Poultry weight monitoring and health early warning system and method, and storage medium

Abstract

The invention relates to the technical field of Internet of Things, and discloses a poultry weight monitoring and health early warning system and method, and a storage medium. The system mainly comprises a distributed monitoring terminal, an edge computing node, a cloud management platform and a user interaction terminal. Wherein the distributed monitoring terminal is deployed in a free-ranging area of poultry (such as chicken), and is used for automatically collecting data in a non-contact mode under the condition that normal activities of the poultry are not interfered. According to the invention, through deep integration of technologies of Internet of Things, edge calculation, big data prediction, artificial intelligence natural language processing and the like, the problems of inaccurate data, response lag, experience dependence, extensive management and the like in the traditional breeding industry are solved; finally, precise monitoring of poultry weight, active early warning of health risk and intelligence and humanization of breeding decision are realized.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a poultry weight monitoring and health early warning system and method, and a storage medium. Background Technology

[0002] In modern large-scale chicken farming, especially in free-range models, precision farming faces numerous challenges. Traditional monitoring methods mainly rely on manual sampling and weighing, which is not only labor-intensive but also results in poor data timeliness and low individual coverage. Particularly in free-range settings, chickens roam a wide area, making continuous and comprehensive monitoring impossible, which makes it difficult for farmers to grasp the real-time growth status of each poultry.

[0003] Furthermore, existing monitoring systems typically focus on single-dimensional indicators, such as monitoring only ambient temperature or only the average weight of the flock, lacking the ability to integrate and analyze environmental data (such as temperature, humidity, and light) with biological data (such as weight and body temperature). This makes the status assessment incomplete and difficult to accurately determine the health status of poultry.

[0004] Meanwhile, in terms of decision support, existing technologies mostly provide simple data displays and lack the ability to predict dynamic growth trends based on individual historical data. Furthermore, farming experience often resides in the minds of seasoned experts and is difficult to digitize. Novice farmers lack effective guidance and struggle to make precise feeding (such as adjusting feed conversion ratios) and disease control decisions based on complex data. Current system response times are typically on the order of hours or even days, failing to meet the immediate needs of early disease warnings.

[0005] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0006] This invention provides a poultry weight monitoring and health early warning system, method, and computer-readable storage medium, aiming to solve technical problems such as incomplete data collection, delayed decision-making, and lack of multi-dimensional data fusion analysis in existing free-range farming models.

[0007] To achieve the above objectives, the present invention provides a poultry weight monitoring and health early warning system, comprising: Distributed monitoring terminals are deployed in free-range areas and configured to automatically collect individual identification information and biological data of poultry in a non-contact manner, while simultaneously collecting environmental parameters. Edge computing nodes are connected to the distributed monitoring terminal and configured to clean, time-series align, and perform multi-task concurrent scheduling processing on the collected data, and upload the data through a wireless communication network. A cloud management platform, which is communicatively connected to the edge computing node, is used to receive and store processed data; the cloud management platform includes a multi-dimensional data fusion and prediction module, configured to build an association mapping model based on historical data, and predict the growth trend of poultry based on real-time environmental parameters and biological ontology data; The intelligent decision generation module is configured to use a large language model and retrieval enhancement technology to call a pre-set aquaculture knowledge base and generate aquaculture decision suggestions in natural language form based on the growth trend. The user interaction terminal is configured to display the real-time status of the poultry, the prediction results of growth trends, and the breeding decision suggestions.

[0008] The following are preferred technical solutions of the present invention, but are not intended to limit the technical solutions provided by the present invention. The purpose and beneficial effects of the present invention can be better achieved and realized through the following preferred technical solutions.

[0009] As a preferred technical solution, in the poultry weight monitoring and health early warning system, the distributed monitoring terminal includes: The weighing unit integrates a pressure sensor and is configured to collect weight data when poultry stand up independently. The radio frequency identification unit integrates an RFID read / write module and is configured to read electronic tags worn on poultry to obtain individual identity information. The distributed monitoring terminal is configured to trigger the radio frequency identification unit to work when the weight data is stable, thereby binding the individual identity with the weight data.

[0010] As a preferred technical solution, in the poultry weight monitoring and health early warning system, the edge computing node runs on a real-time operating system and is configured to execute the following multi-task scheduling logic: The radio frequency identification task is executed in a first preset period; the weight data collection task is executed in a second preset period; wherein the first preset period is shorter than the second preset period, so as to realize the rapid identification of poultry in dynamic scenarios.

[0011] As a preferred technical solution, the poultry weight monitoring and health early warning system includes environmental parameters such as temperature, humidity, and light intensity; and biological data such as weight and body temperature.

[0012] As a preferred technical solution, the poultry weight monitoring and health early warning system, wherein the construction process of the intelligent decision generation module includes: Construct an aquaculture knowledge base, which includes scientific literature data, historical aquaculture case data, and real-time monitoring data; The aquaculture knowledge base is vectorized and stored in the retrieval system; in response to user queries or system alerts, relevant knowledge fragments are retrieved from the retrieval system. The retrieved knowledge fragments and current monitoring data are used as prompt words and input into the large language model to generate personalized growth reports or dynamic feeding plans.

[0013] As a preferred technical solution, in the poultry weight monitoring and health early warning system, the edge computing node is configured to send a high-priority early warning signal to the cloud management platform within a minute-level time window when abnormal poultry body temperature or growth data is detected, and the cloud management platform responds to the signal by immediately pushing an abnormality prompt to the user interaction terminal.

[0014] As a preferred technical solution, the poultry weight monitoring and health early warning system further includes: Business interaction module: The business interaction module is configured to open a data interface to authorized third-party user terminals; the third-party user terminals are configured to remotely access the real-time weight, activity trajectory and image data of specific poultry individuals through an application, and support full life-cycle traceability viewing based on blockchain technology.

[0015] Secondly, a method for monitoring poultry weight and providing early warning of health conditions, comprising: Using distributed monitoring terminals deployed in free-range areas, individual identification information and biological data of poultry are collected, and environmental parameters are collected simultaneously. The collected data is cleaned, time-series aligned, and processed by multi-task concurrent scheduling through edge computing nodes connected to the distributed monitoring terminal, and then uploaded through a wireless communication network. The cloud management platform receives and stores data uploaded via wireless communication network, and uses a multi-dimensional data fusion prediction module to predict the growth trend of poultry based on the correlation mapping model built from historical data and real-time environmental parameters and biological ontology data. Using the intelligent decision generation module, based on the large language model and retrieval enhancement technology, and calling the pre-built aquaculture knowledge base, aquaculture decision suggestions in natural language form are generated according to the growth trend. The real-time status of the poultry, growth trend prediction results, and breeding decision suggestions are displayed through the user interaction terminal.

[0016] Thirdly, a computer-readable storage medium stores a poultry weight monitoring and health early warning program, which, when executed by a processor, implements the steps of the poultry weight monitoring and health early warning method described above.

[0017] Beneficial effects: Compared to existing technologies, this invention achieves automated, non-contact, continuous monitoring of poultry data in free-range environments through the collaboration of distributed monitoring terminals and edge computing nodes, overcoming the spatiotemporal limitations of data collection and improving data integrity. Utilizing a multi-dimensional data fusion and prediction module, combined with environmental and biological ontological data, it can more accurately assess poultry status; combining RAG technology and large language models, it digitizes expert knowledge, providing targeted natural language decision-making suggestions and lowering the barrier to entry for professional poultry farming. Through edge-side cleaning and priority scheduling, it achieves minute-level anomaly early warning response, significantly improving the timeliness of disease prevention and control. Attached Figure Description

[0018] Figure 1 This is a general architecture diagram of a preferred embodiment of the poultry weight monitoring and health early warning system of the present invention; Figure 2 This is a system topology diagram of a preferred embodiment of the poultry weight monitoring and health early warning system of the present invention; Figure 3 This is a schematic diagram of the poultry weight monitoring and health early warning method of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] like Figure 1 and Figure 2 As shown in the figure, this embodiment of the invention provides a poultry weight monitoring and health early warning system. The system mainly includes a distributed monitoring terminal 100, an edge computing node 200, a cloud management platform 300, an intelligent decision generation module 400, and a user interaction terminal 500. The distributed monitoring terminal is deployed in the free-range area of ​​poultry (e.g., chickens) to automatically collect data in a non-contact manner without disturbing the normal activities of the poultry.

[0021] Specifically, the distributed monitoring terminal 100 includes a weighing unit and a radio frequency identification (RFID) unit. The weighing unit integrates a high-precision pressure sensor and is configured to collect weight data when poultry stand independently (e.g., at feeding or resting points). The RFID unit integrates an RFID read / write module and is configured to read electronic tags worn on the poultry's leg bands or wings to obtain individual identification information. In addition, the terminal also simultaneously collects environmental parameters, including but not limited to temperature, humidity, and light intensity; and in addition to weight, biological data can also be collected via an infrared sensor to collect body temperature data.

[0022] Edge computing node 200 is connected to distributed monitoring terminal 100 via wired or wireless means. To adapt to free-range environments, edge computing nodes can be developed based on embedded systems, such as running the FreeRTOS real-time operating system. The edge computing node is configured to clean (remove noise), time-align, and perform multi-task concurrent scheduling processing on the collected raw data, and upload the processed data to the cloud via a wireless communication network (such as 4G / 5G / NB-IoT).

[0023] The cloud management platform (e.g., built on public or private clouds like Huawei Cloud) communicates with edge computing nodes to receive and store massive amounts of data. The core of the cloud platform consists of two modules: Multidimensional data fusion prediction module: Configured to construct an association mapping model based on historical data (including the historical growth curves of the individual and the group). In one specific embodiment, a machine learning model (such as XGBoost) is used to establish a dynamic correlation between environmental parameters and growth performance, and the future growth trend of poultry is predicted based on real-time environmental parameters and biological ontology data.

[0024] Intelligent decision generation module: configured to use a large language model (such as the Qwen series large model) and retrieval enhancement technology (RAG) to call a pre-built aquaculture knowledge base and generate aquaculture decision suggestions in natural language form based on the growth trend.

[0025] User interaction terminals (such as mobile apps, WeChat mini programs, or PC dashboards) are configured to display the real-time status of poultry (body temperature, weight), growth trend prediction results, and system-generated breeding decision suggestions (such as "It is recommended to adjust the feed ratio" or "Pay attention to the impact of temperature difference").

[0026] In a preferred embodiment, the system further includes a business interaction module. This module is configured to open a data interface to authorized third-party user terminals (e.g., consumers). Third-party users can remotely access real-time weight, activity trajectory, and image data of specific poultry individuals through an application. Preferably, this data supports full lifecycle traceability based on blockchain technology, thereby realizing a "cloud farming" business model.

[0027] The beneficial effects of this embodiment are as follows: Through a collaborative architecture of edge-cloud, intelligent monitoring of the entire lifecycle of free-range poultry is achieved. The non-contact data collection method avoids stress reactions in the poultry, and the distributed architecture adapts to the free-range model, ensuring the comprehensiveness and integrity of the data. The business interaction module extends the value chain of poultry farming and enhances consumer trust.

[0028] This embodiment provides a detailed description of the scheduling logic of the edge computing nodes in Embodiment 1. Because poultry are frequently active in free-range environments, simply collecting data at set intervals may lead to a mismatch between identity information and weight data.

[0029] Therefore, in this embodiment, the edge computing nodes run on a real-time operating system (such as FreeRTOS) and are configured to execute specific multi-task scheduling logic. Specifically, the system sets a first preset period for executing RFID tasks and a second preset period for executing weight data acquisition tasks. The first preset period is shorter than the second preset period.

[0030] In a specific application scenario, the cycle of the RFID task can be set to 20ms, while the cycle of the weight acquisition task can be set to 50ms. When the weighing unit of the distributed monitoring terminal detects stable weight data (e.g., several consecutive readings fluctuating within a threshold), it immediately triggers the high-frequency RFID unit to operate. Due to the extremely short RFID cycle (20ms), the system can quickly lock the poultry's identity ID the instant before it leaves the weighing platform, thereby achieving high-precision binding between individual identity and weight data.

[0031] In addition, edge computing nodes are configured to perform anomaly detection. When anomalies are detected in poultry body temperature or growth data (such as a sudden drop in weight), the edge node does not wait for the regular batch upload cycle, but instead sends a high-priority warning signal to the cloud management platform within a minute-level time window. The cloud management platform responds to this signal by immediately pushing anomaly alerts to the user's interactive terminal.

[0032] The beneficial effects of this embodiment are as follows: by using differentiated multi-task scheduling cycles (RFID is faster than weighing), it effectively solves the problem of difficult identification of live targets in dynamic scenarios, ensuring accurate data association. At the same time, the anomaly priority mechanism on the edge side reduces the early warning delay from hours to minutes (e.g., less than 1 minute), significantly improving the response speed of disease prevention and control.

[0033] This embodiment provides a detailed description of the intelligent decision generation module and its construction process. To address the issues of gaps in aquaculture experience and insufficient intelligent decision-making, this embodiment introduces large AI models and RAG technology. The construction process of the intelligent decision generation module includes the following steps: First, a knowledge base for aquaculture is constructed. The data sources for this knowledge base are extensive, including scientific literature data (such as papers on poultry nutrition), historical aquaculture case data (success or failure records of past batches), and real-time monitoring data.

[0034] Secondly, the aquaculture knowledge base is vectorized and stored in a vector retrieval system.

[0035] When responding to a user query (such as "What should I do if my chickens have a poor appetite lately?") or an automatically triggered warning (such as "The average body temperature in area 3 is detected to be high"), the system first retrieves relevant knowledge fragments from the retrieval system.

[0036] Finally, the retrieved knowledge fragments (as background knowledge) and the current monitoring data (as context) are combined as prompts and input into a large language model (such as Qwen3). Based on these inputs, the large language model generates highly personalized growth reports or dynamic feeding plans (such as "It is recommended to increase the ventilation of area 3 by 10% and supplement with vitamin C").

[0037] The beneficial effects of this embodiment are as follows: by connecting the general large model with the professional aquaculture knowledge base through RAG technology, the generated decision suggestions not only have the reasoning ability of artificial intelligence, but also conform to the professional aquaculture scientific logic, effectively accumulating expert experience, helping farmers reduce the difficulty of decision-making and the cost of trial and error, and actual tests can optimize the feed conversion ratio by 5-10%.

[0038] See Figure 3 This invention also provides a method for monitoring poultry weight and providing health early warning, the method comprising the following steps: S101: Using distributed monitoring terminals deployed in free-range areas, collect individual identification information and biological data of poultry, and simultaneously collect environmental parameters.

[0039] S102: The collected data is cleaned, time-series aligned, and processed by multi-task concurrent scheduling through the edge computing node connected to the distributed monitoring terminal, and the data is uploaded through the wireless communication network.

[0040] S103: The cloud management platform receives and stores data uploaded through the wireless communication network, and uses the multi-dimensional data fusion prediction module to predict the growth trend of poultry based on the correlation mapping model built from historical data and real-time environmental parameters and biological ontology data.

[0041] S104: Using the intelligent decision generation module, based on the large language model and retrieval enhancement technology, the pre-set aquaculture knowledge base is invoked to generate aquaculture decision suggestions in natural language form according to the growth trend.

[0042] S105: Display the real-time status of the poultry, the predicted growth trend, and the breeding decision suggestions through the user interaction terminal.

[0043] This embodiment deeply integrates technologies such as the Internet of Things, edge computing, big data prediction, and artificial intelligence natural language processing to solve the pain points of traditional breeding industry, such as inaccurate data, slow response, reliance on experience, and extensive management. Ultimately, it achieves accurate monitoring of poultry weight, proactive early warning of health risks, and intelligent and humanized breeding decisions.

[0044] Furthermore, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a program for monitoring poultry weight and providing health warnings, and the poultry weight monitoring and health warning program, when executed by a processor, implements the steps of the poultry weight monitoring and health warning method as described above.

[0045] In summary, this invention provides a poultry weight monitoring and health early warning system and method, and a storage medium. The system includes: a distributed monitoring terminal deployed in a free-range area, configured to automatically collect individual identification information and biological data of poultry in a non-contact manner, and simultaneously collect environmental parameters; an edge computing node connected to the distributed monitoring terminal, configured to clean, time-series align, and perform multi-task concurrent scheduling processing on the collected data, and upload the data via a wireless communication network; and a cloud management platform communicatively connected to the edge computing node, used to receive and store the processed data. The cloud management platform includes: a multi-dimensional data fusion and prediction module configured to construct an association mapping model based on historical data, and predict the growth trend of poultry based on real-time environmental parameters and biological data; an intelligent decision generation module configured to use a large language model and retrieval enhancement technology to call a pre-set breeding knowledge base and generate breeding decision suggestions in natural language form based on the growth trend; and a user interaction terminal configured to display the real-time status of the poultry, the predicted growth trend results, and the breeding decision suggestions. This invention achieves automated, non-contact, continuous monitoring of poultry data in free-range environments through the collaboration of distributed monitoring terminals and edge computing nodes, overcoming the spatiotemporal limitations of data collection and improving data integrity. Utilizing a multi-dimensional data fusion and prediction module, combined with environmental and biological ontological data, it can more accurately assess poultry status. By combining RAG technology and a large language model, expert knowledge is digitized, providing targeted natural language decision-making suggestions and lowering the barrier to entry for professional poultry farming. Through edge-side cleaning and priority scheduling, it achieves minute-level anomaly early warning response, significantly improving the timeliness of disease prevention and control.

[0046] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes that element.

[0047] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The computer-readable storage medium can be a memory, magnetic disk, optical disk, etc.

[0048] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A poultry weight monitoring and health early warning system, characterized in that, include: Distributed monitoring terminals are deployed in free-range areas and configured to automatically collect individual identification information and biological data of poultry in a non-contact manner, while simultaneously collecting environmental parameters. Edge computing nodes are connected to the distributed monitoring terminal and configured to clean, time-series align, and perform multi-task concurrent scheduling processing on the collected data, and upload the data through a wireless communication network. A cloud management platform, which communicates with the edge computing nodes, is used to receive and store processed data; The cloud management platform includes a multi-dimensional data fusion and prediction module, configured to build an association mapping model based on historical data, and predict the growth trend of poultry based on real-time environmental parameters and biological ontology data. The intelligent decision generation module is configured to use a large language model and retrieval enhancement technology to call a pre-set aquaculture knowledge base and generate aquaculture decision suggestions in natural language form based on the growth trend. The user interaction terminal is configured to display the real-time status of the poultry, the prediction results of growth trends, and the breeding decision suggestions.

2. The poultry weight monitoring and health early warning system according to claim 1, characterized in that, The distributed monitoring terminal includes: The weighing unit integrates a pressure sensor and is configured to collect weight data when poultry stand up independently. The radio frequency identification unit integrates an RFID read / write module and is configured to read electronic tags worn on poultry to obtain individual identity information. The distributed monitoring terminal is configured to trigger the radio frequency identification unit to work when the weight data is stable, thereby binding the individual identity with the weight data.

3. The poultry weight monitoring and health early warning system according to claim 1, characterized in that, The edge computing nodes run on a real-time operating system and are configured to execute the following multi-task scheduling logic: The radio frequency identification task is executed in a first preset period, and the weight data collection task is executed in a second preset period, wherein the first preset period is shorter than the second preset period, so as to realize the rapid identification of poultry in dynamic scenarios.

4. The poultry weight monitoring and health early warning system according to claim 1, characterized in that, The environmental parameters include temperature, humidity, and light intensity; the biological data include weight and body temperature.

5. The poultry weight monitoring and health early warning system according to claim 1, characterized in that, The construction process of the intelligent decision generation module includes: Construct an aquaculture knowledge base, which includes scientific literature data, historical aquaculture case data, and real-time monitoring data; The aquaculture knowledge base is vectorized and stored in the retrieval system; in response to user queries or system alerts, relevant knowledge fragments are retrieved from the retrieval system. The retrieved knowledge fragments and current monitoring data are used as prompt words and input into the large language model to generate personalized growth reports or dynamic feeding plans.

6. The poultry weight monitoring and health early warning system according to claim 3, characterized in that, The edge computing node is configured to send a high-priority warning signal to the cloud management platform within a minute-level time window when it detects abnormal poultry body temperature or growth data. The cloud management platform responds to the signal by immediately pushing an anomaly notification to the user interaction terminal.

7. The poultry weight monitoring and health early warning system according to any one of claims 1-6, characterized in that, The system also includes: Business interaction module: The business interaction module is configured to open a data interface to authorized third-party user terminals; the third-party user terminals are configured to remotely access the real-time weight, activity trajectory and image data of specific poultry individuals through an application, and support full life-cycle traceability viewing based on blockchain technology.

8. A method for monitoring poultry weight and providing early warning of health conditions, characterized in that, include: Using distributed monitoring terminals deployed in free-range areas, individual identification information and biological data of poultry are collected, and environmental parameters are collected simultaneously. The collected data is cleaned, time-series aligned, and processed by multi-task concurrent scheduling through edge computing nodes connected to the distributed monitoring terminal, and then uploaded through a wireless communication network. The cloud management platform receives and stores data uploaded via wireless communication network, and uses a multi-dimensional data fusion prediction module to predict the growth trend of poultry based on the correlation mapping model built from historical data and real-time environmental parameters and biological ontology data. Using the intelligent decision generation module, based on the large language model and retrieval enhancement technology, and calling the pre-built aquaculture knowledge base, aquaculture decision suggestions in natural language form are generated according to the growth trend. The real-time status of the poultry, growth trend prediction results, and breeding decision suggestions are displayed through the user interaction terminal.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a poultry weight monitoring and health early warning program, which, when executed by a processor, implements the steps of the poultry weight monitoring and health early warning method as described in claim 8.

Images