Intelligent regulation and control system for livestock and poultry breeding environment based on internet of things

Abstract

This invention relates to the field of intelligent management and control technology for livestock and poultry farming, specifically an intelligent environmental control system for livestock and poultry farming based on the Internet of Things (IoT). The system includes a data acquisition module, a coupled field analysis module, and a decision optimization module. The data acquisition module collects environmental parameters such as air temperature, air humidity, harmful gas concentration, and light intensity in real time through a sensor array within the livestock shed. The coupled field analysis module inputs these parameters into a pre-trained environmental coupled field model, simulates the dynamic interaction and propagation relationships between parameters, and outputs an environmental disturbance map reflecting the degree of mutual influence between parameters. The decision optimization module identifies key influencing parameters based on the environmental disturbance map and generates optimization instructions containing control priorities by combining them with preset environmental target ranges. This system optimizes the environmental management and control logic for livestock and poultry farming through environmental parameter coupling analysis and priority control design, adapting to the needs of refined environmental control in livestock sheds.

Description

Technical Field

[0001] This invention relates to the field of intelligent management and control technology for livestock and poultry farming, and in particular to an intelligent control system for the livestock and poultry farming environment based on the Internet of Things. Background Technology

[0002] Traditional livestock and poultry farming environmental management mostly relies on independent sensors to collect parameters such as air temperature, humidity, harmful gas concentration, and light intensity within the barn. The management system triggers control actions based solely on preset thresholds for a single parameter. Existing IoT-based environmental monitoring systems can only simultaneously collect and independently display multiple environmental parameters, without quantitative analysis or modeling the inherent relationships between them. Environmental parameters within the barn exhibit natural dynamic interaction characteristics. Temperature changes directly affect humidity distribution and the diffusion efficiency of harmful gases, while fluctuations in light intensity also have a ripple effect on temperature and humidity levels. Furthermore, these parameters can propagate environmental disturbances within the space, which cannot be reflected by single-parameter monitoring.

[0003] Existing environmental control methods all rely on individual parameters for regulation, failing to simulate the dynamic interactions and propagation relationships between various environmental parameters. They cannot generate quantitative representations reflecting the degree of mutual influence between parameters, nor can they identify the key parameters with the greatest impact on the overall environment from among multiple parameters. Furthermore, control commands lack execution priorities, and the control behavior does not match the actual environmental changes in the livestock sheds, resulting in insufficient targeting and rationality in environmental control. This solution addresses the shortcomings of existing technologies in achieving coupled field analysis of environmental parameters and generating priority-based control commands based on the degree of parameter influence by establishing a corresponding intelligent control system. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing an intelligent control system for livestock and poultry farming environments based on the Internet of Things.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent control system for livestock and poultry farming environment based on the Internet of Things, comprising: The data acquisition module acquires environmental parameters in real time through a sensor array deployed in the breeding shed. These environmental parameters include air temperature, air humidity, concentration of harmful gases, and light intensity. The coupled field analysis module inputs the environmental parameters into a pre-trained environmental coupled field model. The environmental coupled field model simulates the dynamic interaction and propagation relationship between the environmental parameters and outputs an environmental disturbance map that reflects the degree of mutual influence between the environmental parameters. The decision optimization module identifies the key impact parameters that have the greatest impact on the overall environment based on the environmental disturbance map, and generates optimization instructions that include the priority of the key impact parameters in combination with the preset environmental target range.

[0006] As a further aspect of the present invention, the real-time acquisition of environmental parameters through a sensor array deployed within the breeding shed includes: Temperature sensors, humidity sensors, gas sensors, and light sensors are evenly distributed across multiple vertical height layers and horizontal zones in the breeding shed. Set a unified sampling clock to drive all sensors to collect environmental parameters at the same timestamp; The collected environmental parameters are preprocessed, including removing abrupt abnormal values ​​caused by instantaneous sensor failures, smoothing parameters with minor disturbances using median filtering, and binding all sensor data with timestamps and location information, packaging them into environmental parameter data frames.

[0007] As a further aspect of the present invention, the environmental parameters are input into a pre-trained environmental coupled field model, including: Construct a virtual field network with the spatial location of the breeding shed as nodes, and assign the environmental parameter values ​​collected by each sensor to the corresponding location node; In the virtual field network, the propagation rules and coupling strength of environmental parameters between adjacent nodes are defined. The propagation rules include the laws of heat conduction, gas diffusion and light attenuation. Using the current environmental parameter values ​​of all nodes as the initial state, an iterative evolution simulation is performed in the virtual field network. During the evolution process, the fluctuations of each environmental parameter are recorded to determine how they trigger changes in other environmental parameters. After the evolution is complete, the sum of disturbances caused by each environmental parameter in the virtual field network is calculated, normalized, and an environmental disturbance map is drawn with environmental parameter type as the dimension and disturbance influence as the value.

[0008] As a further aspect of the present invention, the construction of a virtual field network with the spatial location of the breeding shed as nodes, and the assignment of environmental parameter values ​​collected by each sensor to the corresponding location node, includes: Based on the actual three-dimensional physical structure of the breeding house, a virtual field network is established in the digital space in a gridded manner with equal spacing, where each grid intersection represents a virtual location node; Establish a mapping relationship between the physical location of the sensor and the virtual location node, and match the three-dimensional coordinates of the sensor deployment to the corresponding virtual node; At each sampling moment, the location information of each sensor and its corresponding measured environmental parameter values ​​are extracted from the environmental parameter data frames collected by the sensor array. The extracted measured values ​​are assigned to the corresponding location nodes with mapping relationships in the virtual field network. For virtual location nodes that are not directly covered by sensors, their environmental parameter values ​​are marked as states to be inferred.

[0009] As a further aspect of the present invention, in the virtual field network, the propagation rules and coupling strength of environmental parameters between adjacent nodes are defined. The propagation rules include the laws of heat conduction, gas diffusion, and light attenuation, including: For air temperature parameters, their propagation between adjacent nodes is defined to follow the law of heat conduction, and their coupling strength is calculated based on the thermal conductivity of air, the distance between nodes, and the presence of obstacles. For the concentration parameter of harmful gas, its propagation between adjacent nodes is defined to follow the gas diffusion law, and its coupling strength is set based on the gas diffusion coefficient, the pressure gradient between nodes and the ventilation airflow direction. For the light intensity parameter, its propagation between adjacent nodes is defined to follow the light attenuation law, and its coupling strength is calculated based on the linear propagation characteristics of light, whether there are obstructions between nodes, and the characteristics of the reflective surface. For the air humidity parameter, its propagation is defined as coupled with air temperature and gas flow state, and the coupling strength is determined by the temperature-humidity correlation model and the air flow velocity of adjacent nodes.

[0010] As a further aspect of the present invention, the step of taking the environmental parameter values ​​of all nodes at the current moment as the initial state and performing iterative evolution simulation in the virtual field network, recording how the fluctuation of each environmental parameter triggers changes in other environmental parameters during the evolution process, includes: After the virtual field network is initialized, the current environmental parameter values ​​of all nodes are loaded to the initial state; Initiate a discrete iterative evolution simulation based on time steps. Within each time step, calculate the fluctuation increment of each node's environmental parameter values ​​due to the changes in the parameter values ​​of its neighboring nodes, according to the defined propagation rules and coupling strength. During the calculation process, for each original environmental parameter type, the initial fluctuation is recorded, and the changes in all related environmental parameter types caused by it at each node in the network are recorded, and the source parameters of the changes are traced. The influence relationships between parameters recorded in each iteration are accumulated until the simulation reaches the preset evolution time or the network state changes tend to stabilize, thus completing a single evolution process.

[0011] As a further aspect of the present invention, after the evolution is completed, the sum of the disturbances caused by each environmental parameter in the virtual field network is statistically analyzed, normalized, and an environmental disturbance map is drawn with environmental parameter type as the dimension and disturbance influence as the numerical value, including: After the evolution simulation is completed, the sum of the absolute values ​​of the changes in all other environmental parameter types caused by the initial values ​​of each environmental parameter type during the entire evolution process is obtained to obtain the original perturbation sum of the environmental parameter types. The sum of the original disturbances for all environmental parameter types is normalized. Each original disturbance sum is divided by the maximum value of the sum of the original disturbances for all environmental parameter types to obtain the normalized disturbance influence for each environmental parameter type, with a value ranging from 0 to 1. Using the type of environmental parameter as the horizontal axis and the corresponding normalized disturbance influence as the vertical axis, a visual environmental disturbance map in the form of a bar chart or line graph is generated. Each bar or data point represents a type of environmental parameter and a quantitative indicator of its impact on the overall environment of the breeding house.

[0012] As a further aspect of the present invention, based on the environmental disturbance map, the key impact parameters that have the greatest impact on the overall environment are identified, including: In the environmental disturbance map, the disturbance influence values ​​corresponding to each environmental parameter are sorted in descending order; The environmental parameter ranked first is marked as the key influencing parameter, and the difference between the measured value of the key influencing parameter and the preset environmental target range is calculated as the current core control deviation. Analyze the environmental disturbance map to identify one or more dependent influence parameters whose coupling strength with the key influence parameter exceeds a preset threshold.

[0013] As a further aspect of the present invention, the step of generating optimization instructions that include the priority of key influencing parameters by combining a preset environmental target range includes: Generate core optimization instructions for the key influencing parameters. The goal of the core optimization instructions is to adjust their values ​​to within the environmental target range. The instructions include the expected adjustment amount for the corresponding actuator calculated based on the current core control deviation. To generate follow-up optimization instructions for subordinate influencing parameters that are significantly coupled with key influencing parameters, the adjustment direction and magnitude of the follow-up optimization instructions are determined by the coupling relationship between the subordinate influencing parameters and the key influencing parameters, in order to offset or compensate for the adverse linkage effects on the subordinate influencing parameters when the core optimization instructions are executed; The core optimization instruction is combined with one or more follow-up optimization instructions in order of priority to generate the final optimization instruction.

[0014] As a further aspect of the present invention, the system further includes: The control execution module sends the optimization instructions to the execution device corresponding to the environmental parameters, driving the execution device to adjust the environmental parameters; The feedback adjustment module synchronously acquires the adjusted environmental parameters through the sensor array during the operation of the actuator, compares the adjusted environmental parameters with the preset environmental target range, and if the requirements are met, maintains the current operating state of the actuator; otherwise, it recalculates new optimization instructions. The optimization command is sent to the execution device corresponding to the environmental parameters, driving the execution device to adjust the environmental parameters, specifically including: Analyze and optimize instructions to identify the control objectives and adjustment values ​​for different environmental parameters within the instructions; The control target and adjustment amount are converted into control signals that can be recognized by the corresponding actuator. The control signals include commands for the fan speed and damper opening of the ventilation equipment, commands for the power level of the heating or cooling equipment, and commands for the illumination period and intensity of the lighting equipment. The control signal is sent to the corresponding actuator controller via the Internet of Things communication network, and the actuator controller drives the ventilation equipment, temperature control equipment, and lighting equipment to adjust according to the instructions.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: The collected parameters, such as air temperature, humidity, harmful gas concentration, and light intensity in the livestock shed, are input into a pre-trained environmental coupling field model. The model can fully reproduce the dynamic interaction process of various environmental parameters in the livestock shed space, accurately depict the interaction forms and transmission paths between parameters, fully present the propagation law of environmental parameters in different spatial locations, and the output environmental disturbance map can directly quantify the degree of mutual influence between various environmental parameters. It fully preserves the environmental change characteristics under the coupling effect of multiple parameters, breaks through the limitations of traditional single-parameter independent analysis, and fully presents the real correlation state of environmental parameters in the livestock shed. This ensures that the analysis results of environmental parameters are consistent with the actual environmental change state in the livestock shed, and allows the inherent correlation characteristics of environmental parameters to be fully presented and quantitatively expressed.

[0016] Based on the parameter influence relationships reflected in the environmental disturbance map, the key parameters that have the most significant impact on the overall environmental state of the livestock shed can be accurately identified. Combined with the preset environmental target range, optimized instructions with control priorities are generated. The control execution sequence can be planned according to the weight ranking of key influence parameters, so that the control actions are implemented in an orderly manner according to the actual impact of environmental parameters, avoiding disorder and arbitrariness in control behavior. The logic of the control instructions is aligned with the coupling change characteristics of the livestock shed environmental parameters, reducing unnecessary control actions and ensuring that the control behavior is consistent with the transmission direction of environmental disturbances. The execution logic of the control instructions is optimized, so that the implementation process of environmental control is consistent with the actual change needs of the livestock shed environment, improving the adaptability of environmental control to the real environmental state of the livestock shed, making the control actions more in line with the coupling mechanism of environmental parameters, and making the overall environmental control execution method more in line with the actual management status of the livestock and poultry breeding environment. Attached Figure Description

[0017] Figure 1This is a timing diagram of the intelligent control system for livestock and poultry farming environment based on the Internet of Things as described in this invention. Figure 2 A flowchart for real-time acquisition of environmental parameters; Figure 3 A flowchart for constructing a virtual field network and assigning environmental parameter values. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] See Figure 1This invention provides an intelligent control system for livestock and poultry farming environment based on the Internet of Things (IoT). The overall implementation scheme of the system is as follows: The core of the IoT-based intelligent control system for livestock and poultry farming environment lies in achieving coordinated optimization of the farming environment through a closed loop of data perception, coupled analysis, and intelligent decision-making. The system operation begins with a data acquisition module, which continuously and synchronously captures multi-dimensional environmental parameters, including air temperature, air humidity, harmful gas concentration, and light intensity, through a sensor array deployed in the farming shed. This real-time data is transmitted to a coupled field analysis module. This module has a pre-set environmental coupled field model. This model does not analyze a single parameter independently, but focuses on simulating the complex dynamic interactions and spatial propagation relationships between parameters such as temperature, humidity, gas, and light. After the environmental parameters are input into the model, calculations are performed, and a result called an environmental disturbance diagram is output. This diagram quantitatively reveals the comprehensive impact of each environmental parameter on the overall environment of the farming shed through interaction. The decision optimization module receives this environmental disturbance map, analyzes the data in the map, accurately identifies the key parameters that have the greatest impact on the overall environment at the current moment, and generates a set of optimization instructions with clear priorities by combining them with a pre-set environmental target range suitable for livestock and poultry growth. This set of instructions not only adjusts the key parameters but also considers their cascading effects on other parameters, thus realizing the transformation from single-parameter adjustment to multi-parameter collaborative optimization.

[0021] In one embodiment of the present invention, see [reference] Figure 2 Temperature, humidity, gas, and light sensors were evenly distributed across multiple vertical levels and horizontal zones within the breeding shed. A unified sampling clock was set to drive all sensors to collect environmental parameters at identical timestamps. The raw environmental parameters were preprocessed, including removing outliers caused by momentary sensor malfunctions and smoothing parameters with minor disturbances using median filtering. After preprocessing, all sensor data was bound to its corresponding timestamp and three-dimensional spatial location information, and packaged into structured environmental parameter data frames.

[0022] A virtual field network is constructed, with the spatial locations of the aquaculture sheds as nodes. Environmental parameter values ​​collected by each sensor are assigned to the corresponding location node. Within this virtual field network, the propagation rules and coupling strengths of environmental parameters between adjacent nodes are defined. The propagation rules include heat conduction, gas diffusion, and light attenuation. Using the current environmental parameter values ​​of all nodes as the initial state, an iterative evolution simulation is performed within the virtual field network. During the evolution, how fluctuations in each environmental parameter trigger changes in other environmental parameters is recorded. After the evolution is complete, the sum of disturbances caused by each environmental parameter in the virtual field network is calculated, normalized, and an environmental disturbance map is plotted, with environmental parameter type as the dimension and disturbance influence as the numerical value.

[0023] In the specific implementation, a closed broiler chicken farm measuring 20 meters long, 10 meters wide, and 3 meters high was selected as the implementation scenario. The farm was horizontally divided into four equal rectangular sections, and vertically, three height levels were set at 0.5 meters, 1.5 meters, and 2.5 meters above the ground. Temperature, humidity, gas, and light sensors were evenly distributed at the intersections of these sections and height levels, totaling 36 sensors. A unified sampling clock with a 1-minute cycle was set. This master clock synchronously sent sampling pulses to all sensors via an IoT communication network, driving the 36 sensors to collect environmental parameter values ​​for air temperature, air humidity, ammonia concentration, and light intensity at identical millisecond-level timestamps. The collected raw environmental parameter values ​​are preprocessed. The preprocessing operation includes removing abrupt abnormal values ​​caused by instantaneous sensor failures. For example, if a humidity sensor value drops sharply from 65% to 10% and then recovers to 64% within three consecutive sampling periods, this 10% jump value is identified as abnormal and removed. At the same time, parameters with small disturbances are smoothed by median filtering. For example, a set of continuous air temperature readings [25.1, 25.3, 25.0, 25.4, 25.2] is output as [25.1, 25.1, 25.3, 25.2, 25.2] after being filtered by a median filter with a window size of 3. After preprocessing, the data from all sensors are bound to their corresponding timestamps and three-dimensional spatial coordinates and packaged into structured environmental parameter data frames.

[0024] In some embodiments, the process of constructing a virtual field network with the spatial location of the breeding shed as nodes is as follows: The three-dimensional space of the breeding shed is divided into a grid in digital space with a spacing of 0.5 meters, generating a 41×21×7 grid. Each grid intersection represents a virtual location node in the virtual field network. A mapping relationship between the physical location of the sensor and the virtual location node is established. For example, a sensor located at coordinates (5.0, 2.5, 1.5) is matched to the nearest virtual node (5.0, 2.5, 1.5). At each sampling time, the location information of each sensor and its corresponding measured environmental parameter values ​​are extracted from the environmental parameter data frame. The extracted measured values ​​of air temperature, air humidity, harmful gas concentration, and light intensity are assigned to the corresponding location nodes with mapping relationships in the virtual field network. For virtual location nodes not directly covered by sensors, the environmental parameter values ​​of that node are marked as a state to be inferred.

[0025] In the virtual field network, the propagation rules and coupling strengths of environmental parameters between adjacent nodes are defined. The propagation rules include heat conduction, gas diffusion, and light attenuation. For air temperature, its propagation between adjacent nodes is defined as following the law of heat conduction, and its coupling strength is calculated based on the thermal conductivity of air, the distance between nodes, and the presence of obstacles determined by the cage layout within the livestock shed. For harmful gas concentration, its propagation between adjacent nodes is defined as following the law of gas diffusion, and its coupling strength is set based on the gas diffusion coefficient, the pressure gradient between nodes, and the direction of ventilation airflow. For light intensity, its propagation between adjacent nodes is defined as following the law of light attenuation, and its coupling strength is calculated based on the rectilinear propagation characteristics of light, the presence or absence of obstructions between nodes, and the characteristics of reflective surfaces. For air humidity, its propagation is defined as coupling with air temperature and gas flow state, and its coupling strength is determined through a temperature-humidity correlation model and the airflow velocity between adjacent nodes.

[0026] Optionally, the environmental parameter values ​​of all nodes at the current moment are loaded as the initial state, and an iterative evolution simulation is performed in the virtual field network. During the evolution, the fluctuations of each environmental parameter are recorded to determine how they trigger changes in other environmental parameters. After the virtual field network is initialized, the current environmental parameter values ​​of all nodes are loaded as the initial state. A discrete iterative evolution simulation based on time steps is initiated. Within each time step, according to the defined propagation rules and coupling strength, the fluctuation increment of each node's environmental parameter values ​​due to changes in the parameter values ​​of its neighboring nodes is calculated. The calculation process can be expressed in the following form: in: Indicates the first Virtual nodes at each evolutionary step size The fluctuation increment of a specific type of environmental parameter This represents the mapping function determined by the propagation rules of this type of environment parameter. Represents virtual nodes The set of all adjacent nodes, Indicates from node To the node Regarding the coupling strength weights for this specific type of environmental parameter, and These represent the nodes of the previous evolution step. and nodes The value of the environmental parameter is set. During the calculation, for the initial fluctuation of each original environmental parameter type, the changes in all related environmental parameter types triggered by it at each node in the network are recorded, and the source parameters of the changes are traced. The influence relationships between parameters recorded in each iteration are accumulated until the simulation reaches the preset evolution time or the network state changes tend to stabilize, thus completing a single evolution process.

[0027] Understandably, after the evolutionary simulation, the sum of disturbances caused by each environmental parameter in the virtual field network is statistically analyzed and normalized. The sum of the absolute values ​​of the changes in all other environmental parameter types caused by the initial values ​​of each environmental parameter type throughout the entire evolutionary process is then calculated to obtain the original disturbance sum for each environmental parameter type. This original disturbance sum for all environmental parameter types is normalized by dividing each original disturbance sum by the maximum value of the original disturbance sums for all environmental parameter types, yielding the normalized disturbance influence for each environmental parameter type, with a value ranging from 0 to 1. A visual environmental disturbance map is generated using the environmental parameter type as the x-axis and the corresponding normalized disturbance influence as the y-axis, in the form of a bar chart or line graph. Each bar or data point represents a type of environmental parameter and its quantitative indicator of its impact on the overall environment of the livestock shed. For example, the normalized disturbance influence data obtained after one simulation are: air temperature 0.95, air humidity 0.72, ammonia concentration 0.61, and light intensity 0.30. These data are used to generate the environmental disturbance map.

[0028] In one embodiment of the present invention, see [reference] Figure 3Based on the actual three-dimensional physical structure of the livestock shed, a virtual field network is established in digital space using an equally spaced grid, where each grid intersection represents a virtual location node. A mapping relationship is established between the physical locations of sensors and the virtual location nodes, associating the three-dimensional coordinates of the sensor deployments with the corresponding virtual nodes using a coordinate matching algorithm. At each sampling time, the location information of each sensor and its corresponding measured environmental parameter values ​​are extracted from the environmental parameter data frames collected by the sensor array. The extracted measured values ​​are assigned to the corresponding location nodes with mapping relationships in the virtual field network. For virtual location nodes not directly covered by sensors, their environmental parameter values ​​are marked as pending inference.

[0029] In the virtual field network, the propagation rules and coupling strengths of environmental parameters between adjacent nodes are defined. The propagation rules include heat conduction, gas diffusion, and light attenuation laws. For air temperature, its propagation between adjacent nodes is defined as following the heat conduction law, and its coupling strength is calculated based on the thermal conductivity of air, the distance between nodes, and the presence of obstacles determined by the layout of the livestock shed. For harmful gas concentration, its propagation between adjacent nodes is defined as following the gas diffusion law, and its coupling strength is set based on the gas diffusion coefficient, the pressure gradient between nodes, and the direction of ventilation airflow. For light intensity, its propagation between adjacent nodes is defined as following the light attenuation law, and its coupling strength is calculated based on the linear propagation characteristics of light, the presence or absence of obstructions between nodes, and the characteristics of reflective surfaces. For air humidity, its propagation is defined as coupling with air temperature and gas flow state, and its coupling strength is determined through a temperature-humidity correlation model and the airflow velocity of adjacent nodes.

[0030] In the specific implementation, a closed gestation pig house measuring 30 meters long, 12 meters wide, and 3.5 meters high was selected as the implementation scenario. Based on the actual three-dimensional physical structure of the pig house, a virtual field network was established in the digital space by dividing the space into grids at 0.5-meter intervals in the length, width, and height directions. After division, 61 nodes were obtained in the length direction, 25 nodes in the width direction, and 8 nodes in the height direction, with a total of 61 × 25 × 8 nodes in the virtual field network. Each grid intersection represents a virtual location node. A mapping relationship between the physical locations of sensors and virtual location nodes was established. The physical three-dimensional coordinates of the 48 sensors deployed in the pig house were associated with the nearest virtual node in the virtual field network using a nearest neighbor matching algorithm. For example, a temperature sensor located at physical coordinates (8.7, 5.2, 1.1) was matched to the virtual node coordinates (8.5, 5.0, 1.0). At each sampling moment, the location information of each sensor and its corresponding measured environmental parameter values ​​are extracted from the environmental parameter data frames collected and packaged by the sensor array. The extracted measured values ​​of air temperature, air humidity, carbon dioxide concentration and light intensity are assigned to the corresponding location nodes with mapping relationships in the virtual field network. For virtual location nodes that are not directly covered by sensors, all environmental parameter values ​​of the virtual location node are marked as a state to be inferred. For example, the virtual node with coordinates (10.0, 6.0, 2.0) is marked as a state to be inferred because there are no sensors within 1 meter around it.

[0031] In some embodiments, the process of defining the propagation rules and coupling strength of environmental parameters between adjacent nodes in a virtual field network involves the digital expression of various physical laws. For the air temperature parameter, its propagation between adjacent nodes is defined to follow the law of heat conduction. Its coupling strength is calculated based on the thermal conductivity of air, the distance between nodes, and the presence of obstacles. The existence of obstacles is determined by Boolean judgment based on the layout of the enclosures, feeding passages, and drainage ditches within the livestock shed. For the harmful gas concentration parameter, taking carbon dioxide as an example, its propagation between adjacent nodes is defined to follow the law of gas diffusion. Its coupling strength is set based on the diffusion coefficient of carbon dioxide, the pressure gradient between nodes, and the dominant airflow direction determined by the ventilation system design. For the light intensity parameter, its propagation between adjacent nodes is defined to follow the law of light attenuation. Its coupling strength is calculated based on the rectilinear propagation characteristics of light, the presence or absence of obstructions between nodes, and the reflective surface characteristics of walls and the ground. For the air humidity parameter, its propagation is defined as coupled with air temperature and gas flow state. The coupling strength is determined by the temperature-humidity correlation model and the airflow velocity of adjacent nodes. The airflow velocity is estimated from the initial flow field or inlet wind speed in the computational fluid dynamics simulation. Optionally, the coupling strength of air temperature between adjacent nodes can be quantified using a specific formula that considers the basic principles of heat conduction. in: Indicates from virtual node To virtual node The coupling strength coefficient for air temperature parameters has the physical meaning of "thermal conductivity" in relation to heat conduction. It is the thermal conductivity of the air inside the breeding shed under specific temperature and humidity conditions; it is a physical property constant. It is a virtual node With virtual nodes The assumed equivalent heat transfer area is usually related to the mesh size. It is a virtual node With virtual nodes The three-dimensional spatial distance between them It is an obstacle attenuation factor between 0 and 1, which is used when there is a solid obstacle completely blocking the view between two nodes. Indicates complete blockage, with no obstacles. When there are porous or partially obstructed areas, the intermediate value is taken. This formula ensures that the propagation intensity of the air temperature parameter in the virtual field network matches the medium properties, spatial distance, and physical structure.

[0032] It is understandable that the coupling strength of the light intensity parameter is set differently from that of temperature. The coupling strength of light intensity between adjacent nodes is calculated based on the rectilinear propagation characteristics of light, the presence or absence of obstructions between nodes, and the characteristics of reflective surfaces. The calculation considers not only whether the direct light path from the light source to the node is blocked, but also the indirect light influence caused by the relative position and reflective surfaces between adjacent nodes. For example, a virtual node near a white wall, even if not directly on the light path, may still receive some light intensity input from reflected light from the wall. This is reflected in the coupling strength calculation through a factor related to the incident angle and reflectivity. For the air humidity parameter, its coupling strength is determined through a temperature-humidity correlation model and the airflow velocity between adjacent nodes. The transmission of humidity between adjacent nodes depends not only on the humidity gradient itself, but also strongly on the local air temperature (affecting saturated vapor pressure) and airflow velocity. Therefore, the coupling strength of the humidity parameter is a function of temperature difference, humidity difference, and wind speed, rather than a fixed coefficient.

[0033] In one embodiment of the present invention, after the virtual field network is initialized, the current environmental parameter values ​​of all nodes are loaded as the initial state. A discrete iterative evolution simulation based on time steps is initiated. Within each time step, according to the defined propagation rules and coupling strength, the fluctuation increment of each node's environmental parameter values ​​due to changes in the parameter values ​​of its neighboring nodes is calculated. During the calculation, for each initial fluctuation of the original environmental parameter type, the changes in all related environmental parameter types triggered by it at each node in the network are recorded, and the source parameters of the changes are traced. The parameter influence relationships recorded in each iteration are accumulated until the simulation reaches a preset evolution time or the network state changes tend to stabilize, thereby completing a single evolution process.

[0034] After the evolutionary simulation is completed, the sum of the absolute values ​​of the changes in all other environmental parameter types caused by the initial values ​​of each environmental parameter type throughout the entire evolution process is summarized to obtain the original perturbation sum for that environmental parameter type. The original perturbation sums for all environmental parameter types are then normalized by dividing each original perturbation sum by the maximum value of the original perturbation sums for all environmental parameter types, yielding the normalized perturbation influence for each environmental parameter type, with a value ranging from 0 to 1. A visual environmental perturbation graph is generated in the form of a bar chart or line graph, with the environmental parameter type as the x-axis and the corresponding normalized perturbation influence as the y-axis. Each bar or data point represents a type of environmental parameter and its quantitative indicator of the degree of influence on the overall environment of the livestock shed.

[0035] In practical implementation, an environmental control system for egg-laying hen farming begins executing a coupled-field analysis process. After the virtual field network completes initialization based on sensor data, the coupled-field analysis module loads the current environmental parameter values ​​of all virtual location nodes as the initial state for the evolutionary simulation. These environmental parameter values ​​include measured values ​​of air temperature, air humidity, ammonia concentration, and light intensity directly obtained from sensors, as well as the state markers for nodes without sensors to be inferred. A discrete iterative evolutionary simulation based on a time step is initiated, with the time step set to 0.1 seconds and the total simulation evolution time preset to 60 seconds. Therefore, 600 iterative calculations with time steps will be performed in the virtual field network. Within each time step, according to predefined propagation rules and coupling strength, the fluctuation increment of each environmental parameter value of each virtual location node due to the changes in the parameter values ​​of its neighboring virtual location nodes is calculated. The calculation process traverses all nodes in the virtual field network and the four parameter types: air temperature, air humidity, ammonia concentration, and light intensity. During the calculation, for each initial fluctuation of the original environmental parameter type, the system records the changes in all related environmental parameter types triggered by the initial fluctuations at each node in the network, and traces the source parameter of the changes. For example, at a certain time step, if the air temperature parameter at the virtual location node at coordinates (5,5,1.5) increases by 0.2 degrees Celsius due to the influence of adjacent nodes, the system records this 0.2-degree Celsius change and analyzes whether it is directly caused by the change in air temperature at adjacent nodes or indirectly caused by the change in air humidity at adjacent nodes through a coupling relationship, while also recording the source parameter type. The influence relationships between parameters recorded in each iteration are accumulated until the simulation reaches the preset evolution time of 60 seconds, thus completing a single evolution process.

[0036] In some embodiments, the data structure recording the influence relationships between parameters during the evolutionary simulation can be embodied as a real-time updated change recording table. This table records the changes in various target environmental parameters caused by changes in each source environmental parameter at each time step and each virtual location node. Referring to Table 1, a portion of the data recorded at a specific time step and a specific node in a simplified simulation is shown: Table 1: Record of parameter changes during the iterative evolution of the virtual field network At each time step of the simulation, the system iterates through all virtual location nodes and all possible source-target parameter pairs, updating records similar to the table above. The "Cumulative Change (Absolute Value)" column records the absolute value of the cumulative impact of a specific source parameter on a specific target parameter at a specific node from the start of the simulation to the current time step. This accumulation process continues until the end of the simulation.

[0037] After the evolution simulation ends, the sum of perturbations caused by each environmental parameter in the virtual field network is calculated and normalized. The sum of the absolute values ​​of the changes in all other environmental parameter types across all virtual location nodes caused by the initial values ​​of each environmental parameter type during the entire 60-second evolution process is obtained to obtain the original perturbation sums for the four environmental parameter types: air temperature, air humidity, ammonia concentration, and light intensity. The calculation of the original perturbation sum can be expressed in the following form: in: Indicates the type of environment parameter The sum of the original perturbations, This represents the total number of evolution time steps (600 in this case). This represents the total number of virtual location nodes in the virtual field network. Indicates except Other environmental parameter types besides Indicates the time step Virtual location nodes Above, determined by environmental parameter type Types of environmental parameters resulting from initial changes The change in this value is obtained by summing the absolute values ​​of this change over all time steps, all nodes, and all other parameter types. The sum of the original disturbances for all environmental parameter types is normalized. Each sum of the original disturbances is divided by the maximum sum of the original disturbances for all environmental parameter types to obtain the normalized disturbance influence for each environmental parameter type, with a value ranging from 0 to 1.

[0038] Optionally, assume that the sum of the original disturbances calculated after one simulation run are as follows: air temperature air humidity ammonia concentration Light intensity The maximum sum of the original disturbances for all environmental parameter types is 2850 (air temperature). Therefore, the normalized disturbance effect of air temperature is... The normalized perturbation effect of air humidity is The normalized perturbation effect of ammonia concentration is The normalized perturbation effect of light intensity is Using environmental parameter types as the x-axis and the corresponding normalized disturbance influence as the y-axis, the system generates a visual environmental disturbance map in the form of a bar chart. The bar representing air temperature has a height of 1.0, the bar representing air humidity has a height of approximately 0.705, the bar representing ammonia concentration has a height of approximately 0.333, and the bar representing light intensity has a height of approximately 0.105. Each bar represents a type of environmental parameter and a quantitative indicator of its impact on the overall environment of the livestock shed.

[0039] It's understandable that generating an environmental disturbance map in the form of a curve is also an implementation method. This involves arranging four parameters—air temperature, air humidity, ammonia concentration, and light intensity—in a specific order on the horizontal axis, plotting their corresponding normalized disturbance influence values ​​on the vertical axis, and then connecting these four data points with a line graph to form a curve showing the comparison of the influence of each parameter. The environmental disturbance map clearly and visually presents the relative influence ranking of different environmental parameters under the current state of the livestock shed under complex parameter coupling, providing direct quantitative input for subsequent decision optimization.

[0040] In one embodiment of the present invention, in the environmental disturbance map, the disturbance influence values ​​corresponding to each environmental parameter are sorted in descending order. The environmental parameter ranked first is marked as a key influence parameter, and the difference between the measured value of the key influence parameter and the preset environmental target range is calculated as the current core control deviation. The environmental disturbance map is analyzed to determine one or more subordinate influence parameters whose coupling strength with the key influence parameter exceeds a preset threshold.

[0041] A core optimization instruction is generated for the key influencing parameter. The goal of this core optimization instruction is to adjust its value to within the environmental target range. The instruction includes the expected adjustment amount for the corresponding actuator, calculated based on the current core control deviation. Follow-up optimization instructions are generated for subordinate influencing parameters that are significantly coupled with the key influencing parameter. The adjustment direction and magnitude of these follow-up optimization instructions are determined by the coupling relationship between the subordinate and key influencing parameters, and are used to offset or compensate for the adverse linkage effects on the subordinate influencing parameters caused by the execution of the core optimization instruction. The core optimization instruction and one or more follow-up optimization instructions are combined in priority order to generate the final issued optimization instruction.

[0042] In practical implementation, the decision optimization module receives an environmental disturbance map generated by the coupled field analysis module. The map displays the normalized disturbance influence values ​​of four environmental parameters, presented in a bar chart format. The horizontal axis represents air temperature, air humidity, ammonia concentration, and light intensity, respectively, with corresponding vertical axis values ​​of 0.95, 0.72, 0.61, and 0.30. The decision optimization module sorts the disturbance influence values ​​of each environmental parameter in descending order within the environmental disturbance map, resulting in the following order: air temperature (0.95), air humidity (0.72), ammonia concentration (0.61), and light intensity (0.30). The environmental parameter ranked first is marked as the key influencing parameter. Air temperature is marked as the key influencing parameter at the current moment. The measured value of the air temperature parameter, obtained through the sensor array, is 28.5 degrees Celsius. The preset target range for air temperature is 22 to 26 degrees Celsius. The difference between the measured value of the key influencing parameter and the preset target range is calculated as the current core control deviation. 28.5 degrees Celsius exceeds the upper limit of the target range of 26 degrees Celsius, so the current core control deviation is +2.5 degrees Celsius. The decision optimization module analyzes the environmental disturbance diagram and queries the parameter coupling relationship matrix from the background data of the environmental coupled field model. It identifies one or more subordinate influencing parameters whose coupling strength with the key influencing parameter air temperature exceeds a preset threshold (e.g., 0.4). The query results show that the coupling strength between air humidity and air temperature is 0.68, and the coupling strength between ammonia concentration and air temperature is 0.52. Both exceed the threshold; therefore, air humidity and ammonia concentration are identified as the current key subordinate influencing parameters.

[0043] In some embodiments, optimization instructions containing the priority of key influencing parameters are generated in combination with a preset environmental target range. The decision optimization module generates core optimization instructions for the key influencing parameters. The goal of the core optimization instructions is to adjust the air temperature value to the environmental target range of 22 degrees Celsius to 26 degrees Celsius. The instructions contain the expected adjustment amount for the corresponding actuator calculated based on the current core control deviation. For example, if the current core control deviation is +2.5 degrees Celsius, by querying the temperature control model, it is calculated that the variable frequency fan speed needs to be increased to 80% of the rated speed and run continuously for 15 minutes. This "fan speed 80%, run for 15 minutes" is the expected adjustment amount and is written into the core optimization instructions. To generate follow-up optimization instructions for subordinate influencing parameters that are significantly coupled with key influencing parameters, the adjustment direction and magnitude of the follow-up optimization instructions are determined by the coupling relationship between the subordinate influencing parameters and the key influencing parameters. This is used to offset or compensate for the adverse linkage effects on subordinate influencing parameters when the core optimization instruction is executed. For example, the execution of the core optimization instruction (increase fan speed to reduce temperature) is expected to accelerate airflow and reduce air humidity, which may conflict with the target range of air humidity (50%-70%). Therefore, a follow-up optimization instruction needs to be generated for the humidifier to "synchronously start and run at low speed for 10 minutes during the fan speed increase" to compensate for the humidity loss. At the same time, the increased fan speed and ventilation also help reduce ammonia concentration, which is consistent with the control target of ammonia concentration. Therefore, it may not be necessary to generate an additional compensation instruction for ammonia concentration, or only a monitoring instruction to "monitor ammonia concentration and adjust if it is below the lower limit" may be generated. The core optimization command is combined with one or more follow-up optimization commands in priority order to generate the final optimization command. Optimization commands typically exist in the form of structured data frames or command lists, clearly indicating the command priority order, target device, control action, and parameters. Optionally, the structured optimization commands generated by the decision optimization module can be internally represented and transmitted in a concise tabular form, as shown in Table 2: Table 2: Structured Optimization Instruction Table Generated by the Decision Optimization Module It is understandable that the generation logic of core optimization instructions and follower optimization instructions relies on a quantitative analysis of the coupling relationship between environmental parameters. A mathematical expression for calculating the adjustment magnitude of follower optimization instructions can be as follows: in: This indicates the adjustment range of the follow-up optimization instruction for a specific subordinate influencing parameter (such as air humidity). This range may correspond to the humidifier's spray volume or operating duration. It is a proportionality coefficient used to map theoretical calculations to control parameters for specific equipment. It is the absolute value (2.5°C) of the current core control deviation of the key influencing parameter (air temperature). It is an adjustable deviation threshold used to determine whether strong follow-up adjustment needs to be activated. This is the coupling strength coefficient between the key influencing parameters and the current subordinate influencing parameters, obtained from the environmental coupled field model. This formula establishes a quantitative correlation between the magnitude of the follow-up optimization instructions and the regulation requirements of the key influencing parameters and their impact on the subordinate parameters. Finally, the decision optimization module will combine the core optimization instructions, follow-up optimization instructions, etc., into a complete set of optimization instructions according to the priority order shown in the table, and send it to the control execution module.

[0044] In one embodiment of the present invention, the system further includes a control execution module and a feedback adjustment module. The control execution module sends the optimization command to the execution device corresponding to the environmental parameters, driving the execution device to adjust the environmental parameters. During the operation of the execution device, the feedback adjustment module synchronously acquires the adjusted environmental parameters through a sensor array, compares the adjusted environmental parameters with a preset environmental target range, and if the requirements are met, maintains the current operating state of the execution device; otherwise, triggers a new round of optimization command calculation.

[0045] The optimization instructions are sent to the corresponding environmental parameter actuators, driving them to adjust the environmental parameters. The specific process includes: parsing the optimization instructions and identifying the control targets and adjustment quantities for different environmental parameters; converting the control targets and adjustment quantities into control signals recognizable by the corresponding actuators, including commands for fan speed and damper opening of ventilation equipment, power settings of heating or cooling equipment, and illumination cycle and intensity of lighting equipment; and sending these control signals to the corresponding actuator controllers via an Internet of Things (IoT) communication network, which then drive the ventilation equipment, temperature control equipment, and lighting equipment to adjust according to the instructions.

[0046] In practical implementation, an intelligent environmental control system for beef cattle farming enters the instruction execution and feedback stage. The optimization instructions generated by the decision optimization module are sent to the control execution module. The optimization instructions exist in the format of structured data frames, which clearly contain multiple sub-instructions and their priorities for air temperature, air humidity, and harmful gas concentration. For example, an optimization instruction set may include: Instruction 1 (priority 1) aims to reduce air temperature, requiring the speed of fan 1 to be increased to 70% and the opening of damper 2 to 60 degrees; Instruction 2 (priority 2) aims to maintain air humidity, requiring the atomizing humidifier to be turned on and the spray cycle to be set to 2 minutes of operation and 3 minutes of interval; Instruction 3 (priority 3) aims to assist in reducing ammonia concentration, requiring the fan on side wall 3 to be turned on in low-speed ventilation mode. The workflow of the control execution module begins with parsing the received optimization instructions. The control execution module parses the optimization instructions and identifies the control objectives and adjustment amounts for different environmental parameters in the instructions. For example, from instruction No. 1, the control objective is to reduce the air temperature, and the adjustment amount is to increase the speed of fan No. 1 to 70% and the opening of damper No. 2 to 60 degrees. From instruction No. 2, the control objective is to maintain the air humidity, and the adjustment amount is for the atomizing humidifier to operate in a cycle of 2 minutes on and 3 minutes off.

[0047] In some embodiments, the control execution module converts the identified control target and adjustment amount into control signals recognizable by the corresponding execution device. These control signals include commands for the fan speed and damper opening of ventilation equipment, power level commands for heating or cooling equipment, and light cycle and intensity commands for lighting equipment. For example, for the adjustment amount of "increasing the speed of fan 1 to 70%", the control execution module converts it into an analog control signal sent to the frequency converter of fan 1, with the signal value being the voltage or current value corresponding to 70% speed. For the adjustment amount of "opening damper 2 to 60 degrees", it converts it into a pulse width modulation (PWM) signal sent to the servo motor of damper 2, with its duty cycle corresponding to the 60-degree opening position. For the periodic operation command of the atomizing humidifier, it converts it into a series of high and low level digital switching signals to achieve a cycle of working for 2 minutes and stopping for 3 minutes. The control execution module sends control signals to the corresponding actuator controllers via the Internet of Things (IoT) communication network. The IoT communication network can be LoRa, Zigbee, or a wired industrial bus. The actuator controllers drive the ventilation equipment, temperature control equipment, and lighting equipment to adjust according to the instructions. After receiving the analog signal, the No. 1 fan inverter adjusts the output frequency to stabilize the fan speed at 70%. The No. 2 damper servo motor rotates to the 60-degree opening position according to the PWM signal. After receiving the digital switch signal sequence, the controller of the atomizing humidifier starts to execute the preset intermittent spray program.

[0048] Optionally, during the control signal conversion process, the control execution module may need to perform feedforward calculations based on equipment characteristics and environmental models. An example of converting airflow demand into a specific fan speed control signal is as follows: in: This represents the digital setpoint corresponding to the wind turbine speed control signal to be issued. This value will be written into a specific register of the wind turbine inverter. The target volumetric air volume is calculated based on the adjustment amount in the optimization instruction, and the unit is cubic meters per hour. It is the radius of the fan outlet, in meters. It is the theoretical output wind speed per unit speed of the fan under the current static pressure. It is a coefficient related to the fan's characteristic curve, and its unit is meters per second. This indicates a rounding down operation to ensure the setpoint is an integer. This calculation process ensures that an abstract adjustment such as "increasing the fan speed" is precisely quantified into a control command that can be executed by the specific fan unit. All generated control signals are packaged into data packets conforming to IoT communication protocols and sent to the respective actuator controllers via wireless or wired networks.

[0049] It is understood that the system also includes a feedback adjustment module. During the operation of actuators such as ventilation equipment, temperature control equipment, and lighting equipment, the feedback adjustment module synchronously acquires adjusted environmental parameters through a sensor array. The sensor array continues to operate at a fixed sampling period, uploading data frames containing parameters such as air temperature, air humidity, and harmful gas concentration. The feedback adjustment module compares the adjusted environmental parameters with the preset environmental target range. If the requirements are met, the current operating state of the actuators is maintained; otherwise, a new optimization command is recalculated. For example, after executing an adjustment operation with the core optimization command of reducing air temperature for 5 minutes, the feedback adjustment module obtains a new average air temperature of 25.8 degrees Celsius, which falls within the preset environmental target range of 22 to 26 degrees Celsius. At the same time, the air humidity is 58%, which is also within the target range of 50% to 70%. The feedback adjustment module then determines that the current environment meets the requirements and sends a signal to the control execution module to maintain the current operating state of the fan, damper, and humidifier. Conversely, if the air temperature remains at 27.5 degrees Celsius after 5 minutes, failing to reach the target range, the feedback adjustment module triggers the decision optimization module to restart the analysis process and generate a new round of optimization instructions based on the latest sensor data.

[0050] In some embodiments, the comparison logic of the feedback adjustment module can employ simple threshold judgment or more complex trend prediction. For example, when determining whether the requirements are met, it may not only check whether the current instantaneous value is within the target range, but also check the moving average of the parameter values ​​over a period of time and its changing trend. If the current value is within the range but the trend shows that it is rapidly deviating from the range, the feedback adjustment module may also determine that the requirements are not met and trigger recalculation. Regardless of the judgment logic used, the output of the feedback adjustment module is binary: maintain the current state or trigger recalculation, thus forming a closed-loop control system of "perception-analysis-decision-execution-feedback", enabling the environment of the breeding house to be dynamically stabilized near the preset target range.

[0051] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An intelligent control system for livestock and poultry farming environment based on the Internet of Things, characterized in that, include: The data acquisition module acquires environmental parameters in real time through a sensor array deployed in the breeding shed. These environmental parameters include air temperature, air humidity, concentration of harmful gases, and light intensity. The coupled field analysis module inputs the environmental parameters into a pre-trained environmental coupled field model. The environmental coupled field model simulates the dynamic interaction and propagation relationship between the environmental parameters and outputs an environmental disturbance map that reflects the degree of mutual influence between the environmental parameters. The decision optimization module identifies the key impact parameters that have the greatest impact on the overall environment based on the environmental disturbance map, and generates optimization instructions that include the priority of the key impact parameters in combination with the preset environmental target range.

2. The intelligent control system for livestock and poultry farming environment based on the Internet of Things as described in claim 1, characterized in that, The method of acquiring environmental parameters in real time through a sensor array deployed in the breeding shed includes: Temperature sensors, humidity sensors, gas sensors, and light sensors are evenly distributed across multiple vertical height layers and horizontal zones in the breeding shed. Set a unified sampling clock to drive all sensors to collect environmental parameters at the same timestamp; The collected environmental parameters are preprocessed, including removing abrupt abnormal values ​​caused by instantaneous sensor failures, smoothing parameters with minor disturbances using median filtering, and binding all sensor data with timestamps and location information, packaging them into environmental parameter data frames.

3. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 2, characterized in that, The environmental parameters are input into the pre-trained environmental coupled field model, including: Construct a virtual field network with the spatial location of the breeding shed as nodes, and assign the environmental parameter values ​​collected by each sensor to the corresponding location node; In the virtual field network, the propagation rules and coupling strength of environmental parameters between adjacent nodes are defined. The propagation rules include the laws of heat conduction, gas diffusion and light attenuation. Using the current environmental parameter values ​​of all nodes as the initial state, an iterative evolution simulation is performed in the virtual field network. During the evolution process, the fluctuations of each environmental parameter are recorded to determine how they trigger changes in other environmental parameters. After the evolution is complete, the sum of disturbances caused by each environmental parameter in the virtual field network is calculated, normalized, and an environmental disturbance map is drawn with environmental parameter type as the dimension and disturbance influence as the value.

4. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 3, characterized in that, The construction of a virtual field network with the spatial location of the breeding shed as nodes, and the assignment of environmental parameter values ​​collected by each sensor to the corresponding location node, includes: Based on the actual three-dimensional physical structure of the breeding house, a virtual field network is established in the digital space in a gridded manner with equal spacing, where each grid intersection represents a virtual location node; Establish a mapping relationship between the physical location of the sensor and the virtual location node, and match the three-dimensional coordinates of the sensor deployment to the corresponding virtual node; At each sampling moment, the location information of each sensor and its corresponding measured environmental parameter values ​​are extracted from the environmental parameter data frames collected by the sensor array. The extracted measured values ​​are assigned to the corresponding location nodes with mapping relationships in the virtual field network. For virtual location nodes that are not directly covered by sensors, their environmental parameter values ​​are marked as states to be inferred.

5. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 3, characterized in that, In the virtual field network, the propagation rules and coupling strength of environmental parameters between adjacent nodes are defined. The propagation rules include the laws of heat conduction, gas diffusion, and light attenuation, including: For air temperature parameters, their propagation between adjacent nodes is defined to follow the law of heat conduction, and their coupling strength is calculated based on the thermal conductivity of air, the distance between nodes, and the presence of obstacles. For the concentration parameter of harmful gas, its propagation between adjacent nodes is defined to follow the gas diffusion law, and its coupling strength is set based on the gas diffusion coefficient, the pressure gradient between nodes and the ventilation airflow direction. For the light intensity parameter, its propagation between adjacent nodes is defined to follow the light attenuation law, and its coupling strength is calculated based on the linear propagation characteristics of light, whether there are obstructions between nodes, and the characteristics of the reflective surface. For the air humidity parameter, its propagation is defined as coupled with air temperature and gas flow state, and the coupling strength is determined by the temperature-humidity correlation model and the air flow velocity of adjacent nodes.

6. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 3, characterized in that, The process involves using the current environmental parameter values ​​of all nodes as the initial state and performing iterative evolution simulations in the virtual field network. During the evolution, the process records how fluctuations in each environmental parameter trigger changes in other environmental parameters, including: After the virtual field network is initialized, the current environmental parameter values ​​of all nodes are loaded to the initial state; Initiate a discrete iterative evolution simulation based on time steps. Within each time step, calculate the fluctuation increment of each node's environmental parameter values ​​due to the changes in the parameter values ​​of its neighboring nodes, according to the defined propagation rules and coupling strength. During the calculation process, for each original environmental parameter type, the initial fluctuation is recorded, and the changes in all related environmental parameter types caused by it at each node in the network are recorded, and the source parameters of the changes are traced. The influence relationships between parameters recorded in each iteration are accumulated until the simulation reaches the preset evolution time or the network state changes tend to stabilize, thus completing a single evolution process.

7. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 3, characterized in that, After the evolution is complete, the sum of disturbances caused by each environmental parameter in the virtual field network is calculated, normalized, and an environmental disturbance map is drawn with environmental parameter type as the dimension and disturbance influence as the numerical value, including: After the evolution simulation is completed, the sum of the absolute values ​​of the changes in all other environmental parameter types caused by the initial values ​​of each environmental parameter type during the entire evolution process is obtained to obtain the original perturbation sum of the environmental parameter types. The sum of the original disturbances for all environmental parameter types is normalized. Each original disturbance sum is divided by the maximum value of the sum of the original disturbances for all environmental parameter types to obtain the normalized disturbance influence for each environmental parameter type, which ranges from 0 to 1. Using the type of environmental parameter as the horizontal axis and the corresponding normalized disturbance influence as the vertical axis, a visual environmental disturbance map in the form of a bar chart or line graph is generated. Each bar or data point represents a type of environmental parameter and a quantitative indicator of its impact on the overall environment of the breeding house.

8. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 3, characterized in that, Based on the environmental disturbance map, the key impact parameters with the greatest influence on the overall environment are identified, including: In the environmental disturbance map, the disturbance influence values ​​corresponding to each environmental parameter are sorted in descending order; The environmental parameter ranked first is marked as the key influencing parameter, and the difference between the measured value of the key influencing parameter and the preset environmental target range is calculated as the current core control deviation. Analyze the environmental disturbance map to identify one or more dependent influence parameters whose coupling strength with the key influence parameter exceeds a preset threshold.

9. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 8, characterized in that, The process of generating optimization instructions that include the priority of key influencing parameters by combining a preset environmental target range includes: Generate core optimization instructions for the key influencing parameters. The goal of these core optimization instructions is to adjust their values ​​to within the target environmental range. The instructions include the expected adjustment amount for the corresponding actuator, calculated based on the current core control deviation. To generate follow-up optimization instructions for subordinate influencing parameters that are significantly coupled with key influencing parameters, the adjustment direction and magnitude of the follow-up optimization instructions are determined by the coupling relationship between the subordinate influencing parameters and the key influencing parameters, in order to offset or compensate for the adverse linkage effects on the subordinate influencing parameters when the core optimization instructions are executed; The core optimization instruction is combined with one or more follow-up optimization instructions in order of priority to generate the final optimization instruction.

10. The intelligent control system for livestock and poultry farming environment based on the Internet of Things according to claim 1, characterized in that, The system also includes: The control execution module sends the optimization instructions to the execution device corresponding to the environmental parameters, driving the execution device to adjust the environmental parameters; The feedback adjustment module synchronously acquires the adjusted environmental parameters through the sensor array during the operation of the actuator, compares the adjusted environmental parameters with the preset environmental target range, and if the requirements are met, maintains the current operating state of the actuator; otherwise, it recalculates new optimization instructions. The optimization command is sent to the execution device corresponding to the environmental parameters, driving the execution device to adjust the environmental parameters, specifically including: Analyze and optimize instructions to identify the control objectives and adjustment values ​​for different environmental parameters within the instructions; The control target and adjustment amount are converted into control signals that can be recognized by the corresponding actuator. The control signals include commands for the fan speed and damper opening of the ventilation equipment, commands for the power level of the heating or cooling equipment, and commands for the illumination period and intensity of the lighting equipment. The control signal is sent to the corresponding actuator controller via the Internet of Things communication network, and the actuator controller drives the ventilation equipment, temperature control equipment, and lighting equipment to adjust according to the instructions.

Images