high-quality breeding cattle and cows can bring additional income. By selecting excellent varieties, improving the breeding efficiency and the quality of beef cattle, the economic benefits can be further increased. In addition to beef, by-products such as cowhide, cow bones, and cow offal also have certain economic value and can be used in fields such as leather manufacturing, bone product processing, and food processing. Cow dung can be processed into organic fertilizer and sold to agricultural producers, which not only increases income but also is beneficial to environmental protection. Beef cattle breeding can drive the development of related industries such as feed production, transportation, slaughtering and processing, and sales, creating more employment opportunities and economic benefits. High-quality beef products are competitive in the international market, and exports can bring foreign exchange income to the country and enterprises.

[0003] During the process of beef cattle feeding, due to the lack of precise nutrition research and advanced detection means, it is difficult to accurately master the nutritional requirements of beef cattle. The understanding of the physiological changes and nutritional metabolism mechanisms during the growth and development process of beef cattle is limited. It is impossible to accurately understand the precise requirements of beef cattle at different growth stages for various nutrients such as protein, energy, minerals, and vitamins. The formulation of feed formulas mostly relies on experience and general reference standards, lacking personalized considerations for specific breeds, regions, and breeding environments. At the same time, there is a lack of effective technologies and equipment for real-time monitoring of the growth status and nutritional status of beef cattle. It is impossible to adjust the nutrient supply in a timely manner according to the actual situation of beef cattle, resulting in insufficient nutrient supply affecting the growth rate, or excessive nutrition causing feed waste. In addition, the evaluation of the nutritional value of feed raw materials is not accurate enough, and it is difficult to accurately calculate the actual content and availability of various nutrients in the feed, which also greatly reduces the scientificity and accuracy of the feed formula. Summary of the Invention

[0004] Based on this, it is necessary to provide a feed regulation method and system for beef cattle breeding in view of the above technical problems.

[0005] In the first aspect, the present invention provides a feed regulation method for beef cattle breeding, including:

[0006] S1. Conduct grouped management according to the individual differences of beef cattle, and collect in real time the feed composition data, growth stage data, and climate environment data of beef cattle breeding in each group.

[0007] S2. Evaluate the nutrient digestion and metabolism ability of beef cattle based on their feed intake and fecal output, and construct a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio;

[0008] S3. Convert the optimal feed ratio into a control command to drive the equipment to perform feed preparation and feeding;

[0009] S4. Real-time monitor the growth status data of beef cattle and optimize the feed ratio according to the actual deviation feedback.

[0010] Furthermore, evaluating the nutrient digestion and metabolism ability of beef cattle based on their feed intake and fecal output, and constructing a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio includes:

[0011] S21. Set a feeding cycle, regularly collect the dry matter mass of feed and the dry matter mass of feces within a single feeding cycle, and calculate the energy utilization efficiency of beef cattle;

[0012] S22. Based on the growth stage and breeding requirements of beef cattle, set multiple types of feeding modes and adapt the types and ratio ranges of feed components under various feeding modes;

[0013] S23. Based on multi-modal data fusion, analyze and quantify the structure-activity relationship between energy utilization efficiency and different feed ratios under multi-factor conditions, and construct a dynamic feeding model;

[0014] S24. According to the energy utilization efficiency and growth stage of beef cattle in the current feeding cycle, match the corresponding feeding mode, and calculate and output the optimal feed ratio through the dynamic feeding model.

[0015] Furthermore, setting a feeding cycle, regularly collecting the dry matter mass of feed and the dry matter mass of feces within a single feeding cycle, and calculating the energy utilization efficiency of beef cattle includes:

[0016] S211. Analyze the digestion and absorption ability of beef cattle for different feed components according to the ratio of the dry matter mass of feed to the dry matter mass of feces within a single feeding cycle to obtain the absorption efficiency;

[0017] S212. Based on the absorption efficiency of beef cattle in the current feeding cycle, obtain the dry matter absorption amount of feed within a single feeding cycle, and combine the energy loss due to feed fermentation and the energy loss due to the physiology of beef cattle within a single feeding cycle to calculate the energy utilization efficiency.

[0018] Furthermore, the calculation formula for the energy utilization efficiency is:

[0019]

[0020] where U eRepresents the energy utilization efficiency; F e Represents the energy loss of feed fermentation; H e Represents the energy of physiological loss of beef cattle; D e Represents the absorption efficiency; D in Represents the dry matter mass of feed; D out Represents the dry matter mass of feces.

[0021] Furthermore, based on multi-modal data fusion, analyze and quantify the structure-activity relationship between energy utilization efficiency and different feed ratios under multi-factor conditions, and construct a dynamic feeding model including:

[0022] S231. Obtain the feed composition data, growth stage data, climate environment data and energy utilization efficiency of beef cattle in the current feeding cycle, and perform cleaning, outlier and standardization preprocessing in turn;

[0023] S232. Construct a multiple linear regression equation, with the energy utilization efficiency as the target variable, and quantify the linear effects of different feed components, growth parameters and environmental parameters on the energy utilization efficiency respectively;

[0024] S233. Use the core variables selected by the multiple linear regression as input features, construct a non-linear prediction model using a support vector machine, map to a high-dimensional space using a radial basis kernel function, and optimize the penalty factor and kernel parameters through grid search to minimize the classification interval error;

[0025] S234. Adopt an improved grey wolf optimization algorithm, combine the optimization objective and constraint conditions, and construct a dynamic feeding model to predict the optimal feed ratio under different input variable conditions.

[0026] Furthermore, adopt an improved grey wolf optimization algorithm, combine the optimization objective and constraint conditions, and construct a dynamic feeding model including:

[0027] S2341. Set the scale of the grey wolf population, generate the initial grey wolf positions using logistic chaos mapping, and linearly map the chaos sequence to the feed ratio parameter range;

[0028] S2342. Use the energy utilization efficiency as the fitness value, take maximizing the energy utilization efficiency as the optimization objective, and introduce cost constraints, nutritional balance constraints and neutral detergent fiber constraints;

[0029] S2343. Dynamically allocate weights according to the fitness values of α, β, δ wolves, combine a non-linear convergence factor to optimize the position update formula, and ordinary grey wolves update their own positions according to the weighted positions of α, β, δ wolves;

[0030] S2344. Limit the feed component ratio within the ratio range, if it exceeds, reset it to the boundary value, and if the feed ratio violates the nutritional balance constraint, add a penalty term to the fitness function;

[0031] S2345. Calculate the fitness values of all gray wolf individuals, reorder the population according to the fitness values, and update the weighted positions of the α, β, and δ wolves.

[0032] S2346. Determine whether the iteration termination condition is reached. If it is reached, stop the iteration and output the feed ratio parameters corresponding to the α wolf as the optimal feed ratio. If not, continue the iteration.

[0033] Furthermore, the calculation formula of the position update formula is:

[0034]

[0035] D = |C·X prey - X(t)|, A = 2a·r1 - a, C = 2·r2;

[0036] In the formula, X(t + 1) represents the position of the gray wolf individual at time t + 1; X(t) represents the position of the gray wolf individual at time t; X α represents the position of the α wolf; X β represents the position of the β wolf; X δ represents the position of the δ wolf; A and C both represent dynamic adjustment coefficients; D represents the distance between the wolf and the prey; r1 and r2 represent random numbers; a represents a non - linear convergence factor; X prey represents the position of the prey.

[0037] Furthermore, according to the energy utilization efficiency and growth stage of beef cattle in the current feeding cycle, match the corresponding feeding mode, and calculate through the dynamic feeding model. The output optimal feed ratio includes:

[0038] S241. Based on the feed composition data, growth stage data, climate environment data, and energy utilization efficiency of beef cattle in the current feeding cycle, use the energy prediction formula to calculate the energy required by beef cattle in the next feeding cycle.

[0039] S242. According to the current growth stage data of beef cattle, match the corresponding feeding mode. The feeding modes include high - protein induction mode, high - roughage restriction mode, transitional balance mode, high - energy intensification mode, and full - stage high - concentrate mode.

[0040] S243. Obtain the types and ratio ranges of the feed components corresponding to the feeding mode, set the input variables of the dynamic feeding model, and solve to calculate the optimal feed ratio for the next feeding cycle.

[0041] Furthermore, the energy prediction formula is:

[0042]

[0043] In the formula, DE represents the energy required for the growth of beef cattle in the next feeding cycle; B w represents the body weight of beef cattle; k represents the energy requirement constant per kilogram of body weight; m represents the climate impact coefficient; T r represents the climate environment data; E ue represents the expected energy utilization efficiency.

[0044] Second, the present invention also provides a feed regulation system for beef cattle breeding, which includes:

[0045] A data acquisition module, which is used to conduct grouped management according to the individual differences of beef cattle and collect in real time the feed composition data, growth stage data, and climate environment data of beef cattle breeding in each group;

[0046] An evaluation and calculation module, which is used to evaluate the nutrient digestion and metabolism ability of beef cattle based on the intake and defecation amount of beef cattle, and construct a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio;

[0047] A feeding control module, which is used to convert the optimal feed ratio into a control instruction to drive the equipment to perform feed preparation and feeding;

[0048] A feedback regulation module, which is used to monitor the growth status data of beef cattle in real time and optimize the feed ratio according to the actual deviation feedback.

[0049] The beneficial effects of the present invention are as follows:

[0050] 1. Based on the individual differences and growth stage characteristics of beef cattle, through the grouped management mechanism, refined differentiation of beef cattle in different physiological states is achieved. Combining multi-dimensional data such as environmental parameters, growth stages, and metabolic ability evaluations, a dynamic feeding model is constructed; by real-time analyzing the matching relationship between nutritional requirements and feed components, the concentrate-roughage ratio and energy-protein ratio are dynamically adjusted to ensure the precise adaptation of nutrient supply to the metabolic ability of beef cattle; breaking the single mode of traditional extensive feeding, through data fusion and algorithm optimization, the transformation of nutritional supply from "experience-driven" to "demand-driven" is realized, significantly improving the feed conversion efficiency and the stability of growth performance. Effective feeding management not only ensures the healthy growth of beef cattle, but also helps to reduce greenhouse gas emissions and environmental pollution, conforms to the concept of green development, and promotes the development of sustainable agriculture.

[0051] 2. By analyzing the energy contributions of different feed ingredients, the feed formula can be effectively adjusted to maximize the energy utilization efficiency, ensuring that beef cattle obtain balanced nutrition at different growth stages. By generating specific feed ratios with high energy utilization efficiency, farmers can flexibly apply the optimal feeding plan to improve the feeding efficiency. It can not only increase the growth rate of beef cattle but also reduce feed waste, thus achieving more economical farming. By converting the feed ratio into specific control instructions and automatically controlling the feed delivery equipment for feed preparation, the error of manual operation is effectively reduced, ensuring the accuracy and timeliness of feed delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The drawings described herein are used to provide a further understanding of the present invention and form a part of the present invention. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0053] Figure 1 is a flowchart of a feed regulation method for beef cattle breeding according to an embodiment of the present invention;

[0054] Figure 2 is a system principle block diagram of a feed regulation system for beef cattle breeding according to an embodiment of the present invention.

[0055] Reference numerals in the drawings: 1, data acquisition module; 2, evaluation and calculation module; 3, feeding control module; 4, feedback adjustment module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0057] Please refer to Figure 1 , which provides a feed regulation method for beef cattle breeding, including:

[0058] S1. Conduct group management according to the individual differences of beef cattle, and collect the feed ingredient data, growth stage data and climate environment data of beef cattle breeding in each group in real time.

[0059] In the description of the present invention, the core of beef cattle group management lies in dynamically dividing groups according to individual differences to achieve precise management. The grouping criteria mainly include the following dimensions: 1. Age and growth stage: Beef cattle are divided into different groups such as calf stage (0 - 6 months old), growing stage (6 - 12 months old), early fattening stage (400 - 600 kg), and late fattening stage (above 600 kg), to match different nutritional requirements. 2. Body weight and breed: Grouping is carried out according to the body weight range (such as 60 - 120 kg in the development stage group and above 450 kg in the fattening stage group) and breed characteristics (such as local breeds being resistant to roughage and imported breeds growing fast).

[0060] For beef cattle in different groups, it is necessary to collect in real time the influencing data during the growth process of beef cattle. The influencing data includes feed composition data, growth stage data of beef cattle, and climate environment data.

[0061] The process of collecting influencing data in real time includes: obtaining feed composition data through sensors and laboratory analysis. The feed composition data includes crude protein, crude fat, crude fiber, and neutral detergent fiber. Using monitoring equipment to track the growth stage data of beef cattle in real time. The growth stage data includes weight and height. Through meteorological sensors, the climate environment data of the beef cattle growth area is collected in real time. The climate environment data includes temperature, humidity, and light intensity data.

[0062] To collect feed composition data, sensors and data collectors need to be configured. The sensors use near-infrared spectroscopy sensors and chemical analysis instruments to analyze the feed composition in real time. The data collector is used to collect sensor data and perform preliminary processing. During the data collection process, samples are regularly collected from the feed storage tank. The near-infrared spectroscopy technology is used to measure the contents of crude protein, crude fat, crude fiber, and neutral detergent fiber in the samples. Chemical analysis is carried out on some samples to correct and verify the sensor data.

[0063] To collect growth stage data, automatic weighing equipment and height measurement equipment need to be configured. The automatic weighing equipment is installed in the beef cattle breeding area to weigh the beef cattle regularly. The height measurement equipment uses devices such as laser rangefinders to record the height changes of beef cattle. During the data collection process, the automatic weighing equipment is used to record the weight of beef cattle regularly. The height of beef cattle is recorded daily, and fixed-point measurement is carried out using a laser rangefinder. The weight and height data are integrated to form the growth stage data of beef cattle.

[0064] To collect climate environment data, meteorological sensors need to be configured. Temperature, humidity, and light intensity sensors are installed to collect environmental data in real time. The meteorological sensors operate 24 hours a day to record the climate environment data at any time. The collected data is filtered for noise and corrected to ensure the accuracy of the data.

[0065] By collecting data on feed ingredients, beef cattle growth stages, and climate environment in real time, it provides an accurate and timely information basis for the entire system. It can help breeding managers quickly identify key factors affecting the growth of beef cattle and adjust feeding programs in a timely manner. Ensure the scientificity and reliability of subsequent analysis, thus laying a solid foundation for improving the growth rate of beef cattle and the utilization efficiency of feed.

[0066] S2. Based on the feed intake and fecal output of beef cattle, evaluate the nutrient digestion and metabolism ability of beef cattle, and construct a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio.

[0067] In the description of the present invention, based on the feed intake and fecal output of beef cattle, evaluating the nutrient digestion and metabolism ability of beef cattle, and constructing a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio includes:

[0068] S21. Set a feeding cycle, regularly collect the dry matter mass of feed and the dry matter mass of feces within a single feeding cycle, and calculate the energy utilization efficiency of beef cattle.

[0069] In the description of the present invention, setting a feeding cycle, setting a feeding cycle, regularly collecting the dry matter mass of feed and the dry matter mass of feces within a single feeding cycle, and calculating the energy utilization efficiency of beef cattle includes:

[0070] S211. According to the ratio of the dry matter mass of feed to the dry matter mass of feces within a single feeding cycle, analyze the digestion and absorption ability of beef cattle for different feed ingredients to obtain the absorption efficiency.

[0071] Specifically, in the process of analyzing the nutrient digestion and metabolism function of beef cattle, it is necessary to first collect the dry matter mass of feed fed to beef cattle per unit time and the dry matter mass of feces excreted per unit time, and according to the ratio of the difference between the dry matter mass of feed and the dry matter mass of feces excreted to the dry matter mass of feed, analyze the digestion and absorption ability of beef cattle for different feed ingredients to obtain the absorption efficiency. The formula is expressed as:

[0072]

[0073] In the formula, D e represents the absorption efficiency; D in represents the dry matter mass of feed; D out represents the dry matter mass of feces.

[0074] S212. Based on the absorption efficiency of beef cattle in the current feeding cycle, obtain the dry matter absorption amount of feed within a single feeding cycle, and combine the energy loss due to feed fermentation and the energy loss due to the physiology of beef cattle within a single feeding cycle to calculate the energy utilization efficiency.

[0075] In the description of the present invention, the calculation formula for energy utilization efficiency is as follows:

[0076]

[0077] In the formula, U e represents the energy utilization efficiency; F e represents the energy loss of feed fermentation; H e represents the energy loss of beef cattle's physiological loss; D e represents the absorption efficiency; D in represents the dry matter mass of the feed; D out represents the dry matter mass of the feces.

[0078] By analyzing the factors affecting the digestion and metabolism of nutrients in beef cattle and calculating the energy utilization efficiency, profound insights into the health and growth of beef cattle are provided. This helps farmers identify high-efficiency and low-efficiency feed ratios, thereby adjusting feeding strategies, optimizing nutrient supply, and ensuring that beef cattle are in the best growth state.

[0079] S22. Based on the growth stage and breeding requirements of beef cattle, set multiple types of feeding modes and adapt the types and ratio ranges of feed ingredients under each feeding mode.

[0080] Specifically, beef cattle need to be fed with feeds of different ingredients and ratios at different growth stages. Multiple feeding modes can be set according to their developmental age. For example, the feeding modes include high-protein induction mode, high-roughage restriction mode, transitional balance mode, high-energy intensification mode, and high-concentrate mode for the whole stage. The basis of each feeding mode and the feed ingredients and ratio ranges can refer to the following aspects:

[0081] I. Calf stage (0 - 6 months old)

[0082] 1.1 Feeding mode: High-protein induction feeding

[0083] 1.2 Basis: The rumen of calves is not fully developed. High-protein feeds (such as breast milk, milk powder) are needed to promote muscle and bone development, and at the same time, easily digestible roughage is gradually introduced to train the feeding ability.

[0084] 1.3 Feed ingredients and ratio ranges

[0085] Concentrate feed: Mainly calf pellets, with protein ≥ 18%. The ratio range includes extruded corn (40% - 45%); soybean meal (20% - 25%); whey powder (8% - 10%); premix (5% - 6%).

[0086] Roughage: High-quality green hay (such as alfalfa hay) accounts for 20% - 30%, chopped to 3 - 5 cm.

[0087] Additives: Vitamin A / D / E (0.5 - 1 kg per ton of feed), probiotics (yeast culture).

[0088] II. Growing stage (6 - 12 months old, body weight 200 - 400 kg)

[0089] 2.1 Feeding mode: High roughage restricted feeding

[0090] 2.2 Basis: At this stage, it is necessary to give priority to building the skeleton, control energy intake to avoid premature obesity, and promote rumen development and the ability to digest roughage.

[0091] 2.3 Feed ingredients and proportion range

[0092] Concentrate feed: Accounting for 0.7% - 0.8% of the daily diet (by body weight), the proportion range includes corn (45% - 50%); soybean meal (15% - 18%); DDGS (corn distillers grains, 15% - 20%); wheat bran (10% - 12%).

[0093] Roughage: Silage corn straw (60% - 70%), ammoniated wheat straw (20% - 30%), neutral detergent fiber (NDF) ≥ 28%.

[0094] Additives: Bone meal (1% - 2%), table salt (0.5% - 0.8%), urea is prohibited.

[0095] III. Early fattening stage (body weight 400 - 600 kg)

[0096] 3.1 Feeding mode: Transitional balanced feeding

[0097] 3.2 Basis: Gradually increase the proportion of energy feed, promote the transition of muscle to fat deposition, and prevent metabolic diseases (such as rumen acidosis).

[0098] 3.3 Feed ingredients and proportion range

[0099] Concentrate feed: Accounting for 1% - 1.2% of the daily diet (by body weight), the proportion range includes corn (55% - 60%); cottonseed cake (12% - 15%); wheat bran (12% - 15%); expanded urea (0.3% - 0.5%).

[0100] Roughage: Silage (50% - 55%), hay (30% - 35%), crude fiber ≥ 20%.

[0101] Additives: Baking soda (1% - 1.5%), trace element premix (zinc, selenium).

[0102] IV. Late fattening stage (body weight above 600 kg)

[0103] 4.1 Feeding mode: High - energy intensive feeding

[0104] 4.2 Basis: Accelerate fat deposition to improve the quality of marbled meat, while controlling costs and shortening the slaughter cycle.

[0105] 4.3 Feed ingredients and proportion range

[0106] Concentrate feed: accounting for 1.5% - 1.8% of the daily ration (by body weight), the proportion range includes corn (65% - 70%); cottonseed cake (18% - 20%); oil (such as calcium palmitate, 2% - 3%).

[0107] Roughage: distillers grains (30% - 40%), chopped straw (10% - 15%), and the dry matter of the daily ration ≥ 50%.

[0108] Additives: rumensin (300 - 360 mg / day), mirabilite (0.4% - 0.5%).

[0109] V. Continuous fattening method (from weaning to slaughter)

[0110] 5.1 Feeding mode: High-concentrate feeding throughout the whole stage

[0111] 5.2 Basis: Suitable for the demand of quick slaughter, with the concentrate feed ratio over 50%, shortening the feeding cycle.

[0112] 5.3 Feed ingredients and proportion range

[0113] Concentrate feed: accounting for 50% - 60% of the daily ration, the proportion range includes corn (60% - 65%); soybean meal (15% - 18%); wheat bran (10% - 12%); premix (3% - 5%).

[0114] Roughage: silage corn (30% - 40%), ammoniated straw (20% - 25%).

[0115] Additives: vitamin AD3 powder (1 kg per ton), ferrous sulfate (0.1% - 0.2%).

[0116] S23. Based on multi-modal data fusion, analyze and quantify the structure-activity relationship between energy utilization efficiency and different feed ratios under multi-factor conditions, and construct a dynamic feeding model.

[0117] In the description of the present invention, based on multi-modal data fusion, analyzing and quantifying the structure-activity relationship between energy utilization efficiency and different feed ratios under multi-factor conditions, constructing a dynamic feeding model includes:

[0118] S231. Obtain the feed ingredient data, growth stage data, climate environment data and energy utilization efficiency of beef cattle in the current feeding cycle, and perform cleaning, outlier and standardization preprocessing in sequence.

[0119] Specifically, for the preprocessing of feed ingredient data, it includes removing duplicate entries of feed raw material test records (such as duplicate values of indicators like crude protein and neutral detergent fiber (NDF)), and retaining the latest results of near-infrared spectroscopy or chemical tests in the laboratory. Unify the units of feed ingredients from different sources (such as unifying the crude protein content to percentage or g / kg), and supplement key indicators such as calcium-phosphorus ratio and acid detergent fiber (ADF) that are missing. According to the beef cattle nutritional requirement standards (such as crude protein ≥ 10% and NDF ≤ 40% during the fattening period), eliminate abnormal data outside the reasonable range (such as straw raw materials with NDF test values higher than 50%). Verify the data rationality through an energy-protein balance model (for example, the combination of high-energy concentrate and low-fiber roughage needs to meet the rumen health threshold). Finally, standardize the nutritional component data using the Z-score or Min-Max method to eliminate the dimension difference (such as mapping the calcium-phosphorus ratio from 1.5 - 2:1 to the 0 - 1 interval) for convenient model input.

[0120] For the preprocessing of growth stage data, it is necessary to integrate fields such as body weight, age, and breed (such as uniformly converting "month age" to "day age"), and eliminate data with incorrect formats or logical contradictions (such as a body weight of 200 kg but marked as the calf stage). Supplement continuous body weight records missing due to equipment failures through interpolation (such as estimating missing values based on the daily weight gain trend of adjacent dates). According to the breed growth curve (such as the daily weight gain of Simmental cattle is 1.2 - 1.8 kg), eliminate abnormal body weight data beyond the physiological limit (such as a single-day weight gain > 3 kg). Combine the health records to exclude abnormal growth stagnation or sudden increase records caused by diseases. Finally, discretize the growth stage into categorical variables (such as encoding the calf stage = 1 and the fattening stage = 2), and unify the timestamp format (such as synchronizing to UTC time).

[0121] For the preprocessing of climate environment data, it is necessary to correct the deviations caused by drift or failure of sensors such as temperature, humidity, and light intensity (such as error correction of temperature data with an error of ±2°C). Match the discrete environmental data (such as collected every 30 minutes) with the time points of feeding records to fill in the breakpoint data. Eliminate extreme temperature and humidity values (such as the cattle shed temperature > 35°C or < -10°C), and verify the rationality by combining historical meteorological data. Set safety thresholds for parameters such as methane concentration and ammonia concentration (such as triggering an alarm when methane ≥ 5%), and mark the data of abnormal periods. Finally, perform Kriging interpolation on multi-point sensor data to generate a heat map of the cattle shed environment distribution and convert it into standardized grid data.

[0122] S232. Construct a multiple linear regression equation, with energy utilization efficiency as the target variable, and quantify the linear effects of different feed ingredients, growth parameters, and environmental parameters on energy utilization efficiency respectively.

[0123] Specifically, taking the linear model between different feed components and energy utilization efficiency as an example, a linear model between feed component data and energy utilization efficiency is constructed through multiple linear regression, and the formula is expressed as:

[0124] E ue = a + b1*C p + b2*C f + b3*E e + b4*N df +…+ b n *x n ;

[0125] In the formula, E ue represents the expected energy utilization efficiency; a represents the intercept of the linear model; b1, b2, …, b n represent the regression coefficients of the effects of each feed component on energy utilization efficiency; C p represents the crude protein component, C f represents the crude fiber component, E e represents the crude fat component, N df represents the neutral detergent fiber component; x n represents other feed components.

[0126] Use program compilation software (such as Python software) to fit the feed component data and the linear model formula, and obtain the values of the regression coefficients through the least squares method, including: setting the independent variable and the dependent variable, the independent variable represents each feed component, including crude protein, crude fiber, crude fat, and neutral detergent fiber, etc., and the dependent variable is the expected energy utilization efficiency.

[0127] Use the pandas library in Python to process the independent variable and the dependent variable, and use the LinearRegression class in scikit-learn for linear regression fitting. Use pandas to load the data and read the data of feed components and energy utilization efficiency into a DataFrame. Extract the independent variable and the dependent variable from the data. The independent variable is a DataFrame containing all feed components; the dependent variable is a Series or array representing the energy utilization efficiency.

[0128] Instantiate an object of the linear regression model. Use the training data to call the fit() method for model fitting. The model calculates the most suitable regression coefficients and intercept to minimize the error between the predicted value and the actual value. After fitting, obtain the regression coefficients and intercept through the attributes of the model. Use the cross-validation method to evaluate the prediction ability of the model.

[0129] By using the above statistical analysis method to study the influence of feed ingredients on energy utilization efficiency, and then constructing a linear relationship, it can help farmers select the best feed formula, achieve the optimal allocation of feed and nutrients, reduce feeding costs, and improve economic benefits. In addition, the clear linear relationship helps to predict the performance of different feed ratios in beef cattle growth and enhance the scientificity of feed use.

[0130] S233. Use the core variables selected by multiple linear regression as input features, construct a non-linear prediction model using a support vector machine, map to a high-dimensional space using a radial basis kernel function, and optimize the penalty factor and kernel parameters through grid search to minimize the classification interval error.

[0131] S234. Adopt an improved grey wolf optimization algorithm, combine the optimization objectives and constraint conditions, and construct a dynamic feeding model to predict the optimal feed ratio under different input variable conditions.

[0132] Specifically, the convergence factor a of the traditional grey wolf algorithm decreases linearly, which may lead to insufficient global search or premature local development. After improvement, a non-linear decreasing strategy based on the cosine law is adopted. In the initial stage, a larger a value is maintained to enhance the global search ability, and in the middle and later stages, a decreases rapidly to accelerate convergence to the optimal solution. For example, in the initial stage of iteration (the first 30% of the iteration times), the a value decreases slowly to ensure full exploration of different ratio combinations; in the later stage (the remaining 70%), it decays rapidly to finely adjust the formula parameters.

[0133] In addition, in the traditional grey wolf algorithm, the guiding weights of the positions of the α, β, and δ wolves for the ordinary wolf (ω) are fixed, which may ignore individual fitness differences. After improvement, the weights are dynamically allocated according to the fitness values of the α, β, and δ wolves (i.e., the energy utilization efficiency η). The higher the fitness of an individual, the greater its weight. At the same time, the Euclidean distance is introduced to calculate the proportional weight, so that the grey wolves closer to the optimal solution have a greater influence on the moving direction of the group. For example, if the fitness of the α wolf is significantly higher than that of the β and δ wolves, its weight ratio is increased to 60%-70% to dominate the search direction.

[0134] In the description of the present invention, adopting the improved grey wolf optimization algorithm, combining the optimization objectives and constraint conditions, the construction of the dynamic feeding model includes:

[0135] S2341. Set the scale of the grey wolf population, generate the initial positions of the grey wolves using logical chaotic mapping, and linearly map the chaotic sequence to the range of feed ratio parameters.

[0136] Specifically, a logical chaotic mapping is used to generate the initial gray wolf population. The ergodicity and randomness of the chaotic sequence are utilized to prevent the algorithm from falling into local optima. The sequence generated by the chaotic mapping (such as the Tent mapping) is linearly transformed and then mapped to the range of feed ratio parameters (such as the concentrate-to-roughage ratio of 10%-70%, protein of 8%-20%, etc.), ensuring that the initial solutions are evenly distributed within the feasible region and laying a foundation for global search.

[0137] This method breaks through the limitations of traditional random initialization, enhances population diversity, and accelerates convergence.

[0138] S2342. Take the energy utilization efficiency as the fitness value, and maximize the energy utilization efficiency as the optimization objective. Introduce cost constraints, nutritional balance constraints, and neutral detergent fiber constraints.

[0139] Specifically, taking the energy utilization efficiency as the fitness value, optimization needs to be achieved through the joint modeling of the objective function and constraint conditions. And the optimization objective is to maximize the energy utilization efficiency within the feeding period.

[0140] Cost constraint: Limit the usage ratio of high-price raw materials (such as soybean meal), and set an upper limit on the total formula cost through a linear or non-linear cost function.

[0141] Nutritional balance constraint: Set thresholds such as the calcium-to-phosphorus ratio (1.5 - 2:1), crude protein content (such as ≥10% during the fattening period), etc., to ensure rumen health and growth requirements.

[0142] Neutral detergent fiber (NDF) constraint: Set a lower limit of NDF (≥28%) to prevent acidosis, and at the same time avoid the decrease in digestibility caused by excessive crude fiber.

[0143] S2343. Dynamically allocate weights according to the fitness values of α, β, and δ wolves, and optimize the position update formula in combination with a non-linear convergence factor. Ordinary gray wolves update their own positions according to the weighted positions of α, β, and δ wolves.

[0144] In the description of the present invention, the calculation formula of the position update formula is:

[0145]

[0146] D = |C·X prey - X(t)|, A = 2a·r1 - a, C = 2·r2;

[0147] In the formula, X(t + 1) represents the position of the gray wolf individual at time t + 1; X(t) represents the position of the gray wolf individual at time t; X α represents the position of the α wolf; X β represents the position of the β wolf; X δIndicates the position of the δ wolf; both A and C represent dynamic adjustment coefficients; D represents the distance between the wolf and the prey; r1 and r2 represent random numbers; a represents a non-linear convergence factor; X prey Indicates the position of the prey.

[0148] In addition, the calculation formula for the non-linear convergence factor is:

[0149]

[0150] In the formula, t represents the current iteration number; t max Indicates the maximum iteration number.

[0151] The calculation formula for the dynamic weight of the α wolf is:

[0152]

[0153] In the formula, w α Indicates the dynamic weight of the wolf; X i Indicates the position of the i-th grey wolf.

[0154] S2344. Limit the feed ingredient ratio within the ratio range. If it exceeds, reset it to the boundary value. If the feed ratio violates the nutrient balance constraint, add a penalty term to the fitness function.

[0155] S2345. Calculate the fitness values of all grey wolf individuals, reorder the population according to the fitness values, and update the weighted positions of the α, β, and δ wolves.

[0156] S2346. Determine whether the iteration termination condition is reached. If it is reached, stop the iteration and output the feed ratio parameters corresponding to the α wolf as the optimal feed ratio. If it is not reached, continue the iteration.

[0157] Among them, the termination condition is to terminate the iteration when any of the following conditions is met:

[0158] 1. Reach the maximum iteration number;

[0159] 2. The fitness change of the α wolf is less than the threshold (such as 0.001%) in 10 consecutive iterations;

[0160] 3. The positions of all grey wolf individuals converge to the same area (the standard deviation of the Euclidean distance < 0.1).

[0161] S24. According to the energy utilization efficiency and growth stage of beef cattle in the current feeding cycle, match the corresponding feeding mode, and calculate through the dynamic feeding model to output the optimal feed ratio.

[0162] In the description of the present invention, according to the energy utilization efficiency and growth stage of beef cattle in the current feeding cycle, matching the corresponding feeding mode, and calculating through the dynamic feeding model to output the optimal feed ratio includes:

[0163] S241. Calculate the energy required by beef cattle in the next feeding cycle using an energy prediction formula based on the feed composition data, growth stage data, climate environment data, and energy utilization efficiency of beef cattle in the current feeding cycle.

[0164] In the description of the present invention, the energy prediction formula is:

[0165]

[0166] In the formula, DE represents the energy required for the growth of beef cattle in the next feeding cycle; B w represents the body weight of beef cattle; k represents the energy requirement constant per kilogram of body weight; m represents the climate impact coefficient; T r represents the climate environment data; E ue represents the expected energy utilization efficiency.

[0167] S242. Match the corresponding feeding mode according to the current growth stage data of beef cattle. The feeding modes include high-protein induction mode, high-roughage restriction mode, transitional balance mode, high-energy intensification mode, and full-stage high-concentrate mode.

[0168] S243. Obtain the types and ratio ranges of feed components corresponding to the feeding mode, set the input variables of the dynamic feeding model, and solve to calculate the optimal feed ratio for the next feeding cycle.

[0169] S3. Convert the optimal feed ratio into a control instruction to drive the equipment to perform feed preparation and feeding.

[0170] In the description of the present invention, the process of performing feed preparation and feeding according to the control instruction includes: converting the feed ratio into a control instruction, and the formula is expressed as:

[0171]

[0172] In the formula, R i represents the control instruction for the feeding ratio of the i-th type of feed; n represents the total number of feed types.

[0173] According to the control instruction, control the feed dispenser to start, put the corresponding amount of feed into the feeding area of beef cattle, and monitor the operation status of the equipment at the same time. Convert the generated feed ratio into a control instruction, and precisely control the feed feeding through intelligent equipment. The automated control process reduces the error of manual intervention, improves the feeding efficiency and accuracy, ensures that beef cattle can obtain scientific and balanced nutrition at each stage, and promotes its healthy growth.

[0174] S4. Real-time monitor the growth status data of beef cattle and optimize the feed ratio according to the actual deviation feedback.

[0175] In the description of the present invention, the growth condition data of beef cattle are monitored in real time, the deviation value between the growth condition data and the expected growth target is analyzed, and the feed ratio is adjusted according to the actual situation.

[0176] The growth condition data include body weight, diet intake, activity level and health condition.

[0177] Specifically, a variety of sensors are installed in the farm, including electronic scales, cameras, temperature and humidity sensors, activity monitors, etc. The body weight of beef cattle is weighed daily using an electronic scale, and the data is automatically uploaded to the system. The feed intake of beef cattle is monitored through the sensors of the feeder, and the quantity of remaining feed after each feeding is recorded to calculate the daily intake. The activity patterns of beef cattle, including the movement frequency and duration, are analyzed by using video monitoring combined with image recognition technology. The physiological state of beef cattle, such as body temperature, respiratory rate, etc., is monitored by combining a thermometer, health monitoring equipment, observation records, etc.

[0178] An early warning system is set up. If the deviation value is greater than the preset deviation threshold, an early warning is generated to prompt the farm manager to take necessary measures for intervention and adjustment.

[0179] After obtaining the original data, regular analysis is carried out on various indicators. For example, the weight gain of beef cattle, the change of feed intake, the activity level and health indicators are quantitatively evaluated. A series of growth standards are set, such as target body weight, optimal feed intake, activity level range, etc., and the actual data is compared with these standards. By using the difference analysis method, the specific aspects deviating from the standards and their potential causes are identified. According to the results of the comparison between the monitoring data and the standards, the effects of the current feed ratio are evaluated, including energy utilization efficiency, nutritional balance, etc. If it is found that the growth of some beef cattle does not meet the standards, the system will analyze whether the corresponding feed components need to be adjusted and put forward optimization suggestions.

[0180] Example: In beef cattle farm A, there are currently 10,000 beef cattle of the same batch. According to the monitoring equipment, the average weight of the beef cattle is 300 KG and the height is 120 CM. A batch of feed is purchased. After analysis, the crude protein content of this feed is 18%, the crude fat content is 5%, the crude fiber content is 12%, and the neutral detergent fiber content is 30%. After being detected by the meteorological sensor, the average temperature of the farm on a single day is 22 °C, the humidity is 60%, and the light intensity is 800 lux.

[0181] The dry matter mass of the feed fed per unit time is 10 kg, and the dry matter mass of the feces excreted per unit time is 2 kg. According to the absorption efficiency calculation formula, the absorption efficiency is obtained as 80%, that is, the absorption amount of the dry matter of the feed fed per unit time is 8 kg. After detection, the energy loss due to feed fermentation is 1.5 MJ, and the energy loss due to the physiology of beef cattle is 0.5 MJ. According to the energy utilization efficiency calculation formula, the energy utilization efficiency can be obtained as 75%.

[0182] Based on the energy utilization efficiency, combined with the weight and energy requirements of beef cattle, generate a feed ratio: Calculate that the daily energy requirement for beef cattle is 20 MJ, establish a linear programming model to ensure that the feeding plan maximizes the energy utilization efficiency. According to the demand of beef cattle for feed components, calculate the feeding amount of each feed: 2 kg of crude protein, 0.5 kg of crude fat, 1 kg of crude fiber, and 3 kg of neutral detergent fiber are required.

[0183] Convert the feed ratio into a control command to control the feed dispenser to perform mixing and feeding.

[0184] Through the monitoring device, real-time monitor the growth status data of beef cattle, and analyze the deviation from the expected growth target: Assume that the monitored weight gain is 5 kg, which is lower than the expected 6 kg. According to the actual situation, adjust the feed ratio and increase the feeding amount of crude protein to improve the growth effect.

[0185] Please refer to Figure 2 , and also provides a feed regulation system for beef cattle breeding, which includes:

[0186] The data acquisition module 1 is used to manage the grouping according to the individual differences of beef cattle, and collect the feed component data, growth stage data and climate environment data of beef cattle breeding in each group in real time.

[0187] The evaluation and calculation module 2 is used to evaluate the nutrient digestion and metabolism ability of beef cattle based on the intake and defecation amount of beef cattle, and construct a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio.

[0188] The feeding control module 3 is used to convert the optimal feed ratio into a control command to drive the equipment to perform feed mixing and feeding.

[0189] The feedback regulation module 4 is used to monitor the growth status data of beef cattle in real time, and optimize the feed ratio according to the actual deviation feedback.

[0190] In summary, by means of the above technical solutions of the present invention, based on the individual differences and growth stage characteristics of beef cattle, a refined differentiation of cattle in different physiological states is achieved through a grouping management mechanism. Combining multi-dimensional data such as environmental parameters, growth stage, and metabolic capacity assessment, a dynamic feeding model is constructed; by real-time analyzing the matching relationship between nutritional requirements and feed ingredients, the concentrate-roughage ratio and energy-protein ratio are dynamically adjusted to ensure the precise adaptation of nutrient supply to the metabolic capacity of cattle; breaking the single mode of traditional extensive feeding, through data fusion and algorithm optimization, the transformation of nutrient supply from "experience-driven" to "demand-driven" is realized, significantly improving the feed conversion efficiency and the stability of growth performance. Effective feeding management not only ensures the healthy growth of beef cattle, but also helps to reduce greenhouse gas emissions and environmental pollution, conforms to the concept of green development, and promotes the development of sustainable agriculture.

[0191] By analyzing the contribution of different feed ingredients to energy, the feed formula is effectively adjusted to maximize the energy utilization efficiency and ensure that beef cattle obtain balanced nutrition at different growth stages; by generating specific feed ratios with high energy utilization efficiency, farmers can flexibly apply the best feeding plan to improve feeding efficiency; it can not only increase the growth rate of beef cattle, but also reduce feed waste, thus achieving more economical breeding; by converting the feed ratio into specific control instructions and automatically controlling the feed delivery equipment for feed preparation, the error of manual operation is effectively reduced, ensuring the accuracy and timeliness of feed delivery. At the same time, farmers can monitor the feed delivery situation in real time, enhancing the convenience and real-time nature of management. By continuously monitoring the growth status data of beef cattle, an effective closed-loop management system is formed. Through real-time monitoring, changes in the growth process of beef cattle, whether in terms of nutrient intake or health status, can be quickly responded to, so as to timely adjust the feed formula or feeding strategy. This dynamic adjustment ensures the adaptability of breeding activities to environmental changes, is more user-friendly and scientific, reduces feed waste and resource consumption, and helps to achieve sustainable beef cattle breeding. Effective feeding management not only ensures the healthy growth of beef cattle, but also helps to reduce greenhouse gas emissions and environmental pollution, conforms to the concept of green development, and promotes the development of sustainable agriculture.

[0192] It should be understood that although the steps in the flowchart of the accompanying drawings are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit and can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

Claims

1. A feed regulation method for beef cattle breeding, characterized in that, Including: S1. Conduct grouped management according to the individual differences of beef cattle, and collect in real time the feed composition data, growth stage data and climate environment data of beef cattle breeding in each group; S2. Based on the feed intake and defecation amount of beef cattle, evaluate the nutrient digestion and metabolism ability of beef cattle, and construct a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio; S3. Convert the optimal feed ratio into a control command to drive the equipment to perform feed preparation and feeding; S4. Monitor the growth status data of beef cattle in real time, and optimize the feed ratio according to the actual deviation feedback.

2. The feed regulation method for beef cattle breeding according to claim 1, characterized in that The step of evaluating the nutrient digestion and metabolism ability of beef cattle based on the feed intake and defecation amount of beef cattle, and constructing a multi-modal data-driven dynamic feeding model to match the feeding mode and the optimal feed ratio includes: S21. Set a feeding cycle, regularly collect the dry matter mass of feed and the dry matter mass of feces within a single feeding cycle, and calculate the energy utilization efficiency of beef cattle; S22. Based on the growth stage and breeding requirements of beef cattle, set multiple types of feeding modes, and adapt the types and ratio ranges of feed components under various feeding modes; S23. Based on multi-modal data fusion, analyze and quantify the structure-activity relationship between energy utilization efficiency and different feed ratios under multi-factor conditions, and construct a dynamic feeding model; S24. According to the energy utilization efficiency and growth stage of beef cattle in the current feeding cycle, match the corresponding feeding mode, and calculate through the dynamic feeding model to output the optimal feed ratio.

3. A feed regulation method for beef cattle breeding according to claim 2, characterized in that, The step of setting a feeding cycle, regularly collecting the dry matter mass of feed and the dry matter mass of feces within a single feeding cycle, and calculating the energy utilization efficiency of beef cattle includes: S211. According to the ratio of the dry matter mass of feed to the dry matter mass of feces within a single feeding cycle, analyze the digestion and absorption ability of beef cattle for different feed components to obtain the absorption efficiency; S212. Based on the absorption efficiency of beef cattle in the current feeding cycle, obtain the dry matter absorption amount of feed within a single feeding cycle, and combine the energy loss due to feed fermentation and the physiological energy loss of beef cattle within a single feeding cycle to calculate the energy utilization efficiency.

4. A feed regulation method for beef cattle breeding according to claim 3, characterized in that The calculation formula of the energy utilization efficiency is: In the formula, U e represents the energy utilization efficiency; F e represents the energy loss of feed fermentation; H e represents the energy loss of beef cattle physiological loss; D e represents the absorption efficiency; D in represents the dry matter mass of feed; D out represents the dry matter mass of feces.

5. A feed regulation method for beef cattle breeding according to claim 2, characterized in that, The step of based on multi-modal data fusion, analyzing and quantifying the structure-activity relationship between energy utilization efficiency and different feed ratios under multi-factor conditions, and constructing a dynamic feeding model includes: S231. Obtain the feed composition data, growth stage data, climate environment data and energy utilization efficiency of beef cattle in the current feeding cycle, and perform cleaning, outlier and standardization preprocessing in sequence; S232. Construct a multiple linear regression equation, with the energy utilization efficiency as the target variable, and respectively quantify the linear effects of different feed components, growth parameters and environmental parameters on the energy utilization efficiency; S233. Use the core variables selected by multiple linear regression as input features, construct a non-linear prediction model using a support vector machine, map to a high-dimensional space using a radial basis kernel function, and optimize the penalty factor and kernel parameters through grid search to minimize the classification interval error; S234. An improved grey wolf optimization algorithm is adopted, combined with the optimization objective and constraint conditions, to construct a dynamic feeding model for predicting the optimal feed ratio under different input variable conditions.

6. The feed regulation method for beef cattle breeding according to claim 5, wherein, The construction of the dynamic feeding model by adopting the improved grey wolf optimization algorithm, combined with the optimization objective and constraint conditions, includes: S2341. Set the scale of the grey wolf population, generate the initial positions of grey wolves using logical chaotic mapping, and linearly map the chaotic sequence to the feed ratio parameter range. S2342. Take the energy utilization efficiency as the fitness value, take maximizing the energy utilization efficiency as the optimization objective, and introduce cost constraints, nutritional balance constraints, and neutral detergent fiber constraints. S2343. Dynamically allocate weights according to the fitness values of the α, β, and δ wolves, combine the non-linear convergence factor to optimize the position update formula, and the ordinary grey wolves update their own positions according to the weighted positions of the α, β, and δ wolves. S2344. Limit the feed composition ratio within the ratio range. If it exceeds, reset it to the boundary value. If the feed ratio violates the nutritional balance constraint, add a penalty term to the fitness function. S2345. Calculate the fitness values of all grey wolf individuals, reorder the population according to the fitness values, and update the weighted positions of the α, β, and δ wolves. S2346. Determine whether the iteration termination condition is reached. If it is reached, stop the iteration and output the feed ratio parameters corresponding to the α wolf as the optimal feed ratio. If not, continue the iteration.

7. A feed regulation method for beef cattle breeding according to claim 6, characterized in that, The calculation formula of the position update formula is: Wherein, X(t + 1) represents the position of the gray wolf individual at the moment of t + 1; X(t) represents the position of the gray wolf individual at the moment of t; X α represents the position of the α wolf; X β represents the position of the β wolf; X δ represents the position of the δ wolf; A and C both represent dynamic adjustment coefficients; D represents the distance between the wolf and the prey; r1 and r2 represent random numbers; a represents a non-linear convergence factor; X prey represents the position of the prey.

8. A feed regulation method for beef cattle breeding according to claim 2, characterized in that, The matching of the corresponding feeding mode according to the energy utilization efficiency and growth stage of beef cattle in the current feeding cycle, and the calculation by the dynamic feeding model to output the optimal feed ratio includes: S241. Based on the feed composition data, growth stage data, climate environment data, and energy utilization efficiency of beef cattle in the current feeding cycle, use the energy prediction formula to calculate the energy required by beef cattle in the next feeding cycle. S242. According to the current growth stage data of beef cattle, match the corresponding feeding mode, and the feeding mode includes high-protein induction mode, high roughage restriction mode, transitional balance mode, high-energy intensive mode, and full-stage high-concentrate mode. S243. Obtain the types and ratio ranges of the feed components corresponding to the feeding mode, set the input variables of the dynamic feeding model, and solve to calculate the optimal feed ratio in the next feeding cycle.

9. A feed regulation method for beef cattle breeding according to claim 8, characterized in that, The energy prediction formula is: Wherein, DE represents the energy required for the growth of beef cattle in the next feeding cycle; B w represents the body weight of beef cattle; k represents the energy requirement constant per kilogram of body weight; m represents the climate impact coefficient; T r represents climate environment data; E ue represents the expected energy utilization efficiency.

10. A feed regulation system for beef cattle breeding, which is used to implement the feed regulation method for beef cattle breeding described in any one of claims 1-9, characterized in that, The system includes: A data acquisition module for grouping management according to the individual differences of beef cattle and real-time collecting the feed composition data, growth stage data, and climate environment data of beef cattle breeding in each group. An evaluation and calculation module for evaluating the nutrient digestion and metabolism ability of beef cattle based on the intake and defecation amount of beef cattle, constructing a multi-modal data-driven dynamic feeding model, and matching the feeding mode and the optimal feed ratio. A feeding control module for converting the optimal feed ratio into a control instruction to drive the equipment to perform feed preparation and feeding. A feedback adjustment module for real-time monitoring of the beef cattle growth status data and optimizing the feed ratio according to the actual deviation feedback.

Images