A remote control system and control method based on digital twin pig farm

Abstract

The present invention relates to the field of intelligent agricultural technology, specifically a remote control system and control method based on a digital twin pig farm, the system includes a behavior monitoring and identification module, a behavior association analysis module, a growth rate analysis module, a digital model construction module, an environmental condition adjustment module, and a pig farm epidemic prevention management module. In the present invention, by real-time monitoring and analyzing the behavior of multiple pigs in a pig farm, combining environmental condition data, analyzing the impact of environmental conditions on pig behavior, achieving targeted environmental adjustment, real-time recording of pig weight and combining environmental data to analyze growth trends, making weight change predictions more accurate, building a digital twin model and simulating the operating status of a pig farm, providing a simulation platform for the adjustment of management measures, enhancing the transparency of pig farm operations and improving the accuracy of decision-making, reducing resource waste and increasing production, optimizing production processes, and achieving cost savings and sustainable development.

Description

Technical Field

[0001] The present invention relates to the field of intelligent agricultural technology, and in particular to a remote control system and a control method based on a digital twin pig farm. Background Art

[0002] The field of smart agriculture technology aims to use information technology, automation technology and data analysis technology to improve the efficiency and sustainability of agricultural production. Through geographic information systems, remote sensing technology, the Internet of Things, artificial intelligence and machine learning, it monitors and manages crop growth and animal husbandry on farms. By using automated data collection and analysis, smart agriculture helps farm managers understand the health of crops and animals in real time, optimize resource allocation such as the use of water and fertilizers, and improve overall productivity, and implement sustainable management strategies, including precision fertilization and irrigation, to reduce impact on the environment and improve the quality and yield of crops.

[0003] Among them, the pig farm remote control system is used to realize remote monitoring and management of the pig farm, reflect and predict the operating status of the pig farm in real time, and through real-time mapping and analysis of the actual situation of the pig farm, including animal health status and feeding environment, managers can adjust the environmental conditions or production activities of the pig farm at multiple times and places, improve the operational efficiency of the pig farm, reduce resource waste, optimize the production process, improve animal welfare, achieve cost savings and increase output, accurately control the feeding environment, and respond to health risks in a timely manner to ensure efficient and sustainable production.

[0004] Traditional pig farm remote control technology is deficient in the real-time and accuracy of animal behavior analysis and environmental condition adjustment. It relies on periodic data collection rather than continuous real-time monitoring, which limits the ability to respond quickly to emergencies and leads to delays in animal health management and environmental regulation, including the spread of disease and increased animal stress due to environmental discomfort, affecting pig growth rate and yield. The lack of real-time data analysis support makes it difficult for farm managers to make accurate resource allocation and environmental adjustments, resulting in excessive use of water and energy, increased costs and environmental burdens. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose a remote control system and control method based on a digital twin pig farm.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solutions: a remote control system based on digital twin pig farm includes: The behavior monitoring and recognition module is based on image acquisition equipment to monitor and analyze the behavior and physiological status of pigs in the pig farm in real time, identify the pigs' action patterns and abnormal behaviors, and generate behavioral health analysis results; Based on the behavioral health analysis results, the behavior association analysis module uses temperature, humidity and gas concentration sensors to collect real-time environmental status information in the pig farm, identify the impact of various environmental conditions on pig behavior, and generate environmental impact assessment information; The growth rate analysis module collects and records the weight data of the pigs in real time based on the environmental impact assessment information, analyzes the growth trend of the pigs in the current environment, predicts the weight change, and generates a growth rate prediction curve; The digital model building module builds a digital twin model of the pig farm based on the growth rate prediction curve, synchronizes the data of the actual pig farm in real time, and generates a model building record; The environmental condition adjustment module builds records based on the model, adjusts the control parameters of the pig house environmental conditions and feed delivery in real time according to growth prediction and environmental data, and generates environmental parameter configuration; The pig farm epidemic prevention management module is based on the environmental parameter configuration, combined with health monitoring and environmental monitoring data, to identify epidemic risks, send early warning information, and generate pig farm management records.

[0007] As a further scheme of the present invention, the behavioral health analysis results include activity frequency statistics, abnormal behavior type information, and a pig physiological health index score list; the environmental impact assessment information includes environmental parameter change trend information, behavioral response sensitivity evaluation data, and the influence of key environmental factors; the growth rate prediction curve includes predicted weight growth trend, expected maturity time, and environmental adaptability growth analysis results; the model construction record includes real-time mapping parameters of pig farm environmental status, synchronous update time points of behavior and growth data, and model verification results; the environmental parameter configuration includes adjusted temperature and humidity ranges, new formula feed delivery cycle, and optimized operation time of the ventilation system; the pig farm management records include the time when epidemic warnings are issued, disinfection measures implementation records, and vaccination and prevention and control measures records.

[0008] As a further solution of the present invention, the behavior monitoring and identification module includes: The image data acquisition submodule is based on the image acquisition device, which collects image data from multiple locations in the pig farm in real time, records various activity scenes of pigs, including moving, eating, and resting, and generates behavioral video records; The pig behavior analysis submodule analyzes images and identifies pig behavior patterns based on the behavior video records, including eating, fighting, and resting, detects abnormal behaviors, including lameness and excessive interaction, and generates behavior pattern data; The pig health assessment submodule evaluates the physiological health status of multiple pigs based on the behavioral pattern data, combined with the pigs' activity frequency and behavioral abnormality indicators, including activity, stress indicators, and physiological health problems, and generates behavioral health analysis results.

[0009] As a further solution of the present invention, the behavior association analysis module includes: Based on the behavioral health analysis results, the environmental real-time monitoring submodule uses a thermometer and ammonia detector to collect real-time environmental status information in the pig farm, including temperature, humidity and harmful gas concentration, and generates real-time environmental status data; The data fusion analysis submodule analyzes the correlation between the pig behavior data and various environmental variables based on the real-time environmental status data, and analyzes the influence of various environmental conditions on the pig behavior to generate behavior correlation data; The environmental impact analysis submodule evaluates the impact of various environmental conditions on pig behavior based on the behavior association data, including the impact of temperature and ammonia concentration on pig frequency, identifies the response of pig behavior patterns to environmental changes, and generates environmental impact assessment information.

[0010] As a further solution of the present invention, the growth rate analysis module includes: The weight information collection submodule uses an electronic scale to regularly measure and record the weight data of the pigs based on the environmental impact assessment information and generates a weight data record; The growth trend analysis submodule analyzes the weight growth pattern of the pigs under the current environmental conditions based on the weight data records, identifies key growth stages and growth-impeding factors, and generates growth pattern analysis results; The pig growth prediction submodule calculates the weight change trend based on the growth pattern analysis results and uses time series analysis to predict the weight change of the pigs and generate a growth rate prediction curve.

[0011] As a further solution of the present invention, the specific formula for calculating the weight change trend is:

[0012] in, represents the predicted pig weight, represents the intercept of the model, which is the basic growth offset, and Represents the linear and quadratic weights of the time factor. These two parameters describe the first-order and second-order effects of time on body weight, respectively, and reflect the nonlinear changes in the weight growth rate of pigs at different growth stages. Represents the age of the pig. represents the environmental adjustment coefficient, which indicates the direct impact of environmental factors on the growth rate of pigs. represents the comprehensive environmental factor score, Represents the feed quality adjustment coefficient, indicating the specific contribution of feed changes to pig growth rate. Stands for feed quality index.

[0013] As a further solution of the present invention, the digital model building module includes: The real-time data synchronization submodule generates a model input data set based on the growth rate prediction curve, combined with real-time environmental status data, pig weight data and behavior data; The model input integration submodule constructs a digital twin model of the pig farm based on the model input data set, simulates the operation of the actual pig farm and the pig growth environment, and generates a digital twin model; The digital twin simulation submodule simulates and predicts the growth and environmental interaction effects of pigs in real time based on the digital twin model, analyzes the optimal growth environment of pigs at multiple growth stages, and generates model construction records.

[0014] As a further solution of the present invention, the environmental condition adjustment module includes: The environmental adjustment submodule builds records based on the model, and adjusts the control parameters of the temperature and humidity adjustment equipment and the feed delivery equipment in the pig house according to the growth stage of the pigs, and obtains the environmental parameter adjustment records; The behavior monitoring submodule monitors the behavior changes of the pigs in real time based on the environmental parameter adjustment records, records the activity level and behavior pattern, and generates a behavior change record; The effect evaluation submodule is based on the behavior change records, evaluates the effects of temperature, humidity and feed adjustments by comparing the behavior change records before and after the adjustments, optimizes parameter settings, and generates environmental parameter configurations.

[0015] As a further solution of the present invention, the pig farm epidemic prevention management module includes: The risk identification submodule is configured based on the environmental parameters, identifies pigs with abnormal behaviors by analyzing the image data of the pigs, records health problems and assesses epidemic risks, sends warning information to managers, including the location and time information of abnormal pigs, and generates epidemic risk warning information; The epidemic prevention response submodule adjusts the ventilation system and starts disinfection in real time according to the epidemic risk warning information, records vaccination information, and generates an epidemic prevention response record; The record management submodule records the management information of the pig farm in real time based on the epidemic prevention response record, including ventilation adjustment, disinfection time and vaccination schedule information, and generates a pig farm management record.

[0016] A remote control method for a pig farm based on a digital twin is implemented based on the remote control system of the pig farm based on the digital twin, and comprises the following steps: S1: Based on image acquisition equipment, the behavior dynamics of pigs are monitored in real time, the behavior and physiological status of pigs are analyzed, and environmental data are monitored in real time to analyze the impact of the environment on pig behavior and generate behavior and environmental analysis results; S2: Based on the behavior and environment analysis results, the weight data of the pigs is collected in real time, the growth trend of the pigs in the current environment is analyzed and the weight change is predicted, and the weight change prediction data is generated; S3: Using the weight change prediction data, a digital twin model of the pig farm is constructed, and the adjustment parameters of the pig farm equipment are optimized by synchronizing the on-site pig, environment and behavior data to the model in real time and performing simulation and analysis, thereby generating model construction and optimization records; S4: According to the model construction and optimization records, adjust the environmental conditions of the pig house in real time, monitor the behavior and growth response of the pigs after the adjustment, evaluate the effect of the environmental adjustment, and generate the results of the implementation of the environmental adjustment; S5: Based on the implementation results of the environmental adjustment, combined with the health monitoring data of the pigs, conduct epidemic risk analysis, identify health risks and send early warning information, record epidemic prevention operations and vaccination information, and generate pig farm management records.

[0017] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by real-time monitoring and analysis of the behaviors of multiple pigs in a pig farm, combined with environmental condition data, the impact of environmental conditions on pig behavior is analyzed, targeted environmental adjustments are achieved, pig weights are recorded in real time, and growth trends are analyzed in combination with environmental data, making weight change predictions more accurate, building a digital twin model and simulating the operating status of the pig farm, providing a simulation platform for the adjustment of management measures, enhancing the transparency of pig farm operations and improving the accuracy of decision-making, reducing resource waste and increasing output, optimizing production processes, and achieving cost savings and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a system flow chart of the present invention; Figure 2 It is a schematic diagram of the system framework of the present invention; Figure 3 It is a flow chart of the behavior monitoring and identification module of the present invention; Figure 4 It is a flow chart of the behavior association analysis module of the present invention; Figure 5 This is a flow chart of the growth rate analysis module of the present invention; Figure 6 A flowchart of a digital model building module of the present invention; Figure 7 This is a flow chart of the environmental condition adjustment module of the present invention; Figure 8 This is a flow chart of the pig farm epidemic prevention management module of the present invention; Fig. 9 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0021] See also Figure 1 to Figure 2 The present invention provides a technical solution, a remote control system based on a digital twin pig farm comprising: The behavior monitoring and recognition module is based on image acquisition equipment to monitor and analyze the behavior and physiological status of pigs in the pig farm in real time, identify the pigs' action patterns and abnormal behaviors, and generate behavioral health analysis results; Based on the behavioral health analysis results, the behavioral association analysis module uses temperature, humidity and gas concentration sensors to collect real-time environmental status information in the pig farm, identify the impact of various environmental conditions on pig behavior, and generate environmental impact assessment information; The growth rate analysis module collects and records the pig's weight data in real time based on the environmental impact assessment information, analyzes the pig's growth trend in the current environment, predicts weight changes, and generates a growth rate prediction curve; The digital model building module builds a digital twin model of the pig farm based on the growth rate prediction curve, synchronizes the data of the actual pig farm in real time, and generates model building records; The environmental condition adjustment module is based on model construction records, and adjusts the control parameters of pig house environmental conditions and feed delivery in real time according to growth forecasts and environmental data to generate environmental parameter configurations; The pig farm epidemic prevention management module is based on environmental parameter configuration, combined with health monitoring and environmental monitoring data, to identify epidemic risks, send early warning information, and generate pig farm management records.

[0022] The behavioral health analysis results include activity frequency statistics, abnormal behavior type information, and a list of pig physiological health indicator scores. Environmental impact assessment information includes environmental parameter change trend information, behavioral response sensitivity evaluation data, and the influence of key environmental factors. The growth rate prediction curve includes the predicted weight growth trend, expected maturity time, and environmental adaptability growth analysis results. Model construction records include real-time mapping parameters of pig farm environmental status, synchronous update time points of behavior and growth data, and model verification results. Environmental parameter configuration includes adjusted temperature and humidity ranges, new formula feed delivery cycle, and optimized operation time of the ventilation system. Pig farm management records include the time when epidemic warnings are issued, disinfection measures implementation records, and vaccination and prevention and control measures records.

[0023] See also Figure 2 and Figure 3 , the behavior monitoring and identification module includes: The image data acquisition submodule is based on the image acquisition device, which collects image data from multiple locations in the pig farm in real time, records various activity scenes of pigs, including moving, eating, and resting, and generates behavioral video records; In the image data acquisition submodule, based on the image acquisition equipment, cameras are installed in key areas of the pig farm, including the feeding area, rest area, and activity area. The cameras continuously collect image data at predetermined time intervals to monitor the daily activities of the pigs in real time. The data is transmitted to the central server for storage and subsequent processing through connected transmission equipment, including Wi-Fi or wired network. The image processing technologies used include inter-frame difference method to detect dynamic targets, optical flow method to track the movement trajectory of pigs, and background subtraction technology to improve the recognition accuracy of foreground targets. The captured scenes include pigs moving, eating, resting and other behaviors, and generate time-stamped behavioral video records. Each record marks the time and place of the behavior.

[0024] The pig behavior analysis submodule analyzes images and identifies pig behavior patterns based on behavioral video records, including eating, fighting, and resting, detects abnormal behaviors, including lameness and excessive interaction, and generates behavior pattern data; In the pig behavior analysis submodule, behavioral video records are received from the data acquisition module, and computer vision and deep learning technologies, including convolutional neural networks and recurrent neural networks, are used to perform in-depth image analysis and behavior recognition. Features are extracted from video frames and specific behavior patterns of pigs, including eating, fighting or resting, are identified. Abnormal behaviors such as lameness or excessive interaction are sensitively detected. Anomaly detection algorithms, including isolated forest or clustering-based anomaly detection, are used to improve the early warning capabilities of health problems. The analyzed data are used to generate behavioral pattern data, which include behavior type, frequency of occurrence and duration. Each behavior is classified by importance and urgency to optimize subsequent health assessment and management strategies.

[0025] The pig health assessment submodule evaluates the physiological health status of multiple pigs based on behavioral pattern data, combined with pig activity frequency and behavioral abnormality indicators, including activity, stress indicators, and physiological health problems, and generates behavioral health analysis results; In the above content, the physiological health status of pigs is evaluated by combining the activity frequency and behavioral abnormality indicators, according to the formula , calculate behavioral health analysis results; In the formula, represents the comprehensive health score, Represents the scores of different health indicators, Represents the impact weight of the corresponding health indicator, represents the number of health indicators; Detailed explanation of the formula and the process of formula calculation and derivation: Assuming that three main health indicators are considered, including activity, stress indicators, and physical health problems, the corresponding scores are The corresponding impact weights are 70 points, 50 points, and 30 points. is 0.4, 0.3, 0.3, calculate :

[0026] The results 52 showed that after combining the scores and weights of activity, stress indicators and physiological health issues, the overall health score of pigs was 52. The score was used to quantify the overall health status of pigs and assist managers in further health management and the formulation of intervention measures.

[0027] See also Figure 2 and Figure 4 , the behavior association analysis module includes: Based on the behavioral health analysis results, the real-time environmental monitoring submodule uses a thermometer, hygrometer and ammonia detector to collect real-time environmental status information in the pig farm, including temperature, humidity and harmful gas concentration, and generates real-time environmental status data; In the real-time environmental monitoring submodule, the equipment configuration includes thermometers and hygrometers and ammonia detectors. The equipment is installed in key areas of the pig farm, including the feeding area and the waste treatment area, for real-time monitoring of key environmental parameters. The thermometers and hygrometers regularly record the temperature and humidity values ​​of the air, and the ammonia detectors are used to monitor the concentration of ammonia. Excessive ammonia concentration will seriously affect the health of pigs. The data is sent to the central processing system in real time via a wireless network. The data transmission protocols used include MQTT, which is a lightweight message transmission protocol suitable for scenarios with low bandwidth. Real-time data processing uses the real-time database system InfluxDB, which optimizes the storage and query of time series data. The real-time environmental status data generated includes the temperature, humidity and ammonia concentration readings of each monitoring point, which are recorded in the form of timestamps to ensure the timeliness and accuracy of the data.

[0028] The data fusion analysis submodule is based on real-time environmental status data. By analyzing the correlation between pig behavior data and multiple environmental variables, it analyzes the impact of multiple environmental conditions on pig behavior and generates behavior correlation data. In the data fusion analysis submodule, the real-time environmental status data and behavioral health analysis results are combined, and statistical analysis software R and Python are used for data processing, including Pearson correlation coefficient analysis and multivariate regression analysis, to quantitatively evaluate the relationship between different environmental factors and pig behavior, including the correlation between temperature, humidity and behavioral patterns such as rest and activity. During the analysis process, the environmental status data is preprocessed, including data cleaning and normalization, to eliminate outliers and problems with different ranges. The machine learning algorithm random forest is used to analyze the impact of environmental variables on pig behavior. The generated behavioral association data shows the probability distribution of pig behavior patterns under each environmental condition, providing a scientific basis for pig farm management.

[0029] The environmental impact analysis submodule evaluates the impact of various environmental conditions on pig behavior based on behavioral association data, including the impact of temperature and ammonia concentration on pig frequency, identifies the response of pig behavior patterns to environmental changes, and generates environmental impact assessment information; In the environmental impact analysis submodule, behavioral association data is used to evaluate the impact of multiple factors such as temperature and ammonia concentration on the activity frequency of pigs through data analysis techniques such as factor analysis and path analysis. The analysis is carried out on the data analysis platform Tableau to achieve data visualization and make the analysis results intuitive and easy to understand. By comprehensively considering the interaction of multiple environmental parameters, the sensitivity of pigs to changes in the target environment is revealed. The generated environmental impact assessment information includes the statistical impact of environmental factors on the behavior frequency of pigs. Each data has passed a rigorous statistical significance test to ensure the scientific nature and operability of the information, providing a basis for formulating more effective environmental control strategies.

[0030] See also Figure 2 and Figure 5 , the growth rate analysis module includes: The weight information collection submodule uses an electronic scale to regularly measure and record the weight data of pigs based on the environmental impact assessment information and generates weight data records; In the weight information collection submodule, electronic scales are used to automatically measure the weight of pigs. The electronic scales are installed in different locations in the feeding area of ​​the pig farm to enable stress-free weighing when the pigs move freely. The system automatically activates the weighing process at a fixed time. The frequency of data collection can be adjusted according to the growth stage of the pigs. It is usually set to measure once a week to monitor the growth progress. The collected weight data is transmitted to the central data server via a wireless network. The data collection technology used includes real-time data stream processing, and IoT technology is used to realize instant data upload and storage. The storage system uses a time series database, which is suitable for processing large amounts of time series data. The generated weight data records include timestamps, weight values, and pig identification information, providing basic data for subsequent growth trend analysis.

[0031] The growth trend analysis submodule analyzes the weight growth pattern of pigs under current environmental conditions based on weight data records, identifies key growth stages and growth-impeding factors, and generates growth pattern analysis results; In the growth trend analysis submodule, based on weight data records, statistical and data analysis techniques, including moving average and exponential smoothing techniques, are used to analyze the weight growth pattern of pigs under current environmental conditions, smooth short-term fluctuations and highlight long-term trends. By comparing weight records at different growth stages, the key stages of pig growth are identified. At the same time, variance analysis and regression models are used to identify growth-hindering factors, including environmental pressure, malnutrition, and disease effects. All analysis steps are performed using SPSS or R to ensure the scientificity and accuracy of data processing. The generated growth pattern analysis results describe the weight change characteristics and growth-limiting factors at each stage, providing a basis for formulating targeted feeding strategies.

[0032] The pig growth prediction submodule uses time series analysis to calculate the weight change trend based on the growth pattern analysis results, predict the weight change of pigs, and generate a growth rate prediction curve; The specific formula for calculating the weight change trend is:

[0033] in, represents the predicted pig weight, represents the intercept of the model, which is the basic growth offset, and Represents the linear and quadratic weights of the time factor. These two parameters describe the first-order and second-order effects of time on body weight, respectively, and reflect the nonlinear changes in the weight growth rate of pigs at different growth stages. Represents the age of the pig. represents the environmental adjustment coefficient, which indicates the direct impact of environmental factors on the growth rate of pigs. represents the comprehensive environmental factor score, Represents the feed quality adjustment coefficient, indicating the specific contribution of feed changes to pig growth rate. Stands for feed quality index.

[0034] formula:

[0035] Detailed explanation of the formula and the process of formula calculation and derivation: The formula is used to calculate the predicted weight of pigs under target environmental and feed conditions, and the results are used to optimize management strategies and resource allocation on pig farms; Parameter meaning and setting value: is the intercept, which indicates the starting offset of the basic body weight without the influence of other factors, and is assumed to be 1.5 kg, reflecting the basic body weight of pigs in the youngest age group; is the first-order time weight, which reflects the direct increase in weight with age. Assuming it is 0.4, it means that the average weight increase is 0.4 kg for every day of age. is the second-order time weight, which describes the acceleration of weight growth with age. It is assumed to be 0.001, indicating the acceleration effect of weight growth; is age, assuming it is 106 days, that is, the target pig grows to 106 days; is the weight of environmental factors, indicating the impact of environmental factors on weight gain, assumed to be 0.05; is the comprehensive environmental factor score, assuming it is 9 points, which represents the current environmental quality score of the pigs; is the feed quality weight, reflecting the effect of feed quality on body weight gain. Assuming it is 0.1, it means that high-quality feed can increase body weight by an additional 0.1 kg per day. is the feed quality index, assuming it is 9.5 points, indicating the quality of feed received by pigs.

[0036] Substitute the parameters into the formula for calculation:

[0037]

[0038] result It shows that at 106 days of age, the predicted body weight is 56.536 kg. The value is used to adjust daily management and feeding plans to ensure efficient operation and sustainable development of the pig farm.

[0039] See also Figure 2 and Figure 6 , the digital model building modules include: The real-time data synchronization submodule generates the model input data set based on the growth rate prediction curve, combined with the real-time environmental status data, pig weight data and behavior data; In the real-time data synchronization submodule, real-time environmental status data, pig weight data and behavior data are collected and integrated, data integration technology is used to ensure the integrity and real-time nature of the information, SQL database management system is used for data storage and access, and real-time data streaming technology Apache Kafka is used to process and synchronize data in real time to generate a model input data set. The data set contains timestamps, temperature, humidity, pig weight changes and behavior patterns. The data set is preprocessed, including data cleaning and normalization, to ensure data quality and provide accurate basic data for building a digital twin model.

[0040] The model input integration submodule builds a digital twin model of the pig farm based on the model input data set, simulates the actual pig farm operation and pig growth environment, and generates a digital twin model; In the model input integration submodule, the system dynamics model and agent model are used to construct a digital twin model of the pig farm. The model is developed in MATLAB / Simulink and Python environments to simulate the actual operation of the pig farm and the pig growth environment, and optimize to ensure the real-time update and accuracy of the model. The digital twin model integrates real-time environmental monitoring data and pig growth data. Through simulation, the model shows the impact of different management decisions on pig growth. The generated digital twin model provides decision makers with a visual and interactive tool for strategy testing and optimization. The generated model records the model's parameter settings, operating status and prediction results, providing a reliable reference framework for further analysis and decision-making.

[0041] The digital twin simulation submodule simulates and predicts the growth and environmental interaction effects of pigs in real time based on the digital twin model, analyzes the optimal growth environment of pigs at multiple growth stages, and generates model construction records; In the digital twin simulation sub-module, based on the constructed digital twin model, real-time simulation and prediction of the growth and environmental interaction effects of pigs are implemented. The predictive analysis tool SAS Visual Analytics is used for data analysis, complex data sets are processed and insight reports are generated. During the simulation process, the module uses machine learning technology neural networks to analyze the data of pigs at different growth stages and predict the optimal growth environment conditions, including ideal temperature, humidity and ammonia concentration. The analysis results help adjust the environmental control system and optimize the growth conditions of pigs. The generated model construction records include environmental settings for each growth stage, predicted growth rates and optimization suggestions, providing pig farms with detailed operational optimization plans.

[0042] See also Figure 2 and Figure 7 , the environmental condition adjustment module includes: The environmental adjustment submodule builds records based on the model and adjusts the control parameters of the temperature and humidity adjustment equipment and feed delivery equipment in the pig house according to the growth stage of the pigs to obtain the environmental parameter adjustment records; In the environmental regulation submodule, based on the model construction records, the optimal environmental requirements of pigs in different growth stages are analyzed, and the environmental management software is used to adjust the control parameters of temperature, humidity and feed delivery equipment. The software integrates automatic control algorithms, fuzzy logic controller and PID controller to ensure that the temperature and humidity regulation equipment can be accurately adjusted according to real-time data. The feed delivery equipment automatically adjusts the delivery amount and frequency according to the growth needs of pigs and predicted data. During operation, the system regularly updates the pig growth data and environmental feedback from the database to implement precise control. The generated environmental parameter adjustment records record the time of each adjustment, the adjusted parameter value and its corresponding pig growth stage. The data provides an empirical basis for continuously optimizing the environmental conditions of pig farms.

[0043] The behavior monitoring submodule adjusts records based on environmental parameters, monitors the behavior changes of pigs in real time, records activity levels and behavior patterns, and generates behavior change records; In the behavior monitoring submodule, behavior recognition technology is used to monitor and record the behavioral changes of pigs in real time, including video analysis and sensor data fusion. The video monitoring system is equipped with motion detection and image recognition algorithms, such as support vector machines and deep learning networks, to identify and classify the behavior patterns of pigs, such as activity, rest, and eating, in real time from video streams. Sensors are used to collect dynamic activity data of pigs. The system synchronously analyzes multi-source data and generates behavioral change records, including activity level data, behavior frequency, and time distribution, providing basic data for subsequent behavioral analysis and health assessment.

[0044] The effect evaluation submodule is based on the behavior change records. By comparing the behavior change records before and after the adjustment, it evaluates the effects of temperature, humidity and feed adjustments, optimizes parameter settings, and generates environmental parameter configurations; In the above content, based on the behavior change records, according to the formula Evaluate the effects of adjustments to environmental parameters; In the formula, represents the quality index of the adjustment effect, represents the adjusted change in activity level, represents the response to feed adjustment, , , are the weight of the overall adjustment effect and the influence coefficient of each variable respectively; Detailed explanation of the formula and the process of formula calculation and derivation: Overall adjustment effect weights for assumptions , the influence coefficients of activity level change and feed adjustment response are and , adjusted for change in activity level Improvement, feed adjustment response Lift, calculate :

[0045] Results 13 showed that the adjustment effect quality index obtained according to the formula reflected that the adjustment of environmental and feed parameters had a positive impact on the pigs' behavior patterns and numerically optimized the environmental parameter settings.

[0046] See also Figure 2 and Figure 8 , the pig farm epidemic prevention management module includes: The risk identification submodule is configured based on environmental parameters. It analyzes the pigs’ image data, identifies pigs with abnormal behaviors, records health problems, and assesses epidemic risks. It sends warning information to managers, including the location and time information of abnormal pigs, and generates epidemic risk warning information. In the above content, the risk identification submodule uses image analysis technology to Assess epidemic risks and generate epidemic risk warning information; In the formula, represents the epidemic risk score, is the frequency of abnormal behaviors identified, is the severity score of the abnormal behavior. is the number of different types of abnormal behaviors identified; Detailed explanation of the formula and the process of formula calculation and derivation: Assume that within a certain period of time, the abnormal behaviors identified by the system and the corresponding severity are: limp times, corresponding to the severity , excessive interaction times, corresponding to the severity , fight times, corresponding to the severity ,calculate :

[0047] result The epidemic risk score is 41, indicating that there may be a higher risk of health or behavioral problems in the pig population. The calculated value is used to send early warnings to managers to ensure that epidemic prevention measures can be taken in time, including adjusting the ventilation system and initiating disinfection procedures to reduce and avoid epidemic risks.

[0048] The epidemic prevention response submodule adjusts the ventilation system and starts disinfection in real time based on the epidemic risk warning information, records vaccination information, and generates epidemic prevention response records; In the epidemic prevention response submodule, the system starts the ventilation system adjustment and disinfection procedures according to the received epidemic risk warning information to reduce the risk of epidemic spread. In the process, the automatic control system PLC and SCADA are used to manage the operation of ventilation and disinfection equipment, adjust the equipment settings in real time to adapt to the current environmental needs, and record the information of each vaccination, including the type of vaccine, vaccination time and pig identification information. The operation records are stored in time series data to ensure the traceability and auditability of the data. The generated epidemic prevention response records provide the pig farm with the epidemic prevention operation history to support health management and epidemic prevention planning.

[0049] The record management submodule records the management information of the pig farm in real time based on the epidemic prevention response records, including ventilation adjustment, disinfection time and vaccination schedule information, and generates pig farm management records; In the record management submodule, the daily operation records of the pig farm are updated and managed, including the adjustment parameters of the ventilation system, the time of disinfection activities, and the detailed schedule of vaccination. Data management is performed through Microsoft SQL Server, and data of key operations and events are recorded and stored in real time to ensure the integrity and security of information. By automatically summarizing data, comprehensive pig farm management records are generated, including operation time, executor, operation content and results, providing a comprehensive management view for the pig farm, supporting decision-making and operation optimization, and conducting health and safety audits to help pig farms meet industry standards and regulatory requirements and improve management efficiency and quality assurance.

[0050] See also Fig. 9 A digital twin pig farm remote control method is provided. The digital twin pig farm remote control method is performed based on the digital twin pig farm remote control system, and includes the following steps: S1: Based on image acquisition equipment, the behavior dynamics of pigs are monitored in real time, the behavior and physiological status of pigs are analyzed, and environmental data are monitored in real time to analyze the impact of the environment on pig behavior and generate behavior and environmental analysis results; S2: Based on the results of behavior and environmental analysis, the weight data of pigs is collected in real time, the growth trend of pigs in the current environment is analyzed and weight changes are predicted, and weight change prediction data is generated; S3: Use the weight change prediction data to build a digital twin model of the pig farm. By synchronizing the on-site pig, environment and behavior data to the model in real time and performing simulation and analysis, the adjustment parameters of the pig farm equipment are optimized and the model construction and optimization records are generated. S4: According to the model construction and optimization records, adjust the environmental conditions of the pig house in real time, monitor the behavior and growth response of the pigs after the adjustment, evaluate the effect of the environmental adjustment, and generate the results of the implementation of the environmental adjustment; S5: Based on the results of environmental adjustment implementation and combined with the health monitoring data of pigs, conduct epidemic risk analysis, identify health risks and send early warning information, record epidemic prevention operations and vaccination information, and generate pig farm management records.

[0051] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A remote control system for a pig farm based on digital twin, characterized in that: The system comprises: The behavior monitoring and recognition module is based on image acquisition equipment to monitor and analyze the behavior and physiological status of pigs in the pig farm in real time, identify the pigs' action patterns and abnormal behaviors, and generate behavioral health analysis results; Based on the behavioral health analysis results, the behavior association analysis module uses temperature, humidity and gas concentration sensors to collect real-time environmental status information in the pig farm, identify the impact of various environmental conditions on pig behavior, and generate environmental impact assessment information; The growth rate analysis module collects and records the weight data of the pigs in real time based on the environmental impact assessment information, analyzes the growth trend of the pigs in the current environment, predicts the weight change, and generates a growth rate prediction curve; The digital model building module builds a digital twin model of the pig farm based on the growth rate prediction curve, synchronizes the data of the actual pig farm in real time, and generates a model building record; The environmental condition adjustment module builds records based on the model, adjusts the control parameters of the pig house environmental conditions and feed delivery in real time according to growth prediction and environmental data, and generates environmental parameter configuration; The pig farm epidemic prevention management module is based on the environmental parameter configuration, combined with health monitoring and environmental monitoring data, to identify epidemic risks, send early warning information, and generate pig farm management records.

2. The digital twin pig farm remote control system according to claim 1 is characterized in that: The behavioral health analysis results include activity frequency statistics, abnormal behavior type information, and a pig physiological health index score list; the environmental impact assessment information includes environmental parameter change trend information, behavioral response sensitivity evaluation data, and the influence of key environmental factors; the growth rate prediction curve includes predicted weight growth trend, expected maturity time, and environmental adaptability growth analysis results; the model construction record includes real-time mapping parameters of pig farm environmental status, synchronous update time points of behavior and growth data, and model verification results; the environmental parameter configuration includes adjusted temperature and humidity ranges, new formula feed delivery cycle, and optimized operation time of the ventilation system; the pig farm management records include the time when epidemic warnings are issued, disinfection measures implementation records, and vaccination and prevention and control measures records.

3. The digital twin pig farm remote control system according to claim 1 is characterized in that: The behavior monitoring and identification module includes: The image data acquisition submodule is based on the image acquisition device, which collects image data from multiple locations in the pig farm in real time, records various activity scenes of pigs, including moving, eating, and resting, and generates behavioral video records; The pig behavior analysis submodule analyzes images and identifies pig behavior patterns based on the behavior video records, including eating, fighting, and resting, detects abnormal behaviors, including lameness and excessive interaction, and generates behavior pattern data; The pig health assessment submodule evaluates the physiological health status of multiple pigs based on the behavioral pattern data, combined with the pigs' activity frequency and behavioral abnormality indicators, including activity, stress indicators, and physiological health problems, and generates behavioral health analysis results.

4. The digital twin pig farm remote control system according to claim 1 is characterized in that: The behavior association analysis module includes: Based on the behavioral health analysis results, the real-time environmental monitoring submodule uses a thermometer and ammonia detector to collect real-time environmental status information in the pig farm, including temperature, humidity and harmful gas concentration, and generates real-time environmental status data; The data fusion analysis submodule analyzes the correlation between the pig behavior data and various environmental variables based on the real-time environmental status data, and analyzes the influence of various environmental conditions on the pig behavior to generate behavior correlation data; The environmental impact analysis submodule evaluates the impact of various environmental conditions on pig behavior based on the behavior association data, including the impact of temperature and ammonia concentration on pig frequency, identifies the response of pig behavior patterns to environmental changes, and generates environmental impact assessment information.

5. The digital twin pig farm remote control system according to claim 1 is characterized in that: The growth rate analysis module comprises: The weight information collection submodule uses an electronic scale to regularly measure and record the weight data of the pigs based on the environmental impact assessment information and generates a weight data record; The growth trend analysis submodule analyzes the weight growth pattern of the pigs under the current environmental conditions based on the weight data records, identifies key growth stages and growth-impeding factors, and generates growth pattern analysis results; The pig growth prediction submodule calculates the weight change trend based on the growth pattern analysis results and uses time series analysis to predict the weight change of the pigs and generate a growth rate prediction curve.

6. The digital twin pig farm remote control system according to claim 5 is characterized in that: The specific formula for calculating the weight change trend is: in, represents the predicted pig weight, represents the intercept of the model, which is the basic growth offset, and Represents the linear and quadratic weights of the time factor. These two parameters describe the first-order and second-order effects of time on body weight, respectively, and reflect the nonlinear changes in the weight growth rate of pigs at different growth stages. Represents the age of the pig. represents the environmental adjustment coefficient, which indicates the direct impact of environmental factors on the growth rate of pigs. represents the comprehensive environmental factor score, Represents the feed quality adjustment coefficient, indicating the specific contribution of feed changes to pig growth rate. Stands for feed quality index.

7. The digital twin pig farm remote control system according to claim 1 is characterized in that: The digital model building module includes: The real-time data synchronization submodule generates a model input data set based on the growth rate prediction curve, combined with real-time environmental status data, pig weight data and behavior data; The model input integration submodule constructs a digital twin model of the pig farm based on the model input data set, simulates the operation of the actual pig farm and the pig growth environment, and generates a digital twin model; The digital twin simulation submodule simulates and predicts the growth and environmental interaction effects of pigs in real time based on the digital twin model, analyzes the optimal growth environment of pigs at multiple growth stages, and generates model construction records.

8. The digital twin pig farm remote control system according to claim 1 is characterized in that: The environmental condition adjustment module comprises: The environmental adjustment submodule builds records based on the model, and adjusts the control parameters of the temperature and humidity adjustment equipment and the feed delivery equipment in the pig house according to the growth stage of the pigs, and obtains the environmental parameter adjustment records; The behavior monitoring submodule monitors the behavior changes of the pigs in real time based on the environmental parameter adjustment records, records the activity level and behavior pattern, and generates a behavior change record; The effect evaluation submodule is based on the behavior change records, evaluates the effects of temperature, humidity and feed adjustments by comparing the behavior change records before and after the adjustments, optimizes parameter settings, and generates environmental parameter configurations.

9. The digital twin pig farm remote control system according to claim 1 is characterized in that: The pig farm epidemic prevention management module includes: The risk identification submodule is configured based on the environmental parameters, identifies pigs with abnormal behaviors by analyzing the image data of the pigs, records health problems and assesses epidemic risks, sends warning information to managers, including the location and time information of abnormal pigs, and generates epidemic risk warning information; The epidemic prevention response submodule adjusts the ventilation system and starts disinfection in real time according to the epidemic risk warning information, records vaccination information, and generates an epidemic prevention response record; The record management submodule records the management information of the pig farm in real time based on the epidemic prevention response record, including ventilation adjustment, disinfection time and vaccination schedule information, and generates a pig farm management record.

10. A remote control method for a pig farm based on digital twins, characterized in that: The digital twin pig farm remote control system according to any one of claims 1 to 9 comprises the following steps: Based on image acquisition equipment, the behavior dynamics of pigs are monitored in real time, the behavior and physiological status of pigs are analyzed, and environmental data are monitored in real time to analyze the impact of the environment on pig behavior and generate behavior and environmental analysis results; Based on the behavior and environment analysis results, the weight data of the pigs is collected in real time, the growth trend of the pigs in the current environment is analyzed and the weight change is predicted, and the weight change prediction data is generated; Using the weight change prediction data, a digital twin model of the pig farm is constructed. By synchronizing the on-site pig, environment and behavior data to the model in real time and performing simulation and analysis, the adjustment parameters of the pig farm equipment are optimized, and model construction and optimization records are generated; According to the model construction and optimization records, the environmental conditions of the pig house are adjusted in real time, and the behavior and growth response of the pigs after the adjustment are monitored, the effect of the environmental adjustment is evaluated, and the implementation results of the environmental adjustment are generated; Based on the results of the implementation of the environmental adjustments, combined with the health monitoring data of the pigs, an epidemic risk analysis is conducted, health risks are identified and early warning information is sent, epidemic prevention operations and vaccination information are recorded, and pig farm management records are generated.

Images