Intelligent control terminal and control method for full-automatic spray disinfection of farm
By collecting environmental parameters through sensors, identifying abnormal areas using cluster analysis and difference matrices, and training models with historical data, precise disinfection of pathogenic microorganisms in farms was achieved, solving the problem of inaccurate drug solution ratios and improving disinfection effectiveness.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-22
- Publication Date
- 2026-03-24
AI Technical Summary
Existing technologies make it difficult to accurately detect the types and quantities of pathogenic microorganisms when spraying disinfectant in farms, leading to inaccurate drug solution ratios and affecting the disinfection effect.
By collecting environmental parameters through sensors, identifying abnormal areas using cluster analysis and difference matrices, and training models based on historical data, the precise formulation of disinfectant solutions can be configured for disinfection.
It improves the disinfection effect against pathogenic microorganisms, achieves precise control of the drug solution ratio, and enhances the targeting and effectiveness of disinfection.
Smart Images
Figure CN121714734A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of control, in particular to a full-automatic spray disinfection intelligent control terminal for a farm and a control method. BACKGROUND
[0002] In the traditional breeding industry, in order to ensure the use of animal husbandry, in order to prevent the livestock from being infected by the virus on a large scale, it is necessary to take weight health management measures and disinfect. The traditional disinfection method mainly relies on manual operation, but such disinfection is often uneven and there are some disinfection, and it also has a great threat to the health of the disinfection personnel. Therefore, today, with the development of intelligence, various sensors are used to detect the environment in the farm. When some indicators exceed the preset warning value, the liquid is selected to spray in the farm, so as to kill the pathogenic microorganisms in the farm.
[0003] In the prior art, when the farm is sprayed and disinfected, the disinfectant is directly configured by comparing whether various environmental parameters exceed the preset threshold, and the pathogenic microorganisms in the farm environment are killed by the preset spraying program. However, in the farm, there are various pathogenic microorganisms, and when the content of the pathogenic microorganisms exceeds a certain order of magnitude, the influence on the environment can be detected. These cause that it is difficult to accurately detect the type and quantity of microorganisms by different environmental parameters, so that it is difficult to accurately control the liquid ratio for these pathogenic microorganisms, and thus it is difficult to accurately kill these pathogenic microorganisms. SUMMARY
[0004] In order to solve the technical problem of poor killing effect caused by inaccurate control of liquid ratio, the present application provides a full-automatic spray disinfection intelligent control terminal for a farm and a control method, and the technical solution adopted is as follows: In a first aspect, the present application provides a full-automatic spray disinfection intelligent control method for a farm, which comprises the following steps: Collecting the parameter values of the environmental parameters of different monitoring points at different times in the farm through various sensors; Based on the clustering of all times for any monitoring point based on the environmental parameters, obtaining similar time and node similarity marker value with the current time; screening the consistent time of the environment based on the node similarity marker value of each time with the current time; Obtaining the environment description vector and the farm environment vector through the parameter values of all environmental parameters of each time monitoring point; obtaining the difference vector by subtracting the farm environment vector from the environment description vector; determining the consistency of the environment difference based on the similarity of the difference vector between the consistent time of the environment and the current time; An environment difference matrix of all monitoring points is constructed based on environment difference consistency, and an abnormal area is divided according to the environment difference matrix; an abnormal area matching degree is obtained by matching each abnormal area at the current time with an abnormal area at an environment consistent time, and a disinfection point is obtained by classifying monitoring points; and a disinfection demand consistency degree is obtained based on the similarity of environment description vectors at the current time and the environment consistent time and the attention degrees of different disinfection points. A model is trained through disinfection demand consistency and historical data, and parameter values of environment parameters at the current time are input into the model to control the proportion of disinfectant liquid for the farm disinfection.
[0005] In the above scheme, the reference time consistent with the current time is first selected by analyzing the basic environment of the farm, and then the fluctuation of a large number of detection parameters of different monitoring points at each reference time is analyzed, so as to obtain the environmental changes caused by the metabolic products of pathogenic microorganisms. Further, the possibility of environmental changes caused by pathogenic microorganisms is obtained by comparing the environmental fluctuations of various abnormalities between each reference time and the current time. Finally, the disinfectant liquid configuration is controlled according to the proportion of disinfectant liquid for the past disinfection of these pathogenic microorganisms, so as to complete the current disinfectant liquid configuration and the disinfection of the farm, and improve the disinfection effect of pathogenic microorganisms.
[0006] In one embodiment, the method for clustering all time points of any one monitoring point to obtain the time similar to the current time and the node similarity label value is as follows: For any one monitoring point, each basic environment point corresponding to each time is constructed by taking each environment parameter as a dimension, and the basic environment points are clustered; Each time in the same cluster as the current time is recorded as a node similar time corresponding to the current time environment, and the corresponding node similarity label value is 1; on the contrary, if it is not in the same cluster, the node similarity label value is recorded as 0.
[0007] In one embodiment, the method for screening the environment consistent time based on the node similarity label value of each time and the current time is as follows: The average value of all node similarity label values of each time and the current time at all monitoring points in the historical time is calculated; if the average value is greater than a preset similarity threshold, the time is recorded as an environment consistent time.
[0008] In one embodiment, the method for obtaining an environment description vector and a farm environment vector by using parameter values of all environment parameters of each monitoring point at each time is as follows: In each monitoring point, for any kind of environment parameter, the parameter values at each time are standardized, and all standardized parameter values at each time in each monitoring point are used to form an environment description vector. For any environmental parameter, the average value of the parameter values of all monitoring points at each time is calculated, the average value is normalized, and all normalized average values at the time form the farm environment vector.
[0009] In one embodiment, the environmental difference consistency is the normalized result of the cosine similarity of the difference vector between the environmental consistent time and the current time in the same monitoring point.
[0010] In one embodiment, the method for constructing the environmental difference matrix of all monitoring points based on the environmental difference consistency is: For each environmental consistent time, the environmental difference consistency value corresponding to each monitoring point is obtained, and the environmental difference matrix corresponding to the two times is constructed according to the positional relationship of each monitoring point.
[0011] In one embodiment, the method for dividing the abnormal area based on the environmental difference matrix is: The monitoring point position with an environmental difference consistency less than a preset value is taken as an initial point, the environmental difference consistency between two monitoring points at the same time is taken as a growth basis, and the region growing method is used to classify these monitoring points, each class being an abnormal area.
[0012] In one embodiment, the method for obtaining the abnormal area matching degree by matching each abnormal area at the current time with the abnormal area at the environmental consistent time, and classifying the monitoring points to obtain the disinfection point is: The similarity between each abnormal area at the current time and the environmental condition of each abnormal area at each environmental consistent time is calculated by the CPD algorithm, and a probability matrix is constructed. The average value of the plurality of elements in the CPD output probability matrix is calculated, the similarity is the element in the corresponding probability matrix output by the CPD, the average value of the element is calculated, and is recorded as the abnormal area matching degree.
[0013] In one embodiment, the method for obtaining the disinfection demand consistency degree based on the similarity of the environmental description vectors of the current time and the environmental consistent time and the attention degree of different disinfection points is: , The environmental description vector of the wth disinfection point at the current time is represented as The environmental description vector of the wth disinfection point at the current time is represented as The environmental description vector of the wth disinfection point at the current time is represented as The environmental description vector of the wth disinfection point at the current time is represented as is used to calculate the cosine similarity between two vectors; is used for normalization, The maximum value of the environmental similarity of each disinfection point at the environmental consistent time The maximum value of the environmental similarity of each disinfection point at the environmental consistent time The maximum value of the environmental similarity of each disinfection point at the environmental consistent time Indicates the number of disinfection points. This indicates the level of attention paid to the w-th disinfection point. This represents the time when the environment is consistent with the current time. The consistency of disinfection requirements; the attention level of the pathogen contamination point is 1, and the attention level of the suspected contamination point is the difference between the number 1 and the matching degree of the abnormal area.
[0014] On the other hand, this application embodiment also provides a fully automatic spray disinfection intelligent control terminal for a farm, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described fully automatic spray disinfection intelligent control methods for a farm.
[0015] The beneficial effects of this application are as follows: This application analyzes the basic environment of the farm, first selecting reference times consistent with the current environment, then analyzing the fluctuations of numerous detection parameters at different monitoring points at each reference time to determine if environmental changes are caused by pathogenic microorganism metabolites. Furthermore, by comparing various abnormal environmental fluctuations between each reference time and the current time, the possibility of these environmental changes being caused by pathogenic microorganisms is determined. Finally, based on this, and combined with the past preparation of disinfectant solutions for these pathogenic microorganisms, the current disinfectant solution ratio is controlled to complete the current disinfectant solution preparation and disinfection of the farm, thereby improving the disinfection effect against pathogenic microorganisms. Attached Figure Description
[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is a flowchart of a fully automated spray disinfection intelligent control method for a farm, provided as an embodiment of this application. Detailed Implementation
[0018] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a fully automatic spray disinfection intelligent control terminal and control method for livestock farms proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0020] An embodiment of a fully automatic intelligent control terminal and control method for spray disinfection in a livestock farm: The following description, in conjunction with the accompanying drawings, details the specific scheme of the fully automatic spray disinfection intelligent control terminal and control method for livestock farms provided in this application.
[0021] Please see Figure 1 The document illustrates a flowchart of an intelligent control terminal and control method for fully automatic spray disinfection in a livestock farm, according to an embodiment of this application. The method includes the following steps: Step S001: Collect the environmental parameter values of different monitoring points at different times in the farm through various sensors.
[0022] To ensure functionality for livestock and veterinary use, disinfection and sterilization are necessary at the farm. Environmental and status sensors are installed throughout the farm to monitor various environmental data at different monitoring points. Environmental sensors include: temperature and humidity sensors, ammonia (NH3) sensors, hydrogen sulfide (H2S) sensors, carbon dioxide (CO2) sensors, and PM2.5 / dust concentration sensors. Status sensors include: flow sensors (monitoring drug flow rate), pressure sensors (monitoring pipeline pressure), level sensors (monitoring tank level), and infrared or ultrasonic sensors (for detecting personnel / vehicle entry / exit).
[0023] Environmental parameter data is collected through the aforementioned sensors. In this embodiment, the data acquisition frequency of each sensor is preset to 6 times per second, and data collected over the past month is used as historical data for learning and analysis. After obtaining the collected data, it is approximated through segmented aggregation. The collected environmental data is cleaned; in this embodiment, the preset window size for data cleaning is 1 second.
[0024] At this point, environmental parameter data for each monitoring point at each moment has been obtained.
[0025] Step S002: Cluster the times with the same environmental parameters at any monitoring point and label and filter the times with consistent environment.
[0026] In livestock farms, the presence of poultry and livestock leads to a diverse range of pathogenic microorganisms. However, due to their living conditions and mechanisms, these pathogenic microorganisms produce different metabolic products, which in turn alter the environment. By observing the distribution and changes of each environmental factor and substance in the farm, we can initially determine the types and amounts of pathogenic microorganisms present. Furthermore, by combining this with the past effectiveness of various disinfectants against these different microorganisms, we can determine the optimal disinfectant ratio for disinfecting the farm.
[0027] Pathogenic microorganisms are highly sensitive to the environment. When the environment is different, they may die in large numbers. Therefore, if two environments are similar, the pathogenic microorganisms present in them will also be similar.
[0028] For any monitoring point, using each environmental parameter as a dimension, basic environmental points are constructed for each time moment, and these basic environmental points are clustered using Dbscan. In this embodiment, the preset minimum number of points (minPts) for clustering is 4, and the preset radius (eps) is determined using the K-distance map method with the minimum number of points (minPts) as K.
[0029] At this point, many moments with similar environments are grouped into the same cluster. Each moment that is in the same cluster as the current moment is recorded as a node similarity moment corresponding to the current environment, and its corresponding node similarity label value is 1; otherwise, the label value is recorded as 0.
[0030] For each monitoring point, a time frame was constructed to find times with similar environments to the current time. However, since farms are often quite large, the gathering and activities of poultry and livestock in the farm can cause local fluctuations in environmental information such as temperature and humidity in some areas. This leads to differences in the environment between the selected nodes. Therefore, when most nodes in the entire farm are similar at two different times, and the fluctuations are consistent with the differences in poultry and livestock activities, it indicates that the environments at these two times are similar.
[0031] Therefore, when the environments observed at various monitoring points in the farm are mostly similar at two different times, it can be effectively concluded that the overall environment in the farm is relatively similar. Accordingly, for any given time, the environmental similarity between each time and the current time is determined by the node similarity marker values at different monitoring points.
[0032] The environmental similarity is the mean of the similarity marker values of all nodes at different monitoring points at each time point and the current time point, with a maximum value of 1 and a minimum value of 0.
[0033] Based on this, the environmental similarity between the farm environment at each historical moment and the current moment can be calculated. The greater the environmental similarity, the more similar the basic environment within the farm at these two moments, meaning the environment conducive to the growth of pathogenic microorganisms is more similar. Therefore, if the environmental similarity between a moment and the current moment is greater than or equal to a similarity threshold, that moment is recorded as an environmentally consistent moment. In this embodiment, the similarity threshold is 0.7. The similarity threshold was obtained through extensive data experiments.
[0034] At this point, all environmental consistent moments have been obtained.
[0035] Step S003: Then, determine the environmental description vector and the farm environment vector through environmental parameters, calculate the vector difference, and determine the consistency of environmental differences based on the similarity of vector differences at different times.
[0036] Using each environmental parameter as a dimension, and the standard score of the corresponding parameter value at a given time as its value in the corresponding dimension, an environmental description vector for each monitoring point at a given time is constructed. In calculating the standard score (z-score), historical time points are used as the sample space. The calculation of the standard score is a well-known technique and will not be described in detail in this embodiment. The above steps yield the environmental description vector for each monitoring point at each time point.
[0037] Then, the mean value of each environmental parameter at all monitoring points at each time point is calculated. This mean value represents the central tendency of the overall environment and is used to approximate the true environment. This value is used as the true fitted value of each environmental parameter at each time point. The standard scores corresponding to the true fitted values of all environmental parameters are used to construct the environmental description vector corresponding to the environment in the farm, which is denoted as the farm environment vector.
[0038] If the environmental conditions detected at the same monitoring point in the farm at two different times are consistent with the differences between the environment and the environment, it indicates that the pathogenic microorganisms and environmental changes caused by livestock farming in the current area are consistent. Therefore, it can be concluded that the environmental differences at the same monitoring point at two different times are consistent.
[0039] For each time step, the difference between the farm environment vector and the environment description vector of each monitoring point is used as the difference vector for each monitoring point at that time step. The difference vector is used to describe the difference between the monitoring parameters of that monitoring point and the general conditions in the farm at that time step.
[0040] Preferably, in this embodiment, the expression for the difference vector is: , This represents the farm environment vector at time c. This represents the environmental description vector of the j-th monitoring point at time c. Let represent the difference vector of the j-th monitoring point at the c-th time.
[0041] Then, the similarity of the difference vector between each environmental consistency moment and the current moment is calculated to obtain the environmental difference consistency between each moment and the current moment at each monitoring point.
[0042] Preferably, in this embodiment, the expression for environmental difference consistency is: , Indicates the first The difference vector of the j-th monitoring point at a time when the environment is consistent; Indicates the current time The difference vector of the j-th monitoring point; Used to calculate the cosine similarity between two vectors; Used for normalization, This represents the difference between time t and the current time. The environmental differences at the j-th monitoring point are consistent. In this embodiment, normalization is performed using maximum and minimum value normalization.
[0043] The greater the consistency of environmental differences, the smaller the difference between that moment and the current moment, meaning the more similar they are.
[0044] Thus, the consistency of environmental differences at any monitoring point at each time point and at the current time has been obtained.
[0045] Step S004: Construct a difference matrix based on the consistency of environmental differences to divide abnormal regions; match the abnormal regions at the current time with those at other times to obtain disinfection points; and obtain the consistency of disinfection requirements based on the different attention levels of the disinfection points and the similarity of the environmental description vectors.
[0046] Since the total number of livestock in a farm is fixed, a decrease in livestock in one area (the monitoring point's location) will inevitably lead to an increase in livestock in other areas. Therefore, when the ambient temperature and humidity change in one area, the environment in other areas will inevitably show the opposite change. Thus, by comparing this movement of environmental differences between numerous areas with significant environmental variations at two different times, abnormal areas can be identified.
[0047] First, the first The environmental consistency moment and the current moment The environmental differences between the monitoring points are taken as values, and these values are used to construct environmental difference matrices for the two time points according to the positional relationship between the monitoring points.
[0048] Then, taking the locations of monitoring points where the environmental difference consistency is less than a preset value as the initial points, and using the environmental difference consistency between two monitoring points at the same time as the growth basis, the region growth method is used to classify these monitoring points; in this embodiment, the preset value is 0.3.
[0049] Each cluster represents a region where environmental data changes differ at two different times, meaning that pathogenic microorganisms and livestock metabolites aggregate differently. These regions are designated as abnormal regions.
[0050] If the abnormal region moves from one place to another at two different times, it means that the abnormal region is caused by the gathering of poultry and livestock in the farm at the current time. Then, the original environmental characteristics of these regions can be obtained based on the matching characteristics of the environmental data between these regions at two different times.
[0051] First, each abnormal region is treated as a data point set, and the environmental description vector corresponding to each monitoring point within it is taken as a data point. The current time is then calculated using the CPD algorithm. The Middle The corresponding data point set of the region and the first The similarity of environmental conditions among the data point sets corresponding to each abnormal region at a time when the environment is consistent is calculated. The similarity is the average value of the elements in the corresponding probability matrix output by CPD, which is denoted as the abnormal region matching degree.
[0052] Then, at the t-th environmentally consistent time, select the anomaly regions that are consistent with the current time. The Middle The abnormal region with the highest matching degree in the corresponding data point set is taken as its matching region. If the matching degree of the abnormal region is greater than or equal to the abnormal threshold, it is considered that the transfer of these two abnormal regions is caused by the transfer of poultry and livestock in the farm, and their matching relationship (i.e., the point correspondence relationship output by CPD) is marked. These points are recorded as suspected contamination points. In this embodiment, the abnormal threshold is set to 0.7. The monitoring points in the remaining unmatched abnormal regions are recorded as pathogen contamination points. Pathogen contamination points and suspected contamination points are recorded together as disinfection points.
[0053] Since each pathogenic microorganism has its unique metabolites, and the production efficiency of metabolites for a pathogenic microorganism is relatively fixed, when the content of pathogenic microorganism products detected at each monitoring point at two different times is the same, it indicates that the types of microorganisms in the farm at these two times are the same and their population sizes are also similar.
[0054] Therefore, the consistency of disinfection requirements is determined based on the consistency of environmental data at the disinfection site at the current time and the level of attention given to different types of disinfection sites. The consistency of disinfection requirements describes the consistency of viral metabolites detected at numerous monitoring points at two different times.
[0055] Preferably, in this embodiment, the expression for the consistency of disinfection requirements is: , Indicates the current time The environmental description vector of the w-th disinfection point. This indicates the time step t under the condition of environmental consistency. An environmental description vector for each disinfection point. Used to calculate the cosine similarity between two vectors; Used for normalization, Indicates the selection of a consistent environment. The following disinfection points and the next disinfection point at the current time The maximum value of environmental similarity at the disinfection points. Indicates the number of disinfection points. This indicates the level of attention paid to the w-th disinfection point. This represents the time when the environment is consistent with the current time. The consistency of disinfection requirements. The attention level of the pathogen contamination point is 1, and the attention level of the suspected contamination point is the difference between the number 1 and the matching degree of the abnormal area.
[0056] At this point, the consistency of disinfection requirements between each environmental moment and the current moment has been obtained.
[0057] Step S005: The consistency of disinfection requirements is combined with historical data to complete model training, and the model is used to control the drug ratio at the current moment for disinfection of the farm.
[0058] The parameters of numerous environmental parameters corresponding to each disinfection point at each consistent environmental moment in history are used as input parameters, and the corresponding disinfection liquid components and concentrations are used as output parameters. The input-output corresponding to each consistent environmental moment is used as a set of learning data to train a convolutional neural network for adjusting the liquid components. In addition, an attention mechanism is introduced during training, and the consistency of disinfection requirements corresponding to each set of data is used as its attention weight.
[0059] Once the model is trained, the environmental parameters corresponding to the numerous disinfection points at the current moment are input into the model to obtain the drug solution ratio for disinfecting the farm at the current moment.
[0060] The disinfection system then prepares the disinfectant solution according to this ratio, and finally uses the prepared solution to disinfect the farm.
[0061] It should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
[0062] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. A fully automated intelligent control method for spray disinfection in a livestock farm, characterized in that, The method includes the following steps: The environmental parameters at different monitoring points in the farm at different times were collected using various sensors. Clustering is performed on all times for any monitoring point based on environmental parameters to obtain times and node similarity labels similar to the current time; times with consistent environment are filtered based on the node similarity labels of each time and the current time. The environmental description vector and the farm environment vector are obtained by measuring the parameter values of all environmental parameters at each monitoring point; the difference vector is obtained by subtracting the farm environment vector from the environmental description vector; the consistency of environmental differences is determined based on the similarity of the difference vectors between the consistent environmental time and the current time. An environmental difference matrix is constructed for all monitoring points based on the consistency of environmental differences. Abnormal areas are divided using the environmental difference matrix. The abnormal area at the current time is matched with the abnormal area at the time of environmental consistency to obtain the abnormal area matching degree. The monitoring points are classified to obtain the disinfection points. The consistency of disinfection requirements is obtained based on the similarity of the environmental description vector at the current time and the time of environmental consistency, as well as the attention of different disinfection points. By training a model based on consistent disinfection needs and historical data, the current environmental parameter values are input into the model to control the drug solution ratio for disinfection of the farm.
2. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 1, characterized in that, The method for clustering all times for any monitoring point to obtain similar times and node similarity markers to the current time is as follows: For any monitoring point, each environmental parameter is used as a dimension to construct the basic environmental points corresponding to each time moment, and these basic environmental points are clustered. Each time point that is in the same cluster as the current time point is recorded as the node similarity time corresponding to the current time environment, and its corresponding node similarity label value is 1; Conversely, nodes not belonging to the same cluster are recorded as having a similarity tag value of 0.
3. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 1, characterized in that, The method for filtering environmentally consistent moments based on the node similarity marker values at each time step and the current time step is as follows: Calculate the average of the similarity marker values of all nodes at all monitoring points for each historical time point and the current time point; if the average value is greater than a preset similarity threshold, then the time point is recorded as the environmental consistency time point.
4. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 1, characterized in that, The method for obtaining the environmental description vector and the farm environment vector by using the parameter values of all environmental parameters at each monitoring point at each time moment is as follows: At each monitoring point, for any environmental parameter, the parameter value at each time moment is standardized, and all standardized parameter values at each monitoring point at that time moment constitute an environmental description vector. For any environmental parameter, calculate the mean value of the parameter at all monitoring points at each time point. After standardizing the mean value, construct the farm environmental vector from all the standardized mean values at each time point.
5. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 1, characterized in that, The environmental difference consistency is the normalized result of the cosine similarity of the difference vectors between the environmental consistency time and the current time at the same monitoring point.
6. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 1, characterized in that, The method for constructing an environmental difference matrix for all monitoring points based on the consistency of environmental differences is as follows: For each consistent environmental moment, the consistency value between it and the environmental differences corresponding to each monitoring point at the current moment is taken as the value, and the environmental differences between the two moments are constructed according to the positional relationship of each monitoring point.
7. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 1, characterized in that, The method for dividing abnormal regions based on the environmental difference matrix is as follows: Starting with the locations of monitoring points where the environmental difference consistency is less than a preset value, and using the environmental difference consistency between two monitoring points at the same time as the growth basis, the region growth method is used to classify these monitoring points, with each category being an abnormal region.
8. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 1, characterized in that, The method for matching each abnormal region at the current moment with the abnormal regions at the same moment in the environment to obtain the abnormal region matching degree, and classifying the monitoring points to obtain the disinfection points, is as follows: The CPD algorithm is used to calculate the similarity between each abnormal region at the current time and the environmental conditions of each abnormal region at the same time. A probability matrix is constructed, and the average value of all elements in the probability matrix output by CPD is used as the abnormal region matching degree.
9. The fully automated intelligent control method for spray disinfection in a livestock farm as described in claim 8, characterized in that, The method for obtaining the consistency of disinfection requirements based on the similarity between the current time and the environmental description vector at the same time as the environment, as well as the attention given to different disinfection points, is as follows: , Indicates the current time The environmental description vector of the w-th disinfection point. Indicates the time when the environment is consistent at the t-th moment. An environmental description vector for each disinfection point. Used to calculate the cosine similarity between two vectors; Used for normalization, Indicates the selection of a consistent environment. The following disinfection points and the next disinfection point at the current time The maximum value of environmental similarity at the disinfection points. Indicates the number of disinfection points. This indicates the level of attention paid to the w-th disinfection point. This represents the time when the environment is consistent with the current time. The consistency of disinfection requirements; the attention level of the pathogen contamination point is 1, and the attention level of the suspected contamination point is the difference between the number 1 and the matching degree of the abnormal area.
10. A fully automatic intelligent control terminal for spray disinfection in a livestock farm, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the fully automatic spray disinfection intelligent control method for a farm as described in any one of claims 1-9.