Ecological cycle agricultural data collection and analysis method and system based on internet of things

Abstract

The application relates to the technical field of data analysis, and discloses an ecological circulation agricultural data collection and analysis method and system based on the Internet of Things. The method comprises the following steps: performing cluster processing on a sensor node through an LMCO algorithm to obtain a heterogeneous device perception network; performing predictive data collection on a UAV RSSI signal according to the perception network to obtain a space-time coupling dataset; performing causal mapping processing on the dataset by a cloud node to obtain causal chain data; performing influence mining on the causal chain data by a circulation effect propagation algorithm to obtain propagation rules and efficiency parameters; and performing intelligent feedback regulation on a collection strategy according to the efficiency parameters to obtain an optimized decision scheme. The application solves the problem that the prior art cannot establish a causal relationship analysis among ecological circulation agricultural links and an intelligent feedback regulation mechanism based on a circulation effect.

Description

Technical Field

[0001] This application relates to the field of data analysis technology, and in particular to a method and system for data collection and analysis in ecological circular agriculture based on the Internet of Things. Background Technology

[0002] In existing technologies, IoT-based agricultural data acquisition and analysis methods mainly employ fixed sensor deployments and timed data acquisition modes. These methods monitor agricultural environmental parameters by deploying sensors such as soil moisture, temperature, and light intensity in farmland, and then use wireless communication technology to transmit the collected data to the cloud for analysis. Traditional methods typically use pre-defined sensor cluster configurations and fixed data acquisition frequencies, combined with simple statistical analysis algorithms to monitor and manage agricultural production processes. Existing agricultural IoT systems also incorporate mobile platforms such as drones for data acquisition, using GPS navigation and pre-defined flight paths to achieve remote sensing monitoring of large areas of farmland.

[0003] However, existing technologies have significant shortcomings: sensor nodes lack intelligent cluster optimization mechanisms and cannot dynamically adjust the network topology according to actual agricultural production needs; UAV data acquisition uses fixed path planning and lacks adaptive adjustment capabilities based on real-time signal quality; data analysis methods are limited to independent analysis of single links and cannot effectively identify and quantify the correlation between various links in agricultural production; acquisition strategies lack intelligent feedback mechanisms based on analysis results and are difficult to dynamically optimize data acquisition parameters according to the system's operating status.

[0004] Planting, animal husbandry, waste treatment, and resource recycling form a network of material flow and energy conversion. Time delays and propagation effects exist between these stages. Traditional methods of analyzing individual stages cannot reveal the causal relationships and cyclical effects between them, making it impossible to accurately assess the overall efficiency and optimization potential of the circular system. Due to a lack of in-depth analysis of the propagation patterns of cyclical effects, existing technologies struggle to establish intelligent feedback control mechanisms based on cyclical efficiency parameters. They also cannot dynamically adjust data acquisition strategies according to the operational status of the circular system, thus affecting the precision management and efficiency optimization of ecological circular agriculture systems. Summary of the Invention

[0005] This application provides a method and system for data collection and analysis in ecological circular agriculture based on the Internet of Things, which solves the problem that existing technologies cannot establish causal relationship analysis between links in ecological circular agriculture and intelligent feedback control mechanisms based on circular effects.

[0006] Firstly, this application provides a method for data collection and analysis of ecological circular agriculture based on the Internet of Things (IoT). The method includes: performing cross-domain collaborative cluster processing on ecological circular agriculture sensor nodes using the LMCO algorithm to obtain a heterogeneous sensing network for circular agriculture; based on the heterogeneous sensing network, performing predictive data collection and processing of UAV RSSI signals on the ecological circular closed loop to obtain a spatiotemporally coupled dataset for circular agriculture; based on the spatiotemporally coupled dataset, performing causal mapping processing on IoT cloud nodes to the agricultural circular chain to obtain ecological circular causal chain data; performing cross-link impact mining processing on the ecological circular causal chain data using a circular effect propagation algorithm to obtain agricultural circular propagation rules and circular efficiency enhancement parameters; and based on the circular efficiency enhancement parameters, performing intelligent feedback regulation processing on the ecological agricultural circular system using IoT collection strategies to obtain an optimized decision-making scheme for circular agriculture.

[0007] Secondly, this application provides an IoT-based ecological circular agriculture data collection and analysis system, which includes:

[0008] The cluster module is used to perform cross-domain collaborative cluster processing of ecological circular agriculture sensor nodes through the LMCO algorithm to obtain a circular agriculture heterogeneous equipment sensing network.

[0009] The acquisition module is used to perform predictive data acquisition and processing of the UAV RSSI signal for the ecological cycle closed loop based on the sensing network of the heterogeneous equipment of the circular agriculture, so as to obtain the spatiotemporal coupled dataset of circular agriculture.

[0010] The mapping module is used to perform causal mapping processing on the IoT cloud nodes to the agricultural cycle chain based on the spatiotemporal coupling dataset of the circular agriculture, so as to obtain ecological cycle causal chain data.

[0011] The mining module is used to perform cross-link impact mining on the ecological cycle causal chain data through the cycle effect propagation algorithm to obtain agricultural cycle propagation rules and cycle efficiency enhancement parameters.

[0012] The control module is used to intelligently feedback control the ecological agricultural cycle system based on the cycle efficiency enhancement parameters and the IoT acquisition strategy to obtain an optimized decision-making scheme for cycle agriculture.

[0013] Thirdly, an IoT-based ecological circular agriculture data acquisition and analysis device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the IoT-based ecological circular agriculture data acquisition and analysis device to execute the aforementioned IoT-based ecological circular agriculture data acquisition and analysis method.

[0014] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored in the computer and, when executed on a computer, cause the computer to perform the above-described method for data collection and analysis of ecological circular agriculture based on the Internet of Things.

[0015] The technical solution provided in this application utilizes the LMCO algorithm for cross-domain collaborative clustering of sensor nodes in ecological circular agriculture, addressing the lack of intelligent optimization configuration in existing sensor networks. The LMCO algorithm combines the group collaboration characteristics of the Lion algorithm with the local search capability of the cat-and-mouse optimization algorithm, comprehensively considering multi-dimensional parameters such as distance between nodes, latency, and energy consumption for cluster optimization. Compared to traditional random deployment or empirical configuration methods, it significantly improves the communication efficiency and energy consumption control of sensor networks. Based on the predictive data acquisition and processing of UAV RSSI signals from the heterogeneous equipment sensing network in circular agriculture, real-time monitoring of signal strength changes and dynamic adjustment of flight parameters overcome the signal blind spots and uneven data acquisition problems caused by fixed paths used by UAVs in existing technologies, ensuring optimal data reception quality in each cluster area. The IoT cloud node establishes a unified data preprocessing and classification labeling mechanism for causal mapping of the agricultural cycle chain, solving the technical challenges of inconsistent formats and varying quality of multi-source heterogeneous data, laying a reliable data foundation for subsequent in-depth analysis.

[0016] The cyclic effect propagation algorithm, through time delay feature identification, impact propagation path construction, and multi-level propagation and diffusion processing, achieves for the first time a quantitative analysis of complex causal relationships among planting, breeding, waste treatment, and resource recycling, overcoming the limitations of existing technologies that can only perform independent analysis of a single link. The algorithm's multi-level propagation mechanism can distinguish between direct, indirect, and feedback effects, accurately capturing the hierarchical propagation characteristics of impacts in the ecological cycle system—a technical effect that traditional linear analysis methods cannot achieve. Based on intelligent feedback control processing using cyclic efficiency enhancement parameters, an adaptive optimization mechanism for data acquisition strategies is established. Through a series of techniques such as threshold judgment, dynamic reconfiguration, and synchronous adjustment, real-time linkage between the acquisition system and the cycle system's operating status is achieved, solving the problem of fixed acquisition strategies and the inability to dynamically adjust according to system status in existing technologies. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of an embodiment of the IoT-based ecological circular agriculture data collection and analysis method in this application.

[0019] Figure 2 This is a schematic diagram of an embodiment of the IoT-based ecological circular agriculture data collection and analysis system in this application.

[0020] Figure 3 This is a schematic block diagram of the structure of the IoT-based ecological circular agriculture data collection and analysis device in this embodiment of the invention. Detailed Implementation

[0021] This application provides a method and system for data collection and analysis in ecological circular agriculture based on the Internet of Things. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0022] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the data collection and analysis method for ecological circular agriculture based on the Internet of Things in this application includes:

[0023] Step S101: Perform cross-domain collaborative cluster processing on the sensor nodes of ecological circular agriculture using the LMCO algorithm to obtain the sensing network of heterogeneous equipment in circular agriculture.

[0024] Step S102: Based on the sensing network of heterogeneous equipment in circular agriculture, the RSSI signal of the UAV is used to perform predictive data acquisition and processing on the ecological cycle closed loop to obtain a spatiotemporal coupled dataset of circular agriculture.

[0025] Step S103: Based on the spatiotemporal coupling dataset of circular agriculture, perform causal mapping processing on the IoT cloud nodes to the agricultural cycle chain to obtain ecological cycle causal chain data.

[0026] Step S104: Perform cross-link impact mining on the ecological cycle causal chain data through the cycle effect propagation algorithm to obtain agricultural cycle propagation rules and cycle efficiency enhancement parameters;

[0027] Step S105: Based on the cyclical efficiency enhancement parameters, the IoT acquisition strategy is used to intelligently feedback and regulate the ecological agricultural cycle system to obtain an optimized decision-making scheme for circular agriculture.

[0028] It is understood that the executing entity of this application can be an IoT-based ecological circular agriculture data collection and analysis system, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiment uses a server as an example for illustration.

[0029] Specifically, the LMCO algorithm is used to perform cross-domain collaborative clustering of sensor nodes in ecological circular agriculture, resulting in a heterogeneous sensing network for circular agriculture. The LMCO algorithm combines the group cooperation characteristics of the Lion algorithm with the local search capability of the cat-and-mouse optimization algorithm to collect coordinates of sensor nodes in planting, breeding, and waste treatment areas, obtaining the spatial location information of each sensor node and forming a set of node spatial coordinates. Based on this coordinate set, individual lions are initialized, mapping the position of each sensor node to the position of a lion in the group, constructing a lion position matrix. Then, the fitness function is calculated using the position matrix, distance between sensor nodes, data transmission delay, and energy consumption parameters to comprehensively evaluate the suitability of each node as the group leader. A cat-and-mouse chasing strategy is used to locally search and optimize the fitness values, simulating the behavior of a cat chasing a mouse, to find a better configuration scheme for the group leader node in the solution space. The group leader node is selected based on the optimized individual lion positions, forming the heterogeneous sensing network for circular agriculture.

[0030] Based on the heterogeneous device sensing network for circular agriculture, the RSSI signals from drones are used for predictive data acquisition and processing of the ecological cycle closed loop, resulting in a spatiotemporally coupled dataset for circular agriculture. Based on the coordinate information of the cluster head nodes in the heterogeneous device sensing network, the drone employs an improved ant colony optimization algorithm for flight path planning, using the shortest path, lowest energy consumption, and optimal signal coverage as objective functions to calculate the optimal flight trajectory sequence. The drone is equipped with a multi-band communication module to measure the RSSI signal strength with each cluster head node in real time, dynamically adjusting its flight altitude and speed based on signal strength changes to ensure optimal signal reception over each cluster area. At the optimal signal reception location, the drone simultaneously collects environmental data from ground sensors and high-resolution multispectral images and infrared thermal imaging data acquired by its own onboard equipment, forming multi-source fusion data. This multi-source fusion data is coupled and labeled with corresponding timestamps and spatial coordinate information to establish a correlation between the data and its spatiotemporal location, generating a spatiotemporally coupled dataset for circular agriculture.

[0031] Based on the spatiotemporal coupling dataset of circular agriculture, IoT cloud nodes are used to perform causal mapping on the agricultural cycle chain to obtain ecological cycle causal chain data. After receiving the spatiotemporal coupling dataset transmitted by the UAV, the base station evaluates the data volume, data type, and processing urgency to generate cloud node selection parameters. A fuzzy comprehensive evaluation method is used to comprehensively score parameters such as the response rate, energy efficiency ratio, and service quality of available IoT cloud nodes, selecting the most suitable cloud computing resources for the current data processing task and obtaining the optimal cloud node identifier. The spatiotemporal coupling dataset is input into the selected cloud nodes, and timestamp alignment and coordinate system unification are performed to eliminate time differences and spatial coordinate deviations between different data sources, forming a standardized dataset. Based on the characteristics of ecological circular agriculture, the standardized dataset is classified and labeled according to planting, breeding, waste treatment, and resource recycling stages. A data quality assessment model is established to automatically identify abnormal, missing, and redundant data and mark them accordingly, generating ecological cycle causal chain data.

[0032] The cyclic effect propagation algorithm is used to mine cross-stage impacts on ecological cycle causal chain data, resulting in agricultural cycle propagation rules and cycle efficiency parameters. The algorithm identifies time delay features in each stage of the causal chain data, analyzing the time difference between organic waste generation in the planting stage and feed demand in the livestock stage, and the time difference between manure generation in the livestock stage and load changes in the waste treatment stage, constructing a time delay matrix between stages. Based on the time delay matrix, a cyclic effect propagation network topology is constructed, with each stage as a network node and material flow and time correlation between stages as connecting edges, forming a propagation path for stage impacts. Weights are calculated for the propagation paths and data changes in each stage, quantifying the impact of parameter changes in one stage on other stages, obtaining inter-stage impact weight coefficients. Based on these weight coefficients, multi-level propagation and diffusion processing of the cyclic effect is performed, distinguishing between direct, indirect, and feedback effects, constructing a cross-stage impact propagation matrix. Based on the propagation matrix, a quantitative assessment of cycle efficiency is conducted, calculating the resource conversion efficiency, energy flow efficiency, and waste utilization rate of the entire ecological cycle system, generating agricultural cycle propagation rules and cycle efficiency parameters.

[0033] Based on the efficiency-enhancing parameters, the IoT data acquisition strategy is applied to the ecological agricultural cycle system through intelligent feedback regulation, resulting in an optimized decision-making scheme for circular agriculture. A three-layer feedback control architecture is established based on the efficiency-enhancing parameters. The real-time feedback layer monitors abnormal changes in key parameters. When a significant decrease in efficiency is detected in a certain link, an automatic data acquisition frequency adjustment command is generated to increase the data acquisition density of relevant sensors. Based on the adjustment command, the heterogeneous device sensing network is dynamically reconfigured, reallocating the cluster affiliation and data transmission paths of sensor nodes to form an optimized sensor network configuration. The optimized network configuration is synchronized with the drone inspection cycle. When soil fertility in the planting area declines, drones are scheduled to focus on monitoring the material flow paths in the waste treatment area and the livestock area, generating a collaborative data acquisition scheduling scheme. Based on the scheduling scheme, the operating parameters of the ecological agricultural cycle system are predicted and analyzed. A long short-term memory network prediction model is used to construct short-term, medium-term, and long-term prediction models to predict the system's operating status at different time scales, obtaining prediction results for cycle benefits. Based on the prediction results, a decision-making scheme is generated and economic benefits are evaluated. Taking into account the input-output ratio and environmental benefit indicators, an optimized decision-making scheme for circular agriculture is generated.

[0034] In one specific embodiment, the process of performing step S101 may specifically include the following steps:

[0035] The coordinates of sensor nodes in the planting area, breeding area, and waste treatment area are collected and processed to obtain a set of node spatial coordinates.

[0036] The lion pride individual initialization process is performed based on the node spatial coordinate set to obtain the lion pride individual position matrix;

[0037] The fitness values ​​of individual lions are obtained by calculating the fitness function of the individual lion position matrix and the sensor node distance, delay, and energy consumption parameters.

[0038] The fitness values ​​of individual lions in the pride are optimized by local search using a cat-and-mouse chasing strategy, resulting in the updated positions of individual lions in the pride.

[0039] Based on the updated individual lion pride locations, cluster head node selection is performed to obtain the sensing network for heterogeneous equipment in circular agriculture.

[0040] Specifically, coordinate acquisition and processing are performed on sensor nodes within the planting, breeding, and waste treatment areas to obtain a set of node spatial coordinates. This step uses a GPS positioning module to acquire the precise geographical location information of each IoT sensor node, converting the sensor positions in three-dimensional space into digital coordinate data. Coordinate acquisition and processing includes measuring and recording the longitude, latitude, and altitude of each sensor node, forming a spatial coordinate set containing the location information of all sensor nodes. The coordinate data is represented using a unified coordinate system to ensure the consistency and accuracy of sensor node location information across different areas. Based on the node spatial coordinate set, individual lion group initialization processing is performed to obtain a lion group individual position matrix. The Lion algorithm part of the LMCO algorithm maps the spatial coordinates of each sensor node to the position of an individual in the lion group. The individual lion group initialization processing converts each coordinate point in the coordinate set into a solution vector in the lion group optimization algorithm, forming the lion group individual position matrix. Each row in the position matrix represents a lion group individual, and each column represents the position value of that individual in the corresponding dimension. The number of rows in the matrix equals the total number of sensor nodes, and the number of columns equals the number of coordinate dimensions. This mapping relationship establishes the correspondence between the physical positions of sensor nodes and the algorithm optimization variables.

[0041] The fitness function is applied to the individual lion population location matrix and sensor node distance, delay, and energy consumption parameters to obtain the individual lion population fitness value. The fitness function calculation comprehensively considers three key parameters: physical distance between sensor nodes, data transmission delay time, and node energy consumption level. The distance parameter is obtained by calculating the Euclidean distance between any two individuals in the lion population location matrix. The delay parameter calculates the data transmission time based on the network topology and communication protocol between sensor nodes. The energy consumption parameter assesses the energy consumption per unit time based on the sensor node's hardware specifications and operating mode. The fitness function weights and sums these three parameters, with weighting coefficients determined according to the application requirements of ecological circular agriculture, to calculate the comprehensive fitness value for each individual lion population. A higher fitness value indicates a greater suitability of the sensor node as the group leader.

[0042] The algorithm employs a cat-and-mouse chase strategy to optimize the fitness values ​​of individual lions within the pride through local search, resulting in updated individual lion positions. This cat-and-mouse chase strategy is the core mechanism of the cat-and-mouse optimization algorithm integrated within the LMCO algorithm, simulating the natural behavior of a cat chasing a mouse. The local search optimization process first selects lions with higher fitness values ​​as the initial cat positions and lions with lower fitness values ​​as the mouse positions. When a cat moves towards a mouse, it updates its own position; the movement distance and direction are calculated based on the distance between the cat and mouse and the fitness difference. During the chase, the algorithm evaluates the fitness value of the new position. If the new position has a higher fitness value, the position is updated; otherwise, the original position is maintained. Through multiple iterative rounds of cat-and-mouse chase, the algorithm gradually optimizes the positional distribution of the lions, obtaining the updated individual lion positions.

[0043] The cluster head node selection process is performed based on the updated individual lion pride locations to obtain the heterogeneous equipment sensing network for circular agriculture. The selection process is based on the updated individual lion pride locations and their corresponding fitness values, choosing the locations of several lion pride individuals with the highest fitness values ​​as the spatial locations of the cluster head nodes. During the selection process, it is necessary to ensure a reasonable spatial distribution of the cluster head nodes, avoiding excessive concentration of multiple cluster head nodes or lack of cluster head coverage in certain areas. By setting minimum spacing constraints and coverage requirements, combinations of cluster head nodes that can effectively cover planting areas, breeding areas, and waste treatment areas are selected. The heterogeneous equipment sensing network for circular agriculture consists of the selected cluster head nodes and ordinary sensor nodes belonging to each cluster head. In the network topology, the cluster head nodes are responsible for collecting sensor data within their cluster and communicating with higher-level nodes, while ordinary sensor nodes are responsible for collecting environmental data and reporting it to the cluster head nodes.

[0044] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0045] The optimal flight trajectory sequence is obtained by processing the flight path planning of UAVs based on the coordinates of the cluster head node of the heterogeneous equipment sensing network in circular agriculture.

[0046] The RSSI signal strength of the UAV and the distance to the cluster head node are measured and processed in real time to obtain the signal strength change parameters.

[0047] The drone's flight altitude and speed are dynamically adjusted based on the signal strength change parameters to obtain the optimal signal reception position;

[0048] Based on the optimal signal receiving location, ground sensor data and aerial remote sensing data are simultaneously acquired and processed to obtain multi-source fused data.

[0049] By coupling and labeling multi-source fused data with spatiotemporal coordinate information, a spatiotemporal coupled dataset for circular agriculture is obtained.

[0050] Specifically, UAV flight path planning is performed based on the coordinates of the cluster head nodes of the heterogeneous equipment sensing network in circular agriculture to obtain the optimal flight trajectory sequence. The UAV flight path planning process employs an improved ant colony optimization algorithm, using the cluster head node coordinates as the target access points for path planning. The algorithm performs multi-objective optimization with shortest path, lowest energy consumption, and optimal signal coverage as objective functions. The path planning process first establishes a set of access points containing the coordinates of all cluster head nodes. Then, it calculates the flight distance and flight time between access points. Considering the UAV's flight speed limitations and battery life constraints, the shortest path traversing all cluster head nodes is found through the pheromone update mechanism of the ant colony algorithm, generating the optimal flight trajectory sequence that includes the flight start point, the access order of each cluster head node, and the flight destination.

[0051] The RSSI signal strength of the UAV and its distance to the cluster head node are measured and processed in real time to obtain signal strength variation parameters. The RSSI signal strength measurement and processing is achieved by the multi-band communication module onboard the UAV, which monitors the wireless signal reception strength indicator values ​​between the UAV and each cluster head node in real time. Real-time measurement and processing includes continuously acquiring the spatial distance between the UAV's current position and the target cluster head node during flight, while simultaneously recording the received RSSI signal strength values, and establishing a data table showing the correspondence between distance and signal strength. The signal strength variation parameters are obtained by analyzing the trend of RSSI values ​​changing with distance, including signal strength attenuation rate, signal stability index, and signal quality evaluation parameters. These parameters reflect the quality of the communication link between the UAV and the cluster head node.

[0052] The drone's flight altitude and speed are dynamically adjusted based on signal strength variation parameters to determine the optimal signal reception position. This dynamic adjustment is based on a flight parameter optimization model established by the signal strength variation parameters. When the RSSI signal strength is detected to be below a preset threshold, the algorithm automatically adjusts the drone's flight altitude, reducing the signal transmission distance and thus enhancing signal strength. Flight speed adjustment is based on signal stability indicators. When signal fluctuations are large, the flight speed is reduced to extend signal acquisition time; when the signal is stable, the flight speed is appropriately increased to improve data acquisition efficiency. The optimal signal reception position is determined through an iterative optimization algorithm. The algorithm searches in three-dimensional space for the drone's position coordinates that maximize the RSSI signal strength, satisfying both signal quality requirements and flight safety and energy consumption control. Based on the optimal signal reception position, ground sensor data and aerial remote sensing data are simultaneously acquired and processed to obtain multi-source fused data. This simultaneous acquisition and processing initiates multiple data acquisition modes after the drone reaches the optimal signal reception position. Ground sensor data acquisition receives environmental monitoring data aggregated by the sensor network, including parameters such as soil moisture, temperature, humidity, pH value, and light intensity, through the wireless communication link between the drone and the cluster head node. Aerial remote sensing data acquisition utilizes high-resolution cameras, multispectral sensors, and infrared thermal imaging equipment mounted on unmanned aerial vehicles (UAVs) to acquire images and spectral information of agricultural areas. Multi-source fusion data processing spatially registers and temporally synchronizes ground-based point sensor data with aerial area remote sensing data, establishing a unified data format and coordinate system to form a comprehensive dataset containing multiple data types.

[0053] By coupling and labeling multi-source fused data with spatiotemporal coordinate information, a spatiotemporally coupled dataset for circular agriculture is obtained. The coupling and labeling process adds precise timestamps and spatial coordinate labels to each data record. The timestamp records the specific time of data collection, including year, month, day, hour, minute, and second information, while the spatial coordinates include the latitude, longitude, and altitude of the data collection point. The labeling process establishes the correlation between data content and spatiotemporal location by assigning unique spatiotemporal identifiers to each sensor and remote sensing data, creating a data traceability and query index. The spatiotemporally coupled dataset for circular agriculture integrates scattered multi-source data into a structured dataset with a unified spatiotemporal reference framework. Each record in the dataset contains complete information such as data value, data type, collection time, collection location, and data quality evaluation.

[0054] In one specific embodiment, the process of executing step S103 may specifically include the following steps:

[0055] The data volume, data type, and processing urgency of the spatiotemporally coupled dataset of circular agriculture are evaluated to obtain cloud node selection parameters.

[0056] Based on cloud node selection parameters, the response rate, energy efficiency ratio, and service quality of IoT cloud nodes are comprehensively scored to obtain the optimal cloud node identifier.

[0057] The spatiotemporal coupled dataset of circular agriculture is input into the optimal cloud node for timestamp alignment and coordinate unification to obtain a standardized dataset.

[0058] Based on the standardized dataset, the planting, breeding, waste treatment, and resource recycling processes are classified and labeled to obtain process classification data.

[0059] Data quality inspection and anomaly labeling are performed based on segment classification data to obtain ecological cycle causal chain data.

[0060] Specifically, the data volume, data type, and processing urgency of the spatiotemporal coupled dataset for circular agriculture are evaluated to obtain cloud node selection parameters. The evaluation quantifies the data volume parameter by statistically analyzing the total number of data records, data file size, and data transmission bandwidth requirements within the dataset. Data type evaluation identifies different formats of data in the dataset, such as sensor numerical data, image data, spectral data, and text data, and analyzes the storage space requirements and computational complexity of each data type. The processing urgency evaluation classifies data according to its timeliness requirements, marking environmental anomaly monitoring data requiring real-time response as high urgency, daily production monitoring data as medium urgency, and historical trend analysis data as low urgency, thus forming cloud node selection parameters that include data volume level, data type weight, and urgency coefficient.

[0061] Based on cloud node selection parameters, a comprehensive scoring process is performed on the response rate, energy efficiency ratio, and service quality of IoT cloud nodes to obtain the optimal cloud node identifier. The comprehensive scoring process employs a fuzzy comprehensive evaluation method to assess available cloud nodes from multiple dimensions. Response rate evaluation measures the timeliness of cloud computing resources by testing the response time and success rate of each cloud node to standard requests. Energy efficiency ratio evaluation evaluates resource utilization efficiency by calculating the ratio of energy consumed per unit of computing task to completion time. Service quality evaluation assesses service level by analyzing the stability, reliability, and data security capabilities of cloud nodes. The scoring process calculates a weighted average of the three evaluation dimensions according to preset weights. The weight allocation is dynamically adjusted based on the characteristics of the current data processing task: a higher weight is given to response rate for high-urgency data processing tasks, a higher weight to energy efficiency ratio for large-volume data processing tasks, and a higher weight to service quality for sensitive data processing tasks. The cloud node with the highest comprehensive score is selected as the optimal cloud node identifier.

[0062] The spatiotemporal coupled dataset of circular agriculture is input into the optimal cloud node for timestamp alignment and coordinate unification processing to obtain a standardized dataset. Timestamp alignment addresses the issue of asynchronous data collection times from different data sources by establishing a unified time benchmark, converting the timestamps of all data records to the same time format and time zone standard. Alignment processing includes identifying the original timestamp format of each data record, calculating the time deviation between different data sources, and applying a time deviation correction formula to adjust all timestamps to a unified time benchmark. Coordinate unification addresses the issue of different sensors and remote sensing equipment using different coordinate systems. Coordinate transformation algorithms convert all spatial location data into a unified geographic coordinate system, including latitude and longitude coordinate system conversion, elevation benchmark unification, and projection standardization, forming a standardized dataset with a consistent spatiotemporal reference frame.

[0063] Based on a standardized dataset, the planting, breeding, waste treatment, and resource recycling stages are categorized and labeled to obtain stage-specific data. The classification and labeling process establishes a stage-specific classification rule base based on the business logic of ecological circular agriculture. This rule base defines the data characteristics and identification standards for each stage. Planting stage data includes soil parameters, crop growth indicators, irrigation records, and other data directly related to crop cultivation. Breeding stage data includes feed input, animal health status, environmental temperature and humidity, and other data related to livestock and poultry farming. Waste treatment stage data includes organic waste production, treatment equipment operating status, waste conversion efficiency, and other data related to waste disposal. Resource recycling stage data includes circulating material flow, energy conversion efficiency, and circulating path monitoring, and other data related to resource recycling. The labeling process automatically identifies data content and assigns corresponding stage labels using a pattern matching algorithm. For data that cannot be automatically identified, manual labeling is used to ensure classification accuracy, generating stage-specific classification data containing stage labels.

[0064] Based on the segmented data, data quality inspection and anomaly labeling are performed to obtain ecological cycle causal chain data. The data quality inspection process establishes a multi-level quality assessment system, including data integrity inspection, data consistency inspection, and data accuracy inspection. Integrity inspection assesses data completeness by statistically analyzing the proportion of missing values ​​and the continuity of data records at each segment. Consistency inspection determines data consistency by comparing relevant data collected by different sensors within the same time period to determine if there are logical conflicts. Accuracy inspection identifies anomalous values ​​by comparing sensor data with known standard values ​​or historical normal ranges. Anomaly labeling classifies and labels the detected anomalous data: missing data is labeled as null value anomalies, data exceeding the normal range is labeled as numerical anomalies, data with abrupt changes in the time series is labeled as trend anomalies, and data conflicting with other data sources is labeled as consistency anomalies, forming ecological cycle causal chain data that includes data quality evaluation and anomaly type identification.

[0065] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0066] Time delay feature identification processing is performed on the planting, breeding, waste treatment and resource recycling links in the ecological cycle causal chain data to obtain the time delay matrix between links;

[0067] Based on the time delay matrix between links, the topology of the cyclic effect propagation network is constructed to obtain the propagation path of the link's influence.

[0068] The influence propagation path of each link and the change in data of each link are weighted and processed to obtain the influence weight coefficient between links;

[0069] Based on the inter-stage influence weighting coefficients, the cyclical effect is processed through multi-level propagation and diffusion to obtain the cross-stage influence propagation matrix.

[0070] Based on the cross-stage influence propagation matrix, the cycle efficiency was quantitatively evaluated to obtain the agricultural cycle propagation rules and cycle efficiency enhancement parameters.

[0071] Specifically, time delay feature identification processing is performed on the planting, breeding, waste treatment, and resource recycling links in the ecological cycle causal chain data to obtain an inter-link time delay matrix. This time delay feature identification processing determines the time lag effect between links by analyzing the causal relationships in the time series of data from each link. The identification process first establishes causal relationship assumptions between links. For example, organic waste generated in the planting link needs to be collected and transported to the waste treatment link, a process with a time delay. Similarly, manure generated in the breeding link needs to be fermented to become organic fertilizer for use in the planting link, a conversion process also with a time delay. The identification algorithm calculates the time offset corresponding to the maximum correlation coefficient between data sequences from different links using cross-correlation analysis. When the organic waste production data from the planting link and the treatment load data from the waste treatment link reach maximum correlation at a specific time offset, this time offset is the time delay between the two links. The processing calculates the time delay for all link pairs, forming an inter-link time delay matrix with links as row and column identifiers and time delay values ​​as matrix elements.

[0072] Based on the time delay matrix between links, a cyclical effect propagation network topology is constructed to obtain the propagation paths of link influence. This topology process abstracts each link in ecological circular agriculture as a network node and the influence relationships between links as network edges, constructing a directed weighted graph structure. The topology process determines the existence and direction of network edges based on the delay values ​​in the time delay matrix. When the time delay value between two links is less than a preset threshold, a direct influence relationship is considered to exist, and a directed edge is established, with the edge pointing from the source link to the affected link. The process also assigns weights to network edges based on the time delay value; a smaller time delay indicates faster propagation of influence, and the corresponding edge weight is larger; a larger time delay indicates slower propagation of influence, and the corresponding edge weight is smaller. The propagation paths of link influence are obtained using a path search algorithm in graph theory. The algorithm identifies all paths from any link to other links, including direct paths and indirect paths through intermediate links, forming a set of paths describing how influence propagates between links.

[0073] The influence propagation path of each stage and the amount of data change in each stage are weighted to obtain the inter-stage influence weight coefficient. This weight calculation process quantifies the impact of data changes in one stage on other stages. The calculation first statistically analyzes the amount of data change in each stage within a specific time window, including indicators such as the magnitude of data increase or decrease, frequency of change, and trend. Then, the propagation intensity of the influence is calculated based on the influence propagation path. The weight calculation considers the path length factor; the influence between directly connected stages has a higher weight, while the influence propagated through multiple intermediate stages has a lower weight. The calculation formula uses the reciprocal of the path length as the basic weight, multiplied by the normalized value of the data change to obtain the standardized influence weight. The process also considers the time decay factor; the influence intensity gradually weakens over time. A time decay function is introduced to adjust the weight coefficients, forming inter-stage influence weight coefficients that reflect the true influence relationship.

[0074] The cyclical effect is processed through multi-level propagation and diffusion based on the inter-link influence weighting coefficients to obtain a cross-link influence propagation matrix. This multi-level propagation and diffusion process simulates the hierarchical propagation process of influence in the ecological cycle system. The propagation and diffusion process divides influence propagation into three levels: primary propagation, secondary propagation, and tertiary propagation. Primary propagation describes the influence relationship between directly adjacent links, secondary propagation describes the indirect influence relationship through an intermediate link, and tertiary propagation describes the feedback influence relationship in the cyclic loop. The processing algorithm first calculates the influence intensity of primary propagation by directly using the inter-link influence weighting coefficients as the primary propagation coefficients. Then, it calculates the influence intensity of secondary propagation by multiplying the primary propagation coefficients by the path attenuation coefficients using matrix multiplication. Finally, it calculates the influence intensity of tertiary propagation, considering the amplification or inhibition effect of cyclic feedback on influence propagation. The cross-link influence propagation matrix is ​​obtained by weighted summation of the propagation coefficients at the three levels. Each element in the matrix represents the comprehensive influence intensity between corresponding link pairs, with rows representing source links and columns representing affected links.

[0075] Based on the cross-stage influence propagation matrix, a quantitative assessment of cycle efficiency is performed to obtain agricultural cycle propagation rules and cycle enhancement parameters. The quantitative assessment evaluates the operational efficiency of the entire ecological cycle system by analyzing the numerical distribution in the influence propagation matrix. The assessment calculates the uniformity index of influence propagation, using the variance of matrix elements to determine whether the influence distribution is balanced. Smaller variance indicates a more uniform distribution of influence across stages and higher cycle efficiency, while larger variance indicates the existence of bottlenecks and lower cycle efficiency. The process also calculates the connectivity index of influence propagation, assessing the connection density between stages by the proportion of non-zero elements in the statistical matrix. High connection density indicates a high degree of diversity in cycle paths and strong system resilience, while low connection density indicates a single cycle path and high system vulnerability. The agricultural cycle propagation rules extract typical influence patterns from the influence propagation matrix using cluster analysis, identifying efficient and inefficient propagation patterns. The cycle enhancement parameters quantify the cycle gain effect of the system by calculating matrix eigenvalues, with the magnitude of the eigenvalues ​​reflecting the cycle amplification factor.

[0076] In one specific embodiment, the process of performing multi-level propagation and diffusion processing of the cyclic effect based on the inter-stage influence weighting coefficients can specifically include the following steps:

[0077] The influence weight coefficients between links are divided into propagation levels to obtain first-level propagation weight, second-level propagation weight and third-level propagation weight.

[0078] The direct influence propagation coefficient is obtained by calculating the influence intensity of directly adjacent links based on the first-level propagation weight.

[0079] The indirect influence propagation coefficient is obtained by quantifying the attenuation of the influence of the secondary propagation weight on the indirect link.

[0080] The feedback influence propagation coefficient is obtained by cumulatively calculating the three-level propagation weights and the feedback influence of the loop.

[0081] A matrix is ​​constructed based on the direct influence propagation coefficient, indirect influence propagation coefficient, and feedback influence propagation coefficient to obtain the cross-stage influence propagation matrix.

[0082] Specifically, the influence weight coefficients between links are processed into three levels of propagation: first-level, second-level, and third-level propagation weights. This classification divides the influence relationships between links into three different levels based on the distance and complexity of influence propagation in ecological circular agriculture. First-level propagation weights correspond to influence relationships between directly adjacent links, indicating a direct flow of materials, energy exchange, or information transmission between the two links without the need for intermediate links. For example, the planting link directly supplies organic waste to the waste treatment link, and the livestock breeding link directly supplies livestock manure to the waste treatment link. Second-level propagation weights correspond to indirect influence relationships propagated through an intermediate link, indicating that the influence needs to pass through a transfer link to spread from the source link to the target link. For example, organic waste from the planting link, after being transformed in the waste treatment link, affects the feed supply in the livestock breeding link. Third-level propagation weights correspond to feedback influence relationships in the circular loop, indicating that the influence returns to the original link after propagating through multiple links, forming a closed-loop feedback. For example, waste generated in the planting link, after being transformed into organic fertilizer through waste treatment and resource recycling, again affects the soil quality in the planting link.

[0083] The influence intensity of directly adjacent links is calculated based on the first-order propagation weight to obtain the direct influence propagation coefficient. The calculation of the direct influence propagation coefficient uses linear correlation analysis to quantify the influence intensity between directly adjacent links. The calculation process first identifies links with direct influence relationships by analyzing whether there is a direct flow of material, energy, or information between the links to determine the direct adjacency. Then, the output changes of the source link and the corresponding changes of the target link are statistically analyzed to establish a mathematical relationship model between them. The influence intensity is calculated by calculating the correlation coefficient between the data changes of the source link and the data changes of the target link. The magnitude of the correlation coefficient is directly used as the direct influence propagation coefficient; the closer the coefficient value is to one, the greater the influence intensity, and the closer the coefficient value is to zero, the smaller the influence intensity. The calculation process also considers the correction for time delay factors. When there is a fixed time delay between two links, the data sequence of the target link needs to be shifted forward by the corresponding time unit before calculating the correlation coefficient.

[0084] The indirect influence propagation coefficient is obtained by quantifying the attenuation of the influence on indirect links based on the second-order propagation weights. The quantification of the indirect influence propagation coefficient considers the attenuation effect of the influence during propagation and the moderating role of intermediate links. The quantification process first identifies the indirect link pairs and their corresponding intermediate propagation links, establishing a three-way propagation path from the source link to the intermediate link and then calculating the direct influence coefficients of the source link on the intermediate link and the intermediate link on the target link, respectively. The indirect influence propagation coefficient is obtained by multiplying the two direct influence coefficients. The product reflects the degree of attenuation of the influence after propagation through intermediate links. A propagation attenuation factor is introduced to simulate the natural attenuation phenomenon of the influence during propagation. The attenuation factor is determined based on the length of the propagation path and the processing characteristics of the intermediate links; the longer the propagation path, the smaller the attenuation factor; the stronger the buffering capacity of the intermediate links, the smaller the attenuation factor. The corrected indirect influence propagation coefficient is obtained by multiplying the attenuation factor by the product.

[0085] The feedback influence of the three-level propagation weights and the feedback loop are cumulatively calculated to obtain the feedback influence propagation coefficient. This cumulative calculation simulates the amplification or suppression effect of the influence in the ecological cycle system. The cumulative calculation first identifies the sequence of links forming a closed loop, such as the cycle path of planting-waste treatment-livestock farming-resource recycling-planting. Then, it calculates the strength of the influence returning to the original link after propagating one loop. The calculation process multiplies the direct influence propagation coefficients of each adjacent link pair in the loop sequentially to obtain the influence propagation strength of one cycle. Considering the cumulative time delay and multiple attenuation effects during the cycle, a cycle attenuation coefficient is introduced to correct the product result. Feedback influence also has a cumulative characteristic; the influence of multiple cycles will superimpose to form a stronger feedback effect. The cumulative calculation uses a geometric series summation formula to calculate the cumulative influence strength of an infinite number of cycles. When the cycle gain coefficient is less than one, the series converges, and the cumulative result is a finite value. When the cycle gain coefficient is greater than one, the series diverges, indicating a positive feedback amplification effect in the cycle system.

[0086] A matrix construction process is performed based on the direct influence propagation coefficient, indirect influence propagation coefficient, and feedback influence propagation coefficient to obtain a cross-stage influence propagation matrix. This matrix construction process integrates the three different levels of influence propagation coefficients into a unified matrix representation. The process establishes a square matrix structure with stages as row and column indices. Rows represent source stages, columns represent target stages, and the values ​​of matrix elements represent the overall influence strength between corresponding stage pairs. The matrix filling process selects the appropriate propagation coefficient based on the propagation level type of the stage pair and fills it in the corresponding position. Directly adjacent stage pairs use the direct influence propagation coefficient, indirectly related stage pairs use the indirect influence propagation coefficient, and stage pairs forming a loop use the feedback influence propagation coefficient. When a stage pair has multiple propagation levels simultaneously, the process uses a weighted summation method to merge the propagation coefficients of different levels into a single overall influence strength value. The weight allocation is determined based on the importance and reliability of each propagation level.

[0087] In one specific embodiment, the process of executing step S105 may specifically include the following steps:

[0088] Based on the cyclic enhancement parameters, a threshold judgment process is performed on the sensor acquisition frequency to obtain the acquisition frequency adjustment command;

[0089] The sensor network configuration of the heterogeneous equipment in circular agriculture is dynamically reconfigured according to the sampling frequency adjustment command to obtain the optimized sensor network configuration.

[0090] The optimized sensor network configuration is synchronized with the UAV inspection cycle to obtain a collaborative data acquisition and scheduling scheme.

[0091] Based on the collaborative data acquisition and scheduling scheme, the operating parameters of the ecological agricultural cycle system are predicted and analyzed to obtain the prediction results of the cycle benefits.

[0092] Based on the prediction results of circular benefits, decision-making schemes are generated and economic benefits are evaluated to obtain optimized decision-making schemes for circular agriculture.

[0093] Specifically, a threshold judgment process is performed on the sensor acquisition frequency based on the cyclic enhancement parameters to obtain acquisition frequency adjustment instructions. This threshold judgment process establishes a dynamic correlation mechanism between the cyclic enhancement parameters and the sensor acquisition frequency. The judgment process first sets multiple threshold ranges for the cyclic enhancement parameters, including a high-efficiency cyclic range, a normal cyclic range, and a low-efficiency cyclic range. When the cyclic enhancement parameters are in the high-efficiency cyclic range, it indicates that the ecological cycle system is operating well; when they are in the normal cyclic range, it indicates that the system is operating stably; and when they are in the low-efficiency cyclic range, it indicates that the system has an operational bottleneck or anomaly. The threshold judgment algorithm compares the current cyclic enhancement parameter value with the preset thresholds. When the parameter value is lower than the low-efficiency cyclic threshold, an adjustment instruction to increase the acquisition frequency is generated; when the parameter value is higher than the high-efficiency cyclic threshold, an adjustment instruction to decrease the acquisition frequency is generated; and when the parameter value is in the normal range, an instruction to maintain the current acquisition frequency is generated. The adjustment instruction includes specific frequency adjustment magnitude and adjustment range information.

[0094] The sensor network of heterogeneous equipment in circular agriculture undergoes dynamic reconfiguration based on the frequency adjustment command, resulting in an optimized sensor network configuration. This dynamic reconfiguration process reallocates the working modes and data transmission strategies of sensor nodes according to the frequency adjustment command. The reconfiguration process includes adjusting the data acquisition interval of sensor nodes. When an instruction to increase the acquisition frequency is received, the sensor acquisition interval is shortened from once per hour to once every half hour. When an instruction to decrease the acquisition frequency is received, the acquisition interval is extended from once per hour to once every two hours. The reconfiguration process also involves reselecting the cluster head node. When the acquisition frequency changes significantly in certain areas, the load capacity and communication efficiency of the cluster head node need to be reassessed. The node fitness is recalculated using the LMCO algorithm, and the cluster partitioning is adjusted to form a sensor network topology adapted to the new acquisition frequency requirements.

[0095] The optimized sensor network configuration is synchronized with the UAV inspection cycle to obtain a collaborative data acquisition scheduling scheme. This synchronization process coordinates the data acquisition activities of the ground sensor network and the aerial UAVs. The adjustment process modifies the UAV flight plan based on the new sensor network configuration. When the sensor acquisition frequency in a certain area increases, the UAV's inspection frequency in that area increases accordingly; when the sensor acquisition frequency decreases, the UAV's inspection interval is extended accordingly. The synchronization adjustment also considers matching the data acquisition time window, ensuring that the time when the UAV arrives above a specific cluster head node is synchronized with the time when the node's data aggregation is completed, avoiding situations where the data is not yet ready or is outdated when the UAV arrives. The collaborative data acquisition scheduling scheme includes the sensor network's acquisition schedule, the UAV's flight path sequence, and the data transmission time arrangement, forming an integrated ground-air data acquisition coordination mechanism.

[0096] Based on a collaborative data acquisition and scheduling scheme, the operational parameters of an ecological agricultural cycle system are predicted and analyzed to obtain predictions of cycle benefits. The prediction and analysis employs time series forecasting methods to conduct forward-looking analysis of key operational parameters of the cycle system. A Long Short-Term Memory (LSTM) network prediction model is established, comprising three components: a short-term prediction engine, a medium-term prediction engine, and a long-term prediction engine. The short-term prediction engine predicts the system's operational status for the next three days based on historical data from the past week; the medium-term prediction engine predicts the operational trend for the next two weeks based on data from the past month; and the long-term prediction engine predicts the system's evolution direction for the next three months based on data from the past year. The prediction model uses changes in data acquisition density from the collaborative data acquisition and scheduling scheme as input variables, combines historical operational data to train neural network parameters, and outputs predicted values ​​for efficiency indicators, resource utilization rates, and cycle gain coefficients for each stage.

[0097] Based on the predicted results of the circular economy, decision-making schemes are generated and economic benefits are evaluated to obtain optimized decision-making schemes for circular agriculture. The decision-making scheme generation process formulates specific management operation suggestions and resource allocation plans based on the prediction results. The generation process establishes a decision rule base containing standardized response strategies for different prediction results. When the prediction shows that the efficiency of a certain link will decrease, a decision suggestion to increase the input of that link is generated; when the prediction shows that the circular gain will decrease, a suggestion to adjust the material flow ratio between links is generated. The economic benefit evaluation process calculates the input costs and expected benefits of each decision-making scheme. Input costs include expenses such as increasing the number of sensors, extending the flight time of drones, and purchasing additional equipment. Expected benefits are calculated based on the increased yield, resource savings, and environmental benefits resulting from improved circular efficiency. The evaluation process uses the net present value (NPV) method to calculate the economic benefit indicators of each decision-making scheme and selects the scheme with the highest NPV as the recommended decision.

[0098] The above describes the data collection and analysis method for ecological circular agriculture based on the Internet of Things (IoT) in the embodiments of this application. The following describes the data collection and analysis system for ecological circular agriculture based on the IoT in the embodiments of this application. Please refer to [link / reference]. Figure 2 One embodiment of the IoT-based ecological circular agriculture data collection and analysis system in this application includes:

[0099] The cluster module is used to perform cross-domain collaborative cluster processing of ecological circular agriculture sensor nodes through the LMCO algorithm to obtain a circular agriculture heterogeneous equipment sensing network.

[0100] The acquisition module is used to perform predictive data acquisition and processing of the UAV RSSI signal for the ecological cycle closed loop based on the sensing network of the heterogeneous equipment of the circular agriculture, so as to obtain the spatiotemporal coupled dataset of circular agriculture.

[0101] The mapping module is used to perform causal mapping processing on the IoT cloud nodes to the agricultural cycle chain based on the spatiotemporal coupling dataset of the circular agriculture, so as to obtain ecological cycle causal chain data.

[0102] The mining module is used to perform cross-link impact mining on the ecological cycle causal chain data through the cycle effect propagation algorithm to obtain agricultural cycle propagation rules and cycle efficiency enhancement parameters.

[0103] The control module is used to intelligently feedback control the ecological agricultural cycle system based on the cycle efficiency enhancement parameters and the IoT acquisition strategy to obtain an optimized decision-making scheme for cycle agriculture.

[0104] above Figure 2 The IoT-based ecological circular agriculture data acquisition and analysis system in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The IoT-based ecological circular agriculture data acquisition and analysis equipment in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0105] Reference Figure 3 This invention also provides an IoT-based ecological circular agriculture data acquisition and analysis device, which can be a server, and its internal structure can be as follows: Figure 3 As shown, the IoT-based ecological circular agriculture data acquisition and analysis device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor in this computer design provides computing and control capabilities. The memory of the IoT-based ecological circular agriculture data acquisition and analysis device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the IoT-based ecological circular agriculture data acquisition and analysis device is used to store the data corresponding to this embodiment. The network interface of the IoT-based ecological circular agriculture data acquisition and analysis device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0106] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the IoT-based ecological circular agriculture data collection and analysis equipment to which the present invention is applied.

[0107] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the Internet of Things-based ecological circular agriculture data collection and analysis method.

[0108] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0109] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an IoT-based ecological circular agriculture data collection and analysis device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0110] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An Internet of Things-based ecological recycling agricultural data collection and analysis method, characterized in that, The method comprises: The method comprises: According to the cycle agricultural heterogeneous equipment perception network, the unmanned aerial vehicle RSSI signal is used for predictive data collection processing on the ecological cycle closed loop, and a cycle agricultural space-time coupling data set is obtained. According to the cycle agricultural space-time coupling data set, the Internet of Things cloud node is used for cause-effect mapping processing on the agricultural cycle chain, and ecological cycle cause-effect chain data is obtained. Through the cycle effect propagation algorithm, the ecological cycle cause-effect chain data is processed for cross-link influence mining, and agricultural cycle propagation rules and cycle efficiency parameters are obtained. According to the cycle efficiency parameters, the Internet of Things collection strategy is used for intelligent feedback regulation and control processing on the ecological agricultural cycle system, and a cycle agricultural optimization decision scheme is obtained. 2.The Internet of Things based ecological cycle agricultural data collection and analysis method according to claim 1, characterized in that, According to the cycle agricultural heterogeneous equipment perception network, the unmanned aerial vehicle RSSI signal is used for predictive data collection processing on the ecological cycle closed loop, and a cycle agricultural space-time coupling data set is obtained. Based on the cluster head node coordinates of the cycle agricultural heterogeneous equipment perception network, unmanned aerial vehicle flight path planning processing is performed, and an optimal flight trajectory sequence is obtained. Real-time measurement processing is performed on the unmanned aerial vehicle RSSI signal strength and the cluster head node distance, and signal strength change parameters are obtained; dynamic adjustment processing is performed on the unmanned aerial vehicle flight height and flight speed according to the signal strength change parameters, and the best signal receiving position is obtained. Based on the best signal receiving position, ground sensor data and aerial remote sensing data are synchronously collected, and multi-source fusion data is obtained. The multi-source fusion data is coupled and labeled with space-time coordinate information, and a cycle agricultural space-time coupling data set is obtained. 3.The Internet of Things based ecological cycle agricultural data collection and analysis method according to claim 1, characterized in that, According to the cycle agricultural space-time coupling data set, the Internet of Things cloud node is used for cause-effect mapping processing on the agricultural cycle chain, and ecological cycle cause-effect chain data is obtained. The data volume, data type and processing urgency in the cycle agricultural space-time coupling data set are evaluated, and cloud node selection parameters are obtained. Based on the cloud node selection parameters, the response rate, energy efficiency ratio and service quality of the Internet of Things cloud node are comprehensively scored, and the optimal cloud node identifier is obtained. The cycle agricultural space-time coupling data set is input into the optimal cloud node for time stamp alignment and coordinate unification processing, and a standardized data set is obtained. According to the standardized data set, a planting link, a breeding link, a waste treatment link and a resource circulation link are classified and labeled to obtain link classification data; Based on the link classification data, data quality detection and abnormal marking processing are performed to obtain ecological circulation causal chain data. 4.The Internet of Things based ecological cycle agricultural data collection and analysis method according to claim 1, characterized in that, The cross-link influence mining processing of the ecological circulation causal chain data is performed by the circulation effect propagation algorithm to obtain agricultural circulation propagation rules and circulation efficiency parameters, including: The time delay feature of the planting link, the breeding link, the waste treatment link and the resource circulation link in the ecological circulation causal chain data is identified to obtain a time delay matrix between links; Based on the time delay matrix between links, a circulation effect propagation network topology is constructed to obtain a link influence propagation path; The link influence propagation path and the link data change amount are calculated by weight to obtain a link influence weight coefficient; According to the link influence weight coefficient, a multi-level propagation and diffusion of the circulation effect is performed to obtain a cross-link influence propagation matrix; based on the cross-link influence propagation matrix, a circulation efficiency quantitative evaluation is performed to obtain agricultural circulation propagation rules and circulation efficiency parameters. 5.The Internet of Things based ecological cycle agricultural data collection and analysis method according to claim 4, characterized in that, The multi-level propagation and diffusion of the circulation effect according to the link influence weight coefficient to obtain the cross-link influence propagation matrix, including: The link influence weight coefficient is divided into propagation levels to obtain a first-level propagation weight, a second-level propagation weight and a third-level propagation weight; Based on the first-level propagation weight, the influence strength of directly adjacent links is calculated to obtain a direct influence propagation coefficient; According to the second-level propagation weight, the influence attenuation of indirectly related links is quantified to obtain an indirect influence propagation coefficient; The third-level propagation weight and the feedback influence of the circulation loop are cumulatively calculated to obtain a feedback influence propagation coefficient; Based on the direct influence propagation coefficient, the indirect influence propagation coefficient and the feedback influence propagation coefficient, a matrix is constructed to obtain a cross-link influence propagation matrix. 6.The Internet of Things based ecological cycle agricultural data collection and analysis method according to claim 1, characterized in that, According to the circulation efficiency parameters, an Internet of Things collection strategy is used to intelligently feedback and regulate the ecological agricultural circulation system to obtain a circulation agricultural optimization decision scheme, including: Based on the circulation efficiency parameters, a threshold judgment is performed on the sensor collection frequency to obtain a collection frequency adjustment instruction; According to the collection frequency adjustment instruction, a dynamic reconfiguration is performed on a circulation agricultural heterogeneous device perception network to obtain an optimized sensor network configuration; The optimized sensor network configuration is synchronously adjusted with a unmanned aerial vehicle inspection cycle to obtain a cooperative collection scheduling scheme; Based on the cooperative collection scheduling scheme, a prediction analysis is performed on the operation parameters of the ecological agricultural circulation system to obtain a circulation benefit prediction result; According to the circulation benefit prediction result, a decision scheme is generated and an economic benefit is evaluated to obtain a circulation agricultural optimization decision scheme.

7. An Internet of Things-based ecological recycling agricultural data collection and analysis system, characterized in that, The Internet of Things-based ecological circulation agricultural data collection and analysis system for implementing the Internet of Things-based ecological circulation agricultural data collection and analysis method according to any one of claims 1 to 6, including: The cluster module is used for cross-domain collaborative cluster processing of the ecological circulation agricultural sensor nodes through the LMCO algorithm, and obtains a circulation agricultural heterogeneous device perception network, and includes: performing coordinate collection processing on the sensor nodes in the planting area, the breeding area and the waste treatment area, to obtain a node spatial coordinate set; performing lion individual initialization processing based on the node spatial coordinate set, to obtain a lion individual position matrix; performing fitness function calculation processing on the lion individual position matrix and sensor node distance, delay and energy consumption parameters, to obtain a lion individual fitness value; performing local search optimization processing on the lion individual fitness value through a cat and mouse pursuit strategy, to obtain an updated lion individual position; performing cluster head node selection processing according to the updated lion individual position, to obtain a circulation agricultural heterogeneous device perception network; The collection module is used for predictive data collection processing of an unmanned aerial vehicle RSSI signal on an ecological circulation closed loop according to the circulation agricultural heterogeneous device perception network, to obtain a circulation agricultural space-time coupling dataset; The mapping module is used for cause-effect mapping processing of an Internet of Things cloud node on an agricultural circulation chain according to the circulation agricultural space-time coupling dataset, to obtain ecological circulation cause-effect chain data; The mining module is used for cross-link influence mining processing of the ecological circulation cause-effect chain data through a circulation effect propagation algorithm, to obtain an agricultural circulation propagation rule and a circulation efficiency parameter; The regulation and control module is used for intelligent feedback regulation and control processing of an ecological agricultural circulation system by an Internet of Things collection strategy according to the circulation efficiency parameter, to obtain a circulation agricultural optimization decision scheme.

8. An Internet of Things-based ecological recycling agricultural data acquisition and analysis device, characterized in that, The computer program is stored in the memory and can be run on the processor, and the processor implements the computer program to realize the method for collecting and analyzing ecological circulation agricultural data based on the Internet of Things according to any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is stored in the memory and can be run on the processor, and the processor implements the computer program to realize the method for collecting and analyzing ecological circulation agricultural data based on the Internet of Things according to any one of claims 1 to 6.

Images