Livestock production environment monitoring method and system based on Internet of Things

Abstract

The invention belongs to the technical field of environment monitoring, and discloses a livestock production environment monitoring method and system based on the Internet of Things, and the method comprises the steps: deploying sensor modules in different regions of a farm, collecting temperature, humidity, illumination and harmful gas concentration data, and carrying out the data preprocessing; performing primary processing on the preprocessed data by adopting an edge computing technology, extracting key features, and transmitting the key features to a cloud platform; a regional differentiation monitoring model is established in the cloud platform, key feature data are monitored in real time, and if abnormal fluctuation of temperature, humidity or harmful gas concentration is found, an early warning mechanism is triggered, and management personnel are notified to handle the abnormal fluctuation in time; and deploying a data fusion algorithm in the cloud platform, integrating multi-region sensor data, and generating a global environment parameter distribution diagram. According to the invention, the problems of data acquisition, transmission and analysis in a complex environment of a farm are effectively solved, refined environment management is realized, and the breeding efficiency and safety are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental monitoring, and in particular relates to a livestock production environment monitoring method and system based on the Internet of Things. Background Art

[0002] In the monitoring of livestock production environment, the deployed sensors need to collect key environmental parameters such as temperature, humidity, light and harmful gas concentration in real time. These sensors are usually installed in different areas of the farm to ensure the comprehensiveness and representativeness of the data. However, due to the complex environment of the farm, the sensors face many technical problems in actual operation. First, the sensors need to work continuously and stably under high humidity, high dust and extreme temperature conditions, which places extremely high demands on the durability and accuracy of the sensors. Secondly, the data collected by the sensors need to be transmitted to the management cloud platform via a wireless network, but inside the farm, the wireless signal may be interfered by buildings, equipment and animal activities, resulting in unstable or delayed data transmission. In addition, the sensor may have low battery or aging problems after a long period of operation, which will affect the continuity and reliability of the data. At the same time, the environmental parameters in different areas of the farm may vary significantly. For example, harmful gases may accumulate in poorly ventilated areas, while areas with sufficient light may have excessive temperatures. This requires sensors to be able to perform differentiated monitoring for different areas, but existing systems often find it difficult to achieve such refined management. Finally, the management cloud platform needs to process and analyze a large amount of sensor data in real time. However, given the large amount of data and the diverse sources, how to efficiently integrate and process this data while avoiding data redundancy and errors is also a technical problem that needs to be solved urgently. These problems together constitute the core challenges in livestock production environment monitoring, which need to be gradually solved through technology and system optimization. Summary of the invention

[0003] In order to solve the problems existing in the prior art, the present invention provides a livestock production environment monitoring method and system based on the Internet of Things, which effectively solves the problems of data collection, transmission and analysis in the complex environment of the farm, realizes refined environmental management, and improves breeding efficiency and safety.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A livestock production environment monitoring method based on the Internet of Things, the method comprising:

[0006] Deploy sensor modules in different areas of the farm to collect temperature, humidity, light and harmful gas concentration data and perform data preprocessing;

[0007] Edge computing technology is used to perform preliminary processing on the preprocessed data, extract key features, and then transmit them to the cloud platform;

[0008] Establish a regional differential monitoring model in the cloud platform to monitor key feature data in real time. If abnormal fluctuations in temperature, humidity, or harmful gas concentration are detected, trigger the early warning mechanism to notify the management staff for timely handling;

[0009] Deploy a data fusion algorithm in the cloud platform to integrate multi-regional sensor data and generate a global environmental parameter distribution map.

[0010] Preferably, the method of deploying sensor modules in different regions of the farm includes:

[0011] Transmit the preprocessed data through low-power wide-area network technology. Specifically, relay nodes are arranged in the farm, and the positions and quantities of the nodes are reasonably planned;

[0012] Establish a data caching mechanism in the cloud platform. If packet loss or delay is detected, automatically reissue the missing data from the cache;

[0013] Deploy multiple types of sensors according to the environmental differences in different regions based on the cached data.

[0014] Preferably, edge computing technology is adopted to perform preliminary processing on the preprocessed data. The method of extracting key features includes: extracting key features by constructing an edge computing network system framework, where the edge computing network system framework includes: service layer, platform layer, transmission layer, edge layer, and perception layer;

[0015] The perception layer is connected to the edge gateway and can transmit various environmental information to the edge layer in real time;

[0016] The edge layer mainly includes an edge gateway, an edge cloud, and an edge controller; among them, the edge gateway is responsible for the management of each sensor and resource calculation; the edge cloud provides storage and computing resources for the monitoring data, enabling data processing and analysis to be carried out locally; the edge controller is responsible for the real-time control and optimization of various sensors;

[0017] The transmission layer is mainly responsible for data transmission and selects a suitable network for information transmission according to different application scenarios;

[0018] The platform layer, as the hub for data collection and processing, analyzes various environmental data;

[0019] The service layer focuses on the analysis of environmental monitoring data. Through in-depth analysis of environmental monitoring data, the service layer can timely discover abnormal information. The service layer also predicts the environmental trend in the next period of time through the mining and analysis of historical data;

[0020] Among them, discrete wavelet transform is introduced to achieve time-frequency domain feature extraction:

[0021] X(a) = ∑(W j，m ·Ψ(a - mj))

[0022] Wherein, X(a) represents the original data, W j,m represents the wavelet coefficient, Ψ(a - mj) represents the wavelet basis function, j represents the scale, and m represents the translation.

[0023] Preferably, the method for establishing a regional differential monitoring model in the cloud platform includes:

[0024] Deploy an edge computing network architecture at the sensor node, obtain environmental data, preprocess the data, and filter out noise information;

[0025] Adopt a feature extraction algorithm to extract key features from the preprocessed data to obtain a feature data set, and transmit the feature data to the cloud platform;

[0026] Receive the feature data on the cloud platform, adopt a data integrity detection algorithm to judge the loss situation during data transmission, and trigger a retransmission mechanism if there is a loss;

[0027] According to the feature data set, use a clustering algorithm to classify the environmental data and determine the distribution law of environmental features in different regions;

[0028] For the distribution law of environmental features, adopt a regional division algorithm to generate a regional differential environmental monitoring model.

[0029] Preferably, the method for establishing a regional differential monitoring model in the cloud platform and performing real-time monitoring on each key feature data includes:

[0030] Establish a conceptual model: Wherein: R, f(H), f(E), f(R), and f(V) are respectively the regional environmental anomaly, anomaly source hazard, receptor exposure, receptor resistance, and receptor vulnerability of the evaluation object;

[0031] Perform dimensionless processing on each evaluation index: x ij (t k ) is the value of the jth index of the object s i at time t k (i = 1, 2,..., n; j = 1, 2,..., m; k = 1, 2,..., N), then there is: Wherein, x' ij (t k ) is the dimensionless value; and σ j (t k ) are respectively the mean and standard deviation of the jth index at time t k ;

[0032] Use the vertical and horizontal grading method to calculate and obtain the risk index of abnormal sources, the receptor exposure index, and the resilience index of each evaluation object at time t; k At the moment, the receptor exposure index and the resilience index;

[0033] If abnormal fluctuations in temperature, humidity, or harmful gas concentration are detected, trigger the early warning mechanism and notify the management staff for timely handling.

[0034] Preferably, the method of deploying a data fusion algorithm in the cloud platform to integrate multi-region sensor data and generate a global environmental parameter distribution map includes:

[0035] For each region, use the K-nearest neighbor classification algorithm to match each sensor data information after dimensionality reduction; select the data information in the storage that is more matched with the preset sample, and then perform precise matching on the filtered data through the triple + neural network with Res Net as the backbone network, so as to obtain the fused sensor data of each region;

[0036] Transmit the fused sensor data of each region to the central fusion module, and perform tracking, association, and estimation in the fusion module to generate a global environmental parameter distribution map. Among them, the central fusion module adopts the FOP-MOC model.

[0037] The present invention also provides an Internet of Things-based livestock production environment monitoring system, which is used to implement any one of the above methods. The system includes: a preprocessing module, a feature extraction module, an anomaly detection module, and a data fusion module;

[0038] The preprocessing module is used to deploy sensor modules in different regions of the farm, collect temperature, humidity, light, and harmful gas concentration data, and perform data preprocessing;

[0039] The feature extraction module is used to use edge computing technology to perform preliminary processing on the preprocessed data, extract key features, and then transmit them to the cloud platform;

[0040] The anomaly detection module is used to establish a regional differentiation monitoring model in the cloud platform, perform real-time monitoring on each key feature data. If abnormal fluctuations in temperature, humidity, or harmful gas concentration are detected, trigger the early warning mechanism and notify the management staff for timely handling;

[0041] The data fusion module is used to deploy a data fusion algorithm in the cloud platform to integrate multi-region sensor data and generate a global environmental parameter distribution map.

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] The present invention discloses an Internet of Things-based monitoring method and system for livestock production environment. In view of the characteristics of high humidity, high dust and large temperature changes in the farm, sensor modules resistant to extreme environments are used to collect environmental data in different areas. Data preprocessing is carried out by integrating an adaptive filtering algorithm, and low-power wide-area network technology and relay nodes are used to optimize the transmission. The present invention establishes a data caching and reissuing mechanism on the cloud platform to ensure data continuity. Combining edge computing technology and regional differential monitoring models, real-time anomaly monitoring and early warning are realized. At the same time, through equipment health monitoring and data fusion algorithms, a global environmental parameter distribution map is provided. The present invention effectively solves the problems of data collection, transmission and analysis in the complex environment of the farm, realizes refined environmental management, and improves breeding efficiency and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the present invention, the following briefly introduces the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0045] Figure 1 Schematic flowchart of an Internet of Things-based monitoring method for livestock production environment according to an embodiment of the present invention;

[0046] Figure 2 Schematic diagram of the edge computing network architecture framework according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the protection scope of the present invention.

[0048] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0049] Embodiment 1

[0050] As Figure 1 shown, the present invention provides an Internet of Things-based monitoring method for livestock production environment, and the method includes:

[0051] Deploy sensor modules in different areas of the farm, collect data on temperature, humidity, light and harmful gas concentration, and perform data preprocessing;

[0052] Adopt edge computing technology to preliminarily process the preprocessed data, extract key features and then transmit them to the cloud platform;

[0053] Establish a regional differentiation monitoring model in the cloud platform to monitor the key feature data in real time. If abnormal fluctuations in temperature, humidity or harmful gas concentration are found, trigger the early warning mechanism to notify the management personnel for timely handling;

[0054] Deploy a data fusion algorithm in the cloud platform to integrate multi-region sensor data and generate a global environmental parameter distribution map.

[0055] In this embodiment, an adaptive filtering algorithm is integrated in the sensor module to perform data preprocessing for the high-humidity and high-dust environment, remove noise interference and improve data accuracy, including:

[0056] The sensor module is deployed in the high-humidity and high-dust area of the farm to collect temperature, humidity, light and gas data;

[0057] Use the adaptive filtering algorithm to preprocess the original data, remove noise interference, and obtain a filtered data set;

[0058] According to the preset threshold, judge whether the filtered temperature value exceeds the threshold;

[0059] If the temperature value exceeds the threshold, trigger a temperature anomaly warning, and at the same time obtain the filtered humidity value and gas value for correlation analysis;

[0060] If neither the humidity value nor the gas value exceeds the preset threshold, only record the temperature anomaly information and update the data set;

[0061] If the humidity value or the gas value exceeds the preset threshold, trigger a comprehensive warning and generate a comprehensive anomaly report;

[0062] Use the K-means clustering algorithm to classify the comprehensive anomaly report, determine the anomaly type and occurrence frequency, and form the environmental monitoring results of the farm.

[0063] In this embodiment, the method of deploying sensor modules in different regions of the farm includes:

[0064] Transmit the preprocessed data through low-power wide-area network technology. Specifically, relay nodes are arranged in the farm, and the node positions and quantities are reasonably planned;

[0065] Establish a data caching mechanism in the cloud platform. If packet loss or delay is detected, automatically reissue the missing data from the cache;

[0066] According to the cached data, deploy multi-type sensors according to the environmental differences in different regions.

[0067] Specifically, relay nodes are arranged in the farm, and the location and number of nodes are reasonably planned, including:

[0068] According to the location and number of nodes, calculate the signal coverage and strength, and determine whether there is a signal blind spot. If there is a blind spot, adjust the node location and number; obtain signal coverage and strength data, and optimize the relay node layout in combination with the distribution of buildings and equipment to reduce signal interference; through the optimized relay node layout, transmit filtered data, monitor the signal transmission quality, and determine whether there is data loss or delay; if data loss or delay is found, readjust the relay node layout and monitor the signal transmission quality again until the preset transmission standard is met; use a clustering algorithm to classify the transmission quality data, determine the distribution pattern of signal coverage and interference, and generate a signal coverage and interference distribution map; based on the signal coverage and interference distribution map, update the relay node layout to continuously optimize data transmission efficiency and signal coverage.

[0069] Specifically, according to the content of the data packet in the cache area, multiple types of sensors are deployed in the transmission chain to obtain the environmental data of each area; for the acquired environmental data, the detection point is used to monitor the integrity and timeliness of the data packet, and the loss and delay are recorded; if the detection point finds that the data packet is lost or delayed, the missing value is extracted from the cache area through the retransmitter and resent to the transmission chain; according to the operating data of the retransmitter, the environmental monitoring status of each area is updated, and the regional differentiated environmental standards are clarified; a clustering algorithm is used to classify the loss and delay to obtain the data anomaly distribution law in different areas; according to the data anomaly distribution law, the storage strategy of the cache area is adjusted, and the regional differentiated monitoring model is optimized; through the optimized monitoring model, the deployment strategy of the sensor is reconfigured to achieve refined environmental management.

[0070] In this embodiment, if Figure 2 As shown, edge computing technology is used to perform preliminary processing on the preprocessed data, and the method for extracting key features includes: extracting key features by constructing an edge computing network system framework, wherein the edge computing network system framework includes: a service layer, a platform layer, a transmission layer, an edge layer, and a perception layer;

[0071] The perception layer is connected to the edge gateway and can transmit various environmental information to the edge layer in real time;

[0072] The edge layer mainly includes edge gateway, edge cloud and edge controller. The edge gateway is responsible for sensor management and resource calculation. The edge cloud provides storage and computing resources for monitoring data, so that data processing and analysis can be performed locally. The edge controller is responsible for real-time control and optimization of various sensors.

[0073] The transport layer is mainly responsible for data transmission and selects a suitable network for information transmission according to different application scenarios;

[0074] As the hub for data collection and processing, the platform layer analyzes various environmental data;

[0075] The service layer focuses on the analysis of environmental monitoring data. Through in-depth analysis of environmental monitoring data, the service layer can timely detect abnormal information. It can also predict the environmental trends in the future for a period of time through the mining and analysis of historical data;

[0076] Among them, discrete wavelet transform is introduced to realize time-frequency domain feature extraction:

[0077] X(a) = ∑(W j，m ·Ψ(a - mj))

[0078] In the formula, X(a) represents the original data, W j，m represents the wavelet coefficient, Ψ(a - mj) represents the wavelet basis function, j represents the scale, and m represents the translation.

[0079] In this embodiment, the method for establishing a regional differential monitoring model in the cloud platform includes:

[0080] Deploy an edge computing network architecture framework at the sensor node, obtain environmental data, preprocess the data, and filter out noise information;

[0081] Adopt a feature extraction algorithm to extract key features from the preprocessed data to obtain a feature data set, and transmit the feature data to the cloud platform;

[0082] Receive the feature data in the cloud platform, adopt a data integrity detection algorithm to judge the loss situation during data transmission, and trigger a retransmission mechanism if there is a loss;

[0083] According to the feature data set, use a clustering algorithm to classify the environmental data and determine the distribution law of environmental characteristics in different regions;

[0084] Aiming at the distribution law of environmental characteristics, adopt a regional division algorithm to generate a regional differential environmental monitoring model.

[0085] In this embodiment, the method for establishing a regional differential monitoring model in the cloud platform and performing real-time monitoring on each key feature data includes:

[0086] Establish a conceptual model: In the formula: R, f(H), f(E), f(R), and f(V) are respectively the regional environmental anomaly, the danger of the anomaly source, the receptor exposure, the receptor resistance, and the receptor vulnerability of the evaluation object;

[0087] Make the dimensionless treatment of each evaluation index: x ij (t k ) is the value of the j-th index of object s i at time t k (i = 1, 2,..., n; j = 1, 2,..., m; k = 1, 2,..., N), then there is: In the formula, x' ij (t k ) is the dimensionless value; and σ j (t k ) are the mean value and standard deviation of the j-th index at time t k respectively;

[0088] Use the vertical and horizontal grading method to calculate and obtain the abnormal source risk index, receptor exposure index and resilience index of each evaluation object at time t k ;

[0089] If abnormal fluctuations in temperature, humidity or harmful gas concentration are found, trigger the early warning mechanism and notify the management staff for timely handling.

[0090] Specifically, the method of using the vertical and horizontal grading method to calculate and obtain the abnormal source risk index, receptor exposure index and resilience index of each evaluation object at time t k includes:

[0091] (1) For the m indexes x k of n evaluation objects at a given time t 1 , x 2 ,..., x m values (already dimensionless), represented by matrix A k , that is:

[0092]

[0093] (2) Calculate the m×m real symmetric matrix H k , H k = A k T A k ;

[0094] (3) Calculate the largest eigenvalue λ k of H max and its corresponding weight coefficient vector, and perform normalization processing to calculate the weight ω j (t k );

[0095] (4) Calculate the function:

[0096]

[0097] Where: y i (t k ) is the risk index of the abnormal source, the exposure index of the receptor, and the resilience index of the receptor of the evaluation object s i at time t k The moment of the abnormal source risk index, receptor exposure index, and receptor resilience index.

[0098] (5) Calculate the regional environmental anomaly index Calculate the regional environmental anomaly index R i of the evaluation object s at time t k The moment of the regional environmental anomaly index R i (t k ) and the receptor vulnerability index V i (t k ).

[0099] In this embodiment, the method of deploying a data fusion algorithm in the cloud platform, integrating multi-region sensor data, and generating a global environmental parameter distribution map includes:

[0100] For each region, use the K-nearest neighbor classification algorithm to match each sensor data information after dimensionality reduction; select the data information in the storage that is relatively matched with the preset sample, and then perform precise matching on the filtered data through a triple + neural network with Res Net as the backbone network, so as to obtain the fused sensor data of each region;

[0101] Transfer the fused sensor data of each region to the central fusion module, and perform tracking, association, and estimation in the fusion module to generate a global environmental parameter distribution map, where the central fusion module adopts the FOP-MOC model.

[0102] The present invention adopts a sensor module that is resistant to high humidity, high dust, and extreme temperatures, and deploys it in different regions of the farm to collect data on temperature, humidity, light, and harmful gas concentration; integrates an adaptive filtering algorithm in the sensor module to perform data preprocessing for high humidity and high dust environments, remove noise interference, and improve data accuracy; establishes a data caching mechanism in the cloud platform. If a data packet loss or delay is detected, the missing data is automatically reissued from the cache to ensure data continuity; according to the cached data, deploy multiple types of sensors according to the environmental differences in different regions, establish a regional differential monitoring model in the cloud platform, clarify the environmental standards of each region, and achieve refined management; adopt edge computing technology to perform preliminary processing of data locally at the sensor node, extract key features and then transmit them to the cloud platform to reduce data redundancy. At the same time, through device health monitoring and data fusion algorithms, a global environmental parameter distribution map is provided. The present invention effectively solves the problems of data collection, transmission, and analysis in the complex environment of the farm, realizes refined environmental management, and improves the breeding efficiency and safety.

[0103] Embodiment 2

[0104] The present invention also provides an Internet of Things-based livestock production environment monitoring system for implementing the method described in any one of Embodiment 1. The system includes a preprocessing module, a feature extraction module, an anomaly detection module, and a data fusion module;

[0105] The preprocessing module is used to deploy sensor modules in different areas of the farm, collect temperature, humidity, light, and harmful gas concentration data, and perform data preprocessing;

[0106] The feature extraction module is used to perform preliminary processing on the preprocessed data by using edge computing technology, extract key features, and then transmit them to the cloud platform;

[0107] The anomaly detection module is used to establish a regional differentiation monitoring model in the cloud platform, monitor the key feature data in real time. If abnormal fluctuations in temperature, humidity, or harmful gas concentration are found, it triggers an early warning mechanism to notify the management staff for timely handling;

[0108] The data fusion module is used to deploy a data fusion algorithm in the cloud platform, integrate multi-region sensor data, and generate a global environmental parameter distribution map.

[0109] The above-described embodiments are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A livestock production environment monitoring method based on the Internet of Things, characterized in that: The method comprises: Deploy sensor modules in different areas of the farm to collect temperature, humidity, light and harmful gas concentration data and perform data preprocessing; Edge computing technology is used to perform preliminary processing on the preprocessed data, extract key features, and then transmit them to the cloud platform; Establish regional differentiated monitoring models in the cloud platform to monitor key characteristic data in real time. If abnormal fluctuations in temperature, humidity or concentration of harmful gases are found, the early warning mechanism will be triggered to notify management personnel to handle the situation in a timely manner. Deploy data fusion algorithms in the cloud platform to integrate multi-region sensor data and generate global environmental parameter distribution maps.

2. The method according to claim 1, characterized in that Methods for deploying sensor modules in different areas of the farm include: The pre-processed data is transmitted through low-power wide area network technology. Specifically, relay nodes are arranged in the farm, and the location and number of nodes are reasonably planned; Establish a data cache mechanism in the cloud platform. If data packet loss or delay is detected, the missing data will be automatically resent from the cache. Based on the cached data, multiple types of sensors are deployed to address environmental differences in different areas.

3. The method according to claim 2, characterized in that The edge computing technology is used to perform preliminary processing on the preprocessed data. The method for extracting key features includes: extracting key features by constructing an edge computing network system framework, wherein the edge computing network system framework includes: a service layer, a platform layer, a transmission layer, an edge layer, and a perception layer; The perception layer is connected to the edge gateway and can transmit various environmental information to the edge layer in real time; The edge layer mainly includes edge gateway, edge cloud and edge controller. The edge gateway is responsible for sensor management and resource calculation. The edge cloud provides storage and computing resources for monitoring data, so that data processing and analysis can be performed locally. The edge controller is responsible for real-time control and optimization of various sensors. The transport layer is mainly responsible for data transmission and selects appropriate networks for information transmission according to different application scenarios; The platform layer serves as a hub for data collection and processing, analyzing various environmental data; The service layer focuses on the analysis of environmental monitoring data. Through in-depth analysis of environmental monitoring data, the service layer can timely discover abnormal information. The service layer also predicts environmental trends in the future by mining and analyzing historical data. Among them, discrete wavelet transform is introduced to realize time-frequency domain feature extraction: X(a)=∑(W j，m ·Ψ(a-mj)) In the formula, X(a) represents the original data, W j，m represents the wavelet coefficient, Ψ(a-mj) represents the wavelet basis function, j represents the scale, and m represents the translation.

4. The method according to claim 3, characterized in that Methods for establishing regional differentiated monitoring models in the cloud platform include: Deploy edge computing network framework at sensor nodes to obtain environmental data, pre-process data, and filter noise information; Using feature extraction algorithms, key features are extracted from preprocessed data to obtain feature data sets, which are then transmitted to the cloud platform. Receive feature data on the cloud platform and use data integrity detection algorithms to determine the loss of data during transmission. If there is any loss, a retransmission mechanism is triggered. Based on the characteristic data set, a clustering algorithm is used to classify the environmental data and determine the distribution patterns of environmental characteristics in different regions; According to the distribution law of environmental characteristics, a regional division algorithm is adopted to generate a regional differentiated environmental monitoring model.

5. The method according to claim 4, characterized in that The method of establishing a regional differentiated monitoring model in the cloud platform and performing real-time monitoring of each key characteristic data includes: Build a conceptual model: In the formula, R, f(H), f(E), f(R) and f(V) are the regional environmental anomaly, anomaly source hazard, receptor exposure, receptor resistance and receptor vulnerability of the evaluation object, respectively; Each evaluation index is dimensionless: x ij (t k ) is the object s i The jth index of k The values ​​of the time (i = 1, 2, ..., n; j = 1, 2, ..., m; k = 1, 2, ..., N), then: In the formula, x′ ij (t k ) is the dimensionless value; and σ j (t k ) are the jth index at t k Mean and standard deviation of the moment; The vertical and horizontal grading method is used to calculate the value of each evaluation object at t k Abnormal source hazard index, receptor exposure and resilience index at the moment; If abnormal fluctuations in temperature, humidity or concentration of harmful gases are found, the early warning mechanism will be triggered and the management personnel will be notified to deal with it in time.

6. The method according to claim 1, characterized in that The method of deploying a data fusion algorithm in a cloud platform, integrating multi-region sensor data, and generating a global environmental parameter distribution map includes: For each region, the K nearest neighbor classification algorithm is used to match each sensor data information after dimensionality reduction; the data information in the storage that matches the preset sample is selected, and then the filtered data is accurately matched through the triple + neural network with Res Net as the backbone network, so as to obtain the fused sensor data of each region; The sensor data after fusion in each area are all transmitted to the central fusion module, which is used for tracking, association and estimation to generate a global environmental parameter distribution map. The central fusion module adopts the FOP-MOC model.

7. A livestock production environment monitoring system based on the Internet of Things, the system is used to implement the method described in any one of claims 1 to 6, characterized in that: The system comprises: a preprocessing module, a feature extraction module, an anomaly detection module and a data fusion module; The preprocessing module is used to deploy sensor modules in different areas of the farm to collect temperature, humidity, light and harmful gas concentration data and perform data preprocessing; The feature extraction module is used to use edge computing technology to perform preliminary processing on the preprocessed data, extract key features and transmit them to the cloud platform; The anomaly detection module is used to establish a regional differentiated monitoring model in the cloud platform to monitor key feature data in real time. If abnormal fluctuations in temperature, humidity or concentration of harmful gases are found, an early warning mechanism is triggered to notify management personnel to handle the situation in a timely manner. The data fusion module is used to deploy a data fusion algorithm in the cloud platform, integrate multi-region sensor data, and generate a global environmental parameter distribution map.

Images