AI large model dynamic training method for optimizing feed ratio of laying hens and cloud platform

Abstract

The invention provides a laying hen feed ratio optimized AI large model dynamic training method and a cloud platform, and relates to the technical field of AI model training. The method comprises the following steps: collecting corresponding feed ratio conversion data in different egg laying periods, extracting feed ratio information, and forming corresponding feed ratio conversion extraction information; performing grouping division based on correlation analysis on different feed ratio conversion extraction information to form correlation analysis division data; and performing correlation training aiming at feed ratio conversion on the correlation grouping division data to form a ratio conversion correlation analysis result. According to the method, the feed ratio of laying hen breeding is optimized, the breeding effect is improved, and the egg yield is increased.

Description

Technical Field

[0001] The present invention relates to the field of AI model training technology, and specifically to an AI large-scale dynamic training method and cloud platform for optimizing laying hen feed ratios. Background Art

[0002] With the development of society, people's demand for eggs has increased, and the egg industry has gradually taken shape and become a vital industry in production and daily life. Currently, egg production primarily relies on the centralized and scientific breeding of high-yielding laying hens. With the advancement of industrialization, laying hen breeding has gradually become more streamlined and automated, greatly improving egg production efficiency.

[0003] However, currently, the key to increasing egg production in laying hens lies in the feed supply. Different feed cost levels significantly impact egg production. For example, in conventional farms, the ratio of core feed ingredients like soybean meal directly impacts the growth and development of laying hens. Furthermore, nutritional requirements vary significantly between different growth stages, such as the chickling and laying stages. Current feed rationing technology still faces key bottlenecks: a lack of precise, quantitative dynamic control solutions and a heavy reliance on the experience and judgment of the keeper. This is because laying hen farming is a dynamic system involving growth cycles, physiological states, and environmental changes. Nutrient requirements at each stage must be adjusted in real time based on age and egg production. However, under traditional operation models, rationing plans are often based on empirical estimates, making it difficult to accurately match the actual nutritional needs of the flock at different stages.

[0004] Therefore, designing an AI large-scale dynamic training method and cloud platform for optimizing the feed ratio of laying hens to achieve optimized feed ratio for laying hens, improve the feeding effect and increase egg production is an urgent problem to be solved. Summary of the Invention

[0005] The object of the present invention is to provide an AI large-model dynamic training method for optimizing the feed ratio of laying hens. By collecting feed ratio conversion data of different laying cycles, the proportions of different ratio types in feed, eggs and feces and the weights of the ratios in feed, eggs and feces are extracted, and then a correlation analysis is performed on the ratio weight in terms of ratio and weight. Reference data for reasonable selection of ratio types in ratio and weight are established. On the one hand, important reference data is provided for the optimization and reasonable selection of feed ratios. On the other hand, feed ratios are not limited by experience, and can benefit more breeding units using a big data platform, greatly improving the efficiency of laying hen production, while saving breeding costs and significantly increasing egg production.

[0006] The present invention also aims to provide an AI large-scale dynamic training cloud platform for laying hen feed ratio optimization. The platform completes the collection of feed ratio conversion data through a data acquisition unit, and uses an extraction and analysis unit to complete pre-processing analysis of the collected data before model training to form more targeted feature data information. On this basis, the model training analysis of the analysis and processing unit establishes a correlation model between the ratio type and the ratio weight, which provides important analysis model data for subsequent feed ratio selection and optimization, and is an important material basis for achieving reasonable feed ratio optimization.

[0007] In the first aspect, the present invention provides an AI large-scale dynamic training method for optimizing feed ratios for laying hens, comprising: collecting corresponding feed ratio conversion data in different egg-laying cycles, extracting feed ratio information, and forming corresponding feed ratio conversion extraction information; grouping different feed ratio conversion extraction information based on correlation analysis to form correlation analysis division data; performing correlation training on the correlation grouping division data for feed ratio conversion to form ratio conversion correlation analysis results.

[0008] In the present invention, the method collects feed ratio conversion data of different egg-laying cycles, extracts the proportions of different ratio types in feed, eggs and feces, and the weights of the ratio weights in feed, eggs and feces, and then conducts a correlation analysis on the ratio weight in the two aspects of the ratio ratio and weight, and establishes reference data for the reasonable selection of the ratio type in the ratio ratio and the ratio weight. On the one hand, it provides important reference data for the optimization and reasonable selection of feed ratios. On the other hand, the feed ratio is not limited by experience, and can benefit more breeding units by using the big data platform, greatly improving the efficiency of laying hen production, saving breeding costs and significantly increasing egg production.

[0009] As a possible implementation method, the corresponding feed ratio conversion data in different egg-laying cycles are collected, and the feed ratio information is extracted to form the corresponding feed ratio conversion extraction information, including: extracting the corresponding feed ratio content information from the feed ratio conversion data in different egg-laying cycles. ,in, , m represents the number of different egg-laying cycles, n represents the number of different feed ratios determined based on the ratio with the most ratios in all egg-laying cycles, Indicates the weight percentage of the ratio type numbered n in the feed in the egg-laying cycle numbered m; for the feed ratio conversion data of different egg-laying cycles, extract the corresponding average feed supply ratio for each egg and the average total food intake ratio of a single ; Convert feed ratio data for different egg laying cycles and extract the corresponding average egg ratio content information ,in, , Indicates the average weight percentage of the ratio type numbered n in the eggs in the egg-laying cycle numbered m; extracts the corresponding average feces ratio content information for the feed ratio conversion data of different egg-laying cycles ,in, , Indicates the average weight percentage of the ratio type numbered n in the feces in the egg-laying cycle numbered m; for the feed ratio conversion data of different egg-laying cycles, extract the corresponding average single egg ratio weight and the average feces weight of a single ; For different egg-laying cycles, collect the corresponding feed ratio content information , Average feed supply ratio per head , Average total food intake ratio for a single bird , average egg ratio information , average feces content information , Average egg weight per egg and the average feces weight of a single , generating corresponding feed ratio conversion and extraction information. For example, feed ratio data is collected through IoT (Internet of Things) devices. Adjustments to feed ratio types and proportions based on farm needs require scheduled data updates. Data on the proportions of different feed types in eggs and feces can be obtained through sampling surveys to ensure representativeness.

[0010] In the present invention, the extracted feed ratio conversion data corresponding to different egg-laying cycles is used for subsequent reasonable data segmentation and model training. Therefore, it is necessary to extract feature information about the correlation effect of the ratio types from the feed ratio conversion data. Here, the extracted feature information mainly includes the ratio ratio information of different ratio types in feed, the ratio ratio information of different ratio types in eggs, and the ratio ratio information of different ratio types in feces. Here, the ratio ratio information reflects the distribution of different ratio types from feed to eggs and feces in terms of proportion, and to a certain extent reflects the absorption of the ratio types. This absorption is affected by the situation of the laying hens themselves on the one hand, and on the other hand, it is also affected by the interaction between different ratio types. Of course, for the ratio types, the ratio types that have an impact on the egg production of laying hens are selected. Such ratio types have direct and easy-to-identify correlation information. It is understandable that the proportion of different feed types has a correlation effect on the egg production of laying hens, and the amount of feed types also has a certain influence. This influence needs to reach a certain level or the impact on egg production is different under different body sizes. Therefore, the total weight of feed types, the total weight of feed types consumed by laying hens, and the total weight of feed types in eggs and feces are extracted. On this basis, the weight combined with the proportion of feed types can form important correlation data affecting egg production. It should be noted that for egg production, it is mainly based on the laying hen group, so the weight data obtained is also reflected in the form of an average for a single laying hen. Only in this way can the data formed be representative of the feeding significance, making the correlation analysis of the impact of feed types more accurate and effective.

[0011] As a possible implementation method, different feed ratio conversion and extraction information is grouped and divided based on correlation analysis to form correlation analysis division data, including: dividing different feed ratio conversion and extraction information into two different groups, forming a correlation fitting analysis data group and a correlation adjustment analysis data group, and satisfying: the number of feed ratio content information corresponding to the feed ratio conversion and extraction information in the correlation fitting analysis data group is not less than X There are n types of feed ratios; there are no less than Y feed ratio content information corresponding to the feed ratio conversion and extraction information in the correlation adjustment analysis data group The number of feed ratio types reaches n; the number of feed ratio conversion and extraction information in the correlation adjustment analysis data group is less than the number of feed ratio conversion and extraction information in the correlation fitting analysis data group.

[0012] In the present invention, when performing correlation analysis on the conversion extraction information of the extracted ratio types, it is considered that the data of different egg-laying cycles do not necessarily have consistent variability in correlation, that is, the feed ratio conversion extraction information corresponding to the egg-laying cycle does not present a simple gradual increase or decrease in egg production in the time dimension or other dimensions. Therefore, direct correlation analysis of all feed ratio conversion extraction information does not have the condition for determining whether the analysis results meet the accuracy requirements of the correlation data. Therefore, this application divides the different feed ratio conversion extraction information into reasonable groups, one part of which is used as the basic data for correlation fitting analysis, and the other part is used as reference data for verifying and adjusting whether the correlation data meets the required accuracy. Of course, considering that the ratio types will change under different egg-laying cycles, this change may affect the results of the correlation analysis. Therefore, in the different data groups divided, there must be a certain number of feed ratio conversion extraction information with the most complete ratio types to ensure that the fitting analysis of the correlation data covers all ratio types. The specific number of X and Y can be determined according to actual needs. The purpose of Y being less than X is to provide more data for the fitting analysis data set to improve the effect of the fitting analysis. On the other hand, it is to reduce the complexity of the adjustment analysis. After all, the adjustment verification after the fitting analysis is basically a fine-tuning of the accuracy of the correlation data and will not be applied to a large amount of data information.

[0013] As a possible implementation method, correlation training for feed ratio conversion is performed on the correlation grouping data to form a ratio conversion correlation analysis result, including: performing correlation relationship training for feed ratio conversion based on the correlation fitting analysis data group to form corresponding correlation initial fitting relationship data; performing relationship adjustment training with correlation accuracy as the goal based on the correlation initial fitting relationship data in combination with the correlation adjustment analysis data group to form feed ratio correlation target relationship data.

[0014] In the present invention, the correlation training performed on the correlation fitting analysis data group is mainly to establish the correlation information of the proportion and weight of the ratio types, and the adjustment training performed on the correlation adjustment analysis data is mainly to adjust the fitting accuracy of the correlation relationship data formed by the correlation fitting analysis data group to achieve the desired correlation accuracy.

[0015] As a possible implementation method, according to the correlation fitting analysis data group, the correlation relationship training for feed ratio conversion is carried out to form the corresponding initial correlation fitting relationship data, including: feed ratio content information from the correlation fitting analysis data group The feed ratio conversion extraction information with the highest comprehensive conversion rate is determined from the n different feed ratio conversion extraction information, and is calibrated as the ratio conversion benchmark information; based on the ratio conversion benchmark information, the correlation fitting analysis data group is subjected to a correlation fitting analysis based on the influence of the ratio change amount to form the correlation initial fitting relationship data.

[0016] In the present invention, the correlation training performed on the correlation fitting analysis data group is mainly to establish a reasonable correlation in the proportion of the proportion of the ratio type and the weight information, thereby being able to provide reference data for optimizing the feed ratio and weight. However, considering that the proportion and weight information of the ratio types in different egg-laying cycles do not have a unified reference, the analysis of the correlation has a certain independence in different egg-laying cycles, which will lead to a decrease in the accuracy of the analysis results. Therefore, this application uses the feed ratio conversion extraction information with the most complete ratio types as the data benchmark, obtains the variation of the feed ratio conversion extraction information of other feed ratio conversion extraction information relative to the benchmark, and performs correlation analysis on the variation data, that is, it is possible to standardize and unify the data of different egg-laying cycles, and also to improve the accuracy of the correlation analysis. Of course, there will be multiple feed ratio conversion extraction information with the most complete ratio types. Here, the ratio with the highest comprehensive conversion rate is used as the benchmark. This is because the comprehensive conversion rate of the ratio reflects the situation of the egg in absorbing the components of the ratio types to a certain extent. It can be understood that the more the ratio types that affect egg production are converted by the eggs, the more the egg production is affected. It should be noted that the conversion of feed ratios can be based on elements, that is, using chemical elements as the unit of feed ratio. This allows for better confirmation of feed ratio conversion and migration. Therefore, the key to feed ratio is to accurately control the daily feed intake of laying hens. Of course, if molecular or even material-level objects can be directly measured in eggs and feces, these material or molecular-level objects can also be used as feed ratio types.

[0017] As a possible implementation method, the feed ratio content information in the correlation fitting analysis data set is The feed ratio conversion extraction information with the highest comprehensive conversion rate is determined from the n different feed ratio conversion extraction information, and marked as the ratio conversion benchmark information, including: the feed ratio content information in the correlation fitting analysis data group The ratio of all feed ratios of n types is converted and extracted, and the corresponding feed ratio content information is obtained. Average egg content ratio information , determine the corresponding ratio comprehensive conversion rate ,in, ; The comprehensive conversion rate of the ratio The maximum feed ratio conversion extraction information is calibrated as the ratio conversion benchmark information.

[0018] In the present invention, the comprehensive conversion rate of the ratio is the sum of the proportions of different ratio types in eggs relative to the proportions of the ratio types in feed. To a certain extent, it reflects the volume of the ratio types migrating into eggs, and reflects the degree of impact on egg production and even egg quality.

[0019] As a possible implementation method, based on the ratio conversion benchmark information, the correlation fitting analysis data group is subjected to a correlation fitting analysis based on the influence of the ratio change amount to form the correlation initial fitting relationship data, including: for the feed ratio conversion extraction information other than the ratio conversion benchmark information in the correlation fitting analysis data group, the following ratio change amount data are determined respectively: based on the feed ratio content information corresponding to the feed ratio conversion extraction information, the feed ratio benchmark change amount information relative to the feed ratio content information in the ratio conversion benchmark information is determined. , where k represents the number of different feed ratio conversion extraction information in the correlation fitting analysis data except the ratio conversion benchmark information and k<m, , The weight percentage difference between the weight percentage of the ratio type numbered n in the feed ratio conversion extraction information numbered k and the weight percentage of the ratio type numbered n in the ratio conversion benchmark information; based on the average total amount of feed supply ratio for a single head corresponding to the feed ratio conversion extraction information, determine the single feed supply benchmark change amount information relative to the average total amount of feed supply ratio for a single head in the ratio conversion benchmark information. ,in, The difference between the total amount of the average feed supply ratio of a single bird corresponding to the feed ratio conversion extraction information and the total amount of the average feed supply ratio of a single bird in the ratio conversion benchmark information; based on the average total feed ratio of a single bird corresponding to the feed ratio conversion extraction information, determine the single bird feed benchmark change amount information relative to the average total feed ratio of a single bird in the ratio conversion benchmark information ,in, The difference between the average total feed ratio of a single chicken corresponding to the feed ratio conversion extraction information and the average total feed ratio of a single chicken in the ratio conversion benchmark information; based on the average egg ratio content information corresponding to the feed ratio conversion extraction information, determine the egg ratio benchmark change information relative to the average egg ratio content information in the ratio conversion benchmark information. ,in, , The weight percentage difference between the average weight percentage of the ratio type numbered n in the feed ratio conversion extraction information numbered k and the average weight percentage of the ratio type numbered n in the ratio conversion benchmark information; based on the average feces ratio content information corresponding to the feed ratio conversion extraction information, determine the egg ratio benchmark change information relative to the average feces ratio content information in the ratio conversion benchmark information. ,in, , The weight percentage difference between the average weight percentage of the ratio type numbered n in the feed ratio conversion extraction information numbered k and the average weight percentage of the ratio type numbered n in the ratio conversion benchmark information; based on the average single egg ratio weight corresponding to the feed ratio conversion extraction information, determine the single egg benchmark change weight information relative to the average single egg ratio weight in the ratio conversion benchmark information ,in, The difference between the average egg ratio weight of a single chicken corresponding to the feed ratio conversion extraction information and the average egg ratio weight of a single chicken in the ratio conversion benchmark information; the average feces ratio weight of a single chicken corresponding to the feed ratio conversion extraction information is determined, and the average feces ratio weight of a single chicken relative to the average feces ratio weight in the ratio conversion benchmark information is determined. ,in, The difference between the average feces ratio weight of a single pig corresponding to the feed ratio conversion extraction information and the average feces ratio weight of a single pig in the ratio conversion benchmark information; the feed ratio benchmark change information corresponding to different feed ratio conversion extraction information , Single feed supply benchmark change information , Single feeding baseline change information , Egg ratio benchmark change information , Egg ratio benchmark change information , Single egg benchmark weight information , Single feces baseline weight change information , conduct correlation fitting analysis on configuration ratio and proportion weight, and establish the following correlation fitting relationship group: ; ; ; ;in, Indicates the weight percentage difference between the weight percentage of the numbered n compounding type in the feed and the weight percentage of the numbered n compounding type in the feed in the compounding conversion benchmark information; Indicates the weight percentage difference between the average weight percentage of the ratio type numbered n in the eggs and the weight percentage of the ratio type numbered n in the eggs in the ratio conversion benchmark information; Indicates the weight percentage difference between the average weight percentage of the ratio type numbered n in feces and the weight percentage of the ratio type numbered n in feces in the ratio conversion benchmark information; Indicates the egg ratio conversion related items corresponding to the ratio type numbered n; Indicates the feces ratio conversion related items corresponding to the ratio type numbered n; The correlation function representing the weight percentage of different proportions of species in the feed; Indicates the weight difference between the total amount of average feed supply ratio for a single head and the total amount of average feed supply ratio for a single head in the ratio conversion benchmark information; Indicates the weight difference between the average total food intake ratio of a single bird and the average total food intake ratio of a single bird in the ratio conversion benchmark information; Indicates feed utilization rate; Indicates the weight difference between the average single egg ratio weight and the average single egg ratio weight of the ratio conversion benchmark information; It represents the weight difference between the average feces ratio weight of a single egg and the average feces ratio weight of a single egg in the ratio conversion benchmark information; U represents the influence correlation formula of the ratio weight of a single egg; V represents the influence correlation formula of the ratio weight of a single egg.

[0020] In the present invention, the relationship equations in the correlation fitting relationship group reflect the correlation relationship between the change in feed ratio conversion extraction information relative to the baseline data. The correlation fitting relationship group includes the correlation relationship between the conversion of each ratio type from feed to eggs and feces, the correlation relationship between the ratios of different ratio types, the correlation relationship between the ratio weight of feed supply and the ratio weight of food intake, and the correlation relationship between the weight change of the ratio weight of food intake from feed to eggs and feces. Among them, the correlation relationship of the conversion of each ratio type from feed to eggs and feces shows the change in the ratio of the ratio of different ratio types to eggs and feces relative to the corresponding baseline. That is, the increase or decrease in the ratio of feed relative to the baseline corresponds to the increase or decrease in the ratio of eggs relative to the baseline in eggs, and the increase or decrease in the ratio of feces relative to the baseline in feces. In this way, the amount of migration that can occur due to the change in the ratio type can be determined, and the direction of this migration ratio can also be determined, that is, whether it flows to eggs or feces, providing reference information for subsequent nutritional and cost analysis. By focusing on the correlation between the ratios of different feed types, we can determine the limiting influence relationship between the changes in the ratios of different feed types relative to the baseline ratio. It is understandable that the increase or decrease in the ratios of different feed types is not independent, but rather mutually restrictive. This is manifested macroscopically in the directional migration of different feed types into laying hens. For example, increasing the proportion of feed types that affect egg production weight will cause the proportions of other feed types to decrease, further affecting the proportions of feed types related to factors such as laying hen weight and sleep. This can lead to a single promotion of egg production weight increase while ignoring the growth and development of laying hens, which will also reduce egg production overall. Therefore, the limiting influence relationship between feed types can be established based on big data training to establish a good correspondence, comprehensively focusing on multiple factors affecting laying hens' egg production to achieve reasonable allocation. The correlation between the weight of the feed supply and the weight of the food intake primarily reflects the control of the laying hens' feed intake when a given feed type is increased or decreased. This relationship ensures the rationality of feed supply, avoids feed waste that increases costs, and prevents situations where insufficient feed supply leads to an inability to prepare feed in advance. The correlation between the weight of the feed intake and the weight changes from feed to eggs and feces shows the absorption of the feed type by volume when a given feed type is increased or decreased. This helps to rationally control the total weight of the feed types based on the feed intake of laying hens, avoiding waste or insufficient supply. The different relationship equations in the fitted relationship group are interconnected and constrained, establishing a comprehensive relationship between the influence of feed types on feed ratio and weight. This provides a reasonable reference for adding feed types to the feed, and is more accurate and effective than empirical feed analysis.Of course, the corresponding correlation and other relationship parameters in the relationship can be defined according to the actual need for accuracy. They can be simple linear constants or multiple power relationship expressions. As long as they can be determined through correlation fitting analysis data, they are acceptable.

[0021] As a possible implementation method, based on the initial correlation fitting relationship data, combined with the correlation adjustment analysis data group, relationship adjustment training with correlation accuracy as the goal is carried out to form feed ratio correlation target relationship data, including: extracting information on different feed ratios in the correlation adjustment analysis data group, determining the corresponding comprehensive conversion rate of the ratio, and sorting the different feed ratio conversion extraction information in order from small to large according to the comprehensive conversion rate of the ratio to form an ordered adjustment analysis data group; performing deviation-based verification adjustment analysis on the correlation fitting relationship group in the order of different feed ratio conversion extraction information in the ordered adjustment analysis data group to form feed ratio correlation target relationship data.

[0022] In the present invention, the relationship adjustment training of the initial correlation fitting relationship data using the correlation adjustment analysis data group takes into account a certain adjustment direction. Therefore, the order of extraction is sorted and adjusted with the comprehensive conversion rate of the ratio as a reference until the correlation relationship data meets the required accuracy.

[0023] As a possible implementation method, the correlation fitting relationship group is sequentially subjected to a verification adjustment analysis based on deviation from the target in the order of different feed ratio conversion extraction information in the ordered adjustment analysis data group to form feed ratio correlation target relationship data, including: sequentially extracting different feed ratio conversion extraction information in the ordered adjustment analysis data group, and performing the following verification adjustment analysis on the correlation fitting relationship group: obtaining the following data information of the feed ratio conversion extraction information: obtaining the average weight percentage of the corresponding different ratio types in the feed in the feed ratio conversion extraction information relative to the weight percentage of the corresponding ratio type in the feed in the ratio conversion benchmark information. , i represents the sequence number of different feed ratio conversion extraction information in the ordered adjustment analysis data group; obtain the difference between the average weight percentage of different ratio types in eggs corresponding to the feed ratio conversion extraction information and the weight percentage of the corresponding ratio type in eggs in the ratio conversion benchmark information ; Obtain the difference between the average weight percentage of the different ratio types in the feces corresponding to the feed ratio conversion extraction information and the weight percentage of the corresponding ratio types in the feces in the ratio conversion benchmark information ; Obtain the feed supply benchmark change information of the average feed supply ratio of a single head in the feed ratio conversion extraction information relative to the average feed supply ratio of a single head in the ratio conversion benchmark information ; Obtain the adjusted feed benchmark change information of the average total feed ratio of a single bird in the feed ratio conversion extraction information relative to the average total feed ratio of a single bird in the ratio conversion benchmark information ; Get the adjusted egg benchmark weight change information of the average egg ratio weight of a single egg in the feed ratio conversion extraction information relative to the average egg ratio weight of a single egg in the ratio conversion benchmark information ; Obtain the adjusted feces benchmark change weight information of the average feces ratio weight of a single head in the feed ratio conversion extraction information relative to the average feces ratio weight of a single head in the ratio conversion benchmark information ; According to the weight percentage difference of eggs corresponding to different ratios , feces weight percentage difference and adjust feed supply benchmark change information , Adjust the egg base weight information and adjust stool baseline change weight information , combined with the correlation fitting relationship group, determine the feed ratio correlation percentage difference corresponding to the feed ratio conversion extraction information , adjust the difference in correlation between meal ratio and adjust feed supply related variance ; According to the feed ratio conversion extraction information, the weight percentage difference of the feed ratio of different types corresponding to the feed ratio , Adjust the food intake baseline change information and adjust feed supply benchmark change information , and combined with the feed ratio correlation percentage difference , adjust the difference in correlation between meal ratio and adjust feed supply related variance , respectively make the following adjustment conditions judgment: the weight percentage difference of feed ratio corresponding to different ratio types Percent difference in correlation with feed ratio :If all meet , then the feed ratio compliance information is formed, among which, Indicates the allowable deviation of the proportion of the proportion type numbered n; if there is , then the feed ratio does not meet the standard information, and according to the feed ratio conversion and extraction information, the relationship in the correlation fitting relationship group is , make adjustments so that the newly formed feed ratio correlation percentage difference All satisfied , and the relationship in the correlation fitting relationship group is fitted according to the proportional relationship of different ratio types Update so that the mapping relationship in the relationship envelope feed ratio conversion extraction information corresponding to the ratio type ratio relationship; adjust the feeding benchmark change information Difference in correlation with adjusted meal ratio :If satisfied , then the conversion target information is formed, among which, Indicates the allowable deviation of a single feeding; if not satisfied , then the conversion does not meet the standard information, and according to the feed ratio conversion extraction information, the relationship in the correlation fitting relationship group is Adjustments are made so that the newly formed adjustment of the feeding ratio correlation difference satisfy ; Information on changes in feed supply benchmarks and adjusted feed supply related differences :If satisfied , then the supply compliance information is formed, among which, Indicates the allowable deviation of a single supply; if not satisfied , then the information of substandard supply is formed, and the relationship in the correlation fitting relationship group is extracted according to the feed ratio conversion information. Adjustments are made so that the newly formed feed supply correlation difference satisfy ; If the adjustment conditions for the feed ratio conversion extraction information are all met, the adjustment analysis is stopped; if the adjustment conditions are not met, the feed ratio conversion extraction information in the ordered adjustment analysis data group is continuously extracted in sequence for adjustment analysis until the adjustment conditions are all met or all the feed ratio conversion extraction information in the ordered adjustment analysis data group is extracted.

[0024] In the present invention, the adjustment training of the correlation fitting relationship group is mainly to provide new feed ratio conversion extraction information to verify whether the deviation between the predicted data formed by the correlation fitting relationship group and the actual data is within the required allowable deviation range. Of course, because the relationship formulas in the correlation fitting relationship group are mutually correlated, each relationship formula has the independence of data verification, that is, the accuracy can be verified for each relationship formula. In this way, there are three adjustment judgment conditions for whether the accuracy meets the standard. One is whether the correlation of different ratio types in the proportion migration meets the standard. This also involves the proportion mapping relationship between different ratios. After all, the migration of different ratios will also be affected by the interaction and influence between different ratio types, that is, different chemical elements have mutuality in the influence of biological relationships. Therefore, when the correlation does not meet the standard for adjustment, it is also necessary to make an overlay adjustment of the mapping relationship function between different ratio types, that is, to envelop the new mapping relationship for adjustment. The second is the relationship between feed intake and supply. Different ratios will result in different actual feed weights, and it is impossible to ensure that the provided feed is eaten up immediately during feeding. There is a certain amount of consumption, and this consumption and the type of ratio will affect the feeding habits of laying hens. For example, certain chemical elements affect the appetite of laying hens, and different ratios will result in different feed intakes, so adjustments and analysis are needed. The third is the conversion of the total ratio into eggs and feces, which reflects the impact of volume on egg production. If the standard is not met, the relationship needs to be adjusted. It should be noted that the adjustment methods adopted for the three relationship formulas are diverse. It can be to introduce a unified proportional value, and after the conditions are met, the proportional value is assigned to each related item to form a new related item. It can also be to directly increase or decrease the constant term of the related item to make adjustments, or to increase or decrease the power formula in the related item to make adjustments.

[0025] In the second aspect, the present invention provides an AI large-model dynamic training cloud platform for optimizing laying hen feed ratios, comprising: a data acquisition unit for collecting corresponding feed ratio conversion data in different egg-laying cycles; an extraction and division unit for extracting feed ratio information from different feed ratio conversion data collected by the data acquisition unit to form corresponding feed ratio conversion extraction information, and performing grouping and division based on correlation analysis to form correlation analysis division data; an analysis and processing unit for performing correlation training on the correlation analysis division data formed by the extraction and division unit to form a ratio conversion correlation analysis result.

[0026] In the present invention, the platform completes the collection of feed ratio conversion data through the data collection unit, and uses the extraction and analysis unit to complete the pre-processing analysis of the collected data before model training to form more targeted feature data information. On this basis, the model training analysis of the analysis and processing unit establishes a correlation model between the ratio type and the ratio weight, which provides important analysis model data for subsequent feed ratio selection and optimization, and is an important material basis for achieving reasonable feed ratio optimization.

[0027] The AI ​​large-model dynamic training method and cloud platform for optimizing laying hen feed ratio provided by the present invention have the following beneficial effects: This method collects feed ratio conversion data for different egg-laying cycles, extracts the proportions of different ratio types in feed, eggs and feces, as well as the weight of the ratios in feed, eggs and feces, and then conducts a correlation analysis on the ratio weight in the ratio and weight, and establishes reference data for the reasonable selection of ratio types in the ratio and weight. On the one hand, this method provides important reference data for the optimization and reasonable selection of feed ratios. On the other hand, feed ratios are not limited by experience, and can benefit more breeding units using the big data platform, greatly improving the efficiency of laying hen production, saving breeding costs and significantly increasing egg production.

[0028] The platform completes the collection of feed ratio conversion data through the data acquisition unit, and uses the extraction and analysis unit to complete the pre-processing analysis of the collected data before model training to form more targeted feature data information. On this basis, the model training and analysis of the analysis and processing unit establishes a correlation model between the ratio type and the ratio weight, which provides important analysis model data for subsequent feed ratio selection and optimization, and is an important material basis for achieving reasonable feed ratio optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 A step diagram of a dynamic training method for an AI large model for optimizing laying hen feed ratio provided by an embodiment of the present invention; Figure 2 A schematic structural diagram of an AI large-model dynamic training cloud platform for optimizing laying hen feed ratios provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.

[0032] With the development of society, people's demand for eggs has increased, and the egg industry has gradually taken shape and become a vital industry in production and daily life. Currently, egg production primarily relies on the centralized and scientific breeding of high-yielding laying hens. With the advancement of industrialization, laying hen breeding has gradually become more streamlined and automated, greatly improving egg production efficiency.

[0033] However, currently, the key to increasing egg production in laying hens lies in the feed supply. Different feed cost levels significantly impact egg production. For example, in conventional farms, the ratio of core feed ingredients like soybean meal directly impacts the growth and development of laying hens. Furthermore, nutritional requirements vary significantly between different growth stages, such as the chickling and laying stages. Current feed rationing technology still faces key bottlenecks: a lack of precise, quantitative dynamic control solutions and a heavy reliance on the experience and judgment of the keeper. This is because laying hen farming is a dynamic system involving growth cycles, physiological states, and environmental changes. Nutrient requirements at each stage must be adjusted in real time based on age and egg production. However, under traditional operation models, rationing plans are often based on empirical estimates, making it difficult to accurately match the actual nutritional needs of the flock at different stages.

[0034] refer to Figure 1-Figure 2 , an embodiment of the present invention provides an AI large-model dynamic training method for optimizing feed ratios for laying hens. The method collects feed ratio conversion data for different laying cycles, extracts the proportions of different ratio types in feed, eggs and feces, and the weights of the ratios in feed, eggs and feces, and then conducts a correlation analysis on the ratio weight in terms of ratio and weight, and establishes reference data for the reasonable selection of ratio types in terms of ratio and weight. On the one hand, it provides important reference data for the optimization and reasonable selection of feed ratios. On the other hand, feed ratios are not limited by experience, and can benefit more breeding units using a big data platform, greatly improving the efficiency of laying hen production, saving breeding costs, and significantly increasing egg production.

[0035] The dynamic training method of the AI ​​large model for optimizing laying hen feed ratio specifically includes the following steps: S1: Collect the corresponding feed ratio conversion data in different egg-laying cycles, extract the feed ratio information, and form the corresponding feed ratio conversion extraction information.

[0036] Collect the corresponding feed ratio conversion data in different egg-laying cycles, extract the feed ratio information, and form the corresponding feed ratio conversion extraction information, including: extracting the corresponding feed ratio content information from the feed ratio conversion data in different egg-laying cycles ,in, , m represents the number of different egg-laying cycles, n represents the number of different feed ratios determined based on the ratio with the most ratios in all egg-laying cycles, Indicates the weight percentage of the ratio type numbered n in the feed in the egg-laying cycle numbered m; for the feed ratio conversion data of different egg-laying cycles, extract the corresponding average feed supply ratio for each egg and the average total food intake ratio of a single ; Convert feed ratio data for different egg laying cycles and extract the corresponding average egg ratio content information ,in, , Indicates the average weight percentage of the ratio type numbered n in the eggs in the egg-laying cycle numbered m; extracts the corresponding average feces ratio content information for the feed ratio conversion data of different egg-laying cycles ,in, , Indicates the average weight percentage of the ratio type numbered n in the feces in the egg-laying cycle numbered m; for the feed ratio conversion data of different egg-laying cycles, extract the corresponding average single egg ratio weight and the average feces weight of a single ; For different egg-laying cycles, collect the corresponding feed ratio content information , Average feed supply ratio per head , Average total food intake ratio for a single bird , average egg ratio information , average feces content information , Average egg weight per egg and the average feces weight of a single , generating corresponding feed ratio conversion and extraction information. For example, feed ratio data is collected through IoT (Internet of Things) devices. Adjustments to feed ratio types and proportions based on farm needs require scheduled data updates. Data on the proportions of different feed types in eggs and feces can be obtained through sampling surveys to ensure representativeness.

[0037] The extracted feed ratio conversion data corresponding to different egg-laying cycles is used for subsequent reasonable data segmentation and model training. Therefore, it is necessary to extract feature information about the correlation effect of ratio types from the feed ratio conversion data. Here, the extracted feature information mainly includes the ratio information of different ratio types in feed, the ratio information of different ratio types in eggs, and the ratio information of different ratio types in feces. The ratio information here reflects the distribution of different ratio types from feed to eggs and feces in terms of proportion, and to a certain extent reflects the absorption of the ratio types. This absorption is affected by the conditions of the laying hens themselves on the one hand, and the interaction between different ratio types on the other hand. Of course, for the ratio types, the ratio types that affect the egg production of laying hens are selected. Only such ratio types have direct and easy-to-identify correlation information. It is understandable that the proportion of different feed types has a correlation effect on the egg production of laying hens, and the amount of feed types also has a certain influence. This influence needs to reach a certain level or the impact on egg production is different under different body sizes. Therefore, the total weight of feed types, the total weight of feed types consumed by laying hens, and the total weight of feed types in eggs and feces are extracted. On this basis, the weight combined with the proportion of feed types can form important correlation data affecting egg production. It should be noted that for egg production, it is mainly based on the laying hen group, so the weight data obtained is also reflected in the form of an average for a single laying hen. Only in this way can the data formed be representative of the feeding significance, making the correlation analysis of the impact of feed types more accurate and effective.

[0038] S2: The extracted information of different feed ratio conversions is grouped and divided based on correlation analysis to form correlation analysis division data.

[0039] The different feed ratio conversion extraction information is grouped and divided based on the correlation analysis to form correlation analysis division data, including: dividing the different feed ratio conversion extraction information into two different groups, forming a correlation fitting analysis data group and a correlation adjustment analysis data group, and satisfying: the number of feed ratio content information corresponding to the feed ratio conversion extraction information in the correlation fitting analysis data group is not less than X There are n types of feed ratios; there are no less than Y feed ratio content information corresponding to the feed ratio conversion and extraction information in the correlation adjustment analysis data group The number of feed ratio types reaches n; the number of feed ratio conversion and extraction information in the correlation adjustment analysis data group is less than the number of feed ratio conversion and extraction information in the correlation fitting analysis data group.

[0040] When performing correlation analysis on the conversion and extraction information of the extracted ratio types, it is considered that the data of different egg-laying cycles do not necessarily have consistent variability in correlation, that is, the feed ratio conversion and extraction information corresponding to the egg-laying cycle does not present a simple gradual increase or decrease in egg production in the time dimension or other dimensions. Therefore, direct correlation analysis of all feed ratio conversion and extraction information does not have the condition to determine whether the analysis results meet the accuracy requirements of the correlation data. Therefore, this application divides the different feed ratio conversion and extraction information into reasonable groups, one part of which is used as the basic data for correlation fitting analysis, and the other part is used as reference data for verifying and adjusting whether the correlation data meets the required accuracy. Of course, considering that the ratio types will change under different egg-laying cycles, this change may affect the results of the correlation analysis. Therefore, there must be a certain number of feed ratio conversion and extraction information with the most complete ratio types in the different data groups divided to ensure that the fitting analysis of the correlation data covers all ratio types. The specific number of X and Y can be determined according to actual needs. The purpose of Y being less than X is to provide more data for the fitting analysis data set to improve the effect of the fitting analysis. On the other hand, it is to reduce the complexity of the adjustment analysis. After all, the adjustment verification after the fitting analysis is basically a fine-tuning of the accuracy of the correlation data and will not be applied to a large amount of data information.

[0041] S3: Conduct correlation training on the correlation grouping data for feed ratio conversion to form a ratio conversion correlation analysis result.

[0042] The correlation grouping data is subjected to correlation training for feed ratio conversion to form ratio conversion correlation analysis results, including: based on the correlation fitting analysis data group, correlation relationship training for feed ratio conversion is performed to form corresponding initial correlation fitting relationship data; based on the initial correlation fitting relationship data, relationship adjustment training with correlation accuracy as the goal is performed in combination with the correlation adjustment analysis data group to form feed ratio correlation target relationship data.

[0043] The correlation training for the correlation fitting analysis data group is mainly to establish the correlation information of the proportion and weight of the proportion types. The adjustment training for the correlation adjustment analysis data is mainly to adjust the fitting accuracy of the correlation relationship data formed by the correlation fitting analysis data group to achieve the expected correlation accuracy.

[0044] According to the correlation fitting analysis data set, the correlation relationship training for feed ratio conversion is carried out to form the corresponding correlation initial fitting relationship data, including: feed ratio content information from the correlation fitting analysis data set The feed ratio conversion extraction information with the highest comprehensive conversion rate is determined from the n different feed ratio conversion extraction information, and is calibrated as the ratio conversion benchmark information; based on the ratio conversion benchmark information, the correlation fitting analysis data group is subjected to a correlation fitting analysis based on the influence of the ratio change amount to form the correlation initial fitting relationship data.

[0045] The correlation training carried out on the correlation fitting analysis data group is mainly to establish a reasonable correlation on the proportion of the proportion of the proportion type and the weight information, so as to provide reference data for optimizing the feed ratio and weight. However, considering that the proportion and weight information of the proportion types in different egg-laying cycles do not have a unified reference, the analysis of the correlation has a certain independence in different egg-laying cycles, which will cause the accuracy of the analysis result to be reduced. Therefore, this application uses the feed ratio conversion extraction information with the most complete proportion types as the data benchmark, obtains the variation of the feed ratio conversion extraction information of other feed ratio conversion extraction information relative to the benchmark, and performs correlation analysis with the variation data, that is, the data of different egg-laying cycles can be standardized and unified, and the accuracy of the correlation analysis can also be improved. Of course, there will be multiple feed ratio conversion extraction information with the most complete proportion types. Here, the one with the highest comprehensive conversion rate of the proportion is used as the benchmark. This is to consider that the comprehensive conversion rate of the proportion reflects the situation of the egg in absorbing the components of the proportion types to a certain extent. It can be understood that these proportion types that affect egg production, the more the eggs are converted, the more they affect egg production. It should be noted that the conversion of feed ratios can be based on elements, that is, using chemical elements as the unit of feed ratio. This allows for better confirmation of feed ratio conversion and migration. Therefore, the key to feed ratio is to accurately control the daily feed intake of laying hens. Of course, if molecular or even material-level objects can be directly measured in eggs and feces, these material or molecular-level objects can also be used as feed ratio types.

[0046] Feed ratio content information from correlation fitting analysis data set The feed ratio conversion extraction information with the highest comprehensive conversion rate is determined from the n different feed ratio conversion extraction information, and marked as the ratio conversion benchmark information, including: the feed ratio content information in the correlation fitting analysis data group The ratio of all feed ratios of n types is converted and extracted, and the corresponding feed ratio content information is obtained. Average egg content ratio information , determine the corresponding ratio comprehensive conversion rate ,in, ; The comprehensive conversion rate of the ratio The maximum feed ratio conversion extraction information is calibrated as the ratio conversion benchmark information.

[0047] The comprehensive conversion rate of the ratio is the sum of the proportion of different ratio types in eggs relative to the proportion of the ratio types in feed. To a certain extent, it reflects the volume of the ratio types migrating into eggs, and reflects the degree of impact on egg production and even egg quality.

[0048] Based on the ratio conversion benchmark information, the correlation fitting analysis data group is subjected to the correlation fitting analysis based on the influence of the ratio change amount to form the correlation initial fitting relationship data, including: for the feed ratio conversion extraction information other than the ratio conversion benchmark information in the correlation fitting analysis data group, the following ratio change amount data are determined respectively: based on the feed ratio content information corresponding to the feed ratio conversion extraction information, the feed ratio benchmark change amount information relative to the feed ratio content information in the ratio conversion benchmark information is determined. , where k represents the number of different feed ratio conversion extraction information in the correlation fitting analysis data except the ratio conversion benchmark information and k<m, , The weight percentage difference between the weight percentage of the ratio type numbered n in the feed ratio conversion extraction information numbered k and the weight percentage of the ratio type numbered n in the ratio conversion benchmark information; based on the average total amount of feed supply ratio for a single head corresponding to the feed ratio conversion extraction information, determine the single feed supply benchmark change amount information relative to the average total amount of feed supply ratio for a single head in the ratio conversion benchmark information. ,in, The difference between the total amount of the average feed supply ratio of a single bird corresponding to the feed ratio conversion extraction information and the total amount of the average feed supply ratio of a single bird in the ratio conversion benchmark information; based on the average total feed ratio of a single bird corresponding to the feed ratio conversion extraction information, determine the single bird feed benchmark change amount information relative to the average total feed ratio of a single bird in the ratio conversion benchmark information ,in, The difference between the average total feed ratio of a single chicken corresponding to the feed ratio conversion extraction information and the average total feed ratio of a single chicken in the ratio conversion benchmark information; based on the average egg ratio content information corresponding to the feed ratio conversion extraction information, determine the egg ratio benchmark change information relative to the average egg ratio content information in the ratio conversion benchmark information. ,in, , The weight percentage difference between the average weight percentage of the ratio type numbered n in the feed ratio conversion extraction information numbered k and the average weight percentage of the ratio type numbered n in the ratio conversion benchmark information; based on the average feces ratio content information corresponding to the feed ratio conversion extraction information, determine the egg ratio benchmark change information relative to the average feces ratio content information in the ratio conversion benchmark information. ,in, , The weight percentage difference between the average weight percentage of the ratio type numbered n in the feed ratio conversion extraction information numbered k and the average weight percentage of the ratio type numbered n in the ratio conversion benchmark information; based on the average single egg ratio weight corresponding to the feed ratio conversion extraction information, determine the single egg benchmark change weight information relative to the average single egg ratio weight in the ratio conversion benchmark information ,in, The difference between the average egg ratio weight of a single chicken corresponding to the feed ratio conversion extraction information and the average egg ratio weight of a single chicken in the ratio conversion benchmark information; the average feces ratio weight of a single chicken corresponding to the feed ratio conversion extraction information is determined, and the average feces ratio weight of a single chicken relative to the average feces ratio weight in the ratio conversion benchmark information is determined. ,in, The difference between the average feces ratio weight of a single pig corresponding to the feed ratio conversion extraction information and the average feces ratio weight of a single pig in the ratio conversion benchmark information; the feed ratio benchmark change information corresponding to different feed ratio conversion extraction information , Single feed supply benchmark change information , Single feeding baseline change information , Egg ratio benchmark change information , Egg ratio benchmark change information , Single egg benchmark weight information , Single feces baseline weight change information , conduct correlation fitting analysis on configuration ratio and proportion weight, and establish the following correlation fitting relationship group: ; ; ; ;in, Indicates the weight percentage difference between the weight percentage of the numbered n compounding type in the feed and the weight percentage of the numbered n compounding type in the feed in the compounding conversion benchmark information; Indicates the weight percentage difference between the average weight percentage of the ratio type numbered n in the eggs and the weight percentage of the ratio type numbered n in the eggs in the ratio conversion benchmark information; Indicates the weight percentage difference between the average weight percentage of the ratio type numbered n in feces and the weight percentage of the ratio type numbered n in feces in the ratio conversion benchmark information; Indicates the egg ratio conversion related items corresponding to the ratio type numbered n; Indicates the feces ratio conversion related items corresponding to the ratio type numbered n; The correlation function representing the weight percentage of different proportions of species in the feed; Indicates the weight difference between the total amount of average feed supply ratio for a single head and the total amount of average feed supply ratio for a single head in the ratio conversion benchmark information; Indicates the weight difference between the average total food intake ratio of a single bird and the average total food intake ratio of a single bird in the ratio conversion benchmark information; Indicates feed utilization rate; Indicates the weight difference between the average single egg ratio weight and the average single egg ratio weight of the ratio conversion benchmark information; It represents the weight difference between the average feces ratio weight of a single egg and the average feces ratio weight of a single egg in the ratio conversion benchmark information; U represents the influence correlation formula of the ratio weight of a single egg; V represents the influence correlation formula of the ratio weight of a single egg.

[0049] The equations in the correlation fitting group represent the correlation between the change in feed ratio conversion information relative to the baseline data. This group includes correlations for each ratio type from feed to eggs and feces, correlations for ratios between different ratio types, correlations between the weight of feed supplied and the weight of food consumed, and correlations for the change in the weight of food consumed from feed to eggs and feces. For each ratio type from feed to eggs and feces, the correlations show how the change in the ratio of each ratio type is transferred to the corresponding baseline in eggs and feces. Specifically, an increase or decrease in the feed ratio relative to the baseline corresponds to an increase or decrease in the egg ratio relative to the baseline in eggs, and an increase or decrease in the feces ratio relative to the baseline in feces. This allows for the determination of the expected transfer of the ratio type change and the direction of this transfer, whether it flows to eggs or feces, providing reference information for subsequent nutritional and cost analysis. By focusing on the correlation between the ratios of different feed types, we can determine the limiting influence relationship between the changes in the ratios of different feed types relative to the baseline ratio. It is understandable that the increase or decrease in the ratios of different feed types is not independent, but rather mutually restrictive. This is manifested macroscopically in the directional migration of different feed types into laying hens. For example, increasing the proportion of feed types that affect egg production weight will cause the proportions of other feed types to decrease, further affecting the proportions of feed types related to factors such as laying hen weight and sleep. This can lead to a single promotion of egg production weight increase while ignoring the growth and development of laying hens, which will also reduce egg production overall. Therefore, the limiting influence relationship between feed types can be established based on big data training to establish a good correspondence, comprehensively focusing on multiple factors affecting laying hens' egg production to achieve reasonable allocation. The correlation between the weight of the feed supply and the weight of the food intake primarily reflects the control of the laying hens' feed intake when a given feed type is increased or decreased. This relationship ensures the rationality of feed supply, avoids feed waste that increases costs, and prevents situations where insufficient feed supply leads to an inability to prepare feed in advance. The correlation between the weight of the feed intake and the weight changes from feed to eggs and feces shows the absorption of the feed type by volume when a given feed type is increased or decreased. This helps to rationally control the total weight of the feed types based on the feed intake of laying hens, avoiding waste or insufficient supply. The different relationship equations in the fitted relationship group are interconnected and constrained, establishing a comprehensive relationship between the influence of feed types on feed ratio and weight. This provides a reasonable reference for adding feed types to the feed, and is more accurate and effective than empirical feed analysis.Of course, the corresponding correlation and other relationship parameters in the relationship can be defined according to the actual need for accuracy. They can be simple linear constants or multiple power relationship expressions. As long as they can be determined through correlation fitting analysis data, they are acceptable.

[0050] Based on the initial correlation fitting relationship data, combined with the correlation adjustment analysis data group, relationship adjustment training with correlation accuracy as the goal is carried out to form feed ratio correlation target relationship data, including: extracting information on different feed ratio conversions in the correlation adjustment analysis data group, determining the corresponding comprehensive conversion rate of the ratios, and sorting the different feed ratio conversion extraction information in order from small to large according to the comprehensive conversion rate of the ratios to form an ordered adjustment analysis data group; and performing deviation-based verification adjustment analysis on the correlation fitting relationship group in the order of different feed ratio conversion extraction information in the ordered adjustment analysis data group to form feed ratio correlation target relationship data.

[0051] The relationship adjustment training for the initial correlation fitting relationship data using the correlation adjustment analysis data group takes into account a certain adjustment direction. Therefore, the order of extraction is sorted and adjusted based on the ratio comprehensive conversion rate as a reference until the correlation relationship data meets the required accuracy.

[0052] According to the order of different feed ratio conversion extraction information in the ordered adjustment analysis data group, the correlation fitting relationship group is sequentially subjected to verification adjustment analysis based on deviation from the target to form feed ratio correlation target relationship data, including: sequentially extracting different feed ratio conversion extraction information in the ordered adjustment analysis data group, and performing verification adjustment analysis on the correlation fitting relationship group in the following manner: obtaining the following data information of the feed ratio conversion extraction information: obtaining the average weight percentage of the corresponding different ratio types in the feed in the feed ratio conversion extraction information relative to the weight percentage of the corresponding ratio type in the feed in the ratio conversion benchmark information; , i represents the sequence number of different feed ratio conversion extraction information in the ordered adjustment analysis data group; obtain the difference between the average weight percentage of different ratio types in eggs corresponding to the feed ratio conversion extraction information and the weight percentage of the corresponding ratio type in eggs in the ratio conversion benchmark information ; Obtain the difference between the average weight percentage of the different ratio types in the feces corresponding to the feed ratio conversion extraction information and the weight percentage of the corresponding ratio types in the feces in the ratio conversion benchmark information ; Obtain the feed supply benchmark change information of the average feed supply ratio of a single head in the feed ratio conversion extraction information relative to the average feed supply ratio of a single head in the ratio conversion benchmark information ; Obtain the adjusted feed benchmark change information of the average total feed ratio of a single bird in the feed ratio conversion extraction information relative to the average total feed ratio of a single bird in the ratio conversion benchmark information ; Get the adjusted egg benchmark weight change information of the average egg ratio weight of a single egg in the feed ratio conversion extraction information relative to the average egg ratio weight of a single egg in the ratio conversion benchmark information ; Obtain the adjusted feces benchmark change weight information of the average feces ratio weight of a single head in the feed ratio conversion extraction information relative to the average feces ratio weight of a single head in the ratio conversion benchmark information ; According to the weight percentage difference of eggs corresponding to different ratios , feces weight percentage difference and adjust feed supply benchmark change information , Adjust the egg base weight information and adjust stool baseline change weight information , combined with the correlation fitting relationship group, determine the feed ratio correlation percentage difference corresponding to the feed ratio conversion extraction information , adjust the difference in correlation between meal ratio and adjust feed supply related variance ; According to the feed ratio conversion extraction information, the weight percentage difference of the feed ratio of different types corresponding to the feed ratio , Adjust the food intake baseline change information and adjust feed supply benchmark change information , and combined with the feed ratio correlation percentage difference , adjust the difference in correlation between meal ratio and adjust feed supply related variance , respectively make the following adjustment conditions judgment: the weight percentage difference of feed ratio corresponding to different ratio types Percent difference in correlation with feed ratio :If all meet , then the feed ratio compliance information is formed, among which, Indicates the allowable deviation of the proportion of the proportion type numbered n; if there is , then the feed ratio does not meet the standard information, and according to the feed ratio conversion and extraction information, the relationship in the correlation fitting relationship group is , make adjustments so that the newly formed feed ratio correlation percentage difference All satisfied , and the relationship in the correlation fitting relationship group is fitted according to the proportional relationship of different ratio types Update so that the mapping relationship in the relationship envelope feed ratio conversion extraction information corresponding to the ratio type ratio relationship; adjust the feeding benchmark change information Difference in correlation with adjusted meal ratio :If satisfied , then the conversion target information is formed, among which, Indicates the allowable deviation of a single feeding; if not satisfied , then the conversion does not meet the standard information, and according to the feed ratio conversion extraction information, the relationship in the correlation fitting relationship group is Adjustments are made so that the newly formed adjustment of the feeding ratio correlation difference satisfy ; Information on changes in feed supply benchmarks and adjusted feed supply related differences :If satisfied , then the supply compliance information is formed, among which, Indicates the allowable deviation of a single supply; if not satisfied , then the information of substandard supply is formed, and the relationship in the correlation fitting relationship group is extracted according to the feed ratio conversion information. Adjustments are made so that the newly formed feed supply correlation difference satisfy ; If the adjustment conditions for the feed ratio conversion extraction information are all met, the adjustment analysis is stopped; if the adjustment conditions are not met, the feed ratio conversion extraction information in the ordered adjustment analysis data group is continuously extracted in sequence for adjustment analysis until the adjustment conditions are all met or all the feed ratio conversion extraction information in the ordered adjustment analysis data group is extracted.

[0053] The adjustment training of the correlation fitting relationship group is mainly to provide new feed ratio conversion and extraction information to verify whether the deviation between the predicted data formed by the correlation fitting relationship group and the actual data is within the required allowable deviation range. Of course, because the relationship formulas in the correlation fitting relationship group are mutually correlated, each relationship formula has the independence of data verification, that is, the accuracy of each relationship formula can be verified. In this way, there are three conditions for adjusting whether the accuracy meets the standard. One is whether the correlation of different ratio types in the proportion migration meets the standard. This also involves the proportion mapping relationship between different ratios. After all, the migration of different ratios will also be affected by the interaction and influence between different ratio types, that is, different chemical elements have mutual influence on biological relationships. Therefore, when the correlation does not meet the standard and is adjusted, it is also necessary to cover the mapping relationship function between different ratio types, that is, to envelop the new mapping relationship for adjustment. The second is the relationship between feed intake and supply. Different ratios will result in different actual feed weights, and it is impossible to ensure that the provided feed is eaten up immediately during feeding. There is a certain amount of consumption, and this consumption and the type of ratio will affect the feeding habits of laying hens. For example, certain chemical elements affect the appetite of laying hens, and different ratios will result in different feed intakes, so adjustments and analysis are needed. The third is the conversion of the total ratio into eggs and feces, which reflects the impact of volume on egg production. If the standard is not met, the relationship needs to be adjusted. It should be noted that the adjustment methods adopted for the three relationship formulas are diverse. It can be to introduce a unified proportional value, and after the conditions are met, the proportional value is assigned to each related item to form a new related item. It can also be to directly increase or decrease the constant term of the related item to make adjustments, or to increase or decrease the power formula in the related item to make adjustments.

[0054] The present invention also provides an AI large-model dynamic training cloud platform for optimizing laying hen feed ratios, which includes: a data acquisition unit for collecting corresponding feed ratio conversion data in different egg-laying cycles; an extraction and division unit for extracting feed ratio information from different feed ratio conversion data collected by the data acquisition unit to form corresponding feed ratio conversion extraction information, and performing group division based on correlation analysis to form correlation analysis division data; an analysis and processing unit for performing correlation training on the correlation analysis division data formed by the extraction and division unit to form a ratio conversion correlation analysis result.

[0055] The platform completes the collection of feed ratio conversion data through the data acquisition unit, and uses the extraction and analysis unit to complete the pre-processing analysis of the collected data before model training to form more targeted feature data information. On this basis, the model training and analysis of the analysis and processing unit establishes a correlation model between the ratio type and the ratio weight, which provides important analysis model data for subsequent feed ratio selection and optimization, and is an important material basis for achieving reasonable feed ratio optimization.

[0056] It should also be noted that since the training method proposed in this application mostly utilizes the basic feed ratio conversion data of the egg-laying cycle, there is no more complicated data processing and analysis content. Therefore, it can be used with the help of general AI large models such as deepseek for training and processing. These general AI large models can also accurately establish a reasonable feed ratio optimization model, which provides effective assistance for the improvement of laying hen production capacity and the optimization of feed ratio. The feed ratio analysis and processing of laying hen production capacity based on this greatly reduces the cost of analysis, and on the other hand, it can also accurately and reasonably realize the control of feed ratio.

[0057] In summary, the AI ​​large-scale dynamic model training method and cloud platform for optimizing laying hen feed ratio provided by the embodiments of the present invention have the following beneficial effects: This method collects feed ratio conversion data for different egg-laying cycles, extracts the proportions of different ratio types in feed, eggs and feces, as well as the weight of the ratios in feed, eggs and feces, and then conducts a correlation analysis on the ratio weight in the ratio and weight, and establishes reference data for the reasonable selection of ratio types in the ratio and weight. On the one hand, this method provides important reference data for the optimization and reasonable selection of feed ratios. On the other hand, feed ratios are not limited by experience, and can benefit more breeding units using the big data platform, greatly improving the efficiency of laying hen production, saving breeding costs and significantly increasing egg production.

[0058] The platform completes the collection of feed ratio conversion data through the data acquisition unit, and uses the extraction and analysis unit to complete the pre-processing analysis of the collected data before model training to form more targeted feature data information. On this basis, the model training and analysis of the analysis and processing unit establishes a correlation model between the ratio type and the ratio weight, which provides important analysis model data for subsequent feed ratio selection and optimization, and is an important material basis for achieving reasonable feed ratio optimization.

[0059] In the embodiment of the present application, "indication" may include direct indication and indirect indication, and may also include explicit indication and implicit indication. The information indicated by a certain information is called information to be indicated. In the specific implementation process, there are many ways to indicate the information to be indicated, such as but not limited to, the information to be indicated can be directly indicated, such as the information to be indicated itself or the index of the information to be indicated. The information to be indicated can also be indirectly indicated by indicating other information, wherein there is an association relationship between the other information and the information to be indicated. It is also possible to indicate only a part of the information to be indicated, while the other parts of the information to be indicated are known or agreed in advance. For example, the indication of specific information can also be achieved by means of the arrangement order of each piece of information agreed in advance (such as specified in the protocol), thereby reducing the indication overhead to a certain extent. At the same time, the common parts of each piece of information can also be identified and indicated uniformly to reduce the indication overhead caused by indicating the same information separately.

[0060] In addition, the specific indication method can also be various existing indication methods, such as but not limited to the above-mentioned indication methods and various combinations thereof. The specific details of the various indication methods can be referred to the prior art and will not be repeated herein. As can be seen from the above, for example, when it is necessary to indicate multiple information of the same type, there may be a situation where the indication methods for different information are different. In the specific implementation process, the required indication method can be selected according to specific needs. The embodiment of the present application does not limit the selected indication method. In this way, the indication method involved in the embodiment of the present application should be understood to cover various methods that can enable the party to be indicated to obtain the information to be indicated.

[0061] It should be understood that the information to be indicated can be sent as a whole or divided into multiple sub-information and sent separately, and the sending period and / or sending time of these sub-information can be the same or different. The specific sending method is not limited in the embodiments of this application. The sending period and / or sending time of these sub-information can be predefined, for example, predefined according to a protocol, or can be configured by the transmitting device by sending configuration information to the receiving device.

[0062] "Pre-definition" or "pre-configuration" can be implemented by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in the device, and the embodiments of the present application do not limit the specific implementation method. Among them, "saving" can mean saving in one or more memories. The one or more memories can be set separately or integrated in an encoder or decoder, a processor, or a communication device. The one or more memories can also be partially set separately and partially integrated in a decoder, a processor, or a communication device. The type of memory can be any form of storage medium, and the embodiments of the present application do not limit this.

[0063] The "protocol" involved in the embodiments of the present application may refer to a protocol family in the communication field, a standard protocol with a similar protocol family frame structure, or a related protocol used in future communication systems. The embodiments of the present application do not make specific limitations on this.

[0064] In the embodiments of the present application, descriptions such as "when...", "in the case of...", "if" and "if" all mean that the device will perform corresponding processing under certain objective circumstances. It does not limit the time, nor does it require the device to perform judgment actions when implemented, nor does it mean that there are other limitations.

[0065] In the description of the embodiments of this application, unless otherwise specified, " / " indicates that the associated objects are in an "or" relationship. For example, A / B can mean A or B. "And / or" in the embodiments of this application is merely a description of the associated relationship between the associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, in the description of the embodiments of this application, unless otherwise specified, "multiple" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural. Furthermore, to facilitate the clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish between identical or similar items with substantially the same function or effect. Those skilled in the art will understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit differences. At the same time, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or design. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way for easy understanding.

[0066] It should be understood that the processor in the embodiments of the present application may be a central processing unit (CPU), but may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0067] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0068] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in the embodiments of this application are fully or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0069] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0070] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0071] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0072] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0073] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0074] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0075] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0076] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0077] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0078] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. The AI ​​large-scale dynamic training method for layer feed ratio optimization is characterized by: include: Collect the corresponding feed ratio conversion data in different egg-laying cycles, extract the feed ratio information, and form the corresponding feed ratio conversion extraction information; performing grouping and division based on correlation analysis on the different feed ratio conversion and extraction information to form correlation analysis division data; The correlation grouping data is subjected to correlation training for feed ratio conversion to form a ratio conversion correlation analysis result.

2. The AI ​​large model dynamic training method for layer feed ratio optimization according to claim 1 is characterized in that, The collecting of corresponding feed ratio conversion data in different egg-laying cycles, extracting feed ratio information, and forming corresponding feed ratio conversion extraction information includes: Extract the corresponding feed ratio content information from the feed ratio conversion data of different egg-laying cycles ,in, , m represents the number of the different egg-laying cycles, n represents the number of the different ratio types determined based on the ratio with the largest number of ratio types among the feed ratios corresponding to all the egg-laying cycles, Indicates the weight percentage of the ratio type numbered n in the feed in the egg-laying cycle numbered m; The feed ratio conversion data of different laying cycles are used to extract the corresponding average feed supply ratio of a single egg. and the average total food intake ratio of a single ; Extract the corresponding average egg ratio content information from the feed ratio conversion data of different egg laying cycles ,in, , Indicates the average weight percentage of the ratio type numbered n in the eggs in the egg-laying cycle numbered m; Extract the corresponding average feces content information from the feed ratio conversion data of different egg-laying cycles ,in, , Indicates the average weight percentage of the ratio type numbered n in the feces during the egg-laying cycle numbered m; The feed ratio conversion data of different laying cycles are used to extract the corresponding single average egg ratio weight and the average feces weight of a single ; For different egg-laying cycles, the corresponding feed ratio content information is collected , the average feed supply ratio for each , the average total food intake ratio of a single , the average egg content information , the average proportion and content information of the feces , the average weight of a single egg And the average feces weight of the single , forming the corresponding feed ratio conversion extraction information.

3. The AI ​​large model dynamic training method for layer feed ratio optimization according to claim 2 is characterized in that, The grouping and dividing of the different feed ratio conversion and extraction information based on correlation analysis to form correlation analysis and division data includes: The different feed ratio conversion and extraction information is divided into two different groups, forming a correlation fitting analysis data group and a correlation adjustment analysis data group, respectively, and satisfying: The correlation fitting analysis data group contains at least X feed ratio content information corresponding to the feed ratio conversion and extraction information. The number of types of medium ratio reaches n; The correlation adjustment analysis data group contains at least Y feed ratio content information corresponding to the feed ratio conversion and extraction information. The number of types of medium ratio reaches n; The amount of the feed ratio conversion and extraction information in the correlation adjustment analysis data group is less than the amount of the feed ratio conversion and extraction information in the correlation fitting analysis data group.

4. The AI ​​large model dynamic training method for layer feed ratio optimization according to claim 3 is characterized in that, The performing correlation training on the correlation grouping data for feed ratio conversion to form a ratio conversion correlation analysis result includes: Based on the correlation fitting analysis data group, correlation relationship training for feed ratio conversion is performed to form corresponding correlation initial fitting relationship data; According to the correlation initial fitting relationship data, in combination with the correlation adjustment analysis data group, relationship adjustment training with correlation accuracy as the goal is performed to form feed ratio correlation target relationship data.

5. The AI ​​large model dynamic training method for optimizing laying hen feed ratio according to claim 4 is characterized in that, The method of performing correlation relationship training for feed ratio conversion based on the correlation fitting analysis data group to form corresponding correlation initial fitting relationship data includes: The feed ratio content information from the correlation fitting analysis data group The feed ratio conversion extraction information having n different types of ratios is determined to obtain the feed ratio conversion extraction information with the highest comprehensive conversion rate, and the information is marked as the ratio conversion benchmark information; Based on the ratio conversion benchmark information, the correlation fitting analysis data group is subjected to a correlation fitting analysis based on the influence of the ratio change amount to form the correlation initial fitting relationship data.

6. The AI ​​large model dynamic training method for optimizing laying hen feed ratio according to claim 5 is characterized in that: The feed ratio content information from the correlation fitting analysis data group The feed ratio conversion extraction information having n different types of ratios is determined to have the highest comprehensive conversion rate among the feed ratio conversion extraction information, and is marked as the ratio conversion benchmark information, including: The feed ratio content information in the correlation fitting analysis data group The ratio of all the feed ratios of n types is converted into extracted information, and the corresponding feed ratio content information is obtained. and the average egg content information , determine the corresponding ratio comprehensive conversion rate ,in, ; The comprehensive conversion rate of the ratio The maximum feed ratio conversion extraction information is calibrated as the ratio conversion benchmark information.

7. The AI ​​large model dynamic training method for optimizing laying hen feed ratio according to claim 6 is characterized in that: The method of performing a correlation fitting analysis on the correlation fitting analysis data group based on the influence of the ratio change amount based on the ratio conversion benchmark information to form the correlation initial fitting relationship data includes: For the feed ratio conversion extraction information other than the ratio conversion benchmark information in the correlation fitting analysis data group, the following ratio change data are determined respectively: Determine the feed ratio base change information relative to the feed ratio content information in the feed ratio conversion base information based on the feed ratio content information corresponding to the feed ratio conversion extraction information , wherein k represents the number of the feed ratio conversion extraction information other than the ratio conversion benchmark information in the correlation fitting analysis data and k<m, , The weight percentage difference between the weight percentage of the ratio type numbered n in the feed ratio conversion extraction information numbered k and the weight percentage of the ratio type numbered n in the ratio conversion reference information; According to the single average feed supply ratio total amount corresponding to the feed ratio conversion extraction information, determine the single feed supply benchmark change amount information relative to the single average feed supply ratio total amount in the ratio conversion benchmark information ,in, The difference between the total amount of the average feed supply ratio for a single head corresponding to the feed ratio conversion extraction information and the total amount of the average feed supply ratio for a single head in the ratio conversion benchmark information; According to the single average total feeding ratio corresponding to the feed ratio conversion extraction information, determine the single feeding benchmark change amount information relative to the single average total feeding ratio in the ratio conversion benchmark information ,in, The difference between the average total feed ratio of a single head corresponding to the feed ratio conversion extraction information and the average total feed ratio of a single head in the ratio conversion benchmark information; According to the egg average ratio content information corresponding to the feed ratio conversion extraction information, determine the egg ratio benchmark change information relative to the egg average ratio content information in the ratio conversion benchmark information ,in, , The weight percentage difference between the average weight percentage of the feed ratio type numbered n in the feed ratio conversion extraction information numbered k and the average weight percentage of the feed ratio type numbered n in the feed ratio conversion benchmark information; Determine the egg ratio reference change information relative to the average feces ratio information in the ratio conversion reference information based on the average feces ratio content information corresponding to the feed ratio conversion extraction information ,in, , The weight percentage difference between the average weight percentage of the feed ratio type numbered n in the feed ratio conversion extraction information numbered k and the average weight percentage of the feed ratio type numbered n in the feed ratio conversion benchmark information; According to the single average egg ratio weight corresponding to the feed ratio conversion extraction information, determine the single egg benchmark change weight information relative to the single average egg ratio weight in the ratio conversion benchmark information ,in, The difference between the average single egg ratio weight corresponding to the feed ratio conversion extraction information and the average single egg ratio weight in the ratio conversion benchmark information; Determine the single feces reference change weight information relative to the single feces reference weight in the feed ratio conversion reference information based on the single average feces ratio weight corresponding to the feed ratio conversion extraction information. ,in, The difference between the average feces ratio weight of a single head corresponding to the feed ratio conversion extraction information and the average feces ratio weight of a single head in the ratio conversion benchmark information; The feed ratio benchmark change information corresponding to the feed ratio conversion and extraction information according to different feed ratios , the single feed supply benchmark change information , the single feeding baseline change information , the egg ratio benchmark change information , the egg ratio benchmark change information , the reference weight change information of a single egg , the single feces baseline change weight information , conduct correlation fitting analysis on configuration ratio and proportion weight, and establish the following correlation fitting relationship group: ； ； ； ； in, Indicates the weight percentage difference between the weight percentage of the numbered n compounding type in the feed and the weight percentage of the numbered n compounding type in the feed in the compounding conversion benchmark information; Indicates the weight percentage difference between the average weight percentage of the ratio type numbered n in the eggs and the weight percentage of the ratio type numbered n in the eggs in the ratio conversion benchmark information; Indicates the weight percentage difference between the average weight percentage of the ratio type numbered n in feces and the weight percentage of the ratio type numbered n in feces in the ratio conversion benchmark information; Indicates the egg ratio conversion related items corresponding to the ratio type numbered n; Indicates the feces ratio conversion related items corresponding to the ratio type numbered n; The correlation function representing the weight percentage of different proportions of species in the feed; The weight difference between the total amount of the average feed supply ratio for a single head and the total amount of the average feed supply ratio for a single head according to the ratio conversion benchmark information; The weight difference between the average total food ratio of a single chicken and the average total food ratio of a single chicken according to the ratio conversion benchmark information; Indicates feed utilization rate; The weight difference between the average ratio of a single egg and the average ratio of a single egg in the ratio conversion benchmark information; It represents the weight difference between the average feces ratio weight of a single egg and the average feces ratio weight of the single egg according to the ratio conversion benchmark information; U represents the influence correlation formula of the ratio weight of a single egg; V represents the influence correlation formula of the ratio weight of a single egg.

8. The AI ​​large model dynamic training method for optimizing laying hen feed ratio according to claim 7 is characterized in that: The initial correlation fitting relationship data is combined with the correlation adjustment analysis data group to perform relationship adjustment training with correlation accuracy as the goal to form feed ratio correlation target relationship data, including: For the different feed ratio conversion and extraction information in the correlation adjustment analysis data group, determining the corresponding ratio comprehensive conversion rate, and sorting the different feed ratio conversion and extraction information in order from small to large according to the ratio comprehensive conversion rate, to form an ordered adjustment analysis data group; According to the order of the different feed ratio conversion and extraction information in the ordered adjustment analysis data group, the correlation fitting relationship group is sequentially subjected to verification adjustment analysis based on deviation from the target to form the feed ratio correlation target relationship data.

9. The AI ​​large model dynamic training method for optimizing laying hen feed ratio according to claim 8 is characterized in that: The step of performing deviation-based verification adjustment analysis on the correlation fitting relationship group in sequence according to the order of the different feed ratio conversion and extraction information in the ordered adjustment analysis data group to form the feed ratio correlation target relationship data includes: Sequentially extract the different feed ratio conversion extraction information in the ordered adjustment analysis data group, and perform verification adjustment analysis on the correlation fitting relationship group in the following manner: Obtain the following data information of the feed ratio conversion and extraction information: Obtain the feed ratio weight percentage difference between the average weight percentage of the different ratio types in the feed corresponding to the feed ratio conversion extraction information and the weight percentage of the corresponding ratio type in the feed in the ratio conversion benchmark information , i represents the sequence number of the different feed ratio conversion and extraction information in the ordered adjustment analysis data group; Obtain the difference in egg ratio weight percentage between the average weight percentage of different ratio types in the eggs corresponding to the feed ratio conversion extraction information and the weight percentage of the corresponding ratio types in the eggs in the ratio conversion benchmark information. ; Obtain the difference in feces weight percentage between the average weight percentage of different ratio types in the feed ratio conversion extraction information and the weight percentage of the corresponding ratio type in the feces in the ratio conversion benchmark information. ; Obtain the feed supply benchmark change information of the average feed supply ratio of a single head in the feed ratio conversion extraction information relative to the average feed supply ratio of a single head in the ratio conversion benchmark information. ; Obtain the adjusted feed reference change amount information of the average total feed ratio of a single head in the feed ratio conversion extraction information relative to the average total feed ratio of a single head in the ratio conversion reference information ; Obtain the adjusted egg benchmark weight change information of the average single egg ratio weight in the feed ratio conversion extraction information relative to the average single egg ratio weight in the ratio conversion benchmark information ; Obtain the adjusted feces reference change weight information of the average feces ratio weight of a single head in the feed ratio conversion extraction information relative to the average feces ratio weight of a single head in the ratio conversion reference information ; The difference in weight percentage of the egg ratio corresponding to different ratio types , the difference in weight percentage of the feces ratio And the information on the change in feed supply benchmark , the adjustment of the egg benchmark weight change information and the adjusted stool baseline change weight information , combined with the correlation fitting relationship group, determine the feed ratio correlation percentage difference corresponding to the feed ratio conversion extraction information , adjust the difference in correlation between meal ratio and adjust feed supply related variance ; The feed ratio weight percentage difference of different ratio types corresponding to the feed ratio conversion and extraction information is obtained according to the feed ratio , the adjustment of the eating reference change information And the information on the change in feed supply benchmark , and combined with the feed ratio correlation percentage difference , the adjustment of the feeding ratio correlation difference and the adjusted feed supply related differential , respectively make the following adjustment condition judgments: The difference in weight percentage of the feed ratio corresponding to different ratio types The difference in percentage of correlation with the feed ratio : If all meet , then the feed ratio compliance information is formed, among which, Indicates the allowable deviation of the proportion of the proportion type numbered n; If exists , then the feed ratio does not meet the standard information is formed, and the relationship formula in the correlation fitting relationship group is matched according to the feed ratio conversion and extraction information. , make adjustments so that the newly formed feed ratio correlation percentage difference All satisfied , and the relationship in the correlation fitting relationship group is fitted according to the proportional relationship of different ratio types The updating is performed so that the mapping relationship in the relational expression encompasses the ratio relationship of the ratio types corresponding to the feed ratio conversion and extraction information; Adjusting the eating standard change amount information and the difference in correlation with the adjusted feeding ratio : If satisfied , then the conversion target information is formed, among which, Indicates the allowable deviation for a single feeding; If not satisfied , then the conversion failure information is formed, and the relationship formula in the correlation fitting relationship group is extracted according to the feed ratio conversion information. Adjustment is made so that the newly formed adjustment feeding ratio correlation difference satisfy ; Adjustment of feed supply benchmark change information and the adjusted feed supply correlation difference : If satisfied , then the supply compliance information is formed, among which, Indicates the allowable deviation of a single supply; If not satisfied , then the supply substandard information is formed, and the relationship formula in the correlation fitting relationship group is matched according to the feed ratio conversion and extraction information. Adjustment is made so that the newly formed adjusted feed supply correlation difference satisfy ; If the adjustment conditions are all met for the feed ratio conversion extraction information, the adjustment analysis is stopped; if any of the adjustment conditions are not met, the feed ratio conversion extraction information in the ordered adjustment analysis data group is continuously extracted in sequence for adjustment analysis until all the adjustment conditions are met or all the feed ratio conversion extraction information in the ordered adjustment analysis data group is extracted.

10. An AI large-scale dynamic training cloud platform for layer feed ratio optimization, using the AI ​​large-scale dynamic training method for layer feed ratio optimization according to any one of claims 1 to 9, characterized in that: include: Data collection unit, used to collect feed ratio conversion data corresponding to different egg-laying cycles; An extraction and division unit is used to extract feed ratio information from the different feed ratio conversion data collected by the data collection unit to form corresponding feed ratio conversion extraction information, and perform grouping and division based on correlation analysis to form correlation analysis division data; The analysis and processing unit is used to perform correlation training on the correlation analysis and division data formed by the extraction and division unit to form a matching conversion correlation analysis result.

Images