A data monitoring method and system for sow behavioral activity

Abstract

The application relates to a pig health monitoring technology field, in particular to a data monitoring method and system for sow behavior activities, a sow single feed scheduling database is deployed, single feed metadata is collected, single feed metadata is analyzed, ingredient attribute data and ingredient correlation data are collected; a feeding monitoring device is deployed in the sow single feed scheduling database, an online compound feed is generated by a feeder based on sow basic data; the online compound feed passing through safety control is analyzed, comprehensive ingredient evaluation parameters are collected, and the feed effect of the sow is monitored, feeding monitoring data is collected, the feeding monitoring data and the comprehensive ingredient evaluation parameters are matched and analyzed, whether reminding data is generated is detected based on the matching analysis result; the feeder responds to the received reminding data and adaptive correction data. The application can effectively correct and feed back the feed compound based on the breeding behavior of the sow after eating the feed, and improve the feed configuration effect.

Description

Technical Field

[0001] This application relates to the field of swine health monitoring technology, and in particular to a data monitoring method and system for sow behavior. Background Technology

[0002] Pig breeding faces an increasingly severe threat of disease, with complex and variable disease patterns, escalating production risks, significantly increased control costs, and incalculable economic losses. With the rapid development of pig breeding and the feed industry, antibiotics, hormones, and synthetic chemicals are added to animal feed to promote growth and reduce disease. However, long-term use of these drugs can lead to drug resistance in pathogens, damage to the animal's immune system, and weakened immunity. Furthermore, the products contain large amounts of drug residues, pollute the environment, and seriously threaten the health of breeding pigs.

[0003] Health monitoring of breeding pigs, especially sows, is particularly important as it plays a crucial role in the reproduction of the herd. The effectiveness of compound feed and the monitoring of sows' behavior after consuming it can help to ensure the rationality of the feed in a timely manner, but there is a lack of relevant technical means in the current technology. Summary of the Invention

[0004] To achieve the above objectives, this application provides the following technical solution: According to a first aspect of the present invention, the present invention claims protection for a data monitoring method for sow behavioral activities, comprising the following steps: Step S1: Deploy the sow single-feed scheduling database, collect single-feed metadata, parse the single-feed metadata, collect component attribute data and component association data, and generate single-feed triples based on the received data; Step S2: Deploy feeding monitoring equipment in the sow single-feed scheduling database, collect basic data of sows based on the feeding monitoring equipment, and have the feeder generate online compound feed based on the basic data of sows; Step S3: Associate the received online compound feed with the single feed ternary set to generate a compound feed association tuple. Based on the compound feed association tuple, detect whether the online compound feed has any risks. Collect reproductive safety control data associated with sows, detect whether online compound feeds without risks meet the requirements for associating with sows, and detect whether the online compound feed has passed safety control based on the detection results. Step S4: Analyze the online compound feed that has passed the safety control, collect comprehensive component evaluation parameters, monitor the feed effect on sows, collect feeding monitoring data, match and analyze the received feeding monitoring data and comprehensive component evaluation parameters, and detect whether to generate reminder data based on the matching analysis results; Step S5: Preset the component monitoring time period, collect the behavioral activity data of the associated sows during the component monitoring time period, deploy adaptive correction data based on the behavioral activity data, and transmit it to the associated feeding identifier. The feeder responds based on the received reminder data and adaptive correction data.

[0005] Pig behavior is further defined as posture (instantaneous behavior) and behavior spanning a time span. Instantaneous behaviors include lying down (side-lying down, prone) and standing; behaviors spanning a time span include walking, eating, exploration (sniffing, etc.), and mating. The four most frequent behaviors in pigs are lying down, standing, walking, and exploration. These behaviors reflect the daily life and health status of pigs, providing useful information for early warning of abnormal behavior, early disease diagnosis, and environmental control in pig farms. Routine monitoring of pig behavior primarily uses image collection for posture (instantaneous behavior) data, while video capture is used for behaviors spanning a time span. Target detection and behavior recognition technologies are used to monitor behavior during pig farming, including both group status and individual behavior.

[0006] Based on the above-mentioned behaviors, the behavior of sows is monitored as a key focus after feeding. Among them, reproductive behavior refers to the postures (instantaneous behaviors) and behaviors with a time span performed by the sows.

[0007] Furthermore, the process of deploying the sow single-feed scheduling database and generating single-feed triplets includes: A sow single-unit feed scheduling database is deployed, and a component input interface is deployed within the sow single-unit feed scheduling database. The component input interface is used to query and collect component metadata based on big data. The component metadata includes component name, component attribute, reproductive data, component interaction data, and component performance data. The received component metadata is parsed, and all component attribute data in the sow's single-component feed scheduling database are collected. A classification system structure is generated based on the component attribute data, and component attribute nodes are deployed based on the associated component attribute data. The mutual effects between each component attribute node are collected based on the mutual influence of components. A single-component feed knowledge graph is generated based on the received mutual effects. The single-component feed triplet includes the mutual influence effects between different component names. The mutual influence effects include positive effects, negative effects, and neutral effects.

[0008] Furthermore, the process of generating online compound feed includes: The sow feed scheduling database is equipped with feeding monitoring equipment, which stores associated feeding identifiers and sow identifiers. Feeders monitor and process the corresponding sow identifiers based on the associated feeding identifiers in the feeding monitoring equipment. The sow submits basic sow data through the sow identifier, which includes individual sow data, veterinary treatment data, and sow symptom data; the basic sow data is then transferred to the feeding identifier based on the veterinary treatment data included therein. The feeder collects and associates basic sow data received from the feeding identifier with the sow identifier, monitors the corresponding sows, and generates sow monitoring results and associated online compound feed based on the monitoring results. The online compound feed includes associated ingredient names and feed standards. The online compound feed associated with the sow identifier and the sow monitoring results are standardized, marked, and saved.

[0009] Furthermore, the process for detecting whether online compound feed poses a risk includes: The sow single-feed scheduling database is equipped with a component analysis device. The component analysis device collects online compound feeds associated with the sow identifier, analyzes them, and detects whether the online compound feeds have any risks. The component analysis device stores the single-component feed ternary sets received in the database. It associates the component names associated with the sow identifier in the online compound feed with the single-component feed ternary sets, and collects the compound feed associated tuples associated with the sow identifier in the online compound feed. The compound feed associated tuples include the component attributes associated with the associated component names and the mutual influence between them. The component attributes in the compound feed associated tuple are randomly arranged. Each arrangement collected by the random arrangement is recorded as a component arrangement subset, and the received component arrangement subset is recorded as a component arrangement set. The mutual influence between the associated components in each component arrangement subset within the component arrangement set is detected. If there is no negative effect association among the component arrangement subsets, the online compound feed passes the safety control. If there is, the component attributes in the component arrangement subset are further analyzed. The component analysis device collects the component names and sow monitoring results within the component arrangement subset. It is equipped with a cloud-based expert database, which includes feeding experts. The component arrangement subset with negative effects and the sow monitoring results are transmitted to the cloud-based expert database for verification by the feeding experts. If the verification is qualified, the online compound feed is not risky; if it is not qualified, it is risky.

[0010] Furthermore, the process of controlling reproductive safety in sows, collecting reproductive safety control data, and matching the non-risk-prone online compound feed with the reproductive safety control data includes: The component analysis device collects online compound feed that is not risky, collects the names of the components contained in the online compound feed, and collects the reproductive data associated with each component name; it collects the associated sow identifiers based on the online compound feed, performs reproductive safety control on sows based on the received sow identifiers and reproductive data, and collects reproductive safety control data. The component analysis device analyzes the reproductive safety control data received by the corresponding sow. When the reproductive data of all component names in the online compound feed matches the associated reproductive safety control data, the online compound feed passes the safety control. If there is a mismatch, the online compound feed fails the safety control. The component analysis device generates compound feed risk data for online compound feeds that fail the safety control or are at risk, and transmits it to the sow single-component feed scheduling database.

[0011] Furthermore, the process of collecting component evaluation parameters and feeding monitoring data, and detecting whether to generate reminder data based on the matching results, includes: Collect online compound feeds associated with sow identifiers, collect component performance data associated with the names of each component in the online compound feeds, extract regular features from the received component performance data, and collect component performance data. The sow single-component feed scheduling database stores the component performance data associated with the component names. Based on big data algorithms, historical component performance datasets associated with historical online compound feeds are collected. Based on the historical component performance datasets, a component evaluation model is generated using deep learning algorithms. The component performance data of the received online compound feeds are input into the component evaluation model to collect comprehensive component evaluation parameters. The sow single-component feed scheduling database is equipped with feeding monitoring equipment. The feeding monitoring equipment is used to collect the comprehensive component evaluation parameters of the associated sows and the online compound feed, and to monitor their feeding. The equipment collects feeding monitoring data and checks whether the received feeding monitoring data meets the comprehensive component evaluation parameters. If it does, no reminder data is generated; if it does not, a monitoring reminder data is generated.

[0012] Furthermore, the process of deploying adaptive correction data includes: Within a preset component monitoring time period, feeding monitoring data and comprehensive component evaluation parameters of associated sows are collected within a preset period. Monitoring fluctuation images and evaluation fluctuation images related to the feeding monitoring data and comprehensive component evaluation parameters are generated for the preset period. Based on these monitoring fluctuation images and evaluation fluctuation images, associated difference information is collected. A preset difference threshold is set, and the absolute value of the received difference information is collected. The received absolute value and the difference threshold are matched and analyzed. When the absolute value is greater than the difference threshold, the direction of the difference information associated with the absolute value is collected. When the direction is positive, it is marked as reproductive safety; when the direction is negative, it is marked as reproductive risk. Based on the preset period proportion of reproductive safety and reproductive risk, the received preset period proportion is recorded as adaptive evaluation data. A preset correction threshold is set. When the adaptive evaluation data is greater than the correction threshold, adaptive correction data is generated.

[0013] Furthermore, the feeding identifier associated with the sow single-feed scheduling database is based on the compound feed risk data, monitoring reminder data, and adaptive correction data associated with the sow identifier, and is responded and corrected by the feeder based on the received data.

[0014] According to a second aspect of the present invention, the present invention claims protection for a data monitoring system for sow behavior, comprising: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, enable the one or more processors to implement the data monitoring method for sow behavior.

[0015] This application relates to the field of sow health monitoring technology, and more particularly to a data monitoring method and system for sow behavior. The method involves deploying a sow single-feed scheduling database, collecting single-feed metadata, parsing the metadata to collect component attribute data and component correlation data, deploying feeding monitoring equipment within the sow single-feed scheduling database, and having the feeder generate online compound feed based on basic sow data. The online compound feed, which passes safety control, is analyzed to collect comprehensive component evaluation parameters and monitor the feed effect on the sows. Feeding monitoring data is collected, and the feeding monitoring data and comprehensive component evaluation parameters are matched and analyzed. Based on the matching analysis results, a notification is generated. The feeder responds based on the received notification data and adaptive correction data. This invention can effectively correct and provide feedback on feed compounding based on the sow's reproductive behavior after consuming feed, improving feed formulation effectiveness. Attached Figure Description

[0016] Figure 1This is a flowchart illustrating a data monitoring method for sow behavior activities claimed in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments accepted by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0018] The terms "first," "second," and "third" in this application are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional effect, movement, etc., between components in a specific orientation (as shown in the figures). If the specific orientation changes, the directional indication will change accordingly. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a neutral or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] like Figure 1 As shown, a data monitoring method for sow behavior includes the following steps: Step S1: Deploy the sow single-feed scheduling database, collect single-feed metadata, parse the single-feed metadata, collect component attribute data and component association data, and generate single-feed triples based on the received data; Step S2: Deploy feeding monitoring equipment in the sow single-feed scheduling database, collect basic data of sows based on the feeding monitoring equipment, and have the feeder generate online compound feed based on the basic data of sows; Step S3: Associate the received online compound feed with the single feed ternary set to generate a compound feed association tuple. Based on the compound feed association tuple, detect whether the online compound feed has any risks. Collect reproductive safety control data associated with sows, detect whether online compound feeds without risks meet the requirements for associating with sows, and detect whether the online compound feed has passed safety control based on the detection results. Step S4: Analyze the online compound feed that has passed the safety control, collect comprehensive component evaluation parameters, monitor the feed effect on sows, collect feeding monitoring data, match and analyze the received feeding monitoring data and comprehensive component evaluation parameters, and detect whether to generate reminder data based on the matching analysis results; Step S5: Preset the component monitoring time period, collect the behavioral activity data of the associated sows during the component monitoring time period, deploy adaptive correction data based on the behavioral activity data, and transmit it to the associated feeding identifier. The feeder responds based on the received reminder data and adaptive correction data.

[0021] In this embodiment, pig behavior is further defined as posture (instantaneous behavior) and behavior spanning a time span. Instantaneous behaviors include lying down (side-lying down, prone) and standing; behaviors spanning a time span include walking, eating, exploration (sniffing, etc.), and mating. The four most frequent pig behaviors are lying down, standing, walking, and exploration. These behaviors reflect the daily life and health status of pigs and can provide useful information for early warning of abnormal pig behavior, early disease diagnosis, and control of the pig farm environment. Daily monitoring of pig behavior primarily uses image acquisition for posture (instantaneous behavior) data and video capture for behaviors spanning a time span. Target detection and behavior recognition technologies are used to monitor pig behavior during the pig farming process, including the group status and individual behavior of pigs.

[0022] Based on these behaviors, the sows' behavior is monitored as a key focus after feeding.

[0023] Among them, reproductive behavior refers to the postures (instantaneous behaviors) performed by the sow and behaviors that span a time span.

[0024] It should be further explained that, in the specific implementation process, the process of deploying the sow single-feed scheduling database, collecting single-feed metadata, parsing the single-feed metadata, collecting component attribute data and component association data, and generating single-feed triples based on the received data includes: A sow single-unit feed scheduling database is deployed, and a component input interface is deployed within the sow single-unit feed scheduling database. The component input interface is used to collect component metadata based on big data queries. The component metadata includes component name, component attribute, reproductive data, component interaction data, and component performance data. The component attribute includes the component's influence mechanism, the component's behavior correction range, and the behavior correction effect. The reproductive data is the reproductive source data associated with the component name. The component interaction data refers to the mutual influence effect of different component names. The received component metadata is classified and processed. All component attribute data in the sow's single-component feed scheduling database are collected. A classification system structure is generated based on the component attribute data, and component attribute nodes are deployed based on the associated component attribute data. The mutual effects between various component attribute nodes are collected based on the mutual influence of components. A single-component feed knowledge graph is generated based on the received mutual effects. The single-component feed triple includes the mutual influence effects between different component metadata. The mutual influence effects include positive effects, negative effects, and neutral effects. The positive effects are mutually promoting effects, the negative effects are mutually opposing effects, and the neutral effects are mutually neutral effects.

[0025] It should be further explained that, in the specific implementation process, the process of deploying feeding monitoring equipment within the sow single-feed scheduling database, collecting basic sow data based on the feeding monitoring equipment, and having the feeder generate online compound feed based on the basic sow data includes: The sow feed scheduling database is equipped with feeding monitoring equipment, which stores associated feeding identifiers and sow identifiers. Feeders monitor and process the corresponding sow identifiers based on the associated feeding identifiers in the feeding monitoring equipment. The sow submits basic sow data through the sow identifier, which includes individual sow data, veterinary treatment data, and sow symptom data; the basic sow data is then transferred to the feeding identifier based on the veterinary treatment data included therein. The feeder collects and associates basic sow data received from the feeding identifier with the sow identifier, monitors the corresponding sows, and generates sow monitoring results and associated online compound feed based on the monitoring results. The online compound feed includes associated ingredient names and feed standards. The online compound feed associated with the sow identifier and the sow monitoring results are standardized, marked, and saved.

[0026] It should be further explained that, in the specific implementation process, the process of associating the received online compound feed with the single-ingredient feed ternary set to generate a compound feed association tuple, detecting whether the online compound feed poses a risk based on the compound feed association tuple, collecting reproductive safety control data associated with sows, detecting whether online compound feeds without risk meet the requirements for associating with sows, and detecting whether the online compound feed has passed safety control based on the detection results includes: The sow single-feed scheduling database is equipped with a component analysis device. The component analysis device collects online compound feeds associated with the sow identifier, analyzes them, and detects whether the online compound feeds have any risks. The component analysis device stores the single-component feed ternary sets received in the database. It associates the component names associated with the sow identifier in the online compound feed with the single-component feed ternary sets, and collects the compound feed associated tuples associated with the sow identifier in the online compound feed. The compound feed associated tuples include the component attributes associated with the associated component names and the mutual influence between them. The component attributes in the compound feed associated tuple are randomly arranged. Each arrangement collected by the random arrangement is recorded as a component arrangement subset, and the received component arrangement subset is recorded as a component arrangement set. The mutual influence between the associated components in each component arrangement subset within the component arrangement set is detected. If there is no negative effect association among the component arrangement subsets, the online compound feed passes the safety control. If there is, the component attributes in the component arrangement subset are further analyzed. The component analysis device collects component names and sow monitoring results within a component arrangement subset. A cloud-based expert database, including feeding experts, is deployed there. Component arrangement subsets with negative effects and sow monitoring results are transmitted to the cloud-based expert database for verification by feeding experts. This involves detecting whether related components influence each other and generating verification results based on these results. If the verification is successful, the online compound feed is considered risk-free; otherwise, it is considered risky.

[0027] The component analysis device collects online compound feed that is not risky, collects the names of the components contained in the online compound feed, and collects the reproductive data associated with each component name; it collects the associated sow identifiers based on the online compound feed, performs reproductive safety control on sows based on the received sow identifiers and reproductive data, and collects reproductive safety control data. The component analysis device analyzes the reproductive safety control data received by the corresponding sow. When the reproductive data of all component names in the online compound feed matches the associated reproductive safety control data, the online compound feed passes the safety control. If there is a mismatch, the online compound feed fails the safety control. The component analysis device generates compound feed risk data for online compound feeds that fail the safety control or are at risk, and transmits it to the sow single-component feed scheduling database.

[0028] It should be further explained that, in the specific implementation process, the process of analyzing the online compound feed that has passed safety control, collecting comprehensive component evaluation parameters, monitoring the feed effect on sows, collecting feeding monitoring data, matching and analyzing the received feeding monitoring data and comprehensive component evaluation parameters, and detecting whether to generate reminder data based on the matching analysis results includes: Collect online compound feeds associated with sow identifiers, collect component performance data associated with the names of each component in the online compound feeds, extract regular features from the received component performance data, and collect component performance data. The sow single-component feed scheduling database stores the component performance data associated with the component names. Based on big data algorithms, historical component performance datasets associated with historical online compound feeds are collected. Based on the historical component performance datasets, a component evaluation model is generated using deep learning algorithms. The component performance data of the received online compound feeds are input into the component evaluation model to collect comprehensive component evaluation parameters. The sow single-component feed scheduling database is equipped with feeding monitoring equipment. The feeding monitoring equipment is used to collect the comprehensive component evaluation parameters of the associated sows and the online compound feed, and to monitor their feeding. The equipment collects feeding monitoring data and checks whether the received feeding monitoring data meets the comprehensive component evaluation parameters. If it does, no reminder data is generated; if it does not, a monitoring reminder data is generated.

[0029] It should be further explained that, in the specific implementation process, the preset component monitoring time period involves collecting behavioral activity data of the associated sows within that time period, deploying adaptive correction data based on the behavioral activity data, and transmitting it to the associated feeding identifier. The process by which the feeder responds based on the received alert data and adaptive correction data includes: Within a preset component monitoring time period, feeding monitoring data and comprehensive component evaluation parameters of associated sows are collected within a preset period. Monitoring fluctuation images and evaluation fluctuation images related to the feeding monitoring data and comprehensive component evaluation parameters are generated for the preset period. The difference information collected from the evaluation fluctuation images within the preset period is subtracted from the monitoring fluctuation images. A preset difference threshold is set, and the absolute value of the received difference information is collected. The received absolute value and the difference threshold are matched and analyzed. When the absolute value is greater than the difference threshold, the direction of the difference information associated with the absolute value is determined. When the value is positive, it is marked as reproductive safety; when the value is negative, it is marked as reproductive risk. A preset periodic percentage is established based on the correlation between reproductive safety and reproductive risk, and the received preset periodic percentage is recorded as adaptive evaluation data. A preset correction threshold is set. When the adaptive evaluation data exceeds the correction threshold, adaptive correction data is generated. This adaptive correction data includes the performance of sows on the effects of the ingredients during ingredient use. Correction suggestions are generated based on the performance of the ingredients on the dosage of the associated ingredients in the online compound feed. The feeding identifier associated with the sow single-component feed scheduling database is based on the compound feed risk data, monitoring reminder data, and adaptive correction data associated with the sow identifier, and is responded and corrected by the feeder based on the received data.

[0030] According to a second embodiment of the present invention, the present invention claims protection for a data monitoring system for sow behavior, comprising: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, enable the one or more processors to implement the data monitoring method for sow behavior.

[0031] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0032] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can be physically separate, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0033] The specific embodiments of the invention have been described in detail above, but they are only examples, and this application is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this application. Therefore, all equivalent changes, modifications, and improvements made without departing from the spirit and principles of this application should be covered within the scope of this application.

Claims

1. A data monitoring method for sow behavior, characterized in that, Includes the following steps: Step S1: Deploy the sow single-feed scheduling database, collect single-feed metadata, parse the single-feed metadata, collect component attribute data and component association data, and generate single-feed triples based on the received data; Step S2: Deploy feeding monitoring equipment in the sow single-feed scheduling database, collect basic data of sows based on the feeding monitoring equipment, and have the feeder generate online compound feed based on the basic data of sows; Step S3: Associate the received online compound feed with the single feed ternary set to generate a compound feed association tuple. Based on the compound feed association tuple, detect whether the online compound feed has any risks. Collect the reproductive safety control data associated with the sows and detect whether the online compound feed that has no risks matches the reproductive safety control data of the associated sows. Based on the detection results, detect whether the online compound feed has passed the safety control. Step S4: Analyze the online compound feed that has passed the safety control, collect comprehensive component evaluation parameters, monitor the feeding effect of sows, collect feeding monitoring data, match and analyze the received feeding monitoring data and comprehensive component evaluation parameters, and detect whether to generate reminder data based on the matching analysis results; Step S5: Preset the component monitoring time period, collect the behavioral activity data of the associated sows during the component monitoring time period, deploy adaptive correction data based on the behavioral activity data, and transmit it to the associated feeding identifier. The feeder responds based on the received reminder data and adaptive correction data. S1 includes: A sow single-unit feed scheduling database is deployed, and a component input interface is deployed within the sow single-unit feed scheduling database. The component input interface is used to query and collect component metadata based on big data. The component metadata includes component name, component attribute, reproductive data, component interaction data, and component performance data. The received component metadata is parsed, and all component attribute data in the sow single-component feed scheduling database are collected. A classification system structure is generated based on the component attribute data, and component attribute nodes are deployed based on the associated component attribute data. The mutual effects between each component attribute node are collected based on the mutual influence of components. A single-component feed knowledge graph is generated based on the received mutual effects. The single-component feed triplet includes the mutual influence effects between different component names. The mutual influence effects include positive effects, negative effects, and neutral effects. The process for detecting whether online compound feed poses a risk includes: The sow single-feed scheduling database is equipped with a component analysis device. The component analysis device collects online compound feeds associated with the sow identifier, analyzes them, and detects whether the online compound feeds have any risks. The component analysis device stores the single-component feed ternary sets received in the database. It associates the sow identifier with the component name associated in the online compound feed to the single-component feed ternary set, and collects the compound feed association tuples associated with the sow identifier and the online compound feed. The compound feed association tuples include the component attributes associated with the associated component names and the mutual influence between them. The component attributes in the compound feed associated tuple are randomly arranged. Each arrangement collected by the random arrangement is recorded as a component arrangement subset, and the received component arrangement subset is recorded as a component arrangement set. The mutual influence between the associated components in each component arrangement subset within the component arrangement set is detected. If there is no negative effect association among the component arrangement subsets, the online compound feed passes the safety control. If there is, the component attributes in the component arrangement subset are further analyzed. The component analysis device collects the component names and sow monitoring results within the component arrangement subset. It is equipped with a cloud-based expert database, which includes feeding experts. The component arrangement subset with negative effects and the sow monitoring results are transmitted to the cloud-based expert database for verification by the feeding experts. If the verification is qualified, the online compound feed is not risky; if it is not qualified, it is risky.

2. The data monitoring method for sow behavior as described in claim 1, characterized in that, The process of generating online compound feed includes: The sow feed scheduling database is equipped with feeding monitoring equipment, which stores associated feeding identifiers and sow identifiers. Feeders monitor and process the corresponding sow identifiers based on the associated feeding identifiers in the feeding monitoring equipment. Sow basic data is submitted through the sow identifier, which includes sow individual data, veterinary treatment data, and sow symptom data; the sow basic data is then transferred to the feeding identifier based on the veterinary treatment data included therein. The feeder collects and associates basic sow data received from the feeding identifier with the sow identifier, monitors the corresponding sows, and generates sow monitoring results and associated online compound feed based on the monitoring results. The online compound feed includes associated ingredient names and feed standards. The online compound feed associated with the sow identifier and the sow monitoring results are standardized, marked, and saved.

3. The data monitoring method for sow behavior as described in claim 2, characterized in that, The S3 step of detecting whether the risk-free online compound feed matches the reproductive safety control data of the associated sow includes: The component analysis device collects online compound feed that is not at risk, collects the names of the components contained in the online compound feed, and collects the reproductive data associated with each component name; based on the sow identifiers associated with the online compound feed, it performs reproductive safety control on the sows based on the received sow identifiers and reproductive data, and collects reproductive safety control data; The component analysis device analyzes the reproductive safety control data received by the corresponding sow. When the reproductive data of all component names in the online compound feed matches the associated reproductive safety control data, the online compound feed passes the safety control. If the match fails, the online compound feed fails the safety control. The component analysis device generates compound feed risk data for online compound feeds that fail the safety control or are at risk, and transmits it to the sow single-component feed scheduling database.

4. The data monitoring method for sow behavior as described in claim 3, characterized in that, The process of collecting comprehensive component evaluation parameters and feeding monitoring data, and detecting whether to generate reminder data based on the matching results, includes: Collect online compound feeds associated with sow identifiers, collect component performance data associated with the names of each component in the online compound feeds, extract regular features from the received component performance data, and collect component performance data. The sow single-component feed scheduling database stores the component performance data associated with the component names. Based on big data algorithms, historical component performance datasets associated with historical online compound feeds are collected. Based on the historical component performance datasets, a component evaluation model is generated using deep learning algorithms. The component performance data of the received online compound feeds are input into the component evaluation model to collect comprehensive component evaluation parameters. The sow single-component feed scheduling database is equipped with feeding monitoring equipment. The feeding monitoring equipment is used to collect feeding monitoring data associated with the associated sow identifier and the online compound feed, and to determine whether the received feeding monitoring data meets the comprehensive component evaluation parameters. If it does, no reminder data is generated; if it does not, monitoring reminder data is generated.

5. The data monitoring method for sow behavior as described in claim 4, characterized in that, The process of deploying adaptive correction data includes: Within a preset component monitoring time period, feeding monitoring data and comprehensive component evaluation parameters of associated sows are collected within a preset period. Monitoring fluctuation images and evaluation fluctuation images related to the feeding monitoring data and comprehensive component evaluation parameters are generated for the preset period. Based on these monitoring fluctuation images and evaluation fluctuation images, associated difference information is collected. A preset difference threshold is set, and the absolute value of the received difference information is collected. The received absolute value and the difference threshold are matched and analyzed. When the absolute value is greater than the difference threshold, the direction of the difference information associated with the absolute value is collected. When the direction is positive, it is marked as reproductive safety; when the direction is negative, it is marked as reproductive risk. Based on the preset period proportion of reproductive safety and reproductive risk, the received preset period proportion is recorded as adaptive evaluation data. A preset correction threshold is set. When the adaptive evaluation data is greater than the correction threshold, adaptive correction data is generated.

6. The data monitoring method for sow behavior as described in claim 5, characterized in that, The feeding identifier associated with the sow single-component feed scheduling database is based on the compound feed risk data, monitoring reminder data, and adaptive correction data associated with the sow identifier, and is responded and corrected by the feeder based on the received data.

7. A data monitoring system for sow behavior, characterized in that, include: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, cause the one or more processors to implement a data monitoring method for sow behavior activities according to any one of claims 1 to 6.

Images