An intelligent monitoring system and method for laying hen breeding environment based on Internet of Things

Abstract

This invention relates to the field of egg-laying hen farming technology, specifically to an intelligent monitoring system and method for the egg-laying hen farming environment based on the Internet of Things (IoT). The invention clusters first and second negative feedback monitoring data that lead to a decline in egg production or quality; establishes a negative feedback data analysis model to analyze the first and second negative feedback monitoring data and positive feedback monitoring parameters, generating first and second abnormal farming monitoring parameters; and calculates the similarity between the egg-laying hen farming environment and the first and second abnormal farming monitoring parameters by real-time monitoring. If the calculated result is not less than the farming environment similarity threshold, a corresponding early warning is issued. This method can summarize and analyze historical abnormal farming parameters, facilitating early warning of these abnormal data when monitoring or adjusting farming environment monitoring parameters in subsequent operations, reducing the workload of relevant personnel, and improving the accuracy of farming environment monitoring parameter research.

Description

Technical Field

[0001] This invention relates to the field of egg-laying hen farming technology, specifically to an intelligent monitoring system and method for egg-laying hen farming environment based on the Internet of Things. Background Technology

[0002] Egg-laying hens are chickens specifically raised for egg production. Unlike broiler chickens, which focus on improving the quality of the chicken meat, egg-laying hens prioritize improving the quality and quantity of their eggs.

[0003] Besides the breed of laying hens themselves affecting egg production and quality, the environmental conditions of the laying hen's rearing also influence egg production and quality, such as ambient temperature, humidity, and feeding methods. To ensure the quality and quantity of eggs produced by laying hens, the main focus is on adjusting the monitoring parameters of the laying hen's rearing environment to improve egg quality and production.

[0004] With the development of emerging technologies such as the Internet of Things and big data, these technologies have also been adopted to adjust environmental monitoring parameters in egg-laying hen farming. However, the problem is that these existing technologies primarily improve the accuracy of environmental monitoring parameters or obtain more parameters affecting egg production and quality. They do not summarize environmental monitoring parameters that lead to a decline in egg production and quality, thus failing to avoid these negative feedback parameters. These negative feedback parameters could help researchers automatically avoid abnormal data when studying and adjusting environmental monitoring parameters, reducing workload and improving the accuracy of research on environmental monitoring parameters.

[0005] To address this, an intelligent monitoring system and method for egg-laying hen farming environment based on the Internet of Things is proposed. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent monitoring system and method for egg-laying hen farming environment based on the Internet of Things (IoT). This system acquires negative feedback monitoring parameters that lead to a decline in egg production or quality; it then uses a clustering algorithm to obtain relevant environmental parameters, generating first and second negative feedback monitoring data. A negative feedback data analysis model is established to analyze abnormal data from the first, second, and positive feedback monitoring parameters, generating first and second abnormal farming monitoring parameters. By monitoring the egg-laying hen farming environment in real time, the similarity to the first and second abnormal farming monitoring parameters is calculated. If the calculated result is not greater than a farming environment similarity threshold, a corresponding warning is issued. This method can summarize historical data and analyze and store abnormal farming parameters, facilitating the automatic avoidance of these abnormal data when monitoring or adjusting farming environment monitoring parameters in subsequent operations, improving the work efficiency of relevant personnel, and enhancing the accuracy of farming environment monitoring parameter research.

[0007] To achieve the above objectives, the present invention provides the following technical solution: An IoT-based intelligent monitoring system for egg-laying hen farming environment includes a negative feedback data acquisition module, a negative feedback data classification module, a positive feedback data acquisition module, a negative feedback data analysis module, a real-time monitoring module for the farming environment, a first farming monitoring and early warning module, and a second farming monitoring and early warning module. The negative feedback data acquisition module acquires negative feedback monitoring parameters of the egg-laying hen farming environment; The positive feedback data acquisition module acquires positive feedback monitoring parameters of the egg-laying hen farming environment; the positive feedback monitoring parameters are time-series data. The negative feedback data classification module performs clustering and classification based on the positive feedback monitoring parameters and the negative feedback results of the negative feedback monitoring parameters; it generates first negative feedback monitoring data and second negative feedback monitoring data; the first negative feedback monitoring data is negative feedback data caused by fowl plague; the second negative feedback monitoring data is negative feedback data not caused by fowl plague; the negative feedback data is time series data; The negative feedback data analysis module establishes a negative feedback data analysis model, analyzes the first negative feedback monitoring data and the second negative feedback monitoring data based on the negative feedback data analysis model, generates and stores the first abnormal aquaculture monitoring parameters, and compares the negative feedback monitoring parameters and the positive feedback monitoring parameters based on the negative feedback data analysis model to generate and store the second abnormal aquaculture monitoring parameters. Furthermore, the negative feedback data analysis model includes a positive and negative feedback data processing unit, a positive and negative feedback data feature extraction unit, a first abnormal aquaculture monitoring parameter analysis unit, a second abnormal aquaculture monitoring parameter analysis unit, an abnormal data output storage unit, and an abnormal similarity calculation unit; The positive and negative feedback data processing unit filters the first negative feedback monitoring data, the second negative feedback monitoring data, and the positive feedback monitoring parameters according to the laying hen breed and feeding type, and preprocesses the data to generate preprocessed feedback data. The positive and negative feedback data feature extraction unit extracts features from the preprocessed feedback data based on Transformer, GRU, dilated convolution and feature fusion layer, and generates first feedback feature data, second feedback feature data and positive feedback feature data; The first abnormal aquaculture monitoring parameter analysis unit generates the first abnormal aquaculture monitoring parameters based on the abnormal feature data of the first feedback feature data and the second feedback feature data. Furthermore, the first feedback feature data or the second feedback feature data is divided equally according to the sliding time window to obtain multiple feedback feature data, which include trend feature data and morphological feature data. The t-th feedback feature data and the (t-1)-th feedback feature data are used to perform a first feedback similarity calculation; the first feedback similarity calculation includes trend feature similarity calculation and morphological feature similarity calculation; The trend feature similarity calculation uses the standard deviation method, empirical statistical method, and dynamic path calculation method; the morphological feature similarity calculation uses the dynamic path calculation method and image recognition analysis method; the specific calculations are as follows: ; ; in, The result of the similarity calculation for the trend features. The standard deviation method, The empirical statistical method is described above. This refers to the dynamic path calculation method. For the t-th feedback feature data, For the (t-1)th feedback feature data, , and The weights calculated for the similarity of the trend features. ; The result is the morphological feature similarity calculation, and C represents the transformation of heatmap data. The image recognition and analysis method is described above. and The weights calculated for the morphological feature similarity; ; If the first feedback similarity calculation result is less than the first feedback similarity threshold, then the t-th feedback feature data is the first abnormal aquaculture monitoring parameter; The second abnormal aquaculture monitoring parameter analysis unit generates the second abnormal aquaculture monitoring parameters by comparing the first feedback feature data or the second feedback feature data with the positive feedback feature data. Furthermore, the first feedback feature data, second feedback feature data, and positive feedback feature data are equally divided according to the sliding time window to obtain multiple negative feedback feature data and multiple positive feedback feature data. The feedback feature data includes trend feature data. A second feedback similarity calculation is performed between the k-th negative feedback feature data and the k-th positive feedback feature data. The second feedback similarity calculation is a trend feature similarity calculation. The trend feature similarity calculation uses the standard deviation method, empirical statistical method, and dynamic path calculation method. The specific calculation is as follows: ; in, For the k-th negative feedback feature data, This refers to the k-th positive feedback feature data. If the second feedback similarity calculation result is less than the second feedback similarity threshold, then the kth negative feedback feature is the second abnormal aquaculture monitoring parameter; The abnormal data output storage unit tags the first abnormal aquaculture monitoring parameter and the second abnormal aquaculture monitoring parameter with corresponding tags, and outputs and stores them into the corresponding database. The anomaly similarity calculation unit compares the real-time monitored egg-laying hen breeding environment with the first and second abnormal breeding monitoring parameters to generate anomaly similarity. The negative feedback data analysis model compares the egg-laying hen breeding environment with the first and second abnormal breeding monitoring parameters, and after data processing and feature extraction, calculates trend feature similarity and morphological feature similarity. The trend feature similarity calculation uses the standard deviation method, empirical statistical method, and dynamic path calculation method. The morphological feature similarity calculation method is a dynamic path calculation method and an image recognition analysis method. The real-time monitoring module for the breeding environment monitors the breeding environment of the laying hens in real time and compares the breeding environment of the laying hens with the first abnormal breeding monitoring parameter and the second abnormal breeding monitoring parameter through the negative feedback data analysis model. If the first abnormal aquaculture monitoring and early warning module detects that the first abnormal aquaculture monitoring parameter or the second abnormal aquaculture monitoring parameter is not less than the aquaculture environment similarity threshold, it will issue a first aquaculture monitoring early warning for adjustment. The second aquaculture monitoring and early warning module issues a second aquaculture monitoring early warning if the first or second abnormal aquaculture monitoring parameter adjusted by relevant personnel is not less than the aquaculture environment similarity threshold. This invention also provides an intelligent monitoring method for the egg-laying hen farming environment based on the Internet of Things, comprising: Obtain negative feedback monitoring parameters of the egg-laying hen breeding environment; perform clustering and classification based on the positive feedback monitoring parameters and the negative feedback results of the negative feedback monitoring parameters; generate first negative feedback monitoring data and second negative feedback monitoring data; Obtain positive feedback monitoring parameters of the egg-laying hen farming environment; establish a negative feedback data analysis model, analyze the first negative feedback monitoring data and the second negative feedback monitoring data according to the negative feedback data analysis model, generate and store the first abnormal farming monitoring parameters; compare the negative feedback monitoring parameters and the positive feedback monitoring parameters according to the negative feedback data analysis model, generate and store the second abnormal farming monitoring parameters; Further, based on the trend feature similarity calculation and morphological feature similarity calculation of the negative feedback data analysis model, the first abnormal aquaculture monitoring parameters are generated; based on the trend feature similarity calculation of the negative feedback data analysis model, the second abnormal aquaculture monitoring parameters are generated; the trend feature similarity calculation is the standard deviation method, the empirical statistical method, and the dynamic path calculation method; the morphological feature similarity calculation is the dynamic path calculation method and the image recognition analysis method. The egg-laying hen breeding environment is monitored in real time, and the egg-laying hen breeding environment is compared with the first abnormal breeding monitoring parameter and the second abnormal breeding monitoring parameter through the negative feedback data analysis model. If the first abnormal aquaculture monitoring parameter or the second abnormal aquaculture monitoring parameter is not less than the aquaculture environment similarity threshold, a first aquaculture monitoring warning will be issued for adjustment. If the first or second abnormal aquaculture monitoring parameter adjusted by relevant personnel is not less than the aquaculture environment similarity threshold, a second aquaculture monitoring warning will be issued.

[0008] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention is based on clustering of abnormal environmental parameters. After establishing a negative feedback data analysis model and reanalyzing the characteristics of each data, in order to ensure the extraction of abnormal data, trend analysis and morphological feature analysis are used to analyze the abnormal characteristics of abnormal data with different results. This method can obtain and store the abnormal characteristics of corresponding environmental parameter data from multiple aspects, which is convenient for subsequent real-time environmental monitoring and comparison. It provides an accurate data basis for the comparison of abnormal environmental parameters, thereby improving the accuracy of aquaculture environmental monitoring parameter research.

[0009] 2. This invention aims to further acquire abnormal data of various environmental parameters and enrich the abnormal characteristics of these abnormal environmental parameters; by comparing and calculating trends with normal data, it can uncover some subtle abnormal characteristics of environmental parameters; this method can further uncover abnormal environmental parameters that lead to a decline in egg production or quality; it can further enrich the amount of stored abnormal environmental characteristic data, provide more accurate abnormal data, facilitate real-time monitoring and parameter adjustment research of environmental parameters, and improve the accuracy of research on monitoring parameters of the breeding environment.

[0010] 3. This invention is based on storing abnormal environmental parameter data that affects egg production or quality. By monitoring environmental parameters in real time, it calculates the similarity to abnormal environmental parameter data from both morphological and trend characteristics, thereby enabling real-time monitoring of current environmental parameters. This method can provide reliable and accurate abnormal data characteristics based on the analysis results of historical abnormal environmental data. It facilitates accurate early warning of abnormal environmental parameters when monitoring or adjusting environmental parameters, thereby reducing the workload of relevant personnel and improving the accuracy of research on aquaculture environmental monitoring parameters. Attached Figure Description

[0011] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a virtual device diagram of the positive and negative feedback data processing unit, positive and negative feedback data feature extraction unit, first abnormal aquaculture monitoring parameter analysis unit, second abnormal aquaculture monitoring parameter analysis unit, abnormal data output storage unit and abnormal similarity calculation unit of the negative feedback data analysis model of the present invention. Figure 3 This is a network structure diagram of the positive and negative feedback data feature extraction unit of the present invention; Figure 4 This is the dynamic path diagram of the present invention; Figure 5 This is a heat map of the present invention; Figure 6 This is a flowchart of the method of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] To ensure the yield and quality of eggs from laying hens, the rearing environment is a crucial factor in egg production, in addition to the breed of the hens. Current technologies primarily focus on improving the accuracy of environmental monitoring parameters or acquiring more relevant parameters for optimization, but they fail to summarize the environmental monitoring parameters that negatively impact egg yield and quality. This neglects the fact that these negative feedback parameters can help researchers reduce workload and improve the accuracy of environmental monitoring parameter research. Therefore, this invention provides an intelligent monitoring system and method for the egg-laying hen rearing environment based on the Internet of Things (IoT), referring to... Figure 1 As shown, the technical solution is as follows: The negative feedback data acquisition module acquires negative feedback monitoring parameters of the egg-laying hen farming environment; The positive feedback data acquisition module acquires positive feedback monitoring parameters of the egg-laying hen farming environment; The negative feedback data classification module performs clustering and classification based on the positive feedback monitoring parameters and the negative feedback results of the negative feedback monitoring parameters; and generates first negative feedback monitoring data and second negative feedback monitoring data. The negative feedback data analysis module establishes a negative feedback data analysis model, analyzes the first negative feedback monitoring data and the second negative feedback monitoring data based on the negative feedback data analysis model, generates and stores the first abnormal aquaculture monitoring parameters, and compares the negative feedback monitoring parameters and the positive feedback monitoring parameters based on the negative feedback data analysis model to generate and store the second abnormal aquaculture monitoring parameters. The real-time monitoring module for the breeding environment monitors the breeding environment of the laying hens in real time and compares the breeding environment of the laying hens with the first abnormal breeding monitoring parameter and the second abnormal breeding monitoring parameter through the negative feedback data analysis model. If the first abnormal aquaculture monitoring and early warning module detects that the first abnormal aquaculture monitoring parameter or the second abnormal aquaculture monitoring parameter is not less than the aquaculture environment similarity threshold, it will issue a first aquaculture monitoring early warning for adjustment. The second aquaculture monitoring and early warning module issues a second aquaculture monitoring early warning if the first or second abnormal aquaculture monitoring parameter adjusted by relevant personnel is not less than the aquaculture environment similarity threshold.

[0014] By accessing previously collected negative feedback monitoring parameters of the egg-laying hen farming environment and classifying these negative feedback data using a clustering algorithm, the main environmental parameter data leading to a decline in egg production or quality are obtained, generating first and second negative feedback monitoring data. This approach narrows the data scope and improves the accuracy of the analysis. A negative feedback data analysis model is established. Based on the first and second negative feedback monitoring data, the abnormal interval characteristics or abnormal point characteristics of each aquaculture environment data are analyzed to obtain and store the corresponding abnormal aquaculture environment data. In order to further improve the accuracy of the abnormal aquaculture environment data, it is compared with the normal aquaculture environment data to obtain those abnormal aquaculture environments with indistinct abnormal characteristics for data storage, so as to further enrich the accuracy of the abnormal aquaculture environment data. By comparing stored abnormal breeding environment data with real-time monitored egg-laying hen breeding environment data and adjusted egg-laying hen breeding environment data, corresponding early warnings are issued once similar situations occur. This method can avoid some environmental data that are prone to negative feedback during the adjustment and monitoring process, reduce the workload of relevant researchers, and improve the accuracy of breeding environment research.

[0015] Example 1 For specific illustration, the following examples will be used to illustrate the point: An IoT-based intelligent monitoring system for egg-laying hen farming environment includes a negative feedback data acquisition module, a negative feedback data classification module, a positive feedback data acquisition module, a negative feedback data analysis module, a real-time monitoring module for the farming environment, a first farming monitoring and early warning module, and a second farming monitoring and early warning module. The negative feedback data acquisition module acquires negative feedback monitoring parameters of the egg-laying hen farming environment; The positive feedback data acquisition module acquires positive feedback monitoring parameters of the egg-laying hen breeding environment (e.g., but not limited to temperature data, humidity data, light duration data, carbon dioxide concentration data, ammonia concentration data, etc.); the positive feedback monitoring parameters are time-series data. The negative feedback data classification module performs clustering and classification based on the positive feedback monitoring parameters and the negative feedback results of the negative feedback monitoring parameters; it generates first negative feedback monitoring data and second negative feedback monitoring data; the first negative feedback monitoring data is negative feedback data caused by chicken plague (e.g., but not limited to temperature data, humidity data, light duration data, carbon dioxide concentration data, and ammonia concentration data during chicken plague, etc.). The second negative feedback monitoring data is negative feedback data of abnormal egg production and egg quality in non-chicken plague cases (e.g., but not limited to temperature data, humidity data, light duration data, carbon dioxide concentration data, and ammonia concentration data during non-chicken plague abnormalities). The negative feedback data is time-series data. Specifically, using existing negative feedback monitoring parameters, relevant technicians mark the negative feedback monitoring parameters when fowl plague occurs and the negative feedback monitoring parameters when fowl plague does not occur; using temperature data as the horizontal axis and other negative feedback parameters as the vertical axis, a feedback breeding parameter coordinate system is generated. The other negative feedback parameters can only be of one type, for example, temperature as the horizontal axis and humidity as the vertical axis; or temperature as the horizontal axis and light duration data as the vertical axis. The parameters during the occurrence of fowl plague were clustered using the density clustering algorithm (DBSCAN algorithm) to determine the clustering range of negative feedback parameters for various fowl plagues; the positive feedback monitoring parameters were clustered to generate the clustering range of various positive feedback parameters; and the negative feedback parameters when fowl plague did not occur were clustered to determine the clustering range of negative feedback parameters for various fowl plagues. The clustering range of feedback parameters during fowl plague is compared with the clustering range of positive feedback parameters. The different clustering ranges are used to determine the clustering ranges of various key negative feedback parameters during fowl plague. The clustering range of negative feedback parameters when fowl plague does not occur is compared with the clustering ranges of positive feedback parameters and the clustering ranges of key negative feedback parameters during fowl plague. The different clustering ranges are used to determine the clustering ranges of various key negative feedback parameters when fowl plague does not occur. The decline in egg production or quality in laying hens can be caused by both excessive hen mortality and non-mortality factors. To differentiate these different monitoring parameters of the breeding environment, the clustering process further subdivides the monitoring parameters of the breeding environment under different conditions. This allows for the screening of a large amount of breeding environment data, eliminating data on breeding environment factors that are not the main influencing factors. By reducing the total amount of these breeding environment data, the accuracy of subsequent analysis can be improved. The negative feedback data analysis module establishes a negative feedback data analysis model, analyzes the first negative feedback monitoring data and the second negative feedback monitoring data based on the negative feedback data analysis model, generates and stores the first abnormal aquaculture monitoring parameters, and compares the negative feedback monitoring parameters and the positive feedback monitoring parameters based on the negative feedback data analysis model to generate and store the second abnormal aquaculture monitoring parameters. Furthermore, referring to Figure 2 As shown, the negative feedback data analysis model includes a positive and negative feedback data processing unit, a positive and negative feedback data feature extraction unit, a first abnormal breeding monitoring parameter analysis unit, a second abnormal breeding monitoring parameter analysis unit, an abnormal data output storage unit, and an abnormal similarity calculation unit. By establishing the model, these time-series data can be further analyzed. By analyzing the abnormal features of these time-series data, the main abnormal features of these characteristic data that cause the decline in egg production can be identified. Storing these abnormal features can ensure that these negative feedback data can reduce the workload of relevant personnel, improve the efficiency of data use, and improve the accuracy of breeding environment research in subsequent breeding environment adjustment or abnormal analysis. The positive and negative feedback data processing unit filters the first negative feedback monitoring data, the second negative feedback monitoring data, and the positive feedback monitoring parameters according to the egg-laying hen breed and feeding type (e.g., but not limited to floor rearing, cage rearing, etc.), and preprocesses the data to generate preprocessed feedback data. The positive and negative feedback data feature extraction unit extracts features from the preprocessed feedback data based on Transformer, GRU, dilated convolution, and feature fusion layer, generating first feedback feature data, second feedback feature data, and positive feedback feature data; refer to Figure 3 As shown, Transformer corresponds to Figure 3 The converter, GRU corresponds to Figure 3The gated loop unit has a feature fusion layer of ReLU; the first abnormal aquaculture monitoring parameter analysis unit generates the first abnormal aquaculture monitoring parameters based on the abnormal feature data of the first feedback feature data and the second feedback feature data. To ensure the accuracy of feature extraction from the positive and negative feedback data feature extraction unit in the first and second feedback feature data (two types of abnormal feedback feature data), an anomaly classification model is generated by connecting a fully connected layer and a Softmax layer after the Transformer, GRU, dilated convolution, and feature fusion layer. In this embodiment, based on a portion of untrained aquaculture parameter data obtained by relevant technicians, anomaly classification and identification are performed using the anomaly classification model generated by the positive and negative feedback data feature extraction unit. The data includes normal data and abnormal data, with a total of 4 sets of data. The identification results are shown in Table 1. Table 1. Accuracy of identifying abnormal aquaculture parameter data Data Number Accuracy of abnormal aquaculture parameter identification EN01 90.22% EN02 91.04% EN03 90.57% EN04 90.89% As can be seen from Table 1, the overall accuracy rate of identification is above 90%, and the identification effect of abnormal breeding parameters is good. Furthermore, the first feedback feature data or the second feedback feature data is divided equally according to the sliding time window to obtain multiple feedback feature data, which include trend feature data and morphological feature data. The first feedback similarity calculation is performed between the t-th feedback feature data and the (t-1)-th feedback feature data. The first feedback similarity calculation includes trend feature similarity calculation and morphological feature similarity calculation. During the analysis, by dividing the data into sliding time windows, further data processing and calculation can be performed on each feedback feature data. By comparing similarity, it is possible to discover whether there are any abnormal change features in these data that cause a significant decrease in egg production or quality, and these features are recorded. This method can discover obvious abnormal feature data in these negative feedback data and record them, so as to avoid the occurrence of the same abnormal environmental features in subsequent studies, thereby reducing the workload of relevant personnel and improving the accuracy of egg production environment research. The trend feature similarity calculation uses the standard deviation method, empirical statistical method, and dynamic path calculation method; the morphological feature similarity calculation uses the dynamic path calculation method and image recognition analysis method; the specific calculations are as follows: ; ; in, The result of the similarity calculation for the trend features. The standard deviation method, The empirical statistical method is described above. The dynamic path calculation method, dynamic path reference Figure 4 As shown, gray represents feature paths when the data is consistent, and black represents feature paths when the data is different. For the t-th feedback feature data, For the (t-1)th feedback feature data, , and The weights calculated for the similarity of the trend features. ; The result is the morphological feature similarity calculation, C represents the heatmap data conversion, and the heatmap reference is... Figure 5 As shown, The image recognition and analysis method (e.g., residual convolution, dilated convolution, Faster R-CNN, etc.) is used. and The weights calculated for the morphological feature similarity; ; The main method is to obtain some characteristic data by relevant personnel in conjunction with negative feedback data; The YOLOv5 algorithm is employed. To ensure accuracy during the research process, this invention analyzes the trend and morphological characteristics of time-series data. Trend characteristics are primarily obtained through standard deviation dispersion, data similarity, and summaries of data characteristics by relevant researchers. Morphological characteristics are mainly obtained through data similarity and image analysis using deep learning. By analyzing from multiple perspectives, the abnormal characteristics of negative feedback data in the egg-laying hen farming environment can be accurately grasped, improving the accuracy of data research. Ensuring the completeness of abnormal data records further enhances the accuracy of research on the egg-laying hen farming environment. If the first feedback similarity calculation result is less than the first feedback similarity threshold, then the t-th feedback feature data is the first abnormal aquaculture monitoring parameter; the first feedback similarity threshold is set based on expert experience, and the default value is 90%; The second abnormal aquaculture monitoring parameter analysis unit generates the second abnormal aquaculture monitoring parameters by comparing the first feedback feature data or the second feedback feature data with the positive feedback feature data. Furthermore, the first feedback feature data, second feedback feature data, and positive feedback feature data are equally divided according to the sliding time window to obtain multiple negative feedback feature data and multiple positive feedback feature data. The feedback feature data includes trend feature data. A second feedback similarity calculation is performed between the k-th negative feedback feature data and the k-th positive feedback feature data. The second feedback similarity calculation is a trend feature similarity calculation. The trend feature similarity calculation uses the standard deviation method, empirical statistical method, and dynamic path calculation method. The specific calculation is as follows: ; in, For the k-th negative feedback feature data, This refers to the k-th positive feedback feature data. This involves feature identification based on the experience of relevant experts; In the study of negative feedback data in the egg-laying hen farming environment, since the data characteristics of some abnormal data are not obvious, in order to further enrich and improve the accuracy of these abnormal data, the abnormal characteristics of negative feedback data are compared with those of positive feedback data to uncover these inconspicuous abnormal data. This will further enrich the database, improve the accuracy of abnormal data storage, facilitate subsequent monitoring and adjustment of the egg-laying hen farming environment, reduce the workload of relevant personnel, and improve the accuracy of egg-laying hen farming environment research. If the calculated similarity of the second feedback is less than the similarity threshold of the second feedback, then the kth negative feedback feature is the second abnormal aquaculture monitoring parameter; the similarity threshold of the second feedback is set based on expert experience, and the default value is 90%; The abnormal data output storage unit tags the first abnormal aquaculture monitoring parameter and the second abnormal aquaculture monitoring parameter with corresponding tags, and outputs and stores them into the corresponding database. The anomaly similarity calculation unit compares the real-time monitored egg-laying hen breeding environment with the first and second abnormal breeding monitoring parameters to generate anomaly similarity. The negative feedback data analysis model compares the egg-laying hen breeding environment with the first and second abnormal breeding monitoring parameters after data processing and feature extraction, and calculates trend feature similarity and morphological feature similarity. To improve the monitoring and adjustment of egg-laying hen breeding environment data and identify any anomalies related to negative feedback data, the same analysis method is used, analyzing trend features and image features. After processing and extracting features using the established model, the data is compared using the same method to ensure consistency in the analysis methods and avoid omissions of anomalies due to different processing and analysis methods, thus ensuring the accuracy of egg-laying hen breeding environment research. This reduces the workload of relevant personnel while improving the accuracy of egg-laying hen breeding environment research. The morphological feature similarity calculation method is a dynamic path calculation method and an image recognition analysis method. The real-time monitoring module for the breeding environment monitors the breeding environment of the laying hens in real time and compares the breeding environment of the laying hens with the first abnormal breeding monitoring parameter and the second abnormal breeding monitoring parameter through the negative feedback data analysis model. If the first abnormal aquaculture monitoring and early warning module detects that the first abnormal aquaculture monitoring parameter or the second abnormal aquaculture monitoring parameter is not less than the aquaculture environment similarity threshold, it will issue a first aquaculture monitoring early warning for adjustment. The second aquaculture monitoring and early warning module issues a second aquaculture monitoring early warning if the first or second abnormal aquaculture monitoring parameter adjusted by relevant personnel is not less than the aquaculture environment similarity threshold. Example 2 This invention also provides a method for intelligent monitoring of the egg-laying hen farming environment based on the Internet of Things, referring to... Figure 6 As shown, it includes: Obtain negative feedback monitoring parameters of the egg-laying hen breeding environment; perform clustering and classification based on the positive feedback monitoring parameters and the negative feedback results of the negative feedback monitoring parameters; generate first negative feedback monitoring data and second negative feedback monitoring data; Obtain positive feedback monitoring parameters of the egg-laying hen farming environment; establish a negative feedback data analysis model, analyze the first negative feedback monitoring data and the second negative feedback monitoring data according to the negative feedback data analysis model, generate and store the first abnormal farming monitoring parameters; compare the negative feedback monitoring parameters and the positive feedback monitoring parameters according to the negative feedback data analysis model, generate and store the second abnormal farming monitoring parameters; Further, based on the trend feature similarity calculation and morphological feature similarity calculation of the negative feedback data analysis model, the first abnormal aquaculture monitoring parameters are generated; based on the trend feature similarity calculation of the negative feedback data analysis model, the second abnormal aquaculture monitoring parameters are generated; the trend feature similarity calculation is the standard deviation method, the empirical statistical method, and the dynamic path calculation method; the morphological feature similarity calculation is the dynamic path calculation method and the image recognition analysis method. In this embodiment, the similarity results of trend feature similarity and morphological feature similarity, as well as combinations of multiple methods, were compared with data whose similarity results had been determined by experts. The comparison results are shown in Table 2. Table 2 Comparison of different similarity schemes

[0016] Table 3 compares the results under different trend similarity schemes: Table 3 Comparison of different schemes with different trend similarities

[0017] The egg-laying hen breeding environment is monitored in real time, and the egg-laying hen breeding environment is compared with the first abnormal breeding monitoring parameter and the second abnormal breeding monitoring parameter through the negative feedback data analysis model. If the first or second abnormal aquaculture monitoring parameter is not less than the aquaculture environment similarity threshold, a first aquaculture monitoring warning is issued for adjustment; if the first or second abnormal aquaculture monitoring parameter after adjustment by relevant personnel is not less than the aquaculture environment similarity threshold, a second aquaculture monitoring warning is issued.

[0018] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent monitoring system for an IoT-based laying hen breeding environment, characterized in that, include: The negative feedback data acquisition module acquires negative feedback monitoring parameters of the egg-laying hen farming environment; The positive feedback data acquisition module acquires positive feedback monitoring parameters of the egg-laying hen farming environment; The negative feedback data classification module performs clustering and classification based on the positive feedback monitoring parameters and the negative feedback results of the negative feedback monitoring parameters; and generates first negative feedback monitoring data and second negative feedback monitoring data. The negative feedback data analysis module establishes a negative feedback data analysis model, analyzes the first negative feedback monitoring data and the second negative feedback monitoring data based on the negative feedback data analysis model, generates and stores the first abnormal aquaculture monitoring parameters. Based on the negative feedback data analysis model, the negative feedback monitoring parameters and the positive feedback monitoring parameters are compared to generate and store the second abnormal aquaculture monitoring parameters; The real-time monitoring module for the breeding environment monitors the breeding environment of the laying hens in real time and compares the breeding environment of the laying hens with the first abnormal breeding monitoring parameter and the second abnormal breeding monitoring parameter through the negative feedback data analysis model. If the first abnormal aquaculture monitoring and early warning module detects that the first abnormal aquaculture monitoring parameter or the second abnormal aquaculture monitoring parameter is not less than the aquaculture environment similarity threshold, it will issue a first aquaculture monitoring early warning for adjustment. The second aquaculture monitoring and early warning module issues a second aquaculture monitoring early warning if the first or second abnormal aquaculture monitoring parameter adjusted by relevant personnel is not less than the aquaculture environment similarity threshold.

2. The intelligent monitoring system for laying hen breeding environment based on Internet of Things according to claim 1, characterized in that, The first negative feedback monitoring data is negative feedback data caused by fowl plague; the second negative feedback monitoring data is negative feedback data not caused by fowl plague. 3.The intelligent monitoring system for laying hen breeding environment based on Internet of Things according to claim 1, characterized in that, Establishing a negative feedback data analysis model includes: The negative feedback data analysis model includes a positive and negative feedback data processing unit, a positive and negative feedback data feature extraction unit, a first abnormal aquaculture monitoring parameter analysis unit, a second abnormal aquaculture monitoring parameter analysis unit, an abnormal data output storage unit, and an abnormal similarity calculation unit. The positive and negative feedback data processing unit filters the first negative feedback monitoring data, the second negative feedback monitoring data, and the positive feedback monitoring parameters according to the laying hen breed and feeding type, and preprocesses the data to generate preprocessed feedback data. The positive and negative feedback data feature extraction unit extracts features from the preprocessed feedback data based on Transformer, GRU, dilated convolution and feature fusion layer, and generates first feedback feature data, second feedback feature data and positive feedback feature data; The first abnormal aquaculture monitoring parameter analysis unit generates the first abnormal aquaculture monitoring parameter based on the abnormal feature data of the first feedback feature data and the second feedback feature data; the second abnormal aquaculture monitoring parameter analysis unit generates the second abnormal aquaculture monitoring parameter by comparing the first feedback feature data or the second feedback feature data with the positive feedback feature data. The abnormal data output storage unit tags the first abnormal aquaculture monitoring parameter and the second abnormal aquaculture monitoring parameter with corresponding tags, and outputs and stores them into the corresponding database. The anomaly similarity calculation unit compares the real-time monitored egg-laying hen breeding environment with the first and second abnormal breeding monitoring parameters to calculate the anomaly similarity and generate an anomaly similarity.

4. The intelligent monitoring system for laying hen breeding environment based on Internet of Things according to claim 1, characterized in that, Analyze the first negative feedback monitoring data and the second negative feedback monitoring data to generate and store the first abnormal aquaculture monitoring parameters; The parameters are calculated and obtained based on the first abnormal aquaculture monitoring parameter analysis unit of the negative feedback data analysis model, including: The first or second feedback feature data is divided equally according to the sliding time window to obtain multiple feedback feature data, which include trend feature data and morphological feature data. The t-th feedback feature data and the (t-1)-th feedback feature data are used to perform a first feedback similarity calculation; the first feedback similarity calculation includes trend feature similarity calculation and morphological feature similarity calculation; if the result of the first feedback similarity calculation is less than the first feedback similarity threshold, then the t-th feedback feature data is the first abnormal aquaculture monitoring parameter.

5. The intelligent monitoring system for laying hen breeding environment based on Internet of Things according to claim 4, characterized in that, The trend feature similarity calculation is performed using the standard deviation method, empirical statistical method, and dynamic path calculation method; the morphological feature similarity calculation is performed using the dynamic path calculation method and image recognition analysis method.

6. The intelligent monitoring system for laying hen breeding environment based on Internet of Things according to claim 1, characterized in that, By comparing the negative feedback monitoring parameters and the positive feedback monitoring parameters, a second abnormal aquaculture monitoring parameter is generated and stored. The parameters are calculated and obtained based on the second abnormal aquaculture monitoring parameter analysis unit of the negative feedback data analysis model, including: The first feedback feature data, the second feedback feature data, and the positive feedback feature data are equally divided according to the sliding time window to obtain multiple negative feedback feature data and multiple positive feedback feature data. The feedback feature data includes trend feature data. The k-th negative feedback feature data and the k-th positive feedback feature data are used to perform a second feedback similarity calculation. The second feedback similarity calculation is a trend feature similarity calculation. If the result of the second feedback similarity calculation is less than the second feedback similarity threshold, then the k-th negative feedback feature is the second abnormal aquaculture monitoring parameter. 7.The intelligent monitoring system for laying hen breeding environment based on Internet of Things according to claim 1, characterized in that, The negative feedback data analysis model compares the egg-laying hen breeding environment with the first abnormal breeding monitoring parameters and the second abnormal breeding monitoring parameters. After data processing and feature extraction, it calculates the similarity based on trend features and morphological features.

8. An intelligent monitoring method for an IoT-based laying hen breeding environment, characterized in that, include: Obtain negative feedback monitoring parameters of the egg-laying hen farming environment; Clustering and classification are performed based on the positive feedback monitoring parameters and the negative feedback results of the negative feedback monitoring parameters; Generate first negative feedback monitoring data and second negative feedback monitoring data; Obtain positive feedback monitoring parameters of the egg-laying hen breeding environment; establish a negative feedback data analysis model, analyze the first negative feedback monitoring data and the second negative feedback monitoring data based on the negative feedback data analysis model, generate and store the first abnormal breeding monitoring parameters; Based on the negative feedback data analysis model, the negative feedback monitoring parameters and the positive feedback monitoring parameters are compared to generate and store the second abnormal aquaculture monitoring parameters; The egg-laying hen breeding environment is monitored in real time, and the egg-laying hen breeding environment is compared with the first abnormal breeding monitoring parameter and the second abnormal breeding monitoring parameter through the negative feedback data analysis model. If the first abnormal aquaculture monitoring parameter or the second abnormal aquaculture monitoring parameter is not less than the aquaculture environment similarity threshold, a first aquaculture monitoring warning will be issued for adjustment. If the first or second abnormal aquaculture monitoring parameter adjusted by relevant personnel is not less than the aquaculture environment similarity threshold, a second aquaculture monitoring warning will be issued. 9.The intelligent monitoring method of the laying hen breeding environment based on the Internet of Things according to claim 8, characterized in that, The first negative feedback monitoring data and the second negative feedback monitoring data are analyzed based on the negative feedback data analysis model to generate and store the first abnormal aquaculture monitoring parameters; Based on the negative feedback data analysis model, the negative feedback monitoring parameters and the positive feedback monitoring parameters are compared to generate and store a second abnormal aquaculture monitoring parameter, including: Based on the trend feature similarity calculation and morphological feature similarity calculation of the negative feedback data analysis model, the first abnormal aquaculture monitoring parameters are generated. The second abnormal aquaculture monitoring parameters are generated based on the trend feature similarity calculation of the negative feedback data analysis model. 10.The intelligent monitoring method of the laying hen breeding environment based on the Internet of Things according to claim 9, characterized in that, The trend feature similarity calculation is performed using the standard deviation method, empirical statistical method, and dynamic path calculation method; the morphological feature similarity calculation is performed using the dynamic path calculation method and image recognition analysis method.

Images