A method and system for evaluating and analyzing the nutrition balance of natural grassland grazing livestock

Abstract

The application discloses a kind of natural grassland grazing livestock nutrition balance evaluation analysis method and system, it is related to livestock intelligent evaluation technical field.The multi-season grassland monitoring data and livestock basic data are obtained;The total amount of pasture nutrition supply in each grazing period is calculated;A dynamic model of livestock nutrition consumption is constructed, and the total demand of livestock dynamic nutrition corresponding to each grazing period is calculated;The total amount of pasture nutrition supply and the total demand of livestock dynamic nutrition are compared in time dimension every time period, and the nutrition balance gap in each grazing period is calculated;With grassland ecological protection threshold and livestock nutrition balance as dual constraints, improved genetic algorithm is used to jointly optimize the transfer node, rotation cycle and carrying capacity, and the optimized grazing timing parameters are output.The application realizes the dynamic coupling evaluation of seasonal pasture grass and livestock nutrition, outputs the grazing control parameters that can be directly landed, and is suitable for the grazing management of various natural seasonal pastures.

Description

Technical Field

[0001] This invention relates to the field of intelligent assessment technology in animal husbandry, and in particular to a method and system for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands. Background Technology

[0002] Seasonal pastures are characterized by drastic fluctuations in forage productivity with the seasons, making it difficult to precisely match grassland forage supply with livestock nutritional needs over time. This is one of the core issues restricting the sustainable development of natural grassland animal husbandry.

[0003] Currently, most grassland-livestock balance assessments employ static carrying capacity calculation methods. On one hand, traditional assessment methods, when calculating forage supply, simply average the total annual or seasonal forage production, failing to consider the impact of time-series dynamic parameters such as migration points, rotational grazing cycles, and grazing intensity on the actual available forage. This leads to a severe disconnect between the forage supply assessment results and the actual grazing rhythm of the pastures. For example, the actual available forage for the same herder may differ by orders of magnitude between the early spring migration and the late autumn rotational grazing period, but the static method cannot reflect this dynamic change. On the other hand, existing static calculation methods only perform quantitative matching of "forage production - number of livestock," failing to couple and analyze the nutritional components of grassland forage (such as crude protein and metabolizable energy content) with the dynamic nutritional needs of grazing livestock at different physiological stages (gestation, lactation, and fattening). While some studies have attempted to incorporate forage nutrient quality parameters to correct carrying capacity in quantifying the nutritional balance of livestock under actual grazing conditions, the assessment dimensions remain limited to static carrying capacity calculations and do not extend to the level of nutrient balance. Furthermore, existing methods only output static carrying capacity values ​​and cannot directly provide actionable grazing control parameters such as adjustments to transhumance schedules and optimization suggestions for rotational grazing cycles. This makes it difficult to support closed-loop pasture management encompassing "carrying capacity assessment - timing adjustment - nutrient supplementation." In practice, even if herders know the theoretical carrying capacity, they still cannot determine specific decision parameters such as "when to transhumance," "whether to shorten or extend the rotational grazing cycle," and "how much nutrient supplementation should be given during the cold season." This can easily lead to mismatch between forage and livestock, resulting in grassland degradation, weight loss in livestock, and decreased productivity.

[0004] Furthermore, although some patents have addressed improvements to grassland-livestock balance assessment methods, such as patent CN109329201A which calculates carrying capacity by season and introduces multiple objectives such as digestible protein and metabolizable energy, it is essentially still a "seasonal static accounting" model, without considering transhumance nodes and rotational grazing cycles as dynamic calculation factors; patent CN106372740B uses a genetic algorithm to optimize the water, soil, grassland, and livestock balance in pastoral areas, but this algorithm optimizes static resource allocation schemes and does not involve joint optimization of grazing time parameters; patent CN122286040A focuses on the "current state assessment" of the nutritional balance of the grazing system and outputs the division of nutritional surplus and deficit areas, without focusing on the optimized output of grazing control parameters.

[0005] In summary, there is an urgent need for a nutritional balance assessment and analysis scheme for livestock grazing on natural grasslands that takes into account seasonal grassland dynamics, grazing sequence regulation, and the dynamic nutritional needs of livestock. Therefore, a method and system for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands is proposed. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing static grassland-livestock balance assessment technology and provide a method and system for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands. This method enables dynamic coupling assessment of grassland forage nutrient supply and livestock nutrient consumption under seasonal pasture conditions, and outputs optimal grazing sequence, rotational grazing cycle and carrying capacity control parameters, providing quantitative basis for seasonal pasture grazing management and subsequent precise supplementary feeding.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands, comprising the following steps: Step S1, Multi-source data acquisition Specifically, acquire multi-season grassland monitoring data, rotational grazing data, grazing intensity data, and basic livestock data of the target natural grassland; Preferably, the grassland monitoring data includes at least the grassland type, forage yield, and forage nutrient composition in different seasons; Preferably, the basic livestock data includes livestock breed, physiological stage, and herd size; Preferably, the grassland monitoring data includes at least the grassland type, forage yield, and forage nutrient composition in different seasons; As a further preferred embodiment, the nutritional components of the forage include at least crude protein content and metabolizable energy content; As a further preferred option, forage yield can be obtained through ground quadrat measurement or remote sensing yield estimation models, and forage nutrient composition can be obtained through laboratory testing.

[0008] Preferably, the rotational grazing data includes at least the rotational grazing cycle, transhumance points, and livestock carrying capacity; Preferably, the grazing intensity data includes the number of livestock grazing per unit area and the foraging rate.

[0009] Preferably, the basic livestock data includes at least the livestock breed, herd size, physiological stage, and seasonal environmental parameters. Among these, the livestock breed determines its basal metabolic rate and nutrient conversion coefficient.

[0010] As a further preferred option, the physiological stages include pregnancy, lactation, and fattening period; As a further preferred option, the seasonal environmental parameters include at least temperature and humidity.

[0011] Step S2, Calculation of available nutrients in the grassland Specifically, based on the multi-season grassland monitoring data, the dynamic changes in grassland yield and the concentration of forage nutrients in each season are calculated. Combined with the grazing time sequence parameters determined based on the rotational grazing data and the grazing intensity data, the total amount of forage nutrients that the target grazing livestock can consume during each grazing period is calculated.

[0012] Preferably, the total amount of pasture nutrient supply is calculated as follows: the pasture yield of each season is multiplied by the concentration of pasture nutrients (crude protein content, metabolizable energy content) to obtain the pasture nutrient content per unit area; then, the time axis is divided according to the rotation grazing cycle and the transition node, and the pasture nutrient supply for each grazing period is dynamically allocated in combination with grazing intensity parameters including the number of livestock grazing per unit area and the foraging rate.

[0013] As a further preferred option, the total amount of nutrients that the grassland can supply The calculation formula is: ; in, For the total amount of nutrients supplied to pasture, Let be the forage yield during grazing period t. The concentration of nutrient components in the forage during grazing period t. Let t be the forage utilization rate during the grazing period. Let t be the area of ​​grassland available during the grazing period.

[0014] Step S3, Calculation of dynamic nutritional requirements for livestock Specifically, based on the livestock breeds, herd sizes, physiological stages, and seasonal environmental parameters, a dynamic model of livestock nutrient consumption is constructed to calculate the dynamic total nutrient requirements of grazing livestock herds during each corresponding grazing period.

[0015] Preferably, the input parameters of the livestock nutrient consumption dynamic model include at least: livestock metabolic weight (based on measured weight values). The calculation includes physiological stage coefficients (nutrient requirement adjustment coefficients corresponding to different physiological stages, with different coefficient values ​​assigned to the gestation, lactation, and fattening periods) and environmental temperature correction coefficients (increasing the maintenance energy consumption coefficient when the environmental temperature deviates from the suitable temperature range for livestock).

[0016] As a further preferred option, the dynamic total nutritional requirements of livestock herds The calculation formula is: ; in, Let be the number of livestock in the i-th physiological stage. This is to meet the maintenance metabolic needs of livestock in the i-th physiological stage. This represents the physiological stage coefficient. This is the ambient temperature correction factor. This refers to the production needs of livestock in the i-th physiological stage (such as the needs of fetal development during pregnancy, lactation, and weight gain).

[0017] Step S4, Grass-Livestock Nutrition Coupling Assessment Specifically, the total supply of pasture nutrients is coupled with the total dynamic nutritional demand of livestock by comparing and matching them over time to calculate the nutritional balance gap of livestock in each grazing period of each season.

[0018] As a preferred method, the comparison is performed time-by-time as follows: the grazing season throughout the year is divided into multiple consecutive grazing periods according to the migration nodes and rotational grazing cycles, and the difference between the total supply of pasture nutrients and the total demand of livestock nutrients is calculated independently for each grazing period.

[0019] As a further preferred embodiment, the formula for calculating the nutritional deficit is: ; in, The nutritional deficit during grazing period t. Let be the total amount of nutrients supplied to the pasture during grazing period t. Let be the dynamic total nutritional requirement of livestock during grazing period t. When this occurs, it indicates that the livestock are nutritionally deficient and that the rotational grazing data needs to be adjusted or supplemental feeding should be provided.

[0020] Step S5: Global optimization of rotational grazing data Specifically, with grassland ecological protection threshold and livestock nutrition balance as dual constraints, an improved genetic algorithm is used to perform global joint optimization of rotational grazing data, including at least the transhumance nodes, rotational grazing cycles, and carrying capacity. Furthermore, the mechanism of grass-livestock coupling is analyzed using structural equation modeling, and the optimized rotational grazing data is verified and corrected by combining the data from the location grazing experiment. Finally, the optimized grazing time sequence parameters are output.

[0021] The specific process of the improved genetic algorithm is as follows: a) Encoding and Population Initialization: Real-number encoding is used to encode transhumance nodes (represented by date or number of days), rotational grazing cycles (represented by number of days), and carrying capacity (represented by standard sheep units or livestock head count) into chromosomal genes. An initial population of N individuals is randomly generated, and the rationality of each individual in the population is checked to ensure that it meets the basic grassland carrying capacity constraints.

[0022] b) Fitness Function Construction: The fitness function of the improved genetic algorithm is constructed with multiple objectives: minimizing the nutrient balance gap and maximizing grassland yield utilization. The fitness function expression is: ; in, These are the weighting coefficients. The fitness value represents the sum of absolute values ​​of the nutrient deficit throughout the grazing period, and η represents the grassland production utilization rate (the ratio of actual forage intake to total forage production). The higher the fitness value, the better the rotational grazing data scheme corresponding to that individual.

[0023] c) Selection, crossover, and mutation operations: Individuals are selected from the parent population using roulette wheel selection; the crossover operation uses simulated binary crossover (SBX), and the mutation operation uses polynomial mutation to maintain population diversity.

[0024] d) Constraint handling: A penalty function strategy is used to handle constraints. For individuals that violate grassland ecological protection thresholds (grass production utilization rate must not exceed a preset upper limit) and livestock nutritional balance constraints (nutritional deficit exceeds the allowable threshold during any grazing period), a penalty term is applied to their fitness value to reduce their probability of being selected.

[0025] e) Termination condition: When the preset maximum number of iterations is reached or the optimal fitness value of the population no longer increases within a certain number of consecutive generations, the algorithm terminates and outputs the optimal solution.

[0026] Preferably, during the optimization process of the improved genetic algorithm, a structural equation model is simultaneously constructed to analyze the mechanism of grassland-livestock coupling. The structural equation model reveals the influence mechanism of rotational grazing data on nutrient balance by introducing causal relationship paths between latent variables (pre-selected as grassland health status and comprehensive factors of livestock production performance) and observed variables (preferably forage yield, forage nutrient composition, daily weight gain of livestock, and changes in livestock weight). The obtained path coefficients can be used as the basis for adjusting the constraint boundaries of the genetic algorithm.

[0027] Preferably, the specific method for verifying the fixed-grazing experiment is as follows: Select representative grasslands to set up fixed-grazing experimental plots, conduct actual grazing experiments according to the optimized rotational grazing data output by the system, and measure the changes in livestock weight and actual forage utilization data. Compare the measured data with the model prediction data, calculate the error, and correct the model parameters (pre-selected as forage intake coefficient and livestock nutrient conversion coefficient). After verification through at least one complete grazing cycle, solidify the corrected model parameters into the system to improve the accuracy of subsequent evaluations.

[0028] Step S6, output the result Specifically, the system outputs grassland carrying capacity assessment results (pre-selected as whether the grassland carrying capacity is overloaded during each grazing period), the livestock nutrient balance gap (pre-selected as the specific profit and loss values ​​for each grazing period), and the optimized grazing timing parameters (optimized transhumance nodes, rotational grazing cycles, and carrying capacity) to guide pasture transhumance and rotational grazing management.

[0029] Preferably, an external data interface is set up to output the livestock nutritional balance gap data, especially the nutritional gap values ​​(pre-selected as lack of crude protein and lack of metabolic energy) to the external downstream supplementary feeding assessment system, so as to serve as the input basis for subsequent precise supplementary feeding programs (pre-selected as supplementary feed ration formula and supplementary feed amount).

[0030] Secondly, the present invention also provides a nutritional balance assessment and analysis system for livestock grazing on natural grasslands, used to implement the above-mentioned assessment method, including: The data acquisition module is used to collect multi-season grassland monitoring data (grassland type, forage yield, forage nutrient composition), rotational grazing data (rotational grazing cycle, transhumance, carrying capacity), grazing intensity data (number of livestock per unit area and forage intake rate), and basic livestock information (livestock breed, physiological stage, herd size) and environmental parameters (temperature, humidity). This module supports multiple data access methods, including manual input, automatic sensor acquisition, and import from external databases.

[0031] Grassland nutrient supply calculation module: Connected to the data acquisition module, this module is used to segment the time axis according to the rotational grazing cycle and transhumance nodes, and calculate the grassland carrying capacity threshold and the total amount of nutrients that forage can supply during each grazing period. This module has a built-in forage yield-nutrient composition fusion calculation unit, which outputs the forage supply in terms of crude protein and metabolizable energy for each time period.

[0032] Livestock Nutritional Requirements Calculation Module: This module is used to construct a dynamic model of livestock nutritional consumption. Based on livestock breed, physiological stage, and environmental parameters, it calculates the dynamic total nutritional requirements of a livestock population during different grazing periods. This module includes a built-in database of basal metabolism for different livestock breeds and a table of nutritional requirement coefficients for different physiological stages.

[0033] Grassland-Livestock Nutrition Coupled Assessment Module: This module connects to both the grassland nutrient supply calculation module and the livestock nutrient demand calculation module. It receives grassland nutrient supply data and livestock nutrient demand data time-by-time, compares and calculates the nutrient balance gap for each grazing period. This module outputs a nutrient surplus / deficit time-series curve and a gap statistics table.

[0034] The rotational grazing data optimization module incorporates an improved genetic algorithm and a structural equation model. With grassland ecological protection and livestock nutritional balance as dual constraints, it jointly optimizes transhumance nodes, rotational grazing cycles, and carrying capacity. It also incorporates data from location-based grazing experiments to verify and correct the optimization results. The structural equation model is used to analyze the mechanism of grassland-livestock coupling and assists in setting the constraint boundaries of the improved genetic algorithm.

[0035] The results output module is used to display grassland carrying capacity assessment results, nutrient deficit time series data, and optimized grazing control parameters (transit points, rotational grazing cycles, and carrying capacity). It also has an external data interface for transmitting nutrient deficit data to external supplementary feeding assessment systems. Beneficial effects

[0036] Compared with the prior art, the present invention has the following beneficial effects: This invention incorporates dynamic parameters of grazing timing, such as transhumance nodes, rotational grazing cycles, and carrying capacity, as constraint factors in the calculation of forage supply into the evaluation process. This enables dynamic quantification of the time axis of seasonal pasture nutrient supply and solves the technical problem of the disconnect between traditional evaluation and the actual grazing rhythm of pastures.

[0037] This invention couples the nutritional components of grassland forage with the dynamic nutritional needs of grazing livestock at different physiological stages for time-by-time calculation, outputting the actual nutritional surplus and deficit values ​​of livestock, rather than simply matching the amount of grass produced with the number of livestock, thus filling the technical gap in the nutritional dimension of existing grassland-livestock balance assessment.

[0038] This invention uses an improved genetic algorithm to achieve global joint optimization of transhumance nodes, rotational grazing cycles, and carrying capacity. The results are verified and corrected using data from location-based grazing experiments, and the output is grazing control parameters that can be directly implemented, thereby guiding the scientific closed-loop management of pastures.

[0039] This invention sets up an external data interface that can connect to downstream supplementary feeding diet assessment systems, outputting the crude protein deficit and metabolizable energy deficit in the nutrient gap, providing an input data source for precise supplementary feeding of grazing livestock, opening up the data link between grazing assessment and supplementary feeding technology, and helping to improve forage conversion efficiency and livestock production performance. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the process for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands, as described in this invention. Figure 2 This is a schematic diagram of the structure of a natural grassland grazing livestock nutrition balance assessment and analysis system according to the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0042] Example 1: Overall System Operation Flow See Figure 1 , Figure 2 This embodiment describes the complete operation process of the natural grassland grazing livestock nutrition balance assessment and analysis system of the present invention in a target natural grassland, specifically including the following stages and steps: Phase 1, Data Acquisition Phase: After system startup, the data acquisition module first loads the pre-configured spatiotemporal database of the target natural grassland. Users import multi-season grassland monitoring data through the human-computer interface, including grassland productivity data (dry matter yield per unit area of ​​forage) obtained through ground quadrat surveys and forage nutrient composition data (at least crude protein content and metabolizable energy content) obtained through laboratory testing. Simultaneously, the data acquisition module receives grazing time-series parameters input by the user, including the number of days in the current rotational grazing cycle, preset pasture transition dates for each season, and the current grazing intensity (number of livestock per unit area). Users also input basic livestock information, including breed (different breeds correspond to different basal metabolic parameters and nutrient conversion coefficients), herd size (number of livestock at each physiological stage), details of each physiological stage (at least distinguishing between gestation, lactation, and fattening periods), and environmental temperature and humidity data during grazing. The data acquisition module performs a reasonableness check on the input data, providing prompts for data exceeding the preset reasonable range, requiring user confirmation or correction.

[0043] Phase 2, Grassland Nutrient Supply Measurement Phase: After the data acquisition module completes data acquisition, it transmits multi-season grassland monitoring data and grazing time sequence parameters to the grassland nutrient supply calculation module.

[0044] The grassland nutrient supply calculation module first calculates the total grass yield of the pasture in each season based on the grass production and corresponding grassland area: ; in, The total hay production (dry matter kg) of the pasture in season j. The yield per unit area of ​​pasture in season j (dry matter kg / mu) is given. Let be the usable grassland area (in mu) of the pasture in season j.

[0045] Subsequently, the total grass yield of pastures in each season was compared with the concentration of forage nutrients (crude protein content) in that season. and metabolizable energy content Multiplying these together, we obtain the total nutrient content of pasture grass in each season: ; in, The total crude protein content (kg) of the pasture in season j. The total metabolizable energy content of the pasture in season j (MJ).

[0046] The grassland nutrient supply calculation module further divides the entire grazing season into multiple consecutive grazing periods according to the transition nodes. For each independent grazing period, based on the dynamic curve of the seasonal pasture yield in that period, combined with the forage intake rate and grazing intensity coefficient of that period, the actual amount of forage nutrients that livestock can graze during that period is calculated: ; in, and These represent the total crude protein supply (kg) and total metabolizable energy supply (MJ) of pasture during grazing period t, respectively; m represents the total number of seasonal pastures covered during grazing period t. The coefficient representing the percentage of time spent in pastures during grazing period t (all) The sum is 1). t represents the forage consumption rate during grazing period t (i.e., the proportion of actual forage consumption by livestock to the total forage yield; this parameter is set according to grassland type, grazing intensity, and literature recommendations). This is the correction coefficient for grazing intensity on feed intake during grazing period t (when grazing intensity exceeds the appropriate threshold, feed intake decreases due to increased trampling and waste, and this coefficient is less than 1).

[0047] The impact of rotational grazing cycles on forage supply is assessed through... The assignment process reflects that when the rotational grazing cycle is shortened, the time livestock spend in a single rotational grazing area per unit time decreases, the time for pasture regeneration and recovery in each area increases, and the total available pasture yield increases accordingly throughout the grazing season. This dynamic process incorporates the rotational grazing cycle as a time axis segmentation parameter into the supply calculation for each period.

[0048] The grassland nutrient supply calculation module outputs the total amount of forage nutrient supply measured in crude protein and metabolizable energy for each grazing period, and transmits it along with the corresponding period label to the grass-livestock nutrient coupling assessment module.

[0049] Phase 3, Livestock Nutritional Requirements Calculation Phase: In parallel with the grassland nutrient supply measurement phase, the livestock nutrient demand calculation module receives basic livestock data from the data acquisition module and initiates the calculation of the dynamic model of livestock nutrient consumption.

[0050] First, the livestock nutrition requirement calculation module calculates the metabolic weight of livestock at each physiological stage based on the measured or estimated weight of the livestock at each physiological stage: ; in, The metabolic weight (kg) of the livestock in the i-th physiological stage. denoted as the average live weight (kg) of livestock at the i-th physiological stage.

[0051] Subsequently, basal maintenance metabolic requirements (BMR) parameters were determined based on livestock breeds. The maintenance metabolic requirement of livestock refers to the net energy and net protein required for basic life activities in a resting state. This parameter is calculated by multiplying metabolic body weight by a breed-specific basal maintenance coefficient to obtain the daily maintenance metabolic requirement per head of livestock. ; in, and These represent the daily crude protein maintenance requirement (kg / head·day) and metabolizable energy maintenance requirement (MJ / head·day) for a single livestock in the i-th physiological stage, respectively. and These are the basic crude protein maintenance coefficient and the basic metabolizable energy maintenance coefficient of the variety, respectively (obtained by consulting the variety nutrition standard database).

[0052] In addition to the basic maintenance needs, livestock also have additional production needs due to different physiological stages. During pregnancy, nutritional requirements for fetal growth and development increase, rising with the number of days of gestation. During lactation, nutritional requirements for milk production increase, related to daily milk yield and composition. During fattening, nutritional requirements for weight gain increase, related to daily weight gain targets and composition. The total daily nutritional requirement for a single livestock animal is the sum of maintenance and production needs. ; in, and These represent the total daily crude protein requirement (kg / head·day) and total metabolizable energy requirement (MJ / head·day) for a single livestock in the i-th physiological stage, respectively. and These represent the daily crude protein production requirement (kg / head·day) and metabolizable energy production requirement (MJ / head·day) for a single livestock in the i-th physiological stage, respectively.

[0053] When the ambient temperature deviates from the suitable temperature range for livestock, the livestock need to expend additional energy to maintain a normal body temperature. The livestock nutrition requirement calculation module incorporates an ambient temperature correction factor. When the ambient temperature Below the lower limit temperature or above the upper limit temperature hour: ; Where λ and μ are the energy consumption increase coefficients under low and high temperature environments, respectively, obtained through literature or experiments. When the ambient temperature is within a suitable range, .

[0054] Based on the above parameters, the dynamic total nutritional requirement of a livestock population at grazing time t is the sum of the products of the individual requirement at each physiological stage and the corresponding number of head, multiplied by the environmental temperature correction factor for that time period: ; in, and These represent the total crude protein requirement (kg) and total metabolizable energy requirement (MJ) of the livestock population during grazing period t, respectively. Let be the number of livestock in the i-th physiological stage, and n be the total number of physiological stages. Let t be the number of days in the grazing period.

[0055] The livestock nutrition requirement calculation module outputs the total crude protein requirement and total metabolizable energy requirement of livestock groups for each grazing period, and transmits them along with the corresponding period labels to the grass-livestock nutrition coupling assessment module.

[0056] Phase 4, Grass-Livestock Nutrition Coupling Assessment Phase: The grass-livestock nutrition coupling assessment module receives pasture nutrient supply data and livestock nutrient demand data for each grazing period on the same time axis from the grassland nutrient supply calculation module and the livestock nutrient demand calculation module, respectively. This module first performs time-period alignment verification on the two sets of data to ensure that the time intervals of the supply and demand sides are completely consistent within the same grazing period.

[0057] After time-period alignment, the grass-livestock nutrition coupling assessment module calculates the nutrient balance gap for each grazing period. The nutrient balance gaps for crude protein and metabolizable energy are calculated as follows: ; in, and These represent the crude protein nutrient balance deficit (kg) and metabolizable energy nutrient balance deficit (MJ) at grazing time t, respectively. When this occurs, it indicates that livestock are not getting enough crude protein and are nutritionally deficient during that grazing period; when When the time is right, it indicates that there is a surplus of crude protein supply. The interpretation of the metabolic energy nutritional balance deficit is similar.

[0058] The grass-livestock nutrition coupling assessment module further calculates the nutrient supply-demand ratio for each grazing period to quantify the severity of the nutrient deficit at each time period: ; when When a severe crude protein deficit risk is identified during that period, an early warning indicator is triggered; when When a mild crude protein deficiency is detected, it is marked as a state of concern; when At this point, it can be determined that the supply of crude protein basically meets the demand. The criteria for determining the supply-demand ratio of metabolizable energy are similar.

[0059] The grass-livestock nutrition coupling assessment module outputs the nutritional balance gap values ​​(crude protein gap, metabolizable energy gap), nutritional supply and demand ratio, and early warning indicators for each grazing period, which are then transmitted to the rotational grazing data optimization module and the results output module along with the time axis labels.

[0060] Phase 5, Global Optimization of Rotational Grazing Data: After receiving the nutrient balance gap data for each grazing period from the grass-livestock nutrition coupling assessment module, the rotational grazing data optimization module initiates the optimization process of the improved genetic algorithm.

[0061] a) Encoding and Population Initialization: The rotational grazing data optimization module uses real number encoding to encode the rotational grazing data to be optimized into chromosome genes.

[0062] Specifically, the transhumance dates for pastures in each season are encoded as the number of days offset from the baseline date (e.g., January 1st or the date the pasture turns green again); the rotational grazing cycle is encoded as a real number in days; and the carrying capacity is encoded as a real number in standard sheep units or livestock head counts. These three sets of parameters together constitute a chromosome. For the k-th chromosome, its gene structure is represented as follows: ; in, For the encoded values ​​of p transition nodes, This is the coded value for the rotational grazing cycle. This is the coded value for carrying capacity.

[0063] During population initialization, the population is randomly generated within the reasonable range of values ​​for each of its variables. (Recommended value: 100-200) chromosomes constitute the initial population. A preliminary feasibility test is performed on each chromosome in the population to ensure that the corresponding rotational grazing data scheme does not exceed the basic physical limits of grassland carrying capacity (e.g., livestock carrying capacity does not exceed the upper limit of grassland production, and the rotational grazing cycle is not shorter than the minimum forage regeneration cycle).

[0064] b) Fitness function construction: The rotational grazing data optimization module employs multiple objectives, including minimizing the nutrient balance gap and maximizing grassland yield utilization, to construct a fitness function for each chromosome. For the rotational grazing data scheme corresponding to the k-th chromosome, the grassland nutrient supply calculation module and the livestock nutrient demand calculation module are first invoked to recalculate the nutrient balance gap for each grazing period under this scheme, resulting in the crude protein gap sequence for each period throughout the entire grazing period. and metabolic energy gap sequence .

[0065] In one possible implementation, the overall nutritional balance error of the kth chromosome is... Defined as the weighted sum of the absolute values ​​of the crude protein deficit and metabolizable energy deficit for each grazing period: ; Where T represents the total number of grazing periods included in the entire grazing period. and Weighting coefficients for crude protein and metabolizable energy deficit (can be adjusted according to the actual focus of the ranch, default values ​​are used). ).

[0066] In one possible implementation, the grassland yield utilization rate of the kth chromosome is... The calculation is the ratio of actual nutrient intake to total available nutrient intake, taking the average of both crude protein and metabolizable energy. ; Fitness value of chromosome k Defined as: ; in, and Weighting coefficients ( ,recommend Add 1 to the denominator to avoid division by zero errors. Fitness value The larger the value, the better the overall performance of the rotational grazing data scheme corresponding to that chromosome in terms of balancing nutrient balance and grassland utilization.

[0067] c) Constraint handling: The rotational grazing data optimization module embeds a penalty function strategy into the fitness function to handle dual constraints.

[0068] In one possible implementation, an upper limit threshold is set for grassland ecological protection constraints to determine grassland yield utilization rate. (The recommended value is 0.65, meaning that forage utilization should not exceed 65%). When scheme k... At that time, ecological penalties are imposed: ; in, This is the ecological penalty coefficient (take the larger positive number).

[0069] In one possible implementation, for livestock nutrient balance constraints, the nutrient supply-demand ratio is set to not fall below a lower threshold during any grazing period. (The recommended value is 0.80, meaning the nutrient deficit in any time period should not exceed 20% of the required amount). When there exists a grazing period t that satisfies: or At that time, apply a nutritional penalty: ; in, This is the nutritional penalty coefficient.

[0070] The corrected fitness value after adding the penalty is: .

[0071] d) Selection, crossover, and mutation operations: Using roulette wheel selection based on modified fitness values Individuals are selected from the parent population based on probability to enter the mating pool. The probability of the k-th individual being selected is proportional to its adjusted fitness value. ; The crossover operation uses simulated binary crossover (SBX). Two parent individuals are randomly selected from the mating pool, with a preset crossover probability. (Recommended value: 0.8) Perform gene recombination. The offspring gene values ​​generated by the SBX operator are distributed within the neighborhood of the parent gene values, with a distribution index of... Control the degree to which offspring deviate from their parents.

[0072] The mutation operation employs polynomial mutation. The offspring individuals produced after crossover are subjected to a preset mutation probability. (Recommended value: 0.05) Perform multinomial mutation on each gene locus, distribution index. Control the range of variation.

[0073] e) Judgment on Elite Retention and Termination: After each generation of evolution, the individual with the highest corrected fitness value in the current population is directly retained into the next generation to prevent the loss of the optimal solution. The number of elite individuals retained is 2% to 5% of the population size.

[0074] The termination condition is: reaching the preset maximum number of iterations. (Recommended value is 200-500 generations), or the optimal corrected fitness value of the population is within a continuous range. The algorithm terminates and outputs the chromosome gene values ​​corresponding to the optimal solution when any condition is met. These values ​​are then decoded into the optimized transhumance date, rotational grazing cycle days, and carrying capacity.

[0075] In the process of improving the genetic algorithm for optimization, the rotational grazing data optimization module simultaneously invokes the built-in structural equation model to analyze the grass-livestock coupling mechanism. The structural equation model constructs a causal path graph using grazing intensity, rotational grazing cycle, and transhumance nodes as exogenous latent variables, forage yield and forage nutrient composition as mediating latent variables, and livestock daily weight gain and nutrient balance as endogenous latent variables. The path coefficients of the structural equation model are used to quantify the strength and path of the influence of rotational grazing data on nutrient balance results. The results are then used to adjust the weight coefficients of each sub-objective in the genetic algorithm's fitness function. and constraint boundary values This achieves the integration of mechanistic knowledge with intelligent optimization.

[0076] f) Verification and parameter correction of the grazing experiment: The rotational grazing data scheme optimized by the genetic algorithm is not directly used as the final output, but is instead used in the location grazing experiment verification stage for validation and correction.

[0077] In one possible implementation, the specific operation for verifying the fixed-grazing experiment is as follows: Representative fixed-grazing experimental plots are selected within the target natural grassland area. Experimental treatments are set up according to the optimized rotational grazing data scheme output by the genetic algorithm, and at least one complete rotational grazing cycle is run under actual grazing conditions. During the experiment, changes in livestock weight are measured regularly (weighed every 15 days), actual forage intake is recorded (using the quadrat comparison method or the nylon bag method), and forage regeneration is monitored (measuring changes in forage yield before and after the experiment).

[0078] The measured daily weight gain of livestock, actual forage intake, and forage utilization data from the grazing experiment were compared with the model predictions item by item. For each model parameter, the relative error between the measured and predicted values ​​was calculated. When the absolute value of the relative error of a certain parameter exceeds a preset threshold (recommended value is 10%), a correction for that parameter is triggered. The correction method uses the feedback error backpropagation method. The ratio of the measured value to the predicted value of this parameter is used as the correction factor: The corrected parameter values ​​are obtained by multiplying the correction coefficient by the original parameter values ​​in the model: ; The parameters that need to be corrected include at least the forage intake coefficient. The parameters include the livestock nutrient conversion coefficient (the efficiency coefficient by which livestock convert ingested nutrients into weight gain or milk production at each physiological stage), environmental temperature correction coefficients λ and μ. If multiple parameters need to be corrected, they are iteratively corrected in descending order of sensitivity to the model output, until the relative errors between the measured and predicted values ​​of each key parameter converge to within the preset threshold.

[0079] The revised model parameters are then embedded into the grassland nutrient supply calculation module and the livestock nutrient requirement calculation module of the system, replacing the original default parameters. These embedded parameters are used for subsequent batch evaluation analysis and optimization to improve the model's localization accuracy. Verification records and parameter correction records from the grazing trials are automatically saved to the system database, allowing users to review and audit them.

[0080] After the location grazing test is verified and the parameters are corrected, the rotational grazing data optimization module determines the optimized and verified grazing time sequence parameters (including the optimized dates of pasture transfer nodes in each season, the optimized number of days in the rotational grazing cycle, and the optimized carrying capacity) as the final output results and transmits them to the result output module.

[0081] Phase 6, Result Output Phase: The results output module receives nutrient balance gap data and nutrient supply-demand ratio data for each grazing period from the grass-livestock nutrition coupling assessment module, as well as optimized grazing time sequence parameters from the rotational grazing data optimization module, and presents them to the user in a visual interface.

[0082] The output module simultaneously transmits the nutritional balance gap data (especially the specific crude protein and metabolizable energy gap values ​​during nutritionally deficient periods) of each grazing period to the external downstream supplemental feeding assessment system in a standardized data format through a preset external data interface. This data serves as the input data source for the subsequent formulation of precision supplemental feeding programs (including supplemental diet formulation design and supplemental feeding amount calculation).

[0083] Example 2: System adaptation for different grassland types The system of this invention can be adapted to different types of natural grassland (including temperate grassland, alpine meadow, desert steppe, mountain meadow, etc.). The dynamic curves of forage productivity and the seasonal variation patterns of forage nutrient components differ among different grassland types.

[0084] For alpine meadow pastures, the data acquisition module adds phenological observation data during the grassland greening and withering periods as additional input, and the grassland nutrient supply calculation module calculates... A phenological correction function was introduced to ensure that the peak yield of grass occurred in the middle and late summer, so as to accurately reflect the characteristics of concentrated growth of alpine meadow forage and rapid nutrient loss after withering.

[0085] For desert steppe type pastures, the forage intake rate in the grassland nutrient supply calculation module... The setting is lower than that of other grassland types to reflect the actual low grazing rate caused by the short and sparse grass in desert steppes and the difficulty for livestock to graze.

[0086] Example 3: System Deployment and Extended Applications In actual deployment, the six core modules of this invention can be integrated and deployed on a local server in the ranch or on a cloud platform, and users can remotely access the system's functional interface through a browser or mobile terminal.

[0087] The data acquisition module supports multiple data access methods, including manual input, mobile terminal photo recognition (scanning the QR code of pasture samples to read test report data), automatic acquisition by IoT sensors (automatic weather station transmits temperature and humidity data), and batch import from external databases, to adapt to pasture environments with different infrastructure conditions.

[0088] The output module uses RESTful API or Web Service standard protocols for its external data interface, supporting the transmission of JSON or XML data formats. This allows the system to be easily embedded into existing ranch management information platforms, forming a complete data exchange ecosystem with herd management systems, supplementary feed formulation systems, and ranch economic accounting systems, thus achieving complete closed-loop management.

[0089] The technical features described in the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. The specific parameter values ​​used in the embodiments (such as population size, crossover probability, mutation probability, number of iterations, weight coefficient, threshold, etc.) are only exemplary settings of the present invention and can be adjusted according to pasture conditions and user needs in actual applications.

[0090] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method and system for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands, characterized in that, include: Acquire multi-season grassland monitoring data, rotational grazing data, grazing intensity data, and basic livestock data for the target natural grassland; Based on the multi-season grassland monitoring data, the dynamic changes in grassland yield and forage nutrient concentration in each season are calculated. Combined with the grazing time sequence parameters determined based on the rotational grazing data and the grazing intensity data, the total amount of forage nutrient supply that the target grazing livestock can consume during each grazing period is calculated. Based on the basic livestock data, a dynamic model of livestock nutrient consumption is constructed to calculate the dynamic total nutrient requirements of grazing livestock groups during each corresponding grazing period. The total supply of pasture nutrients is compared with the total dynamic nutrient demand in time intervals to calculate the nutrient balance gap for each grazing period. With grassland ecological protection threshold and livestock nutritional balance as dual constraints, an improved genetic algorithm is used to jointly optimize the grazing time sequence parameters. The optimization results are then verified and corrected by combining the data from the fixed grazing test, and the optimized grazing time sequence parameters are output to guide the rotational grazing management of pastures.

2. The method for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands according to claim 1, characterized in that, The grassland monitoring data includes at least the grassland type, forage yield, and forage nutrient composition in different seasons; the rotational grazing data includes at least the rotational grazing cycle, transhumance points, and carrying capacity; the grazing intensity data includes the number of livestock grazing per unit area and the forage intake rate; and the basic livestock data includes livestock breed, physiological stage, and herd size. The total amount of pasture nutrient supply is calculated by multiplying the pasture yield of each season by the pasture nutrient composition, and then dynamically allocating it over time according to the rotation grazing cycle and transhumance nodes.

3. The method for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands according to claim 1, characterized in that, The input parameters of the dynamic model of livestock nutritional consumption include at least the livestock metabolic weight, physiological stage coefficient and environmental temperature correction coefficient, wherein the physiological stages include the gestation period, lactation period and fattening period.

4. The method for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands according to claim 1, characterized in that, During the joint optimization process, a structural equation model is simultaneously constructed to analyze the mechanism of grass-livestock coupling. The constraint handling method of the improved genetic algorithm adopts a penalty function strategy, and its fitness function is constructed with multiple objectives of minimizing the nutrient balance gap and maximizing the utilization rate of grass production.

5. The method for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands according to claim 1, characterized in that, The verification and correction method for the location grazing test data is as follows: the changes in livestock weight and forage utilization data measured in the actual grazing test are fed back to the improved genetic algorithm to correct the model parameters in reverse.

6. The method for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands according to claim 1, characterized in that, The nutritional imbalance gap is also used as input data for downstream supplemental feeding assessment systems to guide the development of precise supplemental feeding programs for grazing livestock.

7. A system for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands, used to implement the method described in any one of claims 1-6, characterized in that, include: The data acquisition module is used to collect multi-source data on multi-season grassland monitoring, grazing intensity, livestock base and rotational grazing; The grassland nutrient supply calculation module is connected to the data acquisition module and is used to calculate the total nutrient supply of forage during each grazing period. The livestock nutrition requirement calculation module is used to construct a dynamic model of livestock nutrition consumption and calculate the dynamic total nutrition requirement for each grazing period. The grass-livestock nutrition coupling assessment module is connected to the grassland nutrient supply calculation module and the livestock nutrient demand calculation module, respectively, and is used to compare and calculate the nutrient balance gap in time periods. The rotational grazing data optimization module has an improved genetic algorithm built in, which is used to jointly optimize the transhumance nodes, rotational grazing cycles and carrying capacity under the constraints of grassland ecological protection and livestock nutrition balance, and to verify the data by incorporating location grazing test data. The results output module is used to output the nutrient balance deficit and the optimized grazing timing parameters.

8. A nutritional balance assessment and analysis system for livestock grazing on natural grasslands according to claim 7, characterized in that, The rotational grazing data optimization module also includes a structural equation modeling unit, which is used to analyze the grass-livestock coupling mechanism and assist in setting the constraint boundaries of the improved genetic algorithm.

9. A nutritional balance assessment and analysis system for livestock grazing on natural grasslands according to claim 7, characterized in that, The result output module is equipped with an external data interface for transmitting the nutritional balance deficit data to an external supplemental feeding assessment system.

10. A system for assessing and analyzing the nutritional balance of livestock grazing on natural grasslands according to claim 7, characterized in that, The grassland nutrient supply calculation module has a built-in forage yield-nutrient composition fusion calculation unit, which is used to output the forage supply in terms of crude protein and metabolizable energy at each time period.

Images