A fully automated feeding control method and system for cattle and sheep in multiple areas

Abstract

This invention discloses a fully automated feeding control method and system for cattle and sheep in multiple regions, comprising: acquiring basic information and feeding behavior data of cattle and sheep in each region, preprocessing the data, and then training a feeding speed prediction model using a support vector machine algorithm; constructing a relationship model between feed physical parameters and feeding speed, and obtaining the initial feeding speed for each region based on the relationship model and the feed physical parameters of each region; acquiring real-time feeding behavior data and real-time satiety data of cattle and sheep in each region, and determining whether the average feeding speed of cattle and sheep meets the requirements; if it does not meet the requirements, generating a feeding speed adjustment command based on the changes in the feeding speed of cattle and sheep using a fuzzy control algorithm; acquiring feeding demand data and current feed supply data of cattle and sheep in each region, and calculating the optimal feeding speed. This invention achieves precise feeding, improves feeding efficiency, and effectively solves the problems of supply-demand mismatch and low efficiency in traditional feeding methods.

Description

Technical Field

[0001] This invention belongs to the field of information technology, and in particular relates to a fully automatic feeding control method and system for cattle and sheep in multiple areas. Background Technology

[0002] A key technical challenge in multi-zone fully automated feeding control systems for cattle and sheep is precisely controlling the feeding rate in each zone based on the real-time feeding speed and feed characteristics. The difficulty lies in the fact that different zones contain cattle and sheep of varying breeds, numbers, ages, and body types, resulting in significant differences in their feeding behaviors and speeds. Furthermore, the physical properties of different feeds, such as particle size, density, and flowability, also vary, all of which affect the control of the feeding rate. Improper control of the feeding rate can lead to feed blockages, waste, or competition for food among cattle and sheep, severely impacting feeding efficiency.

[0003] Feeding rate control must consider not only the feeding behavior and feed characteristics of cattle and sheep, but also the dynamic changes in their feeding behavior. For example, during feeding, as the satiety level of cattle and sheep increases, their feeding rate gradually decreases, requiring real-time adjustments to the feeding rate to adapt to this change. Simultaneously, coordination between different areas must be considered: how to ensure feeding efficiency in each area while avoiding insufficient or excessive feed in individual areas. This necessitates an intelligent collaborative control strategy that dynamically adjusts the feeding rate of each area based on its real-time status to achieve optimal feeding results for the entire system. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a fully automated feeding control method and system for cattle and sheep in multiple regions, thereby resolving the issues present in the prior art.

[0005] To achieve the above objectives, the present invention provides a fully automated feeding control method for cattle and sheep in multiple areas, comprising:

[0006] Basic information and feeding behavior data of cattle and sheep in various regions were obtained. After preprocessing the feeding behavior data, a support vector machine algorithm was used to train a model to predict the feeding speed of cattle and sheep.

[0007] A model relating feed physical parameters to feed feeding rate is constructed to obtain feed physical parameters for each region. Based on the model and the feed physical parameters for each region, the initial feed feeding rate for each region is obtained.

[0008] The system acquires real-time feeding behavior data and real-time satiety data of cattle and sheep in each region to determine whether the average feeding speed of cattle and sheep in each region meets the requirements. If the average feeding speed does not meet the requirements, a fuzzy control algorithm is used to generate feeding speed control instructions based on the changes in the feeding speed of cattle and sheep in each region.

[0009] Obtain feeding demand data and current feed supply data for cattle and sheep in each region, and calculate the optimal feed supply rate for each region.

[0010] Optionally, the process of training a model for predicting the feeding speed of cattle and sheep using a support vector machine algorithm includes:

[0011] Basic information on cattle and sheep in each region is obtained, and a basic attribute database of cattle and sheep is constructed based on the basic information. Feeding behavior data of cattle and sheep in each region are collected in real time and preprocessed. Based on the basic attribute database of cattle and sheep, the preprocessed feeding behavior data is divided into several data subsets. Support vector machines are used to train each data subset to obtain the corresponding cattle and sheep feeding speed prediction model, and the trained model parameters are stored in the model library. The preprocessing includes data cleaning and normalization operations.

[0012] Optionally, the feed physical parameters of each region are input into the relational model to obtain the initial feeding rate of each region; if the feed quality changes, the feed physical parameters are obtained in real time to update the initial feeding rate of each region; wherein, the feed physical parameters include feed particle size, feed particle density, and flowability parameters.

[0013] Optionally, the process of determining whether the average feeding rate of cattle and sheep in each region meets the requirements includes:

[0014] The system acquires real-time location information of cattle and sheep in each region, and obtains the number of cattle and sheep in each region based on the real-time location information; it acquires and preprocesses images of the feeding behavior of cattle and sheep in each region, extracts key features of the feeding behavior, and inputs them into a trained convolutional neural network model to obtain the satiety level values ​​of cattle and sheep in each region; it obtains the average feeding speed of each region based on the satiety level data, the number of cattle and sheep, and the key features of the feeding behavior; it obtains a preset normal feeding speed threshold based on the breed, age, and physiological state of cattle and sheep; and it determines whether the average feeding speed of each region meets the requirements based on the preset normal feeding speed threshold and the average feeding speed of each region.

[0015] Optionally, the process of generating feeding speed control instructions includes:

[0016] The real-time average feeding rate data of cattle and sheep in the region is obtained, and the real-time average feeding rate data is input into the fuzzy control algorithm to obtain the feeding rate adjustment value. The feeding rate adjustment value is matched with the real-time average feeding rate data. If they do not match, the feeding rate adjustment value is regenerated by the fuzzy control algorithm. The fuzzy control algorithm continuously optimizes based on the feeding rate control command of each region and the real-time average feeding rate.

[0017] Optionally, the process of calculating the optimal feeding rate for each region includes:

[0018] Feed demand prediction results are obtained based on historical feeding data and the current number of cattle and sheep. Based on the feed supply data and the feed demand prediction results, an optimization algorithm is used to calculate the optimal feeding rate for each region. If the difference between the actual feeding amount and the feed demand prediction result in a region exceeds a preset value, the optimal feeding rate is updated.

[0019] Optionally, the process of calculating the optimal feeding rate for each region using an optimization algorithm based on feed supply data and the feeding demand prediction results includes:

[0020] Historical and real-time data on feed supply and feeding demand for each region are obtained. Based on the historical and real-time data on feed supply, the feed supply for each region is obtained, and based on the historical and real-time data on feeding demand, the feeding demand for each region is obtained. Using the predicted feed supply and feeding demand for each region as constraints, a feed supply rate optimization model is established. The objective function of the feed supply rate optimization model is to minimize the absolute value of the difference between the total feed supply and the total demand. A heuristic algorithm is used to solve the feed supply rate optimization model to obtain the optimal feed supply rate for each region.

[0021] Optionally, it also includes acquiring real-time data on feeding speed and cattle / sheep feeding behavior in each area, performing correlation analysis on the feeding speed and cattle / sheep feeding behavior data to obtain key factors for efficiency evaluation, evaluating feeding efficiency in each area based on the key factors for efficiency evaluation, and optimizing feeding management strategies based on cattle / sheep feeding behavior and feed supply data in the area if the feeding efficiency does not meet preset requirements.

[0022] The present invention also provides a fully automatic feeding control system for cattle and sheep in multiple areas based on the above method, comprising:

[0023] The feeding speed prediction module is used to predict the feeding speed of cattle and sheep in different areas.

[0024] The initial feeding speed determination module is used to determine the initial feeding speed of each area;

[0025] The feeding speed change detection module is used to determine whether the feeding speed of cattle and sheep in different areas has changed.

[0026] The dynamic feed speed adjustment module is used to adjust the feed speed according to changes in the feeding speed of cattle and sheep;

[0027] The multi-region material supply coordination module is used to coordinate the material supply speed between different regions;

[0028] The feeding efficiency assessment and optimization module is used to assess feeding efficiency and optimize feeding management strategies.

[0029] Compared with the prior art, the present invention has the following advantages and technical effects:

[0030] This invention uses pre-established cattle and sheep information and real-time feeding behavior data to predict the feeding speed of cattle and sheep in different areas using a support vector machine algorithm. Combined with the physical characteristics of the feed, an initial feeding rate is determined. During the feeding process, the feeding behavior and satiety levels of cattle and sheep are analyzed in real time, and changes in feeding speed are detected using a convolutional neural network. If changes occur, a fuzzy control algorithm is used to dynamically adjust the feeding speed. Simultaneously, a multi-agent collaborative control algorithm is used to coordinate feeding in different areas to avoid uneven supply. This invention also uses big data analysis to evaluate feeding efficiency, identify inefficient areas, and optimize feeding strategies using data mining techniques. This invention achieves precise feeding and improves feeding efficiency, effectively solving problems such as supply-demand mismatch and low efficiency in traditional feeding methods, providing an advanced intelligent solution for cattle and sheep farming. Attached Figure Description

[0031] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0032] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0033] Figure 2 This is a system schematic diagram according to an embodiment of the present invention. Detailed Implementation

[0034] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0035] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0036] Example 1

[0037] like Figure 1-2 As shown, this embodiment provides a fully automated feeding control method and system for cattle and sheep in multiple areas, including:

[0038] Basic information and feeding behavior data of cattle and sheep in various regions were obtained. After preprocessing the feeding behavior data, a support vector machine algorithm was used to train a model to predict the feeding speed of cattle and sheep.

[0039] In some specific implementations, the process of training a model for predicting the feeding speed of cattle and sheep using a support vector machine algorithm includes:

[0040] Basic information on cattle and sheep in each region is obtained, and a basic attribute database of cattle and sheep is constructed based on this information. Feeding behavior data of cattle and sheep in each region is collected in real time and preprocessed. Based on the basic attribute database of cattle and sheep, the preprocessed feeding behavior data is divided into several data subsets. Support vector machines are used to train each data subset to obtain the corresponding cattle and sheep feeding speed prediction model, and the trained model parameters are stored in the model library. The preprocessing includes data cleaning and normalization operations.

[0041] Furthermore, based on pre-established information on the breed, quantity, age, and body type of cattle and sheep in each region, a basic attribute database for cattle and sheep is constructed to store and manage their basic information. By attaching sensor devices to the cattle and sheep, real-time data on their feeding behavior in each region is collected, including feeding time, frequency, and duration, and the collected data is transmitted to a data processing center. The collected feeding behavior data undergoes preprocessing, including data cleaning and normalization, to remove noise and convert the data into a format suitable for training machine learning algorithms. Based on the breed, age, and body type information in the basic attribute database, the preprocessed feeding behavior data is divided into multiple data subsets according to different attribute dimensions. For each data subset, a support vector machine algorithm is used to train a model to predict the feeding speed of cattle and sheep under the corresponding region and attribute combination. The trained model parameters are stored in a model library. When it is necessary to predict the feeding rate of cattle and sheep in a certain area, the corresponding feeding rate prediction model is retrieved from the model library based on the breed, quantity, age, and body type attributes of the cattle and sheep in that area. Real-time collected feeding behavior data of cattle and sheep in that area is then input into the model for prediction. The predicted feeding rate information for each area is then visualized, generating intuitive charts and reports to facilitate managers' timely understanding of the feeding situation and dynamic adjustments to the feeding plan based on the prediction results.

[0042] Specifically, the basic attribute database for cattle and sheep is used to store and manage basic information about them, such as breed, quantity, age, and size. You can think of this database as an Excel spreadsheet, with each row representing a cow or sheep and each column representing an attribute. For example, it could record a Simmental cow, number 001, 3 years old, large size, located in region A. Simultaneously, it could record a Hu sheep, number 002, 1 year old, medium size, also located in region A. This database is established to provide foundational data for subsequent analysis and prediction of feeding rates. Sensors worn by the cattle and sheep can collect their feeding behavior data in real time. For example, the sensors can record the timestamp when the cattle or sheep begin feeding, the duration of feeding, and the number of chews per unit time. Suppose that Simmental cow number 001 begins feeding at 8:00 AM and continues for 30 minutes, chewing an average of 20 times per minute. This data is wirelessly transmitted to a data processing center. The collected data may contain noise and missing values, requiring data preprocessing. For example, sensors may occasionally malfunction, leading to data anomalies, requiring the identification and removal of this noisy data. Furthermore, different sensors may collect data within different ranges, necessitating normalization to convert the data to a uniform scale, such as between 0 and 1. Suppose cow number 001 experiences a sensor malfunction at 8:15, recording zero chewing counts; this is clearly abnormal data and needs to be removed. Based on the basic attribute database of cattle and sheep, the preprocessed feeding behavior data can be divided into multiple data subsets according to different attribute dimensions. For example, all feeding data of Simmental cattle in region A can be divided into one subset, and all feeding data of Hu sheep in region A into another subset. The purpose of this is to build more accurate feeding speed prediction models for cattle and sheep of different breeds, ages, and body types. For each data subset, a Support Vector Machine (SVM) algorithm can be used for training to obtain a feeding speed prediction model for the corresponding region and attribute combination. The SVM algorithm can be understood as finding an optimal hyperplane in the data to separate samples of different categories. In this example, "fast eating" and "slow eating" can be treated as two categories, and a hyperplane can be found using the support vector machine algorithm to distinguish them. After training, the model parameters are stored in a model library. When it is necessary to predict the eating speed of cattle and sheep in a certain area, the corresponding eating speed prediction model is first obtained from the model library based on the attributes of the cattle and sheep in that area, such as breed, age, and body size. For example, if we want to predict the eating speed of a 3-year-old Simmental cow in area A, we need to find the eating speed prediction model for Simmental cows in area A from the model library. Then, the real-time collected feeding behavior data of this cow, such as feeding time and chewing frequency, is input into the model for prediction. The prediction results can be visualized, such as generating charts or reports to show the eating speed of cattle and sheep in different areas.For example, a bar chart can be generated to show the average feeding rate of different breeds of cattle and sheep in region A. Managers can use these visualizations to understand the feeding situation of cattle and sheep in a timely manner and dynamically adjust feeding programs based on predictions. For instance, if it is predicted that cattle and sheep in a certain area will feed slowly, the nutritional content of the feed can be increased or the feeding time adjusted. Doing so can improve the growth rate and productivity of cattle and sheep.

[0043] A model relating feed physical parameters to feed feeding rate was constructed to obtain feed physical parameters for each region. Based on the model and the feed physical parameters for each region, the initial feed feeding rate for each region was obtained.

[0044] In some specific implementations, the feed physical parameters of each region are input into the relational model to obtain the initial feeding rate of each region; if the feed quality changes, the feed physical parameters are obtained in real time to update the initial feeding rate of each region; wherein, the feed physical parameters include feed particle size, feed particle density, and flowability parameters.

[0045] Furthermore, feed samples currently used in each region are obtained, and feed particle image analysis technology is used to obtain the average particle size parameter. The mass of feed per unit volume is measured using a weighing method to obtain the feed density parameter. A certain amount of feed is placed in a funnel, and the time it takes for it to completely flow out of the funnel is measured to calculate the feed flowability parameter. Based on a pre-established mathematical model relating feed physical properties to feeding speed, the feed particle size, density, and flowability parameters obtained in the above steps are substituted into the model equations to determine the initial feeding speed control parameters for each region. If the feed quality changes during actual production, the feed physical property parameters are re-obtained, and the feeding speed parameters are dynamically adjusted according to the relationship model to adapt to the new feed quality. Based on the initial feeding speed and combined with actual output feedback data from each region, a proportional-integral-derivative (PID) control algorithm is used to dynamically optimize and adjust the feeding speed of each region to achieve precise output control. The optimized feeding speed parameters for each region are recorded and stored as a basis for subsequent production scheduling and used for iterative updates and optimization of the feeding speed relationship model.

[0046] Specifically, feed pellet image analysis technology can be used to obtain the average particle size parameter of feed pellets. For example, images of feed samples used in different areas can be collected, image processing software can be used to identify and measure the diameter of all feed pellets in the image, and then the average value can be calculated to obtain the average particle size. This helps to understand whether the feed pellet size is uniform and whether it is suitable for cattle and sheep of different ages and breeds. Pellet size affects the feeding speed and digestion efficiency of cattle and sheep. The weighing method can be used to measure the mass of feed per unit volume to obtain the feed density parameter. For example, a container of known volume can be filled with feed, weighed, and then the total mass of the feed can be divided by the volume of the container to obtain the feed density. Feed density is an important parameter affecting the feeding speed; too high or too low density will affect the accuracy of feeding. For example, feed with too high density is prone to clogging during transportation, while feed with too low density is prone to waste. By measuring the time it takes for a certain amount of feed to completely flow out of a funnel, the feed flowability parameter can be calculated. For example, 1 kg of feed can be poured into a standard funnel, and the time required for the feed to completely flow out of the funnel can be recorded. The shorter the time, the better the feed flowability. Feed flowability is crucial for the stability and efficiency of automated feeding systems. Poorly flowable feed can easily cause uneven feeding or blockages. A mathematical model of the relationship between feed physical properties and feeding speed can be used to determine the initial feeding speed control parameters. This model can calculate a suitable initial feeding speed based on parameters such as feed particle size, density, and flowability. For example, for feed with smaller particles, higher density, and better flowability, the model might suggest a higher initial feeding speed. Establishing such a model can improve feeding efficiency and reduce feed waste. In actual production, if feed quality changes, such as changing to a different batch, it is necessary to re-obtain the feed physical property parameters and dynamically adjust the feeding speed parameters according to the relationship model. For example, if the new batch of feed has larger particles, lower density, and poorer flowability, then the feeding speed needs to be reduced to ensure feeding stability and uniformity. This avoids feeding problems caused by changes in feed quality and ensures normal feed intake for cattle and sheep. Proportional-Integral-Derivative (PID) control algorithms can dynamically optimize and adjust the feeding speed of each region based on actual output feedback data, achieving precise output control. For example, if the grazing rate of cattle and sheep in a certain area is lower than expected, leading to a decrease in output, the PID controller will automatically increase the feeding rate in that area based on the output deviation until the output reaches the target value. The PID control algorithm can automatically adapt to various changes, keeping the feeding rate at its optimal level. The optimized feeding rate parameters for each area are recorded and stored as a basis for subsequent production scheduling and used for iterative updates and optimization of the feeding rate relationship model. For example, feeding efficiency and output data under different parameters can be analyzed to find the optimal combination of feeding rate parameters and use it to guide future production scheduling.Continuous recording and analysis can help improve material supply strategies and increase production efficiency.

[0047] Obtain real-time feeding behavior data and real-time satiety data of cattle and sheep in each region to determine whether the average feeding speed of cattle and sheep in each region meets the requirements;

[0048] In some specific implementations, the process of determining whether the average feeding rate of cattle and sheep in each region meets the requirements includes:

[0049] The system acquires real-time location information of cattle and sheep in each region, and obtains the number of cattle and sheep in each region based on the real-time location information; it acquires and preprocesses images of the feeding behavior of cattle and sheep in each region, extracts key features of the feeding behavior, and inputs them into a trained convolutional neural network model to obtain the satiety level values ​​of cattle and sheep in each region; it obtains the average feeding speed of each region based on the satiety level data, the number of cattle and sheep, and the key features of the feeding behavior; it obtains a preset normal feeding speed threshold based on the breed, age, and physiological state of cattle and sheep; and it judges whether the average feeding speed of each region meets the requirements based on the preset normal feeding speed threshold and the average feeding speed of each region.

[0050] Furthermore, based on the real-time location information of cattle and sheep in each area, the quantity and distribution of cattle and sheep in each area are obtained. For cattle and sheep in each area, images and video data of their feeding behavior are collected using video monitoring equipment. The collected images and video data of feeding behavior are preprocessed to extract key feature parameters, including feeding frequency and feeding duration. The extracted feeding behavior feature parameters are input into a pre-trained convolutional neural network model, and the model inferences to calculate the satiety level of cattle and sheep in each area. Based on the quantity of cattle and sheep in each area, feeding behavior characteristics, and satiety level, the average feeding speed of each area is calculated. The average feeding speed of each area is compared with a preset normal feeding speed threshold to determine whether there has been a significant change in the feeding speed of cattle and sheep in each area. If the average feeding speed of a certain area is significantly lower than the normal threshold, an early warning message is triggered, prompting the feeders to pay attention to the feeding situation of cattle and sheep in that area and adjust the feeding plan as needed.

[0051] Specifically, feed supply is a complex process that requires precise control to ensure the healthy growth and production efficiency of cattle and sheep. First, it's necessary to understand the real-time location information of the cattle and sheep. A positioning system based on Radio Frequency Identification (RFID) technology can be used, attaching electronic tags to the cattle and sheep and deploying readers in the breeding area to obtain the activity range and location information of each animal in real time. This allows for accurate statistics on the number and distribution of cattle and sheep in each area; for example, area A has 30 cattle, area B has 25 sheep, and so on. After knowing the number of cattle and sheep, it's also necessary to understand their feeding behavior. High-definition video monitoring equipment can be installed in each area to capture images and video data of the cattle and sheep's feeding behavior in real time. After preprocessing this data, key characteristic parameters can be extracted. For example, the number of times cattle and sheep lower their heads to feed per minute can be counted, i.e., feeding frequency; the duration of each feeding session can also be counted, i.e., feeding duration. For example, if a cow lowers its head to feed 10 times in one minute, with each feeding session lasting an average of 5 seconds, then these data can be used as characteristic parameters of that cow's feeding behavior. These extracted feeding behavior feature parameters are then input into a pre-trained convolutional neural network (CNN) model. This model has been pre-trained on a large amount of image and video data of cattle and sheep feeding behavior and can infer the satiety level of the cattle and sheep based on the input feature parameters. Satiety level can be represented as a value between 0 and 1, for example, 0.8 indicates that the cattle and sheep are relatively full, and 0.2 indicates that they are still hungry. Assuming the average satiety level of 30 cattle in area A is 0.6, and the average satiety level of 25 sheep in area B is 0.7, the average feeding rate for each area can be calculated by combining the number of cattle and sheep, feeding behavior features, and satiety level. For example, if the total feed intake of the 30 cattle in area A is X kg / hour, then the average feeding rate is X / 30 kg / hour / head. Similarly, the average feeding rate of the sheep in area B can be calculated. To determine whether the feeding situation of the cattle and sheep is normal, the calculated average feeding rate needs to be compared with a preset normal feeding rate threshold. This threshold is determined based on factors such as the breed, age, and physiological state of the cattle and sheep. Assume the normal feeding rate threshold for beef cattle is Y kg / hour / head. If the average feeding rate of cattle in area A is significantly lower than Y, the system will trigger an early warning, notifying the farmers to monitor the feeding situation of the cattle in area A. Farmers can adjust the feeding plan according to the actual situation, such as increasing feed supply, changing feed types, and checking the health status of the cattle and sheep. By monitoring and analyzing the real-time location, feeding behavior, satiety level, and feeding rate of cattle and sheep, abnormalities can be detected in a timely manner, and corresponding measures can be taken to ensure the healthy growth and production efficiency of the cattle and sheep. This data-driven intelligent feeding management method can effectively improve breeding efficiency and reduce production costs. Recording and storing optimized feed rate parameters provides data support for subsequent production scheduling and is used for iterative updates and optimization of the feed rate relationship model, forming a data-driven virtuous cycle that continuously improves the accuracy and efficiency of feeding management.

[0052] If the average feeding speed does not meet the requirements, a fuzzy control algorithm is used to generate a feeding speed control command based on the changes in the feeding speed of cattle and sheep in each area.

[0053] In some specific implementations, the process of generating the feeding speed control command includes:

[0054] The system acquires real-time feeding speed data of cattle and sheep within the region, inputs this data into a fuzzy control algorithm, and obtains a feed rate adjustment value. The feed rate adjustment value is then matched with the real-time feeding speed data. If they do not match, the fuzzy control algorithm regenerates the feed rate adjustment value. The fuzzy control algorithm continuously optimizes the feed rate control commands for each region based on the real-time average feeding speed.

[0055] Furthermore, based on the location of the cattle and sheep, real-time feeding speed data of those animals within that area is obtained. This data is used as input to a fuzzy control algorithm, which determines the adjustment value for the feeding speed in that area through fuzzy inference. Based on the adjustment value output by the fuzzy control algorithm, the actual feeding speed in that area is dynamically adjusted. It is then determined whether the adjusted feeding speed matches the cattle and sheep feeding speed; if not, the process returns to step two to continue adjusting until a match is achieved. Changes in the cattle and sheep feeding speed in each area are continuously monitored. When a change occurs in the speed of a certain area, the dynamic adjustment process for the feeding speed in that area is triggered. The adjustment status of the feeding speed in each area is fed back to the data analysis module in real time, and the rule base of the fuzzy control algorithm is optimized through big data analysis. Based on the optimized fuzzy control algorithm, the feeding speed in each area is continuously and dynamically adjusted to achieve real-time matching between the feeding speed and the cattle and sheep feeding speed.

[0056] Specifically, the feeding speed of cattle and sheep varies with time, individual differences, and feed type. To ensure the healthy growth of cattle and sheep and the effective utilization of feed, the feeding speed needs to be dynamically adjusted. This is similar to a buffet, where food is replenished in real time according to the pace of consumption to avoid food waste or insufficient supply. First, it is necessary to acquire real-time data on the feeding speed of cattle and sheep. Assuming that cameras and image recognition technology can monitor the feeding frequency and chewing frequency of each animal, and calculate the amount of feed consumed per minute, for example, the average feeding speed of the 10 cattle and sheep in area 1 is 200 grams per minute. Next, the acquired feeding speed data is input into a fuzzy control algorithm. The core of the fuzzy control algorithm is to simulate human experience for control. Unlike traditional control algorithms that require precise mathematical models, it uses fuzzy rules for reasoning. For example, the following rules can be set: if the feeding speed is fast, reduce the feeding speed; if the feeding speed is slow, increase the feeding speed; if the feeding speed is moderate, maintain the current feeding speed. The fuzzy reasoning process can be understood as a weighted averaging process. Suppose the current feeding rate in area 1 is 300 grams per minute, while the feeding rate of cattle and sheep is 200 grams per minute. According to fuzzy rules, the feeding rate needs to be reduced. Assume the membership degrees of the three fuzzy sets "very fast," "moderate," and "very slow" are 0.2, 0.5, and 0.3 respectively, corresponding to feeding rate adjustment values ​​of -50, 0, and +50. Therefore, the final feeding rate adjustment value is 0.2*(-50) + 0.5*0 + 0.3*50 = 5 grams per minute. Then, based on the adjustment value output by the fuzzy control algorithm, the actual feeding rate in this area is dynamically adjusted. In the example above, the feeding rate in area 1 will be adjusted to 300 + 5 = 305 grams per minute. To ensure the feeding rate matches the feeding rate of cattle and sheep, continuous monitoring and adjustment are necessary. For example, after a period of time, if the feeding rate of cattle and sheep in area 1 becomes 350 grams per minute, while the feeding rate remains at 305 grams per minute, fuzzy reasoning and adjustment are performed again until the two match. This is like a continuous fine-tuning process, ensuring the feed rate always follows the feeding speed of the cattle and sheep. Furthermore, the feed rate adjustments in each area are fed back to the data analysis module in real time. Through big data analysis, shortcomings of the fuzzy rules can be identified and optimized. For example, if it's found that the feeding speed of cattle and sheep in a certain area is consistently low during a specific time period, the fuzzy rules can be adjusted for that period to increase the feed rate, thus better meeting the needs of the cattle and sheep. This is similar to optimizing a restaurant's menu based on historical data, predicting customer demand, and preparing in advance. Through these steps, real-time matching of feed rate and cattle / sheep feeding speed can be achieved, improving feed utilization and promoting healthy growth. It's like an intelligent feeding system that automatically adjusts the feed rate according to the needs of the cattle and sheep, achieving precise feeding.

[0057] For example, the process of continuously and dynamically adjusting the feeding speed of each area according to the optimized fuzzy control algorithm to achieve real-time matching between the feeding speed and the grazing speed of cattle and sheep includes: acquiring grazing speed data of each area based on the real-time grazing speed of cattle and sheep in different areas; inputting the acquired grazing speed data of each area into a preset fuzzy control algorithm model for processing; the fuzzy control algorithm model dynamically calculating the optimized feeding speed parameters of each area according to the input grazing speed data and through fuzzy inference rules; determining whether the difference between the calculated feeding speed parameters of each area and the current actual feeding speed parameters exceeds a preset threshold; if it exceeds the preset threshold, the optimized feeding speed parameters are used as the new feeding speed setting value, and the feeding speed of each area is adjusted; if it does not exceed the preset threshold, the current actual feeding speed of each area remains unchanged; continuously repeating the above steps, dynamically optimizing and matching the feeding speed of each area in real time according to the changes in the grazing speed of cattle and sheep, and achieving a dynamic balance between the feeding speed and the grazing speed.

[0058] Obtain feeding demand data and current feed supply data for cattle and sheep in each region, and calculate the optimal feed supply rate for each region.

[0059] In some specific implementations, the process of calculating the optimal feeding rate for each region includes:

[0060] Feed demand forecasts are obtained based on historical feeding data and the current number of cattle and sheep. Based on feed supply data and feed demand forecasts, an optimization algorithm is used to calculate the optimal feeding rate for each region. If the difference between the actual feeding amount and the predicted feeding demand in a region exceeds a preset value, the optimal feeding rate is updated.

[0061] Furthermore, data on the feeding needs and current feed supply of cattle and sheep in each region are acquired and used as input to the multi-agent collaborative control algorithm. Using a feeding demand prediction model, combined with historical feeding data and the current number of cattle and sheep, the trend of feeding demand changes in each region over a future period is predicted. Based on the feed supply data and feeding demand prediction results, an optimization algorithm is used to calculate the optimal feeding rate for each region, ensuring that the total feed supply meets the total feeding demand. While optimizing the feeding rate, a negotiation mechanism among agents coordinates the feeding rates between regions to avoid uneven feeding. Real-time feeding data for each region is monitored; if a significant difference is found between the actual feeding amount and the predicted demand in a region, dynamic adjustments to the feeding rate are triggered. The adjusted feeding rate data is transmitted to the feed supply system to control the actual feeding rate in each region, ensuring dynamic matching between feed supply and feeding demand. The feeding status and changes in the number of cattle and sheep in each region are continuously tracked, and the parameters of the feeding demand prediction model and collaborative control algorithm are updated regularly to maintain the system's optimized performance.

[0062] Specifically, in modern animal husbandry, ensuring a dynamic match between the feeding needs of cattle and sheep and the feed supply is crucial. First, by collecting real-time feeding demand data and current feed supply data for cattle and sheep in various regions, this information can be effectively input into a multi-agent collaborative control algorithm. For example, assuming a region currently has 500 cattle, and each cattle requires an average of 2 kg of feed per day, then the daily feed demand for that region is 1000 kg. Next, using a feeding demand prediction model, combined with historical feeding data and the current number of cattle and sheep, the trend of feed demand changes over a future period can be predicted. For example, if it is predicted that the number of cattle and sheep in the region will increase to 600 within the next week, then the predicted feed demand will increase to 1200 kg / day. This prediction helps decision-makers adjust their feeding strategies in advance to avoid feed shortages or surpluses. Based on the feed supply data and feeding demand prediction results, an optimization algorithm is used to calculate the optimal feeding rate for each region. For example, if the current feed supply is 1100 kg / day and the demand is 1200 kg / day, the optimization algorithm might suggest increasing the feed supply rate to make up for the 100 kg shortfall. While optimizing the feed supply rate, a negotiation mechanism between agents can coordinate the feed supply rates between different regions to avoid supply imbalances. For instance, if neighboring region A has a surplus of feed while region B has a shortage, the agents can coordinate the transfer of some feed from region A to region B. Real-time feeding data for each region is monitored. If a significant discrepancy is found between the actual feed intake and the predicted demand in a region—for example, if the actual consumption in region C suddenly increases to 1300 kg / day—this could be due to the addition of new cattle or sheep or other factors, triggering a dynamic adjustment of the feed supply rate. The adjusted feed supply rate data is then transmitted to the feed supply system to control the actual feed supply rate in each region, ensuring a dynamic match between feed supply and demand. Continuous tracking of the feeding status and changes in the number of cattle and sheep in each region, and regular updates to the parameters of the feed demand prediction model and collaborative control algorithm, maintains the system's optimized performance. This continuous monitoring and updating ensures that the entire system can flexibly respond to changes in the number of cattle and sheep and fluctuations in feeding needs, thereby achieving efficient and sustainable feed management.

[0063] In some specific implementations, the process of calculating the optimal feeding rate for each region using an optimization algorithm based on feed supply data and feeding demand forecasts includes:

[0064] Historical and real-time data on feed supply and demand for each region are obtained. Based on the historical and real-time data on feed supply, the feed supply for each region is obtained, and based on the historical and real-time data on feed demand, the feed demand for each region is obtained. Using the predicted feed supply and demand for each region as constraints, a feed rate optimization model is established. The objective function of the feed rate optimization model is to minimize the absolute value of the difference between the total feed supply and the total demand. A heuristic algorithm is used to solve the feed rate optimization model to obtain the optimal feed rate for each region.

[0065] Furthermore, historical and real-time data on feed supply for each region are acquired. A supply prediction model is established based on the historical data, and the model is calibrated using real-time data to predict the feed supply for each region in the future. Historical and real-time data on feeding demand for each region are also acquired. A demand prediction model is established based on the historical data, and the model is calibrated using real-time data to predict the feeding demand for each region in the future. Using the predicted feed supply and demand for each region as constraints, a feed supply speed optimization model is established. The objective function is to minimize the absolute value of the difference between the total feed supply and the total demand. A heuristic optimization algorithm is used to solve the above optimization model to obtain the optimal feed supply speed for each region, ensuring that the total feed supply is as close as possible to the total demand while meeting the feeding needs of each region. Based on the optimization results, the parameters of the feed supply equipment for each region are set to control the actual feed supply speed in each region, and the feed supply and feeding amount in each region are monitored in real time. If a significant difference is detected between the actual feed supply and the optimization result in a certain region, an early warning is triggered. The cause is analyzed, and measures are taken to adjust the situation, such as adjusting the feed supply speed in that region or coordinating the feed supply in other regions. Regularly evaluate the effectiveness of the optimization model, revise and improve the model based on actual operating data, continuously improve the accuracy and stability of feed supply optimization, and ensure that feeding needs are met in the long term.

[0066] In some specific implementations, the method further includes acquiring data on feeding speed and cattle and sheep feeding behavior in each area in real time, performing correlation analysis on the feeding speed and cattle and sheep feeding behavior data to obtain key factors for efficiency evaluation, evaluating the feeding efficiency of each area based on the key factors for efficiency evaluation, and optimizing the feeding management strategy based on the cattle and sheep feeding behavior and feed supply data in the area if the feeding efficiency does not meet the preset requirements.

[0067] Furthermore, real-time feeding speed data and cattle / sheep feeding behavior data for each region are acquired and transmitted to a big data analytics platform for processing. Data cleaning and preprocessing techniques are employed to denoise and normalize the raw data, improving data quality. Correlation analysis algorithms are used to analyze the relationship between feeding speed and cattle / sheep feeding behavior, identifying key factors affecting feeding efficiency. Based on these key factors, a feeding efficiency evaluation model is established, and support vector machine or random forest algorithms are used for model training and optimization. Real-time data from each region is input into the trained evaluation model to evaluate and score the feeding efficiency of each region in real time. Based on the evaluation results, regions with low feeding efficiency are identified, and efficiency distribution heatmaps are generated using visualization technology to visually display these inefficient areas. For the identified inefficient regions, targeted feeding optimization strategies are developed based on regional characteristics and key factor analysis results. An automated control system is then used to achieve precise feeding, improving overall feeding efficiency.

[0068] Furthermore, the processing for areas with low feeding efficiency includes: acquiring historical data on cattle and sheep feeding behavior and feed supply in these areas; preprocessing the data, including data cleaning, data integration, and data transformation, to obtain a dataset suitable for data mining analysis. Based on attributes such as cattle and sheep breed and feed quality, K-means clustering is used to cluster the dataset, grouping data with similar feeding behavior and feed supply characteristics into the same category, identifying differences in feeding efficiency across different regions and breeds. For each category, association rule mining algorithms are used to analyze the relationship between cattle and sheep feeding behavior and feed supply, identifying key factors affecting feeding efficiency, such as feeding time, feeding frequency, feed type, and feed amount. Based on the association rule mining results, combined with expert knowledge and feeding experience, specific measures to optimize feeding management strategies are determined, such as adjusting feed ratios, changing feeding time and frequency, and improving the feeding environment. The optimized feeding management strategies are applied to areas with low feeding efficiency, and an information management system is used to monitor cattle and sheep feeding behavior and feed supply in real time, dynamically adjusting strategy parameters based on feedback data. Continuously track optimized feeding efficiency indicators, compare data before and after optimization to evaluate the effectiveness of optimization measures, and form a replicable and scalable feeding management optimization model. Establish a feeding efficiency early warning mechanism; when abnormal fluctuations occur in feeding efficiency indicators in a certain area, issue timely warnings, diagnose the causes, and take targeted optimization and adjustment measures to ensure a stable improvement in overall feeding efficiency.

[0069] Specifically, obtaining historical data from areas with low feeding efficiency is crucial for in-depth analysis of the reasons for this inefficiency. For example, the feeding efficiency in area 1 is below average. It's necessary to collect data on cattle and sheep feeding behavior over a period of time, such as one month, including feed intake, feeding time, and feeding frequency per animal, as well as feed supply data, such as feed type, amount, and time. This data will be used for subsequent data mining analysis. In the data preprocessing stage, data cleaning is performed first. For example, missing or abnormal data for certain days may be due to equipment malfunction and needs to be removed or filled in. Then, data integration is performed, combining data from different sources, such as merging cattle and sheep feeding data and feed supply data. Finally, data transformation is performed, such as converting feeding times into time periods, such as morning, noon, and evening. After these processes, a clean, complete, and consistent dataset is obtained. Using the K-means clustering algorithm, the data can be divided into different categories. For example, based on cattle and sheep breeds and feed quality, the data can be divided into different categories such as beef cattle, dairy cattle, and goats, with cattle and sheep in each category exhibiting similar feeding behaviors and feed supply characteristics. This allows for more precise feeding strategies tailored to different categories. For example, cluster analysis might reveal that beef cattle consume less feed at night, while dairy cows consume more in the morning. Association rule mining can uncover the correlation between cattle / sheep feeding behavior and feed supply. For instance, a strong correlation might be found between "feed type A" and "feeding time in the morning," indicating that cattle / sheep prefer to eat feed A in the morning. Similarly, it might be observed that beef cattle consume less feed and drink more water in hot weather, suggesting that high temperatures affect their feeding behavior. Analyzing these association rules can identify key factors influencing feeding efficiency. Based on the association rule mining results and expert experience, optimization strategies can be developed. For example, to address the issue of low feed intake in beef cattle at night, feed formulations could be adjusted, increasing palatable feeds or shifting some feed feeding times to the evening. To address the decrease in feed intake in hot weather, cooling and heat-relieving ingredients could be added to the feed, along with providing ample drinking water. These optimization strategies can then be applied to actual production and continuously monitored and adjusted. For example, the adjusted feed formulation was applied to beef cattle in Area 1, and the feed intake of cattle and sheep was monitored in real time through an information management system. If feed intake still did not improve, further adjustments to the strategy were needed, such as changing the feeding frequency or improving the feeding environment. The optimized feeding efficiency indicators were continuously tracked. For example, after one month, the average feed intake of beef cattle in Area 1 increased by 10%, and feeding efficiency also improved significantly. This indicates that the optimization strategy is effective and can be promoted to other similar areas. An early warning mechanism was established to promptly identify and resolve problems. For example, if the system detected a sudden drop in feed intake in a certain area, it would issue an alert, reminding managers to investigate the cause, which could be a problem with the feed or the cattle and sheep being sick.Timely intervention can prevent greater losses.

[0070] like Figure 2 As shown, this embodiment also provides a fully automatic feeding control system for cattle and sheep in multiple areas based on the above method, including:

[0071] The feeding speed prediction module is used to predict the feeding speed of cattle and sheep in different areas.

[0072] The initial feeding speed determination module is used to determine the initial feeding speed of each area;

[0073] The feeding speed change detection module is used to determine whether the feeding speed of cattle and sheep in different areas has changed.

[0074] The dynamic feed speed adjustment module is used to adjust the feed speed according to changes in the feeding speed of cattle and sheep;

[0075] The multi-region material supply coordination module is used to coordinate the material supply speed between different regions;

[0076] The feeding efficiency assessment and optimization module is used to assess feeding efficiency and optimize feeding management strategies.

[0077] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A full-automatic feeding control method for cattle and sheep multi-regions, characterized in that, The method comprises the following steps: acquiring basic information and feeding behavior data of the cattle and sheep in each region, pre-processing the feeding behavior data, and training a cattle and sheep feeding speed prediction model by using a support vector machine algorithm; constructing a relationship model between the physical parameters of the feed and the feeding speed, acquiring the physical parameters of the feed in each region, and obtaining the initial feeding speed in each region based on the relationship model and the physical parameters of the feed in each region; acquiring real-time feeding behavior data and real-time satiation degree data of the cattle and sheep in each region, and determining whether the average feeding speed of the cattle and sheep in each region meets the requirements; if the average feeding speed does not meet the requirements, generating a feeding speed control instruction by using a fuzzy control algorithm based on the change of the feeding speed of the cattle and sheep in each region; acquiring feeding demand data and current feed supply data of the cattle and sheep in each region, and calculating the optimal feeding speed in each region; the process of training the cattle and sheep feeding speed prediction model by using the support vector machine algorithm comprises the following steps: acquiring the basic information of the cattle and sheep in each region, constructing a cattle and sheep basic attribute database based on the basic information, collecting real-time feeding behavior data of the cattle and sheep in each region and pre-processing the data, dividing the pre-processed feeding behavior data based on the cattle and sheep basic attribute database to obtain a plurality of data subsets, training each data subset by using a support vector machine, obtaining the corresponding cattle and sheep feeding speed prediction model, and storing the model parameters after training in a model library, wherein the pre-processing comprises data cleaning and normalization operation; the process of determining whether the average feeding speed of the cattle and sheep in each region meets the requirements comprises the following steps: acquiring real-time location information of the cattle and sheep in each region, obtaining the number of the cattle and sheep in each region based on the real-time location information, acquiring feeding behavior images of the cattle and sheep in each region and pre-processing the images, extracting feeding behavior key features and inputting the features into a trained convolutional neural network model to obtain satiation degree values of the cattle and sheep in each region, obtaining the average feeding speed of each region based on the satiation degree data, the number of the cattle and sheep, and the feeding behavior key features, obtaining a preset normal feeding speed threshold based on the breed, age, and physiological state of the cattle and sheep, and determining whether the average feeding speed of each region meets the requirements based on the preset normal feeding speed threshold and the average feeding speed of each region; the process of generating the feeding speed control instruction comprises the following steps: acquiring real-time average feeding speed data of the cattle and sheep in each region, inputting the real-time average feeding speed data into a fuzzy control algorithm to obtain a feeding speed adjustment value, matching the feeding speed adjustment value with the real-time average feeding speed data, and if the two do not match, re-generating the feeding speed adjustment value by using the fuzzy control algorithm; wherein the fuzzy control algorithm is continuously optimized based on the feeding speed control instruction of each region and the real-time average feeding speed; the process of calculating the optimal feeding speed in each region comprises the following steps: obtaining a feeding demand prediction result based on historical feeding data and the current number of cattle and sheep, calculating the optimal feeding speed in each region by using an optimization algorithm based on the feed supply data and the feeding demand prediction result, and updating the optimal feeding speed if the difference between the actual feeding amount in each region and the feeding demand prediction result exceeds a preset value. The process of calculating the optimal feed supply speed of each region based on the feed supply data and the feed demand prediction result includes: Obtain the historical data and real-time data of the feed supply and the feeding demand of each region, obtain the feed supply of each region based on the historical data and real-time data of the feed supply of each region, and obtain the feeding demand of each region based on the historical data and real-time data of the feeding demand of each region; take the predicted feed supply and feeding demand of each region as a constraint condition, establish a feed supply speed optimization model, and the objective function of the feed supply speed optimization model is to minimize the absolute value of the difference between the total feed supply and the total demand; solve the feed supply speed optimization model by using a heuristic algorithm to obtain the optimal feed supply speed of each region.

2. The method of claim 1, wherein, The feed physical parameters of each region are input into the relationship model to obtain the initial feed supply speed of each region; if the quality of the feed changes, the feed physical parameters are obtained in real time, and the initial feed supply speed of each region is updated; wherein the feed physical parameters include feed particle size, feed particle density, and flowability parameters.

3. The method of claim 1, further comprising, Further comprising, obtaining the feed supply speed and the feeding behavior data of cattle and sheep in each region in real time, performing correlation analysis on the feed supply speed and the feeding behavior data of cattle and sheep, obtaining efficiency evaluation key factors, evaluating the feeding efficiency of each region based on the efficiency evaluation key factors, and if the feeding efficiency does not meet the preset requirements, optimizing the feeding management strategy according to the feeding behavior and feed supply data of cattle and sheep in the region.

4. A fully automatic feed control system for multi-zone cattle and sheep based on the method of any one of claims 1-3, characterized in that, including: An eating speed prediction module for predicting the eating speed of cattle and sheep in each region; An initial feed supply speed determination module for determining the initial feed supply speed of each region; An eating speed change detection module for determining whether the eating speed of cattle and sheep in each region changes; A feed supply speed dynamic adjustment module for adjusting the feed supply speed according to the change of the eating speed of cattle and sheep; A multi-region feed supply coordination module for coordinating the feed supply speed between regions; A feeding efficiency evaluation and optimization module for evaluating the feeding efficiency and optimizing the feeding management strategy.

Images