Intelligent heating control method and system for livestock farm

Abstract

The invention relates to the technical field of intelligent heating control, in particular to an intelligent heating control method and system for a livestock farm. The method comprises the following steps that behavior activity state monitoring is conducted on livestock in a livestock farm through electronic monitoring equipment, and temperature and humidity fluctuation and greenhouse gas fluctuation data are obtained in real time in combination with a temperature and humidity sensor and a greenhouse gas sensor; performing posture difference analysis on the monitoring video to generate livestock posture difference data; and further performing space heat energy vector loss deconstruction through temperature and humidity and gas data to obtain heat energy loss data, and calculating livestock posture temperature loss data in combination with posture difference data. On the basis of the data, the heating hot steam circulation demand quantity is matched, and a heating self-adaptive adjustment model is constructed. The intelligent heating control technology is optimized, so that the intelligent heating control technology is more perfect.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent heating control, and particularly to an intelligent heating control method and system for livestock farms. Background Art

[0002] In the past, the heating methods in livestock farms often relied on manual adjustment, making it difficult to accurately control environmental factors such as temperature, humidity, and gas concentration, and unable to respond in a timely manner to climate changes and real-time changes in the health status of livestock. The intelligent heating system can, by introducing advanced electronic monitoring, environmental sensing, and data analysis technologies, monitor multiple key indicators such as the behavioral activities of livestock, environmental temperature and humidity, and greenhouse gas concentration in real time, and perform precise analysis and dynamic adjustment in combination with artificial intelligence algorithms. By monitoring and analyzing the behavior of livestock and combining the spatial distribution of environmental heat energy loss, the system can intelligently match the heating demand. However, traditional intelligent heating control methods have low accuracy in analyzing the spatial heat energy loss in livestock farms and the degree of body temperature loss between livestock of different body postures, resulting in large heating adjustment errors. Summary of the Invention

[0003] Based on this, it is necessary to provide an intelligent heating control method and system for livestock farms to solve at least one of the above technical problems.

[0004] To achieve the above object, an intelligent heating control method for livestock farms, the method includes the following steps: Step S1: Monitor the behavioral activity status of livestock in the livestock farm through an electronic monitoring device to obtain a monitoring video of the behavioral activity status of livestock; monitor the temperature and humidity fluctuations and greenhouse gas fluctuations in the livestock farm in extremely cold weather through the temperature and humidity sensors and greenhouse gas sensors in the environmental sensing module, and respectively obtain temperature and humidity fluctuation data and greenhouse gas fluctuation data; Step S2: Perform body posture difference analysis on the monitoring video of the behavioral activity status of livestock to generate livestock body posture difference data; perform spatial heat energy vector loss deconstruction on the temperature and humidity fluctuation data and greenhouse gas fluctuation data to obtain spatial heat energy vector loss deconstruction data; based on the spatial heat energy vector loss deconstruction data, perform body temperature loss carrying measurement integration between livestock of different body postures on the livestock body posture difference data to generate livestock body temperature loss measurement data; Step S3: Based on the livestock body temperature loss learning data, match the heating steam circulation demand between livestock of different body postures to obtain the heating steam circulation demand; construct a heating adjustment model according to the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to perform intelligent heating control of the livestock farm.

[0005] Preferably, step S2 includes the following steps: Step S21: Conduct body posture difference analysis on the monitoring video of livestock behavior activities to generate livestock body posture difference data; Step S22: Analyze the group movement frequency of the monitoring video of livestock behavior activities to obtain livestock group movement frequency data; Step S23: Obtain the structural design data of the livestock farm; Based on the livestock group movement frequency data and the structural design data of the livestock farm, conduct gas disturbance dispersion analysis on the greenhouse gas fluctuation data to obtain greenhouse gas disturbance dispersion structure data; Step S24: Conduct spatial thermal energy vector loss deconstruction on the temperature and humidity fluctuation data and the greenhouse gas disturbance dispersion structure data according to the structural design data of the livestock farm to obtain spatial thermal energy vector loss deconstruction data; Step S25: Based on the spatial thermal energy vector loss deconstruction data, conduct body posture hypothermia carrying measurement integration among livestock with different body postures on the livestock body posture difference data to generate livestock body posture hypothermia measurement data.

[0006] Preferably, step S23 includes the following steps: Step S231: Obtain the structural design data of the livestock farm; Conduct gas flow design path analysis on the structural design data of the livestock farm to obtain the gas flow design path; Step S232: Conduct group gait acceleration frequency analysis on the livestock group movement frequency data to obtain the group gait acceleration frequency; Step S233: Based on the group gait acceleration frequency, conduct continuous pressure difference calculation of the surrounding air flow push to obtain the continuous pressure difference of the surrounding air flow push; Step S234: According to the continuous pressure difference of the surrounding air flow push, conduct deduction of the change in the flow velocity difference in the azimuth of the gas flow cross-section on the gas flow design path to obtain the change data of the flow velocity difference in the azimuth of the flow cross-section; Step S235: Conduct mass / volume regional distribution difference analysis on the greenhouse gas fluctuation data to obtain the mass / volume regional distribution difference of the greenhouse gas; Step S236: According to the change data of the flow velocity difference in the azimuth of the flow cross-section, conduct gas disturbance dispersion analysis on the mass / volume regional distribution difference of the greenhouse gas to obtain greenhouse gas disturbance dispersion structure data.

[0007] Preferably, step S24 includes the following steps: Step S241: Conduct material heat preservation performance loss analysis on the structural design data of the livestock farm to obtain material heat preservation performance loss data; Step S242: Conduct temperature and humidity spatial fluctuation distribution mapping on the temperature and humidity fluctuation data according to the structural design data of the livestock farm to obtain temperature and humidity spatial fluctuation distribution mapping data; Step S243: Based on the data of the reduction in the heat preservation performance of the material, calculate the divergence of the spatial heat energy loss per unit time for the mapped data of the spatial fluctuations in temperature and humidity and the data of the greenhouse gas perturbation dispersion structure, to obtain the spatial heat energy loss divergence data; Step S244: Based on the spatial heat energy loss divergence data, perform a decomposition of the spatial heat energy vector loss to obtain the spatial heat energy vector loss decomposition data.

[0008] Preferably, step S243 includes the following steps: Calculate the increment of the structural pore thermal conductivity for the data of the reduction in the heat preservation performance of the material to obtain the structural pore thermal conductivity increment coefficient; Based on the structural pore thermal conductivity increment coefficient, perform an identification of the geometric progression of the acceleration of the interfacial heat transfer for the data of the reduction in the heat preservation performance of the material to obtain the geometric progression data of the acceleration of the interfacial heat transfer; Calculate the difference in the relative change rates of low temperature / low humidity for the mapped data of the spatial fluctuations in temperature and humidity to generate the difference in the relative change rates of low temperature / low humidity in the spatial distribution; According to the difference in the relative change rates of low temperature / low humidity in the spatial distribution, perform a numerical increment integration of the spatial heat loss density per unit time for the data of the greenhouse gas perturbation dispersion structure to obtain the numerical increment of the spatial heat loss density per unit time; Based on the geometric progression data of the acceleration of the interfacial heat transfer and the numerical increment of the spatial heat loss density per unit time, calculate the divergence of the spatial heat energy loss per unit time to obtain the spatial heat energy loss divergence data.

[0009] Preferably, step S25 includes the following steps: Step S251: Analyze the extreme values of the low temperature tolerance among livestock with different body postures for the livestock body posture difference data to obtain the extreme values of the low temperature tolerance of livestock among different body postures; Step S252: Perform a Fourier heat spectrum transformation on the spatial heat energy vector loss decomposition data to obtain a heat energy loss frequency domain distribution map; Step S253: Based on the heat energy loss frequency domain distribution map and the extreme values of the low temperature tolerance of livestock among different body postures, perform a time-varying non-linear dynamic evolution of the body surface heat energy loss among livestock with different body postures for the livestock body posture difference data to obtain the non-linear time-varying data of the body surface heat energy loss; Step S254: Perform a convergent constraint power series expansion on the non-linear time-varying data of the body surface heat energy loss to obtain a convergent power series of the body surface heat energy loss; Step S255: According to the convergent power series of the body surface heat energy loss, perform a carrying measurement integration of the body temperature loss among livestock with different body postures for the livestock body posture difference data to generate livestock body temperature loss measurement data.

[0010] Preferably, step S254 includes the following steps: Calculate the instantaneous growth index of heat energy loss for the non-linear time-varying data of body surface heat energy loss to obtain the instantaneous growth index of heat energy loss; Conduct relative entropy difference analysis of local mutations of heat energy loss based on the instantaneous growth index of heat energy loss to obtain the relative entropy difference of local mutations of loss; Conduct convergence analysis of mutation points on the non-linear time-varying data of body surface heat energy loss based on the relative entropy difference of local mutations of loss to obtain the convergence data of loss mutation points; Conduct convergent constraint power series expansion on the non-linear time-varying data of body surface heat energy loss according to the convergence data of loss mutation points to obtain the convergent power series of body surface heat energy loss.

[0011] Preferably, step S3 includes the following steps: Step S31: Normalize the deconstructed data of spatial heat energy vector loss to obtain the normalized data of spatial heat energy vector loss; Step S32: Conduct logical learning processing on the livestock body temperature loss measurement data to obtain the livestock body temperature loss learning data; Step S33: Match the spatial heat energy replenishment rate according to the normalized data of spatial heat energy vector loss to obtain the spatial heat energy replenishment rate data; Step S34: Match the heating steam circulation demand among livestock of different body postures based on the livestock body temperature loss learning data to obtain the heating steam circulation demand; Step S35: Construct a heating adjustment model according to the spatial heat energy replenishment rate data and the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to perform intelligent heating control in the livestock farm.

[0012] Preferably, the present invention also provides an intelligent heating control system for a livestock farm, which is used to execute the intelligent heating control method for a livestock farm as described above. The intelligent heating control system for a livestock farm includes: A data monitoring module, which is used to monitor the behavior activity status of livestock in the livestock farm through an electronic monitoring device to obtain a monitoring video of the behavior activity status of livestock; monitor the temperature and humidity fluctuations and greenhouse gas fluctuations in the livestock farm under extremely cold weather through the temperature and humidity sensor and greenhouse gas sensor in the environmental sensing module, and obtain temperature and humidity fluctuation data and greenhouse gas fluctuation data respectively; A body temperature loss measurement module, which is used to conduct body posture difference analysis on the monitoring video of the behavior activity status of livestock to generate livestock body posture difference data; conduct deconstruction of spatial heat energy vector loss on the temperature and humidity fluctuation data and greenhouse gas fluctuation data to obtain deconstructed data of spatial heat energy vector loss; conduct body temperature loss carrying measurement integration among livestock of different body postures on the livestock body posture difference data based on the deconstructed data of spatial heat energy vector loss to generate livestock body temperature loss measurement data; The heating adjustment model construction module is used to match the heating steam circulation demand among livestock of different body postures based on the livestock body posture hypothermia learning data to obtain the heating steam circulation demand; construct a heating adjustment model according to the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to execute the intelligent heating control of the livestock farm.

[0013] The beneficial effects of the present invention are as follows: By using an electronic monitoring device to monitor the behavior activity status of livestock in the livestock farm, a video of the behavior activity status of livestock is obtained. At the same time, the environmental changes under extremely cold weather are monitored through a temperature and humidity sensor and a greenhouse gas sensor. This process can obtain the status information of livestock activities and the fluctuation data of environmental temperature, humidity, and gas concentration in real time, providing accurate data support for subsequent analysis. Timely understanding of livestock behavior and environmental changes helps to discover potential health problems and the impact of temperature fluctuations, providing basic data for intelligent control. By analyzing the body posture differences in the video of the behavior activity status of livestock, livestock body posture difference data is generated to help understand the health status and activity changes of livestock with different body postures in a specific environment. At the same time, the spatial thermal energy vector loss is deconstructed for the temperature and humidity fluctuation data and the greenhouse gas fluctuation data to quantify the thermal energy loss distribution. This process enables the system to accurately understand the impact of environmental factors on livestock body postures, further calculate the hypothermia conditions of livestock with different body postures, and generate body posture hypothermia measurement data, providing a quantitative basis for the subsequent adjustment of heating requirements. Based on the livestock body posture hypothermia learning data, the heating steam circulation demand among livestock of different body postures is matched to determine the heat required for each type of livestock. This process ensures that the heating system can provide accurate steam supply according to the specific needs of livestock with different body postures, while avoiding energy waste. By constructing a heating adjustment model according to the heating steam circulation demand, the intelligent adaptive adjustment of the heating system is realized. The adjustment model can automatically adjust the heating intensity according to different environmental and body posture conditions, thereby improving the heating efficiency and ensuring the healthy growth of livestock. Therefore, the present invention is an optimization of the traditional intelligent heating control method for a livestock farm, solving the problem that the traditional intelligent heating control method for a livestock farm has low accuracy in analyzing the spatial thermal energy loss of the livestock farm and the body posture hypothermia degree among livestock with different body postures, resulting in large heating adjustment errors, improving the accuracy of analyzing the spatial thermal energy loss of the livestock farm and the body posture hypothermia degree among livestock with different body postures, and reducing the heating adjustment error. Description of the Drawings

[0014] Figure 1 It is a schematic diagram of the step flow of an intelligent heating control method for a livestock farm; Figure 2 is Figure 1 The detailed implementation step flow diagram of step S2 in Figure 3 For Figure 1 the detailed implementation step flow schematic diagram of step S3 in Specific implementation manner

[0015] Please refer to Figures 1 to 3 , an intelligent heating control method for a livestock farm, the method comprising the following steps: Step S1: Monitor the behavioral activity status of livestock in the livestock farm through an electronic monitoring device to obtain a monitoring video of the behavioral activity status of livestock; monitor the temperature and humidity fluctuations and greenhouse gas fluctuations in the livestock farm under extremely cold weather through the temperature and humidity sensors and greenhouse gas sensors in the environmental sensing module, respectively obtaining temperature and humidity fluctuation data and greenhouse gas fluctuation data; Step S2: Analyze the body posture differences in the monitoring video of the behavioral activity status of livestock to generate livestock body posture difference data; perform spatial thermal energy vector loss deconstruction on the temperature and humidity fluctuation data and greenhouse gas fluctuation data to obtain spatial thermal energy vector loss deconstruction data; based on the spatial thermal energy vector loss deconstruction data, perform body posture hypothermia carrying measurement integration among livestock with different body postures on the livestock body posture difference data to generate livestock body posture hypothermia measurement data; Step S3: Based on the livestock body posture hypothermia learning data, match the heating steam circulation demand among livestock with different body postures to obtain the heating steam circulation demand; construct a heating adjustment model according to the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to execute intelligent heating control for the livestock farm.

[0016] In the embodiment of the present invention, referring to Figure 1 as described above, it is the step flow schematic diagram of an intelligent heating control method for a livestock farm of the present invention. In this example, the intelligent heating control method for the livestock farm comprises the following steps: Step S1: Monitor the behavioral activity status of livestock in the livestock farm through an electronic monitoring device to obtain a monitoring video of the behavioral activity status of livestock; monitor the temperature and humidity fluctuations and greenhouse gas fluctuations in the livestock farm under extremely cold weather through the temperature and humidity sensors and greenhouse gas sensors in the environmental sensing module, respectively obtaining temperature and humidity fluctuation data and greenhouse gas fluctuation data; In the embodiments of the present invention, under extremely cold weather conditions, to achieve intelligent heating control of livestock farms, the implementation of step S1 is carried out first. In this step, all-weather high-definition infrared cameras with a resolution of not less than 1920×1080 are installed in the central area and the four corner areas of the ceiling of the livestock breeding house to form a full-angle coverage and collect monitoring videos of the behavior and activity status of livestock. The cameras have infrared night vision and thermal imaging functions and can continuously collect data for 24 hours. The video frame rate is set to 25 frames per second to ensure the continuity of dynamic behavior collection. While the video is being collected, environmental parameters are collected through environmental sensing modules arranged at different positions in the breeding house. The temperature and humidity sensors have an accuracy of ±0.1°C and ±1.5%RH, and the sampling period is set to once every 10 seconds; the greenhouse gas sensor uses the sensor model GSS ExplorIR-W, with an accuracy of ±30ppm + 3% reading, and the sampling period is set to once every 10 seconds. The collection positions are arranged at the entrance, the middle part, the intensive livestock activity area, and near the exhaust passage to ensure the representativeness and comprehensiveness of the environmental data. All the collected data are synchronously stored in the local edge data concentration system with time stamps and are accessed to the data processing module through a wired network.

[0017] Step S2: Conduct body posture difference analysis on the monitoring videos of the behavior and activity status of livestock to generate livestock body posture difference data; conduct spatial thermal energy vector loss deconstruction on the temperature and humidity fluctuation data and the greenhouse gas fluctuation data to obtain spatial thermal energy vector loss deconstruction data; based on the spatial thermal energy vector loss deconstruction data, conduct body temperature loss carrying measurement integration among livestock with different body postures on the livestock body posture difference data to generate livestock body temperature loss measurement data; In the embodiment of the present invention, first, the collected monitoring video of the livestock behavior activity state is subjected to structured segmentation of image sequence frames. The external contour of the livestock individual is extracted through edge recognition and contour extraction techniques, and the spatial position of the nodes of the livestock's limbs, trunk, and head is marked by using the human pose recognition method based on the OpenPose algorithm. Subsequently, the Euclidean distance and angle change comparison method is used to quantify the difference in the spatial position of the nodes at different time points, so as to generate livestock body posture difference data. Furthermore, the temperature and humidity fluctuation data and greenhouse gas fluctuation data obtained in step S1 are aligned in time series and spatially mapped in units of minutes. According to the physical structure parameters of the livestock house (including length, width, height, ventilation opening position and size, etc.), a three-dimensional space grid is divided, and the side length of the division unit is 0.5 meters. By using the principle of energy conservation and the principle of heat diffusion, combined with air density, specific heat capacity, and wind speed data, the direction and magnitude of heat energy transfer in each grid unit are analyzed. A spatial heat energy vector transfer matrix is constructed by the finite difference method, and then the directionality and quantity of heat energy loss per unit time are decomposed to obtain spatial heat energy vector loss deconstruction data. Finally, the exposure heat flux of livestock with different body postures in each heat loss area is integrated by using the surface area change rate extracted from the livestock body posture difference data and the activity degree of epidermal movement per unit time, and the trapezoidal integration method is used to sum it to obtain the body posture hypothermia carrying measurement integration between livestock with different body postures, and livestock body posture hypothermia measurement data is generated.

[0018] Step S3: Based on the livestock body posture hypothermia learning data, match the heating steam circulation demand between livestock with different body postures to obtain the heating steam circulation demand; construct a heating adjustment model according to the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to execute intelligent heating control of the livestock farm.

[0019] In the embodiment of the present invention, first, the livestock body temperature loss measurement data obtained in step S2 is normalized to unify the heat flux loss values of different livestock to the range of 0 to 1, eliminate the order-of-magnitude offset caused by body size differences, and use it as an input variable to match the corresponding livestock activity area to estimate the distributed heat energy demand density. The kernel density estimation method is used to fit the heat energy demand change density surface of each area, and then it is compared and analyzed with the spatial heat energy vector loss deconstruction data obtained in step S2 area by area. The regional heating supplement rate matching calculation is carried out according to the heat energy difference to obtain the steam heat energy supply value required for each spatial unit. The parameters of the hot steam circulation heating system include a maximum hourly output heat energy of 120 kW, a maximum steam pipeline pressure of 1.6 MPa. The main control branch pipelines are distributed according to the heat demand priority, and the opening threshold of each pipeline is set according to the unit heat loss density. Through the multi-objective constraint programming algorithm, the heating adjustment model is constructed into a control logic diagram, which includes parameters such as heating start time, steam flow rate, and heater start power, and outputs the heating adaptive adjustment model. This model is encapsulated in the form of the OPC-UA protocol and uploaded to the cloud platform scheduling system through a wired local area network connection to execute remote intelligent control and achieve the differential heating adjustment target for different areas of the livestock farm.

[0020] Step S2 includes the following steps: Step S21: Analyze the body posture differences of the livestock behavior activity state monitoring video to generate livestock body posture difference data; Step S22: Analyze the group movement frequency of the livestock behavior activity state monitoring video to obtain livestock group movement frequency data; Step S23: Obtain the structural design data of the livestock farm; based on the livestock group movement frequency data and the structural design data of the livestock farm, analyze the gas disturbance dispersion of the greenhouse gas fluctuation data to obtain the greenhouse gas disturbance dispersion structure data; Step S24: Deconstruct the spatial heat energy vector loss according to the structural design data of the livestock farm for the temperature and humidity fluctuation data and the greenhouse gas disturbance dispersion structure data to obtain the spatial heat energy vector loss deconstruction data; Step S25: Based on the spatial heat energy vector loss deconstruction data, perform the body temperature loss carrying measurement integration between livestock with different body postures on the livestock body posture difference data to generate livestock body temperature loss measurement data.

[0021] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes: Step S21: Analyze the body posture differences of the livestock behavior activity state monitoring video to generate livestock body posture difference data; In the embodiment of the present invention, in step S21, an infrared high-definition camera with a resolution of 1920×1080 and a frame rate of 30 frames per second is used to continuously collect video of the activities of livestock in a livestock farm, and the collected video images enter the image processing process through an edge server. The YOLOv5 object detection algorithm is used to identify and number livestock individuals, and the bounding box positions of each livestock individual in each frame of image are obtained. By analyzing the time series changes of the bounding box positions in consecutive frames and combining the OpenPose skeleton recognition technology, the position information of the key body joint points of the livestock is extracted. After the feature points are extracted, based on a sliding window with a time window of 2 seconds, the relative position changes of each body joint are analyzed, the body posture similarity is compared through cosine similarity measurement, and the K-means clustering algorithm is used to cluster and label the body movement categories. The posture angle differences and skeleton deformation degrees between different body posture categories are quantified into a feature vector group, and finally, the livestock body posture difference data is constructed. The structure of the body posture difference data is in the form of a two-dimensional matrix, with each row corresponding to a livestock individual and each column being the numerical index of a certain type of posture difference vector. The total dimension is the number of livestock individuals × the number of posture feature dimensions. For example, in actual tests, for 30 beef cattle, 16 key point coordinates are extracted for each cattle to form a feature vector, and finally a 30×16 body posture difference data matrix is formed, which serves as the basic parameter for subsequent precise heating control.

[0022] Step S22: Analyze the group movement frequency of the livestock behavior activity monitoring video to obtain the livestock group movement frequency data; In the embodiment of the present invention, in step S22, based on the livestock individual numbers and the bounding box trajectory sequences obtained in step S21, the optical flow method is used to track the movement trajectories of each livestock in consecutive frames. The optical flow method uses the dense optical flow Lucas-Kanade algorithm to track the pixel movement amount of the center point of the livestock in the image coordinate system, and the total pixel displacement of the movement path of each livestock individual is statistically calculated within each second, and then converted into the actual spatial movement distance (converted according to the camera installation parameters and the site calibration data, and the conversion ratio is 1 pixel corresponding to 2.5 cm). Within each one-minute statistical period, the average movement distances of all livestock are summarized, and fast Fourier transform analysis is performed on them to extract the main components of different frequency components, and the movement frequency distribution curve of the livestock group is constructed based on the main frequency fluctuation range (0.2~1.5Hz). In a specific experiment, for a group of 40 dairy cows, during the high-activity period from 08:00 to 10:00 every day, the monitoring data shows that the main movement frequency is about 0.9Hz, indicating that the group is in a high-frequency walking state. The livestock group movement frequency data finally appears as a movement frequency vector with a timestamp as the main index, and each record includes the time point, the group average frequency value, the standard deviation, and the maximum fluctuation amplitude, constituting the livestock group movement frequency data for subsequent gas disturbance dispersion analysis. Step S23: Obtain the structural design data of the livestock farm; based on the livestock group movement frequency data and the structural design data of the livestock farm, perform gas perturbation dispersion analysis on the greenhouse gas fluctuation data to obtain greenhouse gas perturbation dispersion structure data; In the embodiment of the present invention, first, the structural design data of the livestock farm is retrieved, including the three-dimensional spatial dimensions of the breeding house (length 60 meters, width 15 meters, height 5 meters), the distribution map of doors, windows and ventilation openings, the ventilation equipment parameters (air supply volume 3000 cubic meters per hour, wind speed 0.5 meters per second), and the thermal conductivity coefficients of the ceiling and floor materials (0.23 and 1.14 W / m·K respectively). And the initial boundary conditions of the gas perturbation path are determined by combining the channel distribution and the position of the feed feeding area. The livestock group movement frequency data obtained in step S22 is mapped into the spatial coordinate system, and each livestock position is assigned a perturbation intensity factor according to its movement frequency. Using CFD (Computational Fluid Dynamics) simulation technology, combined with the RNG k-ε turbulence model, a flow field perturbation model of air and greenhouse gases in the breeding house is established. Regarding the livestock individual position perturbation as the perturbation source node, the perturbation energy takes the area within a radius of 1 meter around the cattle body as the perturbation radius, and the initial value of the velocity vector perturbation (the reference perturbation velocity is 0.2 times the livestock movement velocity) is introduced within this range. The perturbation is superimposed on the basis of the flow field caused by the original ventilation. During the simulation period (the simulation time is set to 120 seconds, and the time step is 0.05 seconds), for 、 and other greenhouse gases, multi-time period hierarchical simulation analysis is carried out on the diffusion path under perturbation conditions, the gas concentration change surface under perturbation is extracted, and the gradient boundary extraction of the concentration gradient change in different regions is carried out to form greenhouse gas perturbation dispersion structure data. This data is stored in the form of a three-dimensional grid structure. Each grid node contains coordinate position, perturbation velocity vector, greenhouse gas concentration change value and volatility index, which is used to accurately depict the impact of air flow caused by livestock movement on the gas distribution structure and guide the subsequent linkage control of ventilation and heating.

[0023] Step S24: According to the structural design data of the livestock farm, perform spatial thermal energy vector loss deconstruction on the temperature and humidity fluctuation data and the greenhouse gas perturbation dispersion structure data to obtain spatial thermal energy vector loss deconstruction data; In the embodiment of the present invention, in step S24, first, the structural design data of the livestock farm is called, including the geometric parameters of the internal space of the livestock house (length 60 meters, width 15 meters, height 5 meters), the heat transfer coefficients of the wall and the roof (0.45 W / m·K and 0.32 W / m·K respectively), the ground heat reflectivity of 0.12, the layout of the ventilation ducts, the air velocity at the air inlets of 0.5 m / s, and the distribution coordinates of heat sources in each area (such as the heat exchanger of the drinking fountain and the feed heating equipment). Based on this, a three-dimensional space heat conduction model of the internal space of the livestock house is constructed, and the greenhouse gas perturbation dispersion structure data generated in step S23 is introduced as an interference term to participate in the analysis of the heat diffusion path. The change rate of the greenhouse gas concentration gradient of each grid unit in the perturbation structure is matched with the wind speed perturbation vector. Combining the principles of hot air convection and diffusion, the heat transfer direction and heat loss value in each area of the space are deconstructed. For the temperature and humidity fluctuation data, a spatial temperature distribution map is established according to each sampling time point. Taking every 1 cubic meter as a unit grid, the current temperature value and the environmental humidity of this area are calibrated. According to the law of heat convection, the influence of humidity on the heat exchange efficiency is calculated, and the influence of wind speed perturbation on the heat transfer path is considered additionally. In the specific operation, the temperature and humidity fluctuation data and the gas perturbation structure data are unified and aligned through the spatial grid nodes to form the heat input, output, and loss states in each unit space grid point. Subsequently, the heat flow path analysis is carried out for all grid points, and a heat flux density direction map is established using the finite volume method. Through the boundary conditions, the vector direction of heat flowing from the high-temperature area to the low-temperature area in a certain area is deduced. Finally, the deconstructed data of the spatial heat energy vector loss is output, which is presented as a heat energy flow direction vector field in a three-dimensional coordinate system. Each vector represents the net loss direction and quantity of heat energy per unit time in a specific space unit, with the unit of watt per cubic meter, and is used to guide the subsequent analysis of body temperature loss.

[0024] Step S25: Based on the deconstructed data of the spatial heat energy vector loss, perform the body temperature loss carrying measurement integration between livestock with different body postures for the livestock body posture difference data, and generate the livestock body temperature loss measurement data.

[0025] In the embodiment of the present invention, in step S25, the spatial thermal energy vector loss deconstruction data obtained in step S24 is called, and combined with the livestock body shape difference data obtained in step S21. Taking the position of the livestock individual in space as the reference point, the heat loss value of the thermal energy vector node in the 1 cubic meter where it is located is extracted. Since the key joint displacement amplitude and posture action dimension have been calibrated in the livestock body shape difference data, the heat dissipation efficiency can be affected according to the posture action frequency and dimension. The specific method is to extract the posture activity index (such as the total average displacement per second of the head, torso, and limb joint points) from the body shape difference vector of each livestock individual as the body surface area movement factor per unit time. Then, combined with the direction of the thermal energy loss vector in the space where the individual is located, integral processing of the unit heat flow and the body shape heat-receiving area is performed to estimate the hypothermia intensity suffered by the individual. In an actual experiment, for a Simmental cattle weighing 580 kg, the average displacement of the key points on the head and shoulders was measured to be 3.2 cm / s within 2 minutes of monitoring time, the corresponding posture activity factor was 3.2, the unit volume heat loss value in the area where it was located was 12 W / m³, and based on the average exposed area of the body part being 2.6 m², the total heat loss per unit time was deduced to be 78.4 W. This calculation process is carried out for each individual in the livestock group, and the hypothermia values of all individuals are indexed and output by the unique livestock number to form livestock body shape hypothermia measurement data. The data format is a one-dimensional structured array, and each element contains the livestock number, the thermal loss vector in the space where it is located, the posture factor, and the heat loss value per unit time, which are used as important parameters in the heating control link to participate in the subsequent determination of heat compensation requirements.

[0026] Step S23 includes the following steps: Step S231: Obtain the structural design data of the livestock farm; analyze the gas circulation design path for the structural design data of the livestock farm to obtain the gas circulation design path; Step S232: Perform group gait acceleration frequency analysis on the livestock group movement frequency data to obtain the group gait acceleration frequency; Step S233: Based on the group gait acceleration frequency, perform a calculation of the continuous pressure difference for the surrounding air flow to obtain the continuous pressure difference for the surrounding air flow to push; Step S234: According to the continuous pressure difference for the surrounding air flow to push, perform a deduction of the change in the flow velocity difference in the cross-sectional orientation of the gas circulation design path to obtain the change data of the flow velocity difference in the cross-sectional orientation; Step S235: Analyze the mass / volume regional distribution difference of the greenhouse gas fluctuation data to obtain the mass / volume regional distribution difference of the greenhouse gas; Step S236: According to the change data of the flow velocity difference in the cross-sectional orientation, perform an analysis of the gas disturbance dispersion of the mass / volume regional distribution difference of the greenhouse gas to obtain the greenhouse gas disturbance dispersion structure data.

[0027] In the embodiment of the present invention, in step S231, first, the construction drawings of the livestock farm and the building information modeling data (BIM data) are retrieved. The content includes the overall structural dimensions of the livestock house (the total length is 65 meters, the width is 18 meters, and the eaves height is 5.5 meters), the division of internal functional areas, the distribution of ventilation system components, the specific positions and area distributions of the air inlet and outlet (including 3 air inlets on the west side, each with an area of 0.5 square meters, and 4 air outlets on the east side, each with an area of 0.4 square meters), and the pipeline routing and cross-sectional dimensions (the main air duct has a diameter of 60 cm, and the secondary air duct has a diameter of 30 cm). During the parsing process, a combination of two-dimensional plane layout diagrams and three-dimensional solid structure diagrams is used to extract the boundary nodes of each structure and perform spatial mapping according to the specified coordinate system in the drawings. Based on the spatial finite element mesh method, the entire livestock house is divided into equilateral cubic unit bodies (the unit side length is 1 meter), and the mutual connection relationship between each unit is calibrated according to the gas flow continuity condition and the pipeline distribution path, so as to determine the main gas flow path from the air inlet to the air outlet. Further, the obstacle areas such as walls and brackets are shielded to shield the unit nodes that do not participate in the air flow. By establishing a directed graph structure, path labeling and topological sorting are performed on all gas flow units, and finally, the complete gas flow design path data starting from the air inlet is formed. The data format includes the flow unit number, spatial coordinates, passing state, upstream unit, and downstream unit index, providing a spatial basis for subsequent airflow disturbance modeling.

[0028] In step S232, the livestock group movement frequency data obtained in the foregoing step S22 is called. Taking each livestock as a movement subject, the moving distance between consecutive frames in the monitoring video is subjected to time differentiation to obtain an individual gait speed time series. Based on the speed time series, a first differentiation is performed to extract the acceleration time series, and the maximum value normalization method is used to standardize the acceleration sequence of each individual. Taking one minute as the time window, the maximum acceleration and the acceleration fluctuation frequency per unit time of each livestock are statistically counted, and the main acceleration frequency components are extracted by FFT transformation. In order to eliminate data interference caused by occasional jumps or unnatural behaviors, a median filter is introduced for sequence smoothing processing, and then the acceleration frequencies of each livestock sample are linearly weighted and averaged, and the weight is set to the reciprocal of the residence time of the individual in the central area (that is, the higher the activity, the greater the weight). Finally, the group gait acceleration frequency during the current monitoring period is calculated. In actual tests, 40 beef cattle are subjected to 15 minutes of data collection. The average acceleration frequency of individual gaits is distributed between 0.15 and 0.45 Hz, and the average value of the group acceleration frequency is finally calculated to be 0.31 Hz. This value is used as the input index of the disturbance energy intensity in the air flow disturbance simulation. The mass conservation equation and the momentum conservation equation in continuum mechanics are used to construct a surrounding air flow driving model. Based on the gas flow path obtained in step S231 as the skeleton basis, local disturbance sources caused by livestock movement are introduced into each spatial unit. The intensity of the disturbance source is assigned according to the group gait acceleration frequency in step S232. It is assumed that each livestock is equivalent to a periodic vibration source, and local gas disturbances are formed in the unit volume where it is located. The disturbance frequency is set to 0.31 Hz, and the corresponding disturbance velocity amplitude is 0.07 m per second. In a continuous air medium, this disturbance forms periodic density fluctuations, which further lead to the spatial transmission of pressure fluctuations. For each unit spatial grid point, the finite difference method is used to solve its air pressure change, and then the air pressure gradient field is derived. According to the air pressure gradient value and direction, the acceleration direction and magnitude of the gas per unit volume under this pressure difference are calculated, and then this value is converted into an air flow driving pressure difference represented by a spatial vector field. The specific calculation area is set as the central area of the livestock house (40 m × 10 m × 3 m), and the spatial dissection granularity is a cube with a side length of 0.5 m. Finally, a continuous pressure difference data set including coordinate position, disturbance frequency, instantaneous pressure change value per unit volume, and air flow acceleration direction is formed in each grid, which is used as one of the input parameters for greenhouse gas dispersion modeling. This data reflects the propagation direction and intensity of the spatial air flow disturbance energy caused by the dynamic behavior of the livestock group, providing a dynamic basis for subsequent livestock house environment regulation.

[0029] Based on the continuously varying pressure difference data of the surrounding air flow obtained in step S233, with each spatial cross-section (defined as a unit area cross-section with a size of 1 square meter) in the gas flow design path as the basic analysis unit, a velocity gradient model is established in the adjacent gas flow units upstream and downstream of this cross-section. Based on the pressure gradient and the continuity equation in aerodynamics, the local flow velocity at the center point of the cross-section is deduced. Taking the central passage of the livestock house as an example, there are 9 standard cross-sections along the main gas flow path direction in this area, and the distance between the upstream and downstream is 2 meters. The difference in air pressure values before and after each cross-section is estimated in the form of differences, and combined with the air density of 0.001225 grams per cubic centimeter, the change value of the average flow velocity driven by this pressure difference is calculated. If the pressure difference before and after cross-section A is 15 Pascals, then according to the flow velocity-pressure difference relationship, the average flow velocity generated is 1.4 meters per second, showing an obvious change compared with the flow velocity of 0.8 meters per second at the upstream cross-section. To further extract the directional change characteristics, the velocity vectors of each cross-section are projected according to the air flow design direction, and the component difference of the cross-section air flow velocity in the ventilation design path direction is obtained, and the increment or attenuation trend along the path advancement direction is recorded, thus forming the change data of the cross-section azimuth flow velocity difference. This data set is stored in matrix form, and each item records the cross-section number, coordinates, magnitude of the flow velocity change, included angle of the flow velocity direction, and the velocity difference from the adjacent cross-section. For the greenhouse gas concentration data such as , obtained through the greenhouse gas sensing module, spatial area difference processing is carried out. First, the entire interior of the livestock house is regionally meshed according to the aforementioned spatial division unit, and the monitoring data is mapped according to the spatial unit where each sensor point is located. On this basis, the gas mass value per unit volume of each region is calculated. According to the gas density value (such as the carbon dioxide density is 1.977 kilograms per cubic meter and the methane density is 0.656 kilograms per cubic meter), the actual gas mass is obtained by multiplying the sensor sampling concentration by the unit volume, and then the total gas mass of all monitored regions is statistically calculated. Taking the 10-meter × 10-meter central activity area as an example, there are 16 sensing nodes inside, and the measured concentration values range from 410 ppm to 580 ppm, and the total mass of in this area is approximately 3.25 kilograms after conversion, corresponding to a volume area of 300 cubic meters. The gas mass data in all regions is identified by region number, and then the difference in gas mass between adjacent spatial units is calculated to extract the gradient change trend of gas mass in space. According to the spatial distribution trend diagram, it can be observed that in the northern area of the livestock house The mass distribution is relatively more concentrated, while the mass gradient near the south exhaust port area is significantly reduced, thus forming a mass distribution difference map between regions. All data are uniformly archived as a greenhouse gas mass / volume regional distribution difference data set, and the fields include unit number, center coordinates, gas type, mass value, volume size, and mass difference with adjacent units. The flow section azimuth velocity difference change data obtained in step S234 is used as the driving force basis of gas disturbance, and is fused with the greenhouse gas mass / volume regional distribution difference data in step S235. The processing method adopts a spatial coupling algorithm: first, the spatial section is used as a node to establish a directed channel structure in the direction of gas diffusion, and the velocity change vector of each section is mapped to the adjacent upstream and downstream spatial units, and the disturbance propagation path is determined in combination with its corresponding velocity increment direction. Then, a disturbance transfer coefficient is introduced in each unit body. The coefficient is determined by the section azimuth velocity difference, reflecting the possibility of gas diffusion from a high-mass density area to a low-mass density area. Taking the northwest corner of the livestock house as an example, this is a structurally closed area, and the velocity difference of the ventilation section is negative, which means that the airflow velocity is decreasing, and this area The mass distribution is 1.75 kg, and the adjacent unit is 1.05 kg, with a significant mass gradient. According to the disturbance intensity and flow velocity direction, the disturbance energy is distributed to the three adjacent units downstream, and the disturbance dispersion probability weights are assigned to 0.6, 0.3 and 0.1 respectively. Without introducing the probability calculation framework, the discrete weights are used to describe the disturbance intensity transfer trend. Through the above method, a gas disturbance dispersion structure data map is established in the entire livestock house space. The data fields include unit space number, disturbance direction, disturbance amplitude, target transmission path, disturbance energy distribution structure, and the mass value mapping difference between the source unit and the target unit. The greenhouse gas disturbance dispersion structure data finally obtained is used for subsequent heating thermal energy optimization layout and local air quality regulation modeling input.

[0030] Step S24 includes the following steps: Step S241: performing material thermal insulation performance impairment analysis on the livestock farm structural design data to obtain material thermal insulation performance impairment data; Step S242: mapping the temperature and humidity spatial fluctuation distribution of the temperature and humidity fluctuation data according to the livestock farm structural design data to obtain temperature and humidity spatial fluctuation distribution mapping data; Step S243: based on the thermal insulation performance loss data of the materials, the spatial heat energy loss divergence per unit time is calculated for the temperature and humidity spatial fluctuation distribution mapping data and the greenhouse gas disturbance dispersion structure data to obtain the spatial heat energy loss divergence data; Step S244: performing spatial thermal energy vector loss deconstruction based on the spatial thermal energy loss divergence data to obtain spatial thermal energy vector loss deconstruction data.

[0031] In the embodiments of the present invention, in step S241, an analysis of the reduction in thermal insulation performance is carried out for the information on the materials of the wall, roof, floor, and door and window components involved in the structural design data of the livestock farm. First, according to the design drawings and the engineering material statistics table, the types and parameter data of the building materials are extracted, including the material types (such as polyurethane composite panels, double-layer insulating glass, concrete structures), thickness (in millimeters), thermal conductivity (in watts per meter per kelvin), and structural location. For each type of structural unit, the reduction factor of its thermal insulation performance is quantified according to its exposure environment during the service life. Taking the polyurethane sandwich panel on the roof as an example, the designed thickness is 75 millimeters, and the initial thermal conductivity is 0.023 watts per meter per kelvin. Combining the annual average temperature and humidity in the local environment of 7.5 degrees Celsius and an annual average relative humidity of 82%, the annual increment of the thermal conductivity is calculated using the empirical attenuation function method, and the current thermal conductivity is obtained as 0.031 watts per meter per kelvin. Correspondingly, the thermal insulation ability is processed by taking the difference from the original ability, and the reduction value of the thermal insulation performance of this structural unit is recorded. Similar analysis operations are performed one by one on the 240-millimeter-thick concrete structure of the south wall of the breeding shed, the double-layer insulating glass of the east plastic steel window, and the composite thermal insulation layer on the floor (including 100-millimeter foam board), and a corresponding table of structural partitions and thermal insulation reduction is generated. All data are archived by spatial numbering into a dataset of the reduction in the thermal insulation performance of materials, and the fields include the structural unit number, material type, thickness, initial thermal conductivity, service life, environmental exposure parameters, current thermal conductivity, percentage reduction in thermal insulation ability, etc., which are used for subsequent structured modeling of heat loss. Based on the structural design data of the livestock farm and the time-series data collected by the temperature and humidity sensing nodes arranged at different spatial positions, a spatial distribution mapping process is carried out on the temperature and humidity fluctuation data. First, the internal space of the farm is divided into regular spatial grid units according to the structure diagram, and the size of each unit is set to 2 meters × 2 meters × 2 meters, corresponding to the discrete volume area in three-dimensional space. In each volume unit, the data of the sensor sampling points covered or adjacent to it are collected, and the sampling frequency is once every 10 seconds. The data include temperature (in degrees Celsius) and relative humidity (in percentage). Taking a certain moment (such as 4 am every day) as the unified reference moment, the instantaneous readings of all sensing nodes in the whole field are extracted, and the temperature and humidity data are filled into all spatial units according to the three-dimensional coordinates of the sensors and the positions of the spatial units through nearest neighbor interpolation mapping. For example, if there are two sensors in the north corner area of the farm, with temperatures of 6.8 degrees Celsius and 7.1 degrees Celsius respectively, the grid unit in this area is assigned a value of 6.95 degrees Celsius through the weighted average method. After completing the whole-field mapping, statistical analysis is carried out on all spatial units to identify the interval with the largest temperature and humidity fluctuations. By repeating the above operations for the spatial mapping within 24 hours of a day, 24 daily temperature and humidity distribution maps are generated, and the amplitude of temperature and humidity fluctuations within the same spatial unit is calculated. The temperature fluctuation value is defined as the maximum temperature value minus the minimum temperature value, and the humidity fluctuation is defined as the maximum humidity value minus the minimum humidity value.Finally, the spatio-temporal distribution data of temperature and humidity in all cells are integrated into a temperature and humidity spatial fluctuation distribution mapping dataset, with fields including spatial unit number, coordinate center, hourly temperature value, humidity value, temperature fluctuation amplitude, humidity fluctuation amplitude, etc., providing data support for subsequent thermal energy vector loss analysis.

[0032] In step S243, first, using the obtained data on the reduction of the thermal insulation performance of the material as the basis for the thermal conduction influence factor, by correlating and matching this data with the mapped data of the spatial fluctuation distribution of temperature and humidity, the direction of heat conduction and the starting point of heat loss are established. In specific operations, the three-dimensional structure of the farm is divided into volume units of 1 m × 1 m × 1 m according to spatial grid cells, and each cell correlates the structural material information with the temperature gradient change value in that area. Assume that the wall in the northwest corner area is a 240-mm concrete structure with a thermal insulation reduction rate of 23%. The temperature in this area rises from 6.2 °C in the morning to 13.7 °C at noon, while the adjacent inner area is 14.3 °C. According to the heat conduction direction, it is from the inside to the outside. Combining the heat conduction loss factor, temperature gradient and exposed area, the heat loss amount per unit time is deduced. The spatial unit thermal energy loss divergence is defined as the heat transferred by each grid cell to the external environment per unit time, with the unit of joules per cubic meter per second. At the same time, integrate the air flow disturbance vectors in the greenhouse gas disturbance dispersion structure data, compare the disturbance frequency of the gas in the ventilation channel with the pressure distribution trend chart, and superimpose the heat exchange frequency in the disturbance area to complete the assessment of the auxiliary heat exchange loss caused by gas disturbance. Based on the above analysis, with each minute as the time resolution unit, the thermal energy loss divergence values of each spatial unit are deduced one by one to obtain a complete spatial thermal energy loss divergence data set. The fields of this data set include spatial number, boundary structure thermal conductivity, thermal insulation reduction rate, temperature gradient, thermal energy dissipation rate, heat flow direction vector, disturbance gain coefficient, etc., which are used as the input data for subsequent thermal energy vector deconstruction. In step S244, based on the spatial thermal energy loss divergence data obtained in step S243, further perform the operation of deconstructing the thermal energy vector in the three-dimensional space. First, in each spatial volume unit, according to its thermal energy loss divergence value and the corresponding conduction direction, a thermal energy vector field is established. Each thermal energy vector is defined as having a direction from the inside to the outside, and its magnitude is determined by the heat loss amount per unit volume per unit time. For example, for a volume unit in the middle with a relatively high temperature, the set thermal energy loss divergence is 12.3 joules per cubic meter per second, and the direction of the thermal energy vector points to the low-temperature area in its southeast. For the entire spatial field, construct a three-dimensional thermal conduction structure map of multi-vector superposition. Subsequently, use vector decomposition technology to decompose each thermal energy vector into components along the X, Y, and Z axes, and based on the superposition of the components, the distribution of the thermal energy transmission channels in the entire area is formed. For example, in the southeast corner area of a large cattle shed, after superimposing the thermal energy vectors, it is found that there is a tendency for heat to be concentrated and discharged to the roof. Considering that the thermal insulation reduction rate of the roof structure is 31%, it is confirmed that this channel is the main thermal energy loss path. At the same time, through the vector coupling relationship between adjacent units, the areas where thermal energy is concentrated and discharged and the areas where heat reflux is weak are deduced, and the visualization output of the thermal energy path is carried out.The finally generated deconstructed dataset of spatial thermal energy vector loss, with fields including the direction of the thermal energy vector, vector intensity, vector path number, vector superposition trend, area number, structure-associated material number, etc., provides a structural support basis for subsequent heat compensation and energy-saving control strategies of the heating system.

[0033] Step S243 includes the following steps: Calculate the increment of the structural pore thermal conductivity for the data of the reduction in material insulation performance to obtain the structural pore thermal increment coefficient; Based on the structural pore thermal increment coefficient, perform an isometric identification of the accelerated interfacial heat transfer for the data of the reduction in material insulation performance to obtain the isometric data of the accelerated interfacial heat transfer; Calculate the difference in the relative change rates of low temperature / low humidity for the mapped data of the spatial temperature and humidity fluctuations to generate the relative change rate difference of low temperature / low humidity in the spatial distribution; According to the relative change rate difference of low temperature / low humidity in the spatial distribution, perform a numerical increment integration of the spatial heat loss density per unit time for the greenhouse gas perturbation dispersion structure data to obtain the numerical increment of the spatial heat loss density per unit time; Based on the isometric data of the accelerated interfacial heat transfer and the numerical increment of the spatial heat loss density per unit time, estimate the divergence of the spatial thermal energy loss per unit time to obtain the divergence data of the spatial thermal energy loss.

[0034] In the embodiments of the present invention, the incremental calculation of the structural pore thermal conductivity is carried out for the data of the reduction in the heat preservation performance of materials. According to the structural design data of the livestock farm, the types of building materials, service life, construction methods in each area are extracted, and combined with the known physical aging law, the empirical thermal coefficient growth function is used to correct the original thermal conductivity incrementally in the time dimension. Taking the outer wall as an example, the expanded perlite cement composite board is marked as the thermal insulation layer in the structural design, with an initial thermal conductivity of 0.030 W / (m·K) and a design life of 10 years. Based on the timeliness data of this type of material in the publicly available literature on building thermal performance tests, the growth relationship between the aging years and the porosity function is constructed. The porosity increases by 0.9% per year, and accordingly the thermal conductivity increases proportionally by 0.0025 W / (m·K) per 1% porosity. When the site is built and used for the 7th year, the corresponding porosity increment is 6.3%. Based on this, the thermal conductivity increment is deduced to be 0.01575 W / (m·K). Adding this increment to the original value, the structural pore thermal conductivity increment coefficient is obtained as the conversion value of the total growth rate, which is used to construct the time-varying parameter set of the material thermal conductivity, including the wall number, original thermal conductivity value, service life, theoretical pore growth rate, and thermal conductivity correction value. Based on the above structural pore thermal conductivity increment coefficient, the interface heat transfer acceleration ratio identification operation is carried out on the data of the reduction in the heat preservation performance of materials. According to the markings on the design drawings, the differences in the thermal conductivities of the structural materials at the joint nodes such as wall-ceiling and wall-floor are analyzed, and the equivalent thermal resistance conversion method is used to deduce the central tendency of the heat flux. The nodes with large differences in thermal conductivity are identified as high-risk interfaces, and mathematically, the nodes with a one-way thermal conductivity gradient greater than 0.010 W / (m·K) are set as the acceleration identification targets. Taking the northeast corner connection node as an example, the thermal conductivity of the wall is 0.045 W / (m·K), and that of the ceiling board is 0.085 W / (m·K). Then, a one-way acceleration channel is generated due to the thermal resistance difference. According to the continuous heat transfer boundary conditions, the thermal conductivity rate is quantitatively analyzed. By establishing a geometric progression mapping table, the growth of the heat loss transfer rate from the beginning to the end of the year is summarized into a geometric progression model. If the heat loss at the beginning of the year is q and that at the end of the year is 1.5q, the annual growth ratio factor is 1.5, and an interface heat transfer acceleration ratio data table is generated, including the node number, comparison of material thermal conductivities, thermal conductivity gradient, annual growth rate, and predicted ratio factor. The calculation of the relative change rate difference of low temperature / low humidity is carried out for the mapped data of the spatial fluctuation distribution of temperature and humidity. Using the established temperature and humidity distribution grid data, the entire breeding space is divided into regular grid units. Based on the spatial distribution of the lowest temperature point and the lowest humidity point within a day, the changes in temperature and humidity per unit time within each grid are respectively counted. By calculating the difference in the change gradients of adjacent time periods on the 3-day continuous data, comparing the maximum rate change and the minimum rate change, a spatial change difference matrix is established.Taking the early morning to dawn period of a certain day as an example, in the area near the north exhaust passage, the temperature change rate is the largest, reaching a decrease of 2.1 degrees Celsius per hour, and the humidity decrease rate is 3.8%; while in the central area inside the shed, the temperature change rate is only 1.3 degrees Celsius per hour, and the humidity decreases by 1.1%. The above numerical differences are respectively recorded as the low-temperature change rate difference of 0.8 degrees Celsius per hour and the low-humidity change rate difference of 2.7%. By comparing the rate differences in different regions, the relative change rate difference data of low temperature / low humidity are generated and used as the input parameters of the heat loss gradient. The formed data table structure includes the spatial grid number, daily change period, temperature and humidity change rate, difference gradient, and trend identification flag.

[0035] Based on the obtained spatial distribution of the relative change rate difference of low temperature / low humidity, perform numerical increment integration processing of the spatial heat loss density per unit time for the greenhouse gas perturbation dispersion structure data. Divide the farm space into three-dimensional structural units, with each unit having a basic grid unit of a spatial volume of 1 m × 1 m × 1 m, and label them with numbers respectively. Using the relative change rate difference of low temperature / low humidity as the index variable, set the units with a temperature change rate above 1.5 degrees Celsius per hour and a humidity decrease rate above 2% per hour as high-variation units. For each high-variation unit, extract the perturbation frequency, distribution density, and gas concentration gradient in the corresponding greenhouse gas perturbation dispersion structure data, and estimate the numerical value of the heat perturbation contribution per unit time in combination with the heat capacity constant of the perturbed gas. In this unit, the average frequency of greenhouse gas concentration perturbation is observed to be 12 times per minute, the gas is mainly a mixture of carbon dioxide and ammonia, its average specific heat capacity is about 0.9 kJ per kilogram per Kelvin, the perturbation direction is mainly in the form of upward turbulence, and the volume of the perturbation area is about 0.8 cubic meters. Substitute the product of the perturbation volume and the local temperature gradient into the calculation process, and combine the heat capacity constant and the perturbation frequency to derive an increment of 4.6 kJ per cubic meter for the heat loss density per unit volume per hour. After processing the remaining spatial units in the same area in the same way, accumulate the heat loss density increments of each spatial unit in the time dimension through the integration method to complete the operation of generating the numerical value of the spatial heat loss density increment per unit time, forming a data matrix field including unit number, low temperature / low humidity change rate, perturbation structure characteristics, heat capacity parameters, and integrated cumulative heat value. Based on the interface heat transfer acceleration ratio data and the above-mentioned numerical value of the spatial heat loss density increment per unit time, perform the operation of estimating the spatial thermal energy loss divergence per unit time. First, map the acceleration interface position, ratio coefficient, and thermal conductivity difference index in the interface heat transfer acceleration ratio data to the spatial structure to mark all heat transfer sensitive joint areas. In each joint area, establish a one-way heat transfer flow channel along the interface normal direction, generate a heat flux growth vector for different time sections according to the ratio acceleration factor, and cooperate with the heat loss density increment numerical value of adjacent units, and use the difference recurrence integration method to perform segmented heat transfer amplitude superposition on the heat transfer path. Taking the joint area numbered #P45 as an example, the interface length is 3.2 meters, corresponding to the intersection area of the wall and the ceiling, the thermal conductivity grade difference is 0.042 W per meter per Kelvin, the annual heat flow acceleration ratio factor is 1.4, calculate on an hourly basis, set the basic heat flux to 12 kJ per meter, and combine the adjacent heat loss increment density to be 4.2 kJ per cubic meter to form a heat flux amplification path across the interface. By recursively introducing this acceleration data into five consecutive spatial units along the interface normal direction, set different weight factors in each unit for increment superposition, and finally obtain the direction of thermal energy loss and the total divergence value diverging outward starting from the joint area.The spatial thermal energy loss divergence data generates a data table in the form of a coordinate grid arrangement. Each record item includes an interface number, an acceleration factor, an affected unit number, a superposition value of unit heat, a spatial divergence direction vector, and a quantitative index of total thermal energy divergence, serving as the basic input conditions for subsequent heating path optimization.

[0036] Step S25 includes the following steps: Step S251: Conduct an extreme value analysis of the low-temperature tolerance among livestock with different body postures on the livestock body posture difference data to obtain the extreme values of livestock low-temperature tolerance among livestock with different body postures; Step S252: Perform a Fourier thermal spectrum transformation on the spatial thermal energy vector loss deconstruction data to obtain a thermal energy loss frequency domain distribution map; Step S253: Based on the thermal energy loss frequency domain distribution map and the extreme values of livestock low-temperature tolerance among livestock with different body postures, perform a time-varying non-linear dynamic evolution of the body surface thermal energy loss among livestock with different body postures on the livestock body posture difference data to obtain non-linear time-varying data of body surface thermal energy loss; Step S254: Perform a convergent constraint power series expansion on the non-linear time-varying data of body surface thermal energy loss to obtain a convergent power series of body surface thermal energy loss; Step S255: Perform a body posture hypothermia carrying measurement integration on the livestock body posture difference data according to the convergent power series of body surface thermal energy loss to generate livestock body posture hypothermia measurement data.

[0037] In the embodiment of the present invention, when analyzing the livestock body posture difference data, first, an interval mathematical clustering algorithm is used to classify the body posture data. The body posture data includes, but is not limited to, parameters such as body surface area, subcutaneous fat thickness, back ridge line length, and limb length. Specifically, for example: 180 livestock at different growth cycles are collected, and the recorded body surface area ranges from 0.6 square meters to 1.9 square meters, the subcutaneous fat thickness ranges from 0.8 cm to 4.2 cm, and the back ridge line length ranges from 45 cm to 90 cm. Through interval centroid mean aggregation, 5 typical body posture sections are formed, and each section corresponds to the critical point of heat flux loss in the same temperature environment. By comparing the change in heat flux under the time axis distribution with the actual sampled body surface temperature difference data, the low-temperature response time nodes corresponding to each body posture (such as the time required for the body surface temperature to drop to 34°C) are extracted, and the extreme values of the heat threshold response of each body posture are extracted through the statistical range method, which is defined as the extreme value of low-temperature tolerance. When performing frequency domain processing on the spatial thermal energy vector loss deconstruction data, first, the data is based on the three-dimensional coordinates (x, y, z) and the heat loss value per unit time t Perform discrete calibration and construct a four-dimensional heat tensor matrix. Decompose the thermal energy vector corresponding to each position point into spatial axis components and loss time series. Subsequently, introduce the fast Fourier transform algorithm to perform spectral transformation on the heat loss sequence at each position, extract the frequency response function, and calibrate the energy density distribution in different frequency bands. For example, after transforming the body surface heat loss data, a main peak response appears in the frequency domain spectrum in the range of 0.1 Hz - 0.3 Hz, corresponding to the concentrated response of instantaneous heat loss caused by the room temperature fluctuation frequency at night. The spectral map further constructs a two-dimensional frequency domain map with thermal energy density as the vertical axis and spatial points as the horizontal axis to facilitate subsequent analysis of the heat flow migration trend. When performing dynamic evolution analysis of the time-varying heat loss of livestock based on the combination of the frequency domain distribution map of heat energy loss and the extreme value of low-temperature tolerance of livestock, first establish a non-linear mapping relationship between the body shape parameters and the heat loss data. By constructing a bivariate coupling matrix, with body shape parameters (such as subcutaneous fat thickness) as the main variable and the thermal energy spectral density as the response variable, use the spline function interpolation algorithm to extract the trend of the change in the heat loss rate in each time period. Taking a typical sample with a subcutaneous fat thickness of 1.2 cm and a body surface area of 1.1 square meters as an example, the heat density corresponding to it at a frequency of 0.15 Hz is 2.4 kJ per square decimeter. Combining its low-temperature tolerance extreme value range of 33.5 °C, simulate its heat loss curve per unit time. By fitting the derivative of the heat flow change trend, generate a non-linear time-varying curve of the body surface heat energy loss. Then classify this curve into the corresponding body shape group to form a heat evolution time series data group based on body shape division, and output the final non-linear time-varying data of the body surface heat energy loss.

[0038] In the process of performing a convergent constrained power series expansion on the non-linear time-varying data of body surface heat energy loss, first, the time series change curve in the non-linear time-varying data of heat energy loss is preliminarily smoothed to eliminate the abnormal spike values caused by short-term environmental fluctuations. The smoothing method adopted is piecewise spline interpolation, and taking every 15 minutes as the time node, the change value of heat loss per unit body surface area of livestock in each body posture is extracted. Taking a sample body with a body surface area of 1.2 square meters and a subcutaneous fat thickness of 1.0 centimeter as an example, within the time series from t = 0 to t = 180 minutes, its body surface heat energy loss data points include: 0.11 kJ / dm² at the 1st minute, 0.45 kJ / dm² at the 15th minute, 0.88 kJ / dm² at the 30th minute, and gradually extracted up to 180 minutes. Then, taking this heat energy loss function as f(t), perform a Taylor series expansion on it. Considering the actual environmental stability of the data, the order of expansion is restricted to within 7 orders. The expansion form adopts a convergence constraint condition that the decreasing rate of the coefficient of the power function term is not less than 0.7. By calculating the derivatives of each order of this function at the initial time point, 7 power function terms are respectively constructed, and by verifying that the fitting error between the expanded function and the original non-linear heat loss curve is less than three-thousandths, finally, the convergent power series function form of the body surface heat energy loss of this livestock individual under the current temperature environment and its own body posture conditions is obtained. All body posture samples are respectively executed this power series expansion step to obtain a complete set of power series functions. When performing body posture hypothermia carrying metering integration on the livestock body posture difference data based on the convergent power series of body surface heat energy loss, first, perform a structural alignment process on the power series functions of different body posture samples to ensure that the expansion form is uniformly below 7 orders, and the time parameters of all samples are normalized within 0 to 180 minutes. Subsequently, define the body posture carrying factor, which is composed of three parameters: body surface area, subcutaneous fat thickness, and back ridge line length, and its calculation method is the product of the body surface area multiplied by the subcutaneous fat thickness and the back ridge line length. Taking sample A as an example, with a body surface area of 1.2 square meters, a subcutaneous fat thickness of 1.0 centimeter, and a back ridge line length of 80 centimeters, its carrying factor is 1.2×1.0×80 = 96. Then multiply this factor by the corresponding power series function and perform an integration operation on the power function curve within 0 to 180 minutes to obtain the total body surface heat energy loss of sample A per unit time. All samples are processed in the same way, and finally, the body posture hypothermia metering data of livestock in each body posture under isothermal environment exposure is summarized. The data unit is kJ, and the output structure is the body posture hypothermia amount corresponding to each livestock number at a temperature.

[0039] Step S254 includes the following steps: Calculate the instantaneous growth index of heat energy loss for the non-linear time-varying data of body surface heat energy loss to obtain the instantaneous growth index of heat energy loss; Based on the instantaneous growth index of heat energy loss, perform relative entropy difference analysis on the local mutation of heat energy loss to obtain the relative entropy difference of local mutation of loss; Based on the relative entropy difference of local mutation of loss, perform mutation point convergence analysis on the non-linear time-varying data of body surface heat energy loss to obtain the convergence data of loss mutation points; According to the convergence data of loss mutation points, perform convergent constraint power series expansion on the non-linear time-varying data of body surface heat energy loss to obtain the convergent power series of body surface heat energy loss.

[0040] In the embodiments of the present invention, during the calculation of the instantaneous growth index of heat loss for non-linear time-varying data of body surface heat loss, first, the original time-varying sequence of heat loss is segmented at equally spaced time points. The time interval is set to 5 minutes, and the time range intercepted is from the 0th minute to the 180th minute, and a sequence of heat loss values at 37 time points is extracted in total. Taking a cow with a body surface area of 1.3 square meters and a subcutaneous fat thickness of 0.8 cm as a sample, the heat loss value per unit area of its body surface at 0 minute, 5 minutes, 10 minutes, and 15 minutes is 0.10 kJ / dm², 0.25 kJ / dm², 0.55 kJ / dm², and 0.95 kJ / dm² respectively. The logarithmic growth difference analysis is performed on the heat energy change amount in each adjacent time period. The unit time growth rate is defined as the natural logarithm of the ratio of the heat loss value at the later time point to the heat loss value at the previous time point, and then a set of growth indices in the form of a time series is formed. Taking the section from 10 minutes to 15 minutes as an example, the growth index is the logarithm (0.95 divided by 0.55). Through the time series analysis of this growth index, the heat energy growth fluctuation trend in different time periods is extracted. Further, when it is identified that the growth index in a specific time period, such as from 30 minutes to 45 minutes, is significantly higher than the overall average value, this time period is marked as an early warning node of sudden increase in heat loss. The above operations are repeated for all livestock samples, and their independent sets of growth index time series are respectively formed for subsequent local mutation identification operations. During the relative entropy difference analysis of local mutations in heat loss based on the instantaneous growth index of heat loss, first, the instantaneous growth index sequence of heat loss of each sample mentioned above is subjected to a stationarity test according to the time series structure. A window with a sliding window length of 3 is used to statistically calculate the local distribution probability of the growth index within each window in the entire time series. The statistical method is to discretize the growth index value into equally wide intervals, and the frequency of the growth values falling within each interval is normalized. Subsequently, within two adjacent sliding windows, the normalized probability distribution sequences are respectively extracted, and based on the Shannon entropy theory, the difference in local entropy values within each window is calculated. Further, through the normalization operation of the entropy value difference between the front and rear windows, the relative entropy difference value at this time point is obtained. Taking the section from 45 to 60 minutes of the sample cow as an example, the relative entropy difference reaches the maximum at the 55th minute, which is 0.387, much higher than the average relative entropy difference of 0.114 of the previous overall sequence. It is determined that this point is a mutation point of heat loss. All livestock samples are analyzed for relative entropy difference in the same way to identify their respective local mutation positions in the time series, forming a data set of relative entropy differences of local mutations in loss, and the output form is the livestock number corresponding to the time point with the maximum relative entropy difference and its entropy difference value.

[0041] In the process of analyzing the convergence of mutation points for non-linear time-varying data of body surface heat energy loss, it is first necessary to establish a list of mutation time nodes for each type of livestock sample based on the relative entropy difference data of local mutations of loss obtained in the previous steps. Taking the adult sow with sample number B17 as an example, the relative entropy difference mutation peaks of its heat energy loss growth index appear at the 35th minute, 55th minute, 90th minute, and 110th minute respectively, and the corresponding entropy difference values are 0.412, 0.387, 0.439, and 0.462. Perform a convergence evaluation operation on the time series of such mutation nodes. The method used is to perform a first-order difference on the time interval series of mutation points to generate a series of adjacent mutation point time intervals, and their values are 20 minutes, 35 minutes, and 20 minutes respectively. Subsequently, perform a second-order difference operation on this time interval series. If there are multiple consecutive values tending to zero or the amplitude converges, it is determined that the time position of this type of mutation point has weak convergence in the time domain. On this basis, perform an average trend analysis on the differential data of the mutation time intervals of the samples within the same body posture livestock group. If the second-order difference convergence trends of multiple samples are consistent, it is determined that this livestock body posture category has a stable heat energy mutation rhythm. Finally, output the mutation point convergence data set for each type of livestock. The data structure is a list of mutation time points corresponding to the livestock body posture number and the numerical values of the first-order and second-order difference convergence trends. In the process of performing a convergence constraint power series expansion on the non-linear time-varying data of body surface heat energy loss based on the loss mutation point convergence data, first, for each livestock sample within the time period where the mutation point is located, intercept the time-varying data segment of heat energy loss with a range of ±20 minutes before and after the mutation point centered on the mutation point to construct a local non-linear time series function. Taking the sow with sample number B17 as an example, it mutates at the 55th minute, and intercepts the sequence of heat energy loss values from the 35th minute to the 75th minute, obtaining a total of 9 data points, which are 0.38, 0.42, 0.49, 0.59, 0.68, 0.76, 0.81, 0.83, and 0.85 kilojoules per square decimeter respectively. Perform a power series fitting operation on this local sequence based on the principle of Taylor series. The limiting condition is that the expansion point is set at the 55th minute of the mutation point, and the maximum expansion term is 5th order. During the fitting process, use the convergence trend value in the mutation point convergence data as the weight constraint term to perform a least squares estimation on the coefficients of each order, and solve the coefficient set through constrained optimization. Finally, construct a power series representation expression containing the 0th order constant term to the 5th order time exponential term, representing the non-linear growth characteristics of the heat energy loss of this livestock individual over time during the mutation period. All individuals within the same type of livestock group perform this operation, and output the set of convergence power series function coefficients corresponding to the body posture number for subsequent calls in the hypothermia carrying measurement steps. The data type of this convergence power series is: livestock number, mutation point time, expansion point, coefficient of each order, fitting residual, and convergence degree score value.

[0042] Step S3 includes the following steps: Step S31: Normalize the deconstructed data of spatial thermal energy vector loss to obtain the normalized data of spatial thermal energy vector loss; Step S32: Perform logical learning processing on the livestock body temperature loss measurement data to obtain the livestock body temperature loss learning data; Step S33: Match the spatial thermal energy replenishment rate according to the normalized data of spatial thermal energy vector loss to obtain the spatial thermal energy replenishment rate data; Step S34: Match the heating steam circulation demand among livestock with different body postures based on the livestock body temperature loss learning data to obtain the heating steam circulation demand; Step S35: Construct a heating adjustment model according to the spatial thermal energy replenishment rate data and the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to perform intelligent heating control for the livestock farm.

[0043] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes: Step S31: Normalize the deconstructed data of spatial thermal energy vector loss to obtain the normalized data of spatial thermal energy vector loss; In the embodiment of the present invention, in the process of normalizing the deconstructed data of spatial thermal energy vector loss, it is necessary to first establish the upper and lower limit data of the maximum and minimum values of the spatial thermal energy vector loss value. Taking the deconstructed data of the thermal energy vector loss in the southeast area (numbered A3) within 24 hours in a certain livestock farm as an example, the measured range of the thermal energy vector values at different time points in this area is 1.43 kJ / m² to 7.85 kJ / m². The global maximum value of the deconstructed data of the thermal energy vector loss in this area is statistically obtained as 8.12, and the minimum value is 1.23. The normalization process uses a linear interval mapping method, that is, maps the original thermal energy vector value from the interval [1.23, 8.12] to the standardized interval [0, 1]. The processing process includes point-by-point value taking, interval difference conversion, and mapping output. For example, at a certain moment, the thermal energy vector loss is 6.22 kJ / m², and the normalized value is (6.22 - 1.23) divided by (8.12 - 1.23), and the result is 0.765. This processing operation is carried out point by point for all spatial units and the corresponding time dimension, and finally forms a normalized data matrix of spatial thermal energy vector loss. The matrix dimension corresponds to the spatial coordinates and time series, and the unit element value is between 0 and 1, indicating the relative thermal energy loss intensity level.

[0044] Step S32: Perform logical learning processing on the livestock body temperature loss measurement data to obtain the livestock body temperature loss learning data; In the embodiments of the present invention, during the process of logically learning and processing the livestock body temperature loss measurement data, according to the livestock body temperature loss measurement data obtained from the foregoing steps, the data is grouped according to the classification numbers of livestock body postures (such as adult boars P1, adult sows P2, growing pigs P3, etc.). A classification logic division is performed on the relationship between the hypothermia measurement value and the heat energy loss environmental factor for each group of samples. Taking the body posture of adult sows as an example, during the period from 10:00 to 14:00 every day, the hypothermia measurement data values are concentrated in the range of 0.67 to 0.83. Combining with the fact that the normalized external heat energy vector values in this period are concentrated in the interval of 0.38 to 0.41, a logical relationship set L1 is constructed: when the heat energy vector normalized value is between 0.38 and 0.41, and the livestock is of the P2 body posture type, then the hypothermia measurement value is stable between 0.67 and 0.83. Such logical division adopts a discrete logical Boolean segmentation processing method, and the division conditions include the heat energy normalized value interval, the body posture number, and the day and night time periods. When there are multiple hypothermia-heat energy matching logical linear intervals for samples of the same body posture type at different time periods, multiple logical condition subsets L2, L3, etc. are generated. The finally output logical learning data is a set of multiple condition matching expressions. The expressions are composed of the upper and lower limits of the hypothermia measurement, the upper and lower limits of the normalized heat energy, the body posture number, and the time period. And each condition subset contains the sample support number and the logical consistency confidence level, and the numerical range is set from 0.00 to 1.00.

[0045] Step S33: Perform spatial heat energy replenishment rate matching according to the spatially normalized heat energy vector loss data to obtain spatial heat energy replenishment rate data; In the embodiments of the present invention, during the process of performing spatial heat energy replenishment rate matching according to the spatially normalized heat energy vector loss data, first, the spatial heat source supply area division of the livestock farm is determined. Each area division corresponds to a heat energy supply capacity parameter. For example, the maximum daily heat energy replenishment capacity of Area A3 in the southeast is 12.5 kJ / m². On this basis, according to the normalized heat energy vector loss matrix data, the heat energy shortage amount of each spatial unit at each time period is evaluated. Taking Area A3 at 11:00 as an example, the normalized heat energy loss value is 0.79, then its heat energy replenishment demand rate is the maximum supply rate multiplied by 0.79, that is, 9.875 kJ / (m²·h). And so on, all spatial units in the farm are processed in sequence according to the time period to form a spatial heat energy replenishment rate matrix. To avoid heat disturbance rebound caused by local rate mutations, a third-order moving average smoothing operation is performed on the rate matrix of each time period to reduce the interference of short-term local outliers. Finally, the heat energy replenishment rate value corresponding to each spatial unit at each time period is output, with the unit of kJ / (m²·h). The data structure is a three-dimensional matrix, and the dimensions are the spatial coordinate X-Y axis and the time axis T respectively. The matrix element values are strictly limited within the maximum heat replenishment rate interval of each area.

[0046] Step S34: Based on the livestock body temperature loss learning data, match the heating steam circulation demand among livestock with different body postures to obtain the heating steam circulation demand. In the embodiment of the present invention, in the operation of matching the heating steam circulation demand among livestock with different body postures based on the livestock body temperature loss learning data, it is first necessary to clarify the thermal energy replenishment logic intervals corresponding to the body posture numbers of various livestock. Taking the number P1 representing adult boars, P2 representing adult sows, and P3 representing growing and fattening pigs as examples, the thermal energy replenishment intensity intervals and their confidence levels required for each numbered body posture in different normalized thermal energy environments have been obtained in the aforementioned logic learning data. Taking P2 adult sows as an example, if the logic learning data for the current period matches the normalized thermal energy environment interval of 0.36 to 0.41, the corresponding thermal energy replenishment intensity demand interval is 0.67 to 0.83. Here, the interval median method is used to specifically quantify the demand value, that is, the demand intensity is set as (0.67 + 0.83) divided by 2, which is 0.75. Then, combined with the current space thermal energy replenishment rate data, such as the thermal energy replenishment rate in this area is 11.2 kJ / m² / h, and according to the unit energy output efficiency value of the heating steam being 2.5 kJ / m³, the heating steam circulation demand is obtained by dividing the thermal energy demand by the output efficiency, that is, 11.2 multiplied by 0.75 divided by 2.5, and the hourly heating steam circulation demand is obtained as 3.36 m³ / m². This calculation process is carried out for each type of livestock with a body posture number in the farm, generating a three-dimensional mapping table of body posture number - time period - steam demand, and the output data structure is a two-dimensional matrix of body posture number and spatial area index, and the matrix value is the unit-hour circulation demand of the heating steam, with the unit of m³ / m² / h.

[0047] Step S35: Construct a heating adjustment model based on the space thermal energy replenishment rate data and the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to execute intelligent heating control for the livestock farm.

[0048] In the embodiments of the present invention, in the operation of constructing a heating adjustment model according to the space thermal energy replenishment rate data and the heating hot steam circulation demand, first, the two types of data need to be aligned according to time and space coordinates. Taking the breeding area number A3 as an example, assume that the thermal energy replenishment rate in this area is 9.875 kJ / (m²·h), and the steam circulation demand corresponding to the body state number P2 is 3.36 m³ / (m²·h). According to the heating adjustment logic, the piecewise linear matching method is used to take the thermal energy rate and steam demand as input variables to construct a bivariate thermal energy adjustment correlation table, and a complete adjustment sequence is constructed according to the daily 24-hour cycle data. The adjustment model calculates the required opening duration and steam circulation power of the heating unit at each time point to ensure that the steam output meets the real-time thermal energy demand intensity. The adjustment model uses the discrete-time dynamic control method to handle the fluctuations of input variables, sets the sampling interval to once every 15 minutes, and updates the heating control instruction data. For example, at the time point of 12:15, when the heat replenishment rate in area A3 rises to 10.3 kJ / (m²·h) and the steam demand is 3.88 m³ / (m²·h), the adjustment control parameters output by the model are that the steam valve opening is 72%, the heating duration is 15 minutes, and the steam pressure is set to 0.5 MPa. The control parameters generated by this adjustment model include fields such as spatial area index, timestamp, steam power setting value, heating duration, and valve opening instruction. Finally, the complete heating adaptive adjustment model control parameters are sent to the cloud platform control center through the wired communication protocol MQTT standard interface for the cloud server to complete command scheduling and execution, so as to realize the real-time closed-loop intelligent control of the heating system in the livestock farm.

[0049] The present invention also provides an intelligent heating control system for a livestock farm, which is used to execute the intelligent heating control method for a livestock farm as described above. The intelligent heating control system for a livestock farm includes: A data monitoring module, which is used to monitor the behavior activity status of livestock in the livestock farm through an electronic monitoring device to obtain a monitoring video of the behavior activity status of livestock; and monitor the temperature and humidity fluctuations and greenhouse gas fluctuations in the livestock farm under extremely cold weather through the temperature and humidity sensors and greenhouse gas sensors in the environmental sensing module, and obtain temperature and humidity fluctuation data and greenhouse gas fluctuation data respectively; A body temperature loss measurement module, which is used to perform body state difference analysis on the monitoring video of the behavior activity status of livestock to generate livestock body state difference data; perform spatial thermal energy vector loss deconstruction on the temperature and humidity fluctuation data and greenhouse gas fluctuation data to obtain spatial thermal energy vector loss deconstruction data; and perform body temperature loss carrying measurement integration between livestock of different body states based on the spatial thermal energy vector loss deconstruction data to generate livestock body temperature loss measurement data; The heating adjustment model construction module is used to match the heating steam circulation demand among livestock with different body postures based on the livestock body posture hypothermia learning data to obtain the heating steam circulation demand; construct a heating adjustment model according to the heating steam circulation demand to obtain a heating adaptive adjustment model, and send the heating adaptive adjustment model to the cloud platform to execute the intelligent heating control of the livestock farm.

[0050] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. An intelligent heating control method for livestock farms, characterized in that: Electronic monitoring equipment and environmental sensing modules are deployed in the livestock farm, wherein the environmental sensing modules include temperature and humidity sensors and greenhouse gas sensors. The intelligent heating control method of the livestock farm includes the following steps: Step S1: Monitor the behavior and activity status of livestock in the livestock farm through electronic monitoring equipment to obtain a monitoring video of the behavior and activity status of livestock; monitor the temperature and humidity fluctuations and greenhouse gas fluctuations of the livestock farm under extremely cold weather through the temperature and humidity sensors and greenhouse gas sensors in the environmental sensing module to obtain temperature and humidity fluctuation data and greenhouse gas fluctuation data respectively; Step S2: Performing posture difference analysis on livestock behavior and activity status monitoring videos to generate livestock posture difference data; performing spatial thermal energy vector loss deconstruction on temperature and humidity fluctuation data and greenhouse gas fluctuation data to obtain spatial thermal energy vector loss deconstruction data; performing posture hypothermia carrying measurement integration between livestock with different postures on livestock posture difference data based on the spatial thermal energy vector loss deconstruction data to generate livestock posture hypothermia measurement data; Step S3: Based on the livestock body hypothermia learning data, the heating hot steam circulation demand between livestock of different body shapes is matched to obtain the heating hot steam circulation demand; a heating regulation model is constructed according to the heating hot steam circulation demand to obtain a heating adaptive regulation model, and the heating adaptive regulation model is sent to the cloud platform to execute intelligent heating control of livestock farms.

2. The intelligent heating control method for animal husbandry farms according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: Performing posture difference analysis on the livestock behavior and activity status monitoring video to generate livestock posture difference data; Step S22: performing group movement frequency analysis on the livestock behavior and activity status monitoring video to obtain livestock group movement frequency data; Step S23: Acquire livestock farm structural design data; perform gas disturbance dispersion analysis on greenhouse gas fluctuation data based on livestock group movement frequency data and livestock farm structural design data to obtain greenhouse gas disturbance dispersion structure data; Step S24: performing spatial thermal energy vector loss deconstruction on the temperature and humidity fluctuation data and the greenhouse gas disturbance dispersion structure data according to the livestock farm structural design data to obtain spatial thermal energy vector loss deconstruction data; Step S25: Based on the spatial thermal energy vector loss deconstruction data, the livestock body shape difference data is integrated with the body shape hypothermia measurement between livestock with different body shapes to generate livestock body shape hypothermia measurement data.

3. The intelligent heating control method for animal husbandry farms according to claim 2 is characterized in that: Step S23 includes the following steps: Step S231: Acquire the livestock farm structural design data; perform gas circulation design path analysis on the livestock farm structural design data to obtain the gas circulation design path; Step S232: performing group gait acceleration frequency analysis on the livestock group movement frequency data to obtain the group gait acceleration frequency; Step S233: Calculating the continuous pressure difference of the peripheral airflow based on the group gait acceleration frequency to obtain the continuous pressure difference of the peripheral airflow; Step S234: Deducing the change of the azimuthal velocity difference of the gas flow section on the gas flow design path according to the continuous pressure difference driven by the surrounding airflow, and obtaining the change data of the azimuthal velocity difference of the gas flow section; Step S235: performing mass / volume regional distribution difference analysis on the greenhouse gas fluctuation data to obtain the greenhouse gas mass / volume regional distribution difference; Step S236: performing gas disturbance dispersion analysis on the greenhouse gas mass / volume regional distribution difference according to the flow section azimuthal velocity difference variation data to obtain greenhouse gas disturbance dispersion structure data.

4. The intelligent heating control method for animal husbandry farms according to claim 3 is characterized in that: Step S24 includes the following steps: Step S241: performing material thermal insulation performance impairment analysis on the livestock farm structural design data to obtain material thermal insulation performance impairment data; Step S242: mapping the temperature and humidity spatial fluctuation distribution of the temperature and humidity fluctuation data according to the livestock farm structural design data to obtain temperature and humidity spatial fluctuation distribution mapping data; Step S243: based on the thermal insulation performance loss data of the materials, the spatial heat energy loss divergence per unit time is calculated for the temperature and humidity spatial fluctuation distribution mapping data and the greenhouse gas disturbance dispersion structure data to obtain the spatial heat energy loss divergence data; Step S244: performing spatial thermal energy vector loss deconstruction based on the spatial thermal energy loss divergence data to obtain spatial thermal energy vector loss deconstruction data.

5. The intelligent heating control method for animal husbandry farms according to claim 4 is characterized in that: Step S243 includes the following steps: The structural pore thermal conductivity increment is calculated for the thermal insulation performance loss data of the material to obtain the structural pore thermal conductivity increment coefficient; Based on the incremental coefficient of thermal conductivity of structural pores, the thermal insulation performance loss data of the material is identified by accelerating the proportionality of interface heat transfer, and the accelerated proportionality data of interface heat transfer is obtained; The temperature and humidity spatial fluctuation distribution mapping data is used to calculate the low temperature / low humidity relative change rate difference, and the spatial distribution low temperature / low humidity relative change rate difference is generated; According to the relative change rate difference of low temperature / low humidity in spatial distribution, the spatial discrete heat loss density value increment of greenhouse gas disturbance dispersion structure data per unit time is integrated to obtain the spatial heat loss density increment value per unit time; Based on the interface heat transfer acceleration proportional data and the spatial heat loss density increment value per unit time, the spatial heat energy loss divergence per unit time is estimated to obtain the spatial heat energy loss divergence data.

6. The intelligent heating control method for animal husbandry farms according to claim 4 is characterized in that: Step S25 includes the following steps: Step S251: performing extreme value analysis of low temperature tolerance among livestock with different body shapes on livestock body shape difference data to obtain extreme values ​​of low temperature tolerance among livestock with different body shapes; Step S252: Perform Fourier thermal spectrum transformation on the spatial thermal energy vector loss deconstruction data to obtain a frequency domain distribution spectrum of thermal energy loss; Step S253: performing a time-varying nonlinear dynamic evolution of body surface heat energy loss between livestock with different body shapes on the livestock body shape difference data based on the heat energy loss frequency domain distribution spectrum and the extreme values ​​of livestock low temperature tolerance between livestock with different body shapes, and obtaining nonlinear time-varying data of body surface heat energy loss; Step S254: performing convergence constrained power series expansion on the nonlinear time-varying data of body surface heat energy loss to obtain a convergent power series of body surface heat energy loss; Step S255: integrating the body hypothermia measurement between livestock with different body shapes according to the convergence power series of body surface heat energy loss on the livestock body shape difference data to generate livestock body hypothermia measurement data.

7. The intelligent heating control method for animal husbandry farms according to claim 6 is characterized in that: Step S254 includes the following steps: The instantaneous growth index of heat energy loss is calculated for the nonlinear time-varying data of body surface heat energy loss to obtain the instantaneous growth index of heat energy loss; Based on the instantaneous growth index of heat energy loss, the relative entropy difference of local mutation of heat energy loss is analyzed to obtain the relative entropy difference of local mutation of loss; Based on the relative entropy difference of local mutation of loss, the mutation point convergence analysis is carried out on the nonlinear time-varying data of body surface heat energy loss, and the loss mutation point convergence data is obtained; According to the convergence data of loss mutation point, the nonlinear time-varying data of body surface heat energy loss is expanded with convergence constraint power series to obtain the convergence power series of body surface heat energy loss.

8. The intelligent heating control method for animal husbandry farms according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: normalizing the spatial thermal energy vector loss deconstruction data to obtain spatial thermal energy vector loss normalized data; Step S32: performing logic learning processing on the livestock body hypothermia measurement data to obtain livestock body hypothermia learning data; Step S33: performing spatial heat energy replenishment rate matching according to the spatial heat energy vector loss normalization data to obtain spatial heat energy replenishment rate data; Step S34: matching heating hot steam circulation requirements between livestock of different body shapes based on the livestock body hypothermia learning data to obtain heating hot steam circulation requirements; Step S35: Construct a heating regulation model according to the space heat energy replenishment rate data and the heating hot steam cycle demand to obtain a heating adaptive regulation model, and send the heating adaptive regulation model to the cloud platform to execute intelligent heating control of the livestock farm.

9. An intelligent heating control system for livestock farms, characterized in that: Used to execute the intelligent heating control method for animal husbandry farms as claimed in claim 1, the intelligent heating control system for animal husbandry farms comprises: The data monitoring module is used to monitor the behavior and activity status of livestock in the livestock farm through electronic monitoring equipment to obtain livestock behavior and activity status monitoring videos; the temperature and humidity sensors and greenhouse gas sensors in the environmental sensing module are used to monitor the temperature and humidity fluctuations and greenhouse gas fluctuations in the livestock farm under extremely cold weather to obtain temperature and humidity fluctuation data and greenhouse gas fluctuation data respectively; The body hypothermia measurement module is used to analyze the body difference of livestock behavior and activity status monitoring videos to generate livestock body difference data; perform spatial thermal energy vector loss deconstruction on temperature and humidity fluctuation data and greenhouse gas fluctuation data to obtain spatial thermal energy vector loss deconstruction data; perform body hypothermia carrying measurement integration between livestock with different body shapes on livestock body difference data based on spatial thermal energy vector loss deconstruction data to generate livestock body hypothermia measurement data; The heating regulation model construction module is used to match the heating hot steam circulation demand between livestock of different body shapes based on livestock body hypothermia learning data to obtain the heating hot steam circulation demand; the heating regulation model is constructed according to the heating hot steam circulation demand to obtain the heating adaptive regulation model, and the heating adaptive regulation model is sent to the cloud platform to execute intelligent heating control of livestock farms.

Images