Livestock breeding management system based on digital twinning

Abstract

The invention provides a livestock breeding management system based on digital twin, and relates to the technical field of livestock breeding. The livestock breeding management system based on digital twinning comprises a data detection module used for obtaining site distribution information, livestock information, breeding environment information and production management information of a livestock farm; the digital twin module is used for establishing a virtual digital model of the livestock farm according to the site distribution information, the livestock information, the breeding environment information and the production management information; the data analysis module is used for determining livestock state information and management state information according to the livestock information, the breeding environment information and the production management information, and determining target control data of breeding equipment of the livestock farm based on the virtual digital model and the management state information so as to control operation of the breeding equipment; and the model prompting module is used for determining annotation information of the virtual digital model according to the livestock state information and determining target control data according to the annotation information. The management effect is improved.

Description

Technical Field

[0001] The present application relates to the field of animal husbandry technology, for example, to a animal husbandry management system based on digital twins. Background Art

[0002] A digital twin is a digital model created in virtual space using digital technology to create a fully corresponding physical entity or system. This model can simulate, analyze, and predict the state and behavior of the physical entity in real time, forming a closed-loop system of virtual-physical interaction. Livestock management involves the systematic, scientific planning and control of the feeding, reproduction, health, environment, and production processes of livestock (such as cattle, pigs, sheep, and poultry) to improve production efficiency, ensure animal welfare, and reduce operating costs. With global population growth and rising consumption, the livestock industry faces numerous challenges, including the continued rise in demand for meat and dairy products, which necessitates increased production; fluctuating feed prices and rising labor costs, which drive cost reductions and efficiency gains; and threats to production stability from epidemics (such as African swine fever and avian influenza) and climate change (extreme weather).

[0003] Related technologies typically manage livestock farming through a combination of manual observation and automated equipment. However, current livestock farming management systems consist of multiple subsystems, with production, health, and environmental data dispersed across these subsystems. This fragments the data, making unified management difficult and resulting in poor management effectiveness. Summary of the Invention

[0004] This application aims to provide a livestock breeding management system based on digital twins, which can improve the management effect of livestock breeding.

[0005] According to one aspect of the present application, a livestock breeding management system based on digital twin is proposed, characterized by comprising: Data detection module, used to obtain site distribution information, livestock information, breeding environment information and production management information of livestock farms; The digital twin module is used to build a virtual digital model of the livestock farm based on site distribution information, livestock information, breeding environment information, and production management information; A data analysis module is used to determine livestock status information and management status information based on livestock information, breeding environment information, and production management information, and to determine target control data for breeding equipment on the livestock farm based on the virtual digital model and management status information, so as to control the operation of the breeding equipment; The model prompt module is used to determine the labeling information of the virtual digital model according to the livestock status information, and determine the target control data according to the labeling information.

[0006] According to some embodiments, the digital twin module is specifically used to: Build a three-dimensional model of the livestock farm based on site distribution information; Based on livestock information, a dynamic model is established for each livestock, and the dynamic model is bound to the unique livestock identification and real-time data interface corresponding to each livestock; Determine the environmental parameters of the three-dimensional model based on the breeding environment information; The 3D model is annotated by default according to the production management information.

[0007] According to some embodiments, the livestock information includes livestock identification, physiological data, and reproductive information; and the data analysis module, when determining livestock status information and management status information based on the livestock information, breeding environment information, and production management information, is specifically configured to: Determine preliminary management information based on livestock identification and production management information; Determine livestock status information based on physiological data, reproductive information, and breeding environment information; The preliminary management information is adjusted based on the livestock status information to determine the management status information.

[0008] According to some embodiments, the data analysis module, when determining livestock status information based on physiological data, reproductive information, and breeding environment information, is specifically configured to: Determine the health status of livestock based on physiological data; Determine the reproductive status of livestock based on reproductive information; Determine environmental stress status based on health status, reproductive status and breeding environment information; Health status, reproductive status and environmental stress status are determined as livestock status information.

[0009] According to some embodiments, when the data analysis module determines target control data of breeding equipment of a livestock farm based on a virtual digital model and management status information to control the operation of the breeding equipment, it is specifically used to: Determine the corresponding breeding equipment and equipment target control data based on management status information; In the virtual digital model, the breeding equipment is simulated and operated according to the equipment target control data to determine the target control data; Send the target control data to the breeding equipment to control the operation of the breeding equipment.

[0010] According to some embodiments, the data detection module is specifically configured to: Use pre-set detection equipment to scan the buildings and layout of livestock farms to generate site distribution information and breeding environment information; Determine livestock information of livestock using pre-set monitoring equipment and physiological monitoring equipment pre-installed on livestock; Obtain production management information input by breeding personnel.

[0011] According to some embodiments, the data detection module, when determining livestock information of the livestock using a preset monitoring device and a physiological monitoring device pre-installed on the livestock, is specifically configured to: Obtaining monitoring data collected by monitoring equipment and monitoring data collected by physiological monitoring equipment; Analyze monitoring and surveillance data to determine livestock information.

[0012] According to some embodiments, the data detection module is further used to obtain epidemic prevention information of the livestock farm; The data analysis module is also used to analyze the monitoring data based on the epidemic prevention information to determine the epidemic prevention status of livestock.

[0013] According to some embodiments, the livestock information includes herd dynamics information; wherein the data analysis module is configured to: Analyze livestock group activity patterns and abnormal behaviors based on group dynamics information; Update livestock status information based on group activity patterns and abnormal behavior.

[0014] According to some embodiments, the farming environment information includes pen conditions and external environment; wherein the data analysis module is used to: Compare pen conditions to pre-set internal condition thresholds to identify condition differences; Update management status information based on condition differences and external environment.

[0015] It should be understood that the foregoing general description and the following detailed description are merely illustrative and are not restrictive of the present application.

[0016] Beneficial effects: Through the above-mentioned embodiments provided in this application, a virtual digital model of a livestock farm is constructed through the digital twin module, and real-time synchronization of the physical space and the digital space is achieved. The site layout, livestock distribution and equipment status can be accurately restored, and management personnel are supported to remotely monitor the dynamics of the entire farm through a three-dimensional visualization interface, breaking through the "black box" management limitations of traditional farming. The data detection module integrates site distribution, livestock individual information, environmental parameters and production management data, breaks the information island, and provides full-factor data support for the data analysis module. The model prompt module generates dynamic annotations of the virtual digital model through livestock status information for refined management. The virtual digital model records data throughout the entire life cycle, supports backtracking and optimization analysis of the production process, and provides data basis for breeding process improvement and quality control traceability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without exceeding the scope of protection required by this application.

[0018] Figure 1 A schematic diagram of the structure of a livestock breeding management system based on digital twins provided in an embodiment of the present application; Figure 2 A flowchart of the livestock breeding management method based on digital twins provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. Like reference numerals in the drawings represent like or similar parts, and thus repetitive description thereof will be omitted.

[0020] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner.In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps etc. can be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0021] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0022] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0023] It should be understood that although the terms first, second, third, etc. may be used herein to describe various components, these components should not be limited by these terms. These terms are used to distinguish one component from another. Thus, the first component discussed below could be referred to as the second component without departing from the teachings of the present invention. As used herein, the term "and / or" includes any one and all combinations of one or more of the associated listed items.

[0024] For specific implementation methods, please refer to the following embodiments.

[0025] Figure 1 This is a schematic diagram of the structure of the livestock breeding management system based on digital twins provided in the embodiment of this application. Figure 1 As shown, the digital twin-based livestock breeding management system 10 includes: a data detection module 11, a digital twin module 12, a data analysis module 13 and a model prompt module 14.

[0026] The data detection module 11 is used to obtain site distribution information, livestock information, breeding environment information and production management information of the livestock farm.

[0027] In this application, fixed monitoring equipment can be pre-installed. These devices can be installed at various locations on the livestock farm site, such as cameras and weather observation equipment. Wearable physiological monitoring equipment can also be pre-installed on the livestock on the site. Site distribution information can include the livestock farm's land area, shape, breeding area divisions, breeding area, and the layout of equipment / gates / windows. Production management information can include the operating status of the farm's internal equipment, feeding amounts, and the daily consumption and cost of water, electricity, and feed.

[0028] In some implementations, the site distribution information and production management information of the livestock farm can be directly obtained. The breeding environment information can be obtained through preset fixed monitoring equipment, and the livestock information can be obtained through equipment worn by the livestock.

[0029] The digital twin module 12 is used to establish a virtual digital model of the livestock farm based on site distribution information, livestock information, breeding environment information and production management information.

[0030] This application can establish a 3D virtual model of a livestock farm based on BIM and digital twin technology, map the site distribution information and livestock information in the physical space into the virtual model, and then mark the breeding environment information and production management information at the corresponding positions of the virtual model.

[0031] The data analysis module 13 is used to determine livestock status information and management status information based on livestock information, breeding environment information and production management information, and to determine target control data of breeding equipment of the livestock farm based on the virtual digital model and management status information to control the operation of the breeding equipment.

[0032] In this application, based on livestock information, a preset algorithm model (such as a machine learning classification model) assesses individual health status (healthy / subhealthy / disease), growth stage (young / fattening / pregnant), and estrus cycle. For example, if a livestock's activity level drops suddenly and its body temperature rises, it is marked as "suspected illness." Combining breeding environment information with production management information, a rule engine determines whether current management meets standards (e.g., whether the temperature is within the appropriate range, whether feed is being fed excessively). For example, if the ambient temperature exceeds a threshold and the humidity falls below a lower limit, an "environmental stress risk" alert is generated.

[0033] Livestock status and environmental data collected in real time are mapped to the virtual model. For example, the locations of sick animals or areas of environmental abnormalities are marked with different colors in the digital twin interface. Based on the management status information, the corresponding equipment control plan is matched from the preset control strategy library. For example, if the ammonia concentration exceeds the standard, the ventilation equipment start-up instruction is triggered. Combined with the simulation results of the virtual digital model (such as the predicted drop in ammonia concentration after ventilation), equipment parameters (such as fan speed and spray cooling intensity) are optimized. Control instructions are issued to breeding equipment (such as automatic feeders, fans, and wet curtains) through the IoT platform, and the equipment operating status is continuously monitored to form a closed-loop control.

[0034] The model prompt module 14 is used to determine the labeling information of the virtual digital model according to the livestock status information, and determine the target control data according to the labeling information.

[0035] In this application, collected livestock status information can be parsed and categorized by individual or group, including health status (e.g., healthy, sub-healthy, diseased), growth stage (e.g., young animals, fattening period, gestation), and behavioral characteristics (e.g., activity level, feeding frequency). Abnormal conditions (e.g., abnormal body temperature, sudden drop in activity level) are identified using preset thresholds or machine learning models (e.g., the isolation forest algorithm) and corresponding abnormality labels are generated. Livestock's real-time location information is associated with corresponding pens or areas in the virtual digital model, and individual / group status is annotated in the 3D model using highlighting, color coding, or icons. For example, healthy livestock are displayed with a green icon, while sick livestock are displayed with a red icon and a "fever" label. Based on the annotated information (e.g., "Livestock ID_001 - estrus period"), corresponding livestock management strategies are matched from a knowledge base. For example, livestock in estrus require feed formula adjustments (increasing protein content) or sick animals require isolation, triggering veterinary notification. Management strategies are then translated into specific equipment control parameters, such as formula adjustment instructions to automatic feeders and temperature / humidity adjustment instructions to environmental control systems. If multiple livestock states trigger different control requirements (such as cooling and humidification at the same time), the execution order is optimized through the rule engine.

[0036] This application constructs a virtual digital model of a livestock farm through a digital twin module, achieving real-time synchronization between physical and digital spaces, accurately restoring site layout, livestock distribution, and equipment status, and supporting management personnel to remotely monitor the dynamics of the entire farm through a three-dimensional visualization interface, breaking through the "black box" management limitations of traditional farming. The data detection module integrates site distribution, individual livestock information, environmental parameters, and production management data, breaking down information silos and providing full-factor data support for the data analysis module. The model prompt module generates dynamic annotations for virtual digital models based on livestock status information for refined management. The virtual digital model records data throughout the entire life cycle, supports backtracking and optimization analysis of the production process, and provides a data basis for improving farming techniques and quality control traceability.

[0037] According to some embodiments, the digital twin module 11 is specifically used to: establish a three-dimensional model of a livestock farm based on site distribution information; establish a dynamic model for each livestock based on livestock information, and bind the dynamic model to the unique livestock identification and real-time data interface corresponding to each livestock; determine the environmental parameters of the three-dimensional model based on the breeding environment information; and perform default labeling on the three-dimensional model based on production management information.

[0038] In this application, point cloud data generated by lidar scanning or oblique photography data from drone aerial surveys are received, containing geometric information such as buildings, pens, and equipment within the site. Multi-site cloud data are aligned using an iterative nearest point algorithm to eliminate differences in scanning perspectives. Discrete point clouds are converted into continuous three-dimensional mesh models using Poisson surface reconstruction or triangulation. The model is semantically classified based on a deep learning segmentation model, and functional areas (such as feeding areas, farrowing pens, and manure treatment areas) are labeled. The three-dimensional model is imported into BIM software to supplement structural properties (such as pens material and load-bearing parameters).

[0039] Create parametric model templates based on livestock species (cows, pigs, sheep, etc.), supporting skeletal animation (such as walking and eating movements). Generate a unique instance for each livestock, and associate it with its visual identification ID (such as a hash value of the cow's facial features).

[0040] To create data mapping rules: Temperature data → model color mapping (e.g. 38-39°C is displayed green, >40°C is displayed red) Movement volume → Skeletal animation speed (a 10% increase in step frequency corresponds to a 10% increase in movement speed) Using visual positioning data, the model is dynamically placed at the corresponding coordinate position in the 3D scene, creating a dynamic GLTF model (GL Transmission Format (GLTF) with Dynamic Attributes), which includes a real-time data interface.

[0041] Kriging interpolation or Gaussian process regression is used on discrete environmental sensor data (such as temperature, humidity, and ammonia concentration) to generate a full-field continuous distribution field.

[0042] Map values ​​to particle effects or color gradients: Ammonia concentration: 0-10ppm (transparent), 10-20ppm (light yellow), >20ppm (deep red) Temperature: Isothermal surface rendering (one color level per 0.5℃) Set the data refresh cycle (e.g. every 30 seconds) to reduce rendering load through differential updates.

[0043] Based on the event logs from the production management system (e.g., breeding and vaccination), markers are generated within the 3D model: icons, such as a syringe icon indicating the most recent vaccination location, and a timeline to review historical operations. In some implementations, if the animal is pregnant, a translucent fetal outline appears on the model's abdomen. If the barn has not been disinfected within the past three days, a warning exclamation mark is displayed. Farmers can manually add temporary annotations (e.g., marking maintenance areas) using AR glasses.

[0044] This application creates a three-dimensional model of a livestock farm based on site distribution information, presenting the overall layout and spatial structure of the farm in an intuitive and three-dimensional manner. This allows managers to gain a comprehensive and macroscopic understanding of the farm's actual conditions, facilitating more efficient site planning, facility layout adjustments, and resource allocation. A dynamic model is created for each livestock, linked to a unique livestock identifier and real-time data interface, enabling precise tracking and management of each individual animal. This dynamic model allows managers to obtain real-time information about each animal, such as growth status, health status, and movement patterns, enabling timely detection of anomalies and targeted measures to improve farming efficiency and livestock welfare. Environmental parameters of the three-dimensional model are determined based on the farming environment information, enabling the digital twin module to simulate the actual environmental conditions of the farm. This helps managers understand changes in the farming environment in real time, predict potential environmental issues such as temperature, humidity, and air quality, and promptly adjust the farming environment to create suitable conditions for livestock and ensure their healthy growth. The 3D model is annotated by default according to the production management information, and various key information in the production management process (such as breeding plans, vaccination records, feed delivery information, etc.) are intuitively annotated on the 3D model, allowing managers to quickly and accurately obtain the required production management information, improve the convenience and efficiency of production management, and facilitate comprehensive monitoring and optimization of the breeding production process.

[0045] According to some embodiments, livestock information includes livestock identification, physiological data and reproductive information; when the data analysis module 13 determines livestock status information and management status information based on the livestock information, breeding environment information and production management information, it is specifically used to: determine preliminary management information based on the livestock identification and production management information; determine livestock status information based on physiological data, reproductive information and breeding environment information; adjust the preliminary management information based on the livestock status information to determine the management status information.

[0046] In some implementations, management plans, historical records, and other data related to livestock can be extracted from production management information based on livestock identification. Preliminary management information is generated by combining preset rules or standards in the production management information (such as breeding plans, vaccination cycles, and feed delivery strategies). Pre-set management logic can be relied upon to ensure that management information complies with conventional breeding practices. Physiological data (such as body temperature, weight, feed intake, and so on) and reproductive information (such as estrus cycle and pregnancy status) of livestock are extracted. Combined with breeding environment information (such as temperature, humidity, and air quality), the impact of this data on livestock health and production performance is analyzed. The current status of the livestock, such as health status and reproductive potential, is determined through an established status assessment model (which may be based on statistical methods, machine learning algorithms, or domain knowledge).

[0047] Compare livestock status information with preliminary management information to identify potential mismatches or areas requiring adjustment. Dynamically adjust preliminary management information based on anomalies or special needs in livestock status information. For example, if a livestock is pregnant, feed formulation adjustments or increased inspection frequency may be necessary. This adjusted management information forms the final management status information, ensuring that management measures are more precise and adapted to the actual needs of the livestock.

[0048] This application combines livestock identification with production management information to tailor preliminary management information for each livestock. This personalized management plan fully considers the uniqueness of each livestock and the overall production plan of the farm, helping to improve the targetedness and effectiveness of management, ensuring that each livestock receives the management measures that best suit its growth stage and physiological needs. Using physiological data, reproductive information, and breeding environment information, the data analysis module can comprehensively and accurately assess the status information of livestock. This assessment not only covers the physical health of the livestock, but also includes its reproductive capacity and adaptability to the breeding environment, providing breeders with more detailed and in-depth livestock status information, helping to promptly identify potential problems and take appropriate measures. The data analysis module demonstrates a powerful dynamic management adjustment capability by dynamically adjusting preliminary management information based on livestock status information. This capability enables management measures to be adjusted in a timely manner as the status of livestock changes, ensuring that management measures are always consistent with the actual needs of livestock, improving management flexibility and adaptability, and helping to improve breeding efficiency and livestock welfare.

[0049] According to some embodiments, when the data analysis module determines the livestock status information based on physiological data, reproductive information and breeding environment information, it is specifically used to: determine the health status of the livestock based on the physiological data; determine the reproductive status of the livestock based on the reproductive information; determine the environmental stress status based on the health status, reproductive status and breeding environment information; and determine the health status, reproductive status and environmental stress status as livestock status information.

[0050] In some implementations, physiological data is analyzed using a pre-defined health assessment model or algorithm (possibly based on statistical methods, machine learning, or domain knowledge). Physiological data is compared with normal physiological ranges to identify abnormal indicators. Combined with historical data trends, a comprehensive assessment of the livestock's health status (e.g., healthy, sub-healthy, or sick) is made. Reproductive information is analyzed to identify the livestock's reproductive cycle stage (e.g., estrus, pregnancy, and lactation). Reproductive status (e.g., suitability for breeding, normal pregnancy progression, etc.) is assessed by combining historical reproductive data with current physiological indicators (e.g., hormone levels and behavioral changes).

[0051] Based on health and reproductive status, a comprehensive analysis is conducted in conjunction with information about the livestock environment. The potential impacts of environmental factors (such as extreme temperatures, high humidity, and air pollution) on livestock health and reproduction are assessed. Using established environmental stress models (which may be based on environmental thresholds and livestock tolerance), environmental stress states (such as no stress, mild stress, and severe stress) are determined. Health, reproductive, and environmental stress states are integrated to form a comprehensive livestock status profile. Relevance and consistency between these status information are ensured to provide a comprehensive basis for subsequent management decisions.

[0052] This application achieves a comprehensive assessment of livestock status by integrating physiological, reproductive and environmental data. This assessment not only focuses on the physical health of livestock, but also covers reproductive capacity and environmental adaptability, providing breeders with more complete and in-depth livestock status information. Accurate assessment of health status and reproductive status enables breeders to promptly identify livestock health problems and reproductive opportunities, so as to take targeted management measures. This helps to improve the reproductive success rate and survival rate of livestock and reduce the risk of disease. The determination of environmental stress status enables breeders to identify the potential impact of environmental factors on livestock in advance and take corresponding regulatory measures. This helps to reduce the adverse effects of environmental stress on livestock health and reproduction and improve the comfort of the breeding environment.

[0053] According to some embodiments, the data analysis module 13 determines the target control data of the breeding equipment of the livestock farm based on the virtual digital model and management status information to control the operation of the breeding equipment. Specifically, it is used to: determine the corresponding breeding equipment and equipment target control data according to the management status information; simulate the operation of the breeding equipment in the virtual digital model according to the equipment target control data to determine the target control data; and send the target control data to the breeding equipment to control the operation of the breeding equipment.

[0054] In some implementations, management status information can be parsed first to determine the current farm management needs and objectives. Specifically, this can identify the farming parameters that require adjustment (such as temperature, humidity, and feed amount) and the corresponding farming equipment (such as ventilation equipment, humidifiers, and feeders). Based on this management status information, a mapping relationship is established between the farming equipment and the equipment's target control data. This includes determining the required operating parameters for each piece of equipment (such as fan speed and humidifier on-time) and how these parameters affect the farming environment.

[0055] In a virtual digital model, aquaculture equipment is simulated according to its target control data. This includes simulating the equipment's startup, operation, and its impact on the aquaculture environment. Through simulation, the validity and feasibility of the target control data are verified, ensuring that the control data achieves the intended management effect while avoiding potential operational conflicts or environmental anomalies.

[0056] Based on the simulation results, the equipment's target control data is optimized and adjusted. This may include adjusting parameter values, revising control logic, or optimizing the equipment's operating sequence. Ultimately, the target control data that meets management requirements and has been verified is determined. The determined target control data is then distributed to the actual farming equipment. Upon receiving the control data, the farming equipment operates according to the instructions, automatically adjusting the farming environment and achieving management goals.

[0057] This application achieves precise control of aquaculture equipment through the precise mapping of management status information and equipment target control data, helping to ensure that equipment operating parameters are highly consistent with management requirements and improve control accuracy and effectiveness. Simulation operations in a virtual digital model can verify the validity and feasibility of control data before actual execution. This helps reduce equipment failures or environmental anomalies caused by improper control and reduces operational risks. The automatic issuance and execution of target control data enables automated management of aquaculture equipment. This reduces the need for manual intervention, improves management efficiency, and ensures the timeliness and accuracy of control instructions.

[0058] According to some embodiments, the data detection module 11 is specifically used to: use preset detection equipment to scan the buildings and layout of the livestock farm to generate site distribution information and breeding environment information; use preset monitoring equipment and physiological monitoring equipment pre-installed on the livestock to determine the livestock information; obtain production management information input by the breeding personnel.

[0059] For this application, you can choose 3D laser scanners and drone aerial survey equipment. Ensure these devices are in good working condition and configure the appropriate scanning and survey parameters. Use a 3D laser scanner to comprehensively scan the livestock farm's buildings and layout to obtain high-precision 3D point cloud data. Simultaneously, use drone aerial survey equipment to take aerial photos of the farm, obtaining bird's-eye views and video footage.

[0060] The scanned 3D point cloud data and aerial images are processed, and a 3D reconstruction algorithm is used to generate a 3D model of the farm's site layout. Image recognition technology is also used to extract information about the farming environment from the aerial footage, such as vegetation coverage, water source locations, and topography. Monitoring equipment, such as cameras and sensors, is strategically deployed throughout the farm to ensure comprehensive coverage of livestock activity areas. Physiological monitoring devices, such as ear tags and collars, are pre-installed on each animal to monitor physiological data (such as body temperature, heart rate, and activity level) in real time.

[0061] Monitoring equipment collects real-time information on livestock behavior and location, while physiological monitoring equipment collects real-time physiological data. This data is transmitted to a data processing center for integration and analysis, generating complete information on each animal, including identification and physiological data. Reproductive information can be supplemented through regular inspections and recording of events such as mating and pregnancy.

[0062] A user-friendly input interface can be pre-configured for farmers to enter production management information. This interface can include various input fields, such as breeding plans, feed placement records, and vaccination schedules. Farmers use this interface to enter production management information into the system. The system performs a preliminary review of the entered information to ensure its completeness and accuracy. Approved production management information is stored in the database and regularly updated as needed to maintain currency.

[0063] This application uses three-dimensional laser scanning and drone aerial survey technology to comprehensively and accurately obtain site distribution information and breeding environment information of livestock farms, which helps breeders better understand the actual situation of the farms and provides strong support for the planning, construction and management of the farms. Using monitoring equipment and physiological monitoring equipment, the behavior, position and physiological state of livestock can be monitored in real time and dynamically. This helps to promptly detect abnormal conditions of livestock, such as disease, estrus, etc., so that corresponding management measures can be taken to improve breeding efficiency and livestock welfare. By integrating site distribution information, breeding environment information, livestock information and production management information, a comprehensive breeding data set can be formed. This helps to conduct collaborative analysis, explore potential correlations and patterns between data, and provide strong support for the intelligent management of farms.

[0064] According to some embodiments, the data detection module 11, when determining livestock information of livestock using preset monitoring equipment and physiological monitoring equipment pre-installed on livestock, is specifically used to: obtain monitoring data collected by the monitoring equipment and monitoring data collected by the physiological monitoring equipment; analyze the monitoring data and the monitoring data to determine livestock information.

[0065] In this application, monitoring equipment, such as cameras, infrared sensors, and voice recognition devices, are installed at appropriate locations on the livestock farm to ensure coverage of the livestock's primary activity areas, and equipment parameters are adjusted to optimize data collection. Physiological monitoring equipment, such as wearable sensors, ear tags, and collars, is pre-installed on each livestock. These devices should be able to continuously and stably monitor the livestock's physiological indicators, such as body temperature, heart rate, and activity level.

[0066] Monitoring equipment collects livestock behavioral data (such as movement patterns, feeding behavior, and social interactions) and environmental data (such as temperature and humidity, if the monitoring equipment has this capability) in real time or periodically, according to a preset collection frequency and pattern. Physiological monitoring equipment collects livestock physiological data in real time and transmits the data to a data processing center via wireless transmission technologies (such as Bluetooth, Wi-Fi, and LoRa).

[0067] Ensure that monitoring and surveillance data are accurately and promptly transmitted to the data processing center and synchronized for subsequent analysis. Clean and preprocess the collected monitoring and surveillance data, including removing noise, addressing missing values, and standardizing data formats to improve data quality. Extract livestock behavioral characteristics (such as activity frequency, feeding duration, and rest patterns) from the monitoring data, and extract physiological characteristics (such as body temperature changes and abnormal heart rates) from the surveillance data. Fusion and analysis of behavioral and physiological characteristics are performed, using pre-defined algorithms or models (such as machine learning and behavioral pattern recognition) to identify livestock individual identification, health status, and reproductive status.

[0068] The individual identity of each livestock is determined through behavioral characteristics in monitoring data or unique identifiers from physiological monitoring devices. Based on the results of the fusion analysis, the livestock's health status (e.g., healthy, sub-healthy, sick) and reproductive status (e.g., estrus, pregnancy) are assessed. Individual identification and status assessment results are integrated into complete livestock information, including livestock identification, physiological data, behavioral data, and more.

[0069] This application can achieve real-time dynamic monitoring of livestock through continuous data collection from monitoring equipment and physiological monitoring equipment. This helps to promptly detect abnormal behavior or physiological changes in livestock, providing the possibility of timely intervention. By using the behavioral characteristics in the monitoring data or the unique identifier of the physiological monitoring equipment, the individual identity of each livestock can be accurately identified. This helps to achieve individualized management and monitoring and improve the accuracy of breeding management. By integrating and analyzing monitoring data and monitoring data, the health status and reproductive status of livestock can be comprehensively assessed. This helps breeders understand the physiological and behavioral conditions of livestock more accurately and provides a basis for formulating scientific management strategies.

[0070] According to some embodiments, the data detection module 11 is further used to obtain epidemic prevention information of the livestock farm; the data analysis module 13 is further used to analyze the monitoring data based on the epidemic prevention information to determine the epidemic prevention status of the livestock.

[0071] In this application, the data detection module 11 obtains epidemic prevention information from livestock farms through a pre-set interface or manual input. This information may include vaccination records, disease outbreak history, implementation of epidemic prevention measures, quarantine results, etc. The acquired epidemic prevention information is organized to ensure its completeness and accuracy, and stored in a database for subsequent analysis and use.

[0072] The data detection module 11 continues to acquire monitoring data from the physiological monitoring equipment pre-installed on the livestock. This data includes physiological indicators such as the livestock's body temperature, heart rate, and activity level. The monitoring data is cleaned and pre-processed to remove noise and outliers to ensure data quality.

[0073] Integrate epidemic prevention information with monitoring data to create epidemic prevention profiles for individual or group livestock. Utilize pre-set algorithms or models (such as machine learning and statistical analysis) to analyze abnormal patterns in monitoring data that may be associated with disease symptoms in epidemic prevention information. Based on these analysis results, assess the livestock's epidemic prevention status, including whether it is infected with the disease, whether it is within the vaccine protection period, and whether further quarantine or treatment is required.

[0074] Generate a report based on the results of the epidemic prevention status analysis, including the individual livestock epidemic prevention status, group epidemic prevention level, potential risks, etc. The report is fed back to the livestock breeder or management system so that epidemic prevention measures can be taken in a timely manner or the livestock management strategy can be adjusted.

[0075] By integrating epidemic prevention information and monitoring data, this application can more accurately assess the epidemic prevention status of livestock, help to promptly detect signs of epidemics, prevent the spread of diseases, and improve the pertinence and effectiveness of epidemic prevention management. Real-time monitoring and data analysis enable livestock breeders to quickly identify abnormal physiological indicators of livestock. Combined with epidemic prevention information, they can take timely measures such as isolation and treatment, thereby reducing the risk of disease occurrence and spread. Based on the results of epidemic prevention status analysis, livestock breeders can optimize epidemic prevention strategies, such as adjusting vaccination plans, strengthening disinfection measures in specific areas, etc., to improve the scientific nature and systematic nature of epidemic prevention work.

[0076] According to some embodiments, the livestock information includes group dynamic information; wherein the data analysis module 13 is used to: analyze the group activity patterns and abnormal behaviors of the livestock based on the group dynamic information; and update the livestock status information based on the group activity patterns and abnormal behaviors.

[0077] In some implementations, monitoring equipment (such as cameras and sensors) can be used to continuously collect livestock group dynamics information, including livestock location, movement trajectory, aggregation, feeding behavior, etc. The collected group dynamics information is integrated to form time series data for subsequent analysis.

[0078] Data analysis algorithms (such as cluster analysis and time series analysis) are used to process group dynamics information and identify regular activity patterns of livestock, such as daily foraging times, resting areas, and social interactions. The identified regular activity patterns are stored in a database as a benchmark for subsequent analysis.

[0079] Based on regular activity patterns, thresholds or rules for abnormal behavior are defined, such as sudden gatherings, unusual movement speeds, and high activity outside of feeding times. By comparing current herd dynamics with regular activity patterns in real time, behaviors that meet these abnormal definitions are identified. Based on the analysis of herd activity patterns and abnormal behaviors, the overall status of the livestock herd is assessed, including whether the herd is experiencing stress or potential disease risks. The assessment results are updated in the livestock status information, including herd health status, activity level, and abnormal behavior records.

[0080] When serious abnormal behavior is detected, an alarm mechanism is triggered, notifying livestock personnel for timely intervention. Based on the updated livestock status information, livestock management strategies can be adjusted, such as adjusting feed intake and strengthening monitoring of specific areas.

[0081] By analyzing group activity patterns and abnormal behaviors, this application can provide a more comprehensive understanding of the overall status of livestock groups and improve the scientific and systematic nature of group management. The abnormal behavior detection mechanism can promptly detect abnormal conditions in livestock groups, such as disease transmission and stress reactions, providing the possibility for timely intervention. Based on the analysis of group activity patterns, the allocation of breeding resources can be optimized, such as adjusting feed delivery time, optimizing livestock activity areas, etc., to improve resource utilization efficiency. Timely identification and handling of abnormal behavior can help reduce breeding risks and reduce economic losses caused by abnormal group behavior.

[0082] According to some embodiments, the breeding environment information includes pen conditions and external environment; wherein, the data analysis module 13 is used to: compare the pen conditions with preset internal condition thresholds to determine the condition differences; and update the management status information based on the condition differences and the external environment.

[0083] Sensors deployed within the pens (such as temperature sensors, humidity sensors, and ammonia concentration sensors) can collect real-time environmental data within the pens, including temperature, humidity, and air quality. Weather stations or external sensors can collect external environmental data, such as temperature, humidity, wind speed, and precipitation. The collected pens and external environmental data is cleaned to remove noise and outliers to ensure data quality. These data are then integrated in a time series to form a complete environmental dataset.

[0084] Based on best practices in livestock farming, preset thresholds for internal conditions in the pen are set, such as the appropriate temperature range and humidity limit. Real-time collected pen conditions are compared with the preset internal condition thresholds, the difference in conditions is calculated, and pen conditions that exceed the threshold range are identified.

[0085] The system combines these differences with external environmental data (such as temperature fluctuations and rainfall) to assess the impact of current housing conditions on livestock health and production performance. Based on the assessment results, it updates management status information, including the housing status, environmental parameters requiring adjustment, and recommended improvement measures. When the differences in conditions exceed safe limits, an alarm mechanism is triggered, notifying livestock owners to take timely corrective measures. Based on this updated management status information, livestock management strategies can be adjusted, such as adjusting ventilation equipment or increasing or decreasing feed intake.

[0086] This application can promptly detect and correct unfavorable pen environments by real-time monitoring and comparing pen conditions with preset thresholds, providing livestock with a more comfortable and healthy living environment. An optimized pen environment helps improve livestock production performance, such as increasing feed intake and growth rate, thereby improving overall breeding efficiency. Suitable pen conditions can reduce livestock stress reactions and disease occurrence caused by environmental discomfort, thereby reducing breeding risks. A comprehensive assessment combined with external environmental data can better predict and respond to the impact of external environmental changes on pen conditions, thereby improving the adaptability and resilience of farms.

[0087] The following describes a method embodiment of the present application, which can be used in a system embodiment of the present application.

[0088] Figure 2 Flowchart of the livestock breeding management method based on digital twin provided in the embodiment of this application. Figure 2 As shown, the livestock breeding management method based on digital twins includes: step S20, step S21, step S22 and step S23.

[0089] In step S20, site distribution information, livestock information, breeding environment information and production management information of the livestock farm are obtained; In step S21, a virtual digital model of the livestock farm is established based on site distribution information, livestock information, breeding environment information, and production management information; In step S22, livestock status information and management status information are determined based on the livestock information, breeding environment information, and production management information, and target control data of the breeding equipment of the livestock farm is determined based on the virtual digital model and the management status information to control the operation of the breeding equipment; In step S23, the labeling information of the virtual digital model is determined according to the livestock status information, and the target control data is determined according to the labeling information.

[0090] Optionally, a three-dimensional model of the livestock farm can be established based on the site distribution information; a dynamic model can be established for each livestock based on the livestock information, and the dynamic model can be bound to the unique livestock identification and real-time data interface corresponding to each livestock; the environmental parameters of the three-dimensional model can be determined based on the breeding environment information; and the three-dimensional model can be labeled by default based on the production management information.

[0091] Optionally, livestock information includes livestock identification, physiological data and reproductive information; preliminary management information can be determined based on livestock identification and production management information; livestock status information can be determined based on physiological data, reproductive information and breeding environment information; and preliminary management information can be adjusted based on livestock status information to determine management status information.

[0092] Optionally, the health status of livestock can be determined based on physiological data; the reproductive status of livestock can be determined based on reproductive information; the environmental stress status can be determined based on the health status, reproductive status and breeding environment information; and the health status, reproductive status and environmental stress status can be determined as livestock status information.

[0093] Optionally, the corresponding breeding equipment and equipment target control data can be determined based on the management status information; in the virtual digital model, the breeding equipment is simulated and operated according to the equipment target control data to determine the target control data; the target control data is sent to the breeding equipment to control the operation of the breeding equipment.

[0094] Optionally, preset detection equipment can be used to scan the buildings and layout of the livestock farm to generate site distribution information and breeding environment information; livestock information can be determined using preset monitoring equipment and physiological monitoring equipment pre-installed on livestock; and production management information input by breeding personnel can be obtained.

[0095] Optionally, monitoring data collected by the monitoring device and monitoring data collected by the physiological monitoring device may be obtained; and the monitoring data and the monitoring data may be analyzed to determine livestock information.

[0096] Optionally, epidemic prevention information of the livestock farm can also be obtained; based on the epidemic prevention information, the monitoring data is analyzed to determine the epidemic prevention status of the livestock.

[0097] Optionally, the livestock information includes group dynamic information; the group activity patterns and abnormal behaviors of the livestock can be analyzed based on the group dynamic information; and the livestock status information can be updated based on the group activity patterns and abnormal behaviors.

[0098] Optionally, the breeding environment information includes pen conditions and external environment; the pen conditions can be compared with preset internal condition thresholds to determine condition differences; and the management status information is updated based on the condition differences and the external environment.

[0099] The method performs functions similar to those of the system provided above. For other functions, please refer to the above description and will not be repeated here.

[0100] The embodiments of the present application are described in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only intended to help understand the method and core ideas of the present application. At the same time, changes or modifications made by those skilled in the art based on the ideas of the present application, the specific implementation methods, and the scope of application of the present application, all fall within the scope of protection of the present application. In summary, the contents of this specification should not be construed as limiting the present application.

Claims

1. A livestock breeding management system based on digital twins, characterized by: include: Data detection module, used to obtain site distribution information, livestock information, breeding environment information and production management information of livestock farms; A digital twin module, configured to establish a virtual digital model of the livestock farm based on the site distribution information, the livestock information, the breeding environment information, and the production management information; a data analysis module, configured to determine livestock status information and management status information based on the livestock information, the breeding environment information, and the production management information, and to determine target control data for breeding equipment of the livestock breeding farm based on the virtual digital model and the management status information, so as to control the operation of the breeding equipment; The model prompt module is used to determine the labeling information of the virtual digital model according to the livestock status information, and determine the target control data according to the labeling information.

2. The system according to claim 1, wherein: The digital twin module is specifically used to: Building a three-dimensional model of the livestock farm based on the site distribution information; Building a dynamic model for each livestock based on the livestock information, and binding the dynamic model with a unique livestock identifier and a real-time data interface corresponding to each livestock; Determining environmental parameters of the three-dimensional model according to the breeding environment information; The three-dimensional model is labeled by default according to the production management information.

3. The system according to claim 1, wherein: Livestock information includes livestock identification, physiological data, and reproductive information; when the data analysis module determines livestock status information and management status information based on the livestock information, the breeding environment information, and the production management information, it is specifically used to: Determining preliminary management information based on the livestock identification and production management information; determining the livestock status information based on the physiological data, the reproduction information, and the breeding environment information; The preliminary management information is adjusted according to the livestock status information to determine the management status information.

4. The system according to claim 3, characterized in that When the data analysis module determines the livestock status information based on the physiological data, the reproductive information, and the breeding environment information, the data analysis module is specifically configured to: determining the health status of the livestock based on the physiological data; determining the reproductive status of the livestock based on the reproductive information; determining an environmental stress state according to the health state, the reproductive state, and the breeding environment information; The health status, the reproductive status, and the environmental stress status are determined as the livestock status information.

5. The system according to claim 1, wherein: The data analysis module is specifically used to determine target control data of the breeding equipment of the livestock farm based on the virtual digital model and the management status information to control the operation of the breeding equipment: Determining the corresponding breeding equipment and equipment target control data according to the management status information; In the virtual digital model, the breeding equipment is simulated and operated according to the equipment target control data to determine the target control data; The target control data is sent to the breeding equipment to control the operation of the breeding equipment.

6. The system according to claim 1, wherein: The data detection module is specifically used for: Using preset detection equipment, scanning the buildings and layout of the livestock farm to generate the site distribution information and the breeding environment information; Determining livestock information of the livestock using a preset monitoring device and a physiological monitoring device pre-installed on the livestock; The production management information input by the breeding personnel is obtained.

7. The system according to claim 6, characterized in that The data detection module is specifically used to determine the livestock information of the livestock by using a preset monitoring device and a physiological monitoring device pre-installed on the livestock: Acquiring monitoring data collected by the monitoring device and monitoring data collected by the physiological monitoring device; The monitoring data and the surveillance data are analyzed to determine the livestock information.

8. The system according to claim 7, characterized in that The data detection module is also used to obtain epidemic prevention information of the livestock farm; The data analysis module is further used to analyze the monitoring data based on the epidemic prevention information to determine the epidemic prevention status of the livestock.

9. The system according to claim 1, wherein: The livestock information includes group dynamic information; wherein the data analysis module is used to: analyzing livestock group activity patterns and abnormal behaviors based on the group dynamics information; The livestock status information is updated according to the group activity pattern and the abnormal behavior.

10. The system according to claim 1, wherein: The breeding environment information includes pen conditions and external environment; wherein the data analysis module is used to: Comparing the housing conditions with preset internal condition thresholds to determine condition differences; The management status information is updated according to the condition difference and the external environment.

Images