An irrigation area backwater monitoring system based on an internet of things

Abstract

The application relates to the technical field of water conservancy monitoring, and relates to an irrigation area water recession monitoring system based on the Internet of Things. The system comprises a water recession outlet multi-source data acquisition module, a detection data anomaly elimination module, a filtered data compression strategy module, a compressed data feature analysis module, a water recession state grade analysis module and a water recession state early warning generation module. The flow information, water level information and conductivity information of the water recession outlet of an irrigation area are transmitted to an edge processing node to obtain multi-source detection data. Anomaly data points are eliminated to obtain filtered data. A data compression strategy is constructed to obtain compressed data. An abnormal section and a trend feature are analyzed by piecewise regression to obtain a change trend parameter and a water recession dynamic characteristic parameter. Meteorological environment data, irrigation operation data and the water recession dynamic characteristic parameter are analyzed in cooperation to obtain a water recession state grade index. A water recession state mode is identified to obtain an alarm instruction and a prewarning log tracking link. The application can improve the efficiency of irrigation area water recession monitoring.

Description

Technical Field

[0001] This invention relates to the field of water conservancy monitoring technology, and in particular to an irrigation district drainage monitoring system based on the Internet of Things. Background Technology

[0002] In the field of irrigation district drainage monitoring, existing technologies lack a systematic process design for processing multi-source data such as drainage outlet flow, water level, and conductivity. During the data acquisition stage, various monitoring information is not uniformly timestamped and formatted. Furthermore, no suitable secure communication connection is established during transmission. This results in poor synchronization and inconsistent formats of the acquired monitoring data, making it difficult to form a complete dataset that truly reflects the drainage status of the irrigation district. This lays a foundational defect for subsequent data processing.

[0003] Existing technologies have significant shortcomings in the anomaly removal, compression, and feature analysis stages of irrigation district drainage data. Anomaly removal relies solely on a single numerical threshold judgment without considering the dynamic changes in the sensing network topology to develop adaptive screening strategies. Data compression does not construct a coding scheme based on spatiotemporal correlation rules, and feature analysis fails to perform refined trend node identification and segmentation analysis on anomaly data segments. Ultimately, this leads to significant deviations in the analysis results of drainage status level indicators. The generation of early warning instructions lacks precise rule matching logic, making it impossible to effectively identify abnormal patterns in irrigation district drainage. The overall information processing efficiency and early warning accuracy of the monitoring system are insufficient to meet the actual needs of irrigation district drainage management. Therefore, improving the generation efficiency of alarm information has become an urgent problem to be solved. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides an irrigation district drainage monitoring system based on the Internet of Things, characterized in that the system includes a multi-source data acquisition module for drainage inlets, a module for removing abnormal detection data, a data compression strategy module, a compressed data feature analysis module, a drainage status level analysis module, and a drainage status early warning generation module, wherein:

[0005] The multi-source data acquisition module for the drainage outlet is used to transmit the flow rate information, water level information and conductivity information of the drainage outlet in the irrigation area to the edge processing node to obtain the multi-source detection data of the irrigation area.

[0006] The abnormal data removal module is used to remove abnormal data points from the multi-source detection data based on the real-time characteristics and distribution patterns in the multi-source detection data, thereby obtaining the filtered data of the irrigation area.

[0007] The filtered data compression strategy module is used to construct a data compression strategy for the irrigation area based on the spatiotemporal correlation in the filtered data and the laws of correlation characteristics in the filtered data, so as to obtain compressed data of the irrigation area.

[0008] The compressed data feature analysis module is used to perform piecewise regression analysis on the abnormal segments and trend features in the compressed data according to the threshold boundary conditions and trend change rules in the compressed data, so as to obtain the change trend parameters and water receding dynamic feature parameters of the irrigation area.

[0009] The drainage status level analysis module is used to perform collaborative analysis of the meteorological environment data and irrigation operation data of the irrigation area with the drainage dynamic characteristic parameters to obtain the drainage status level index of the irrigation area.

[0010] The water receding status early warning generation module is used to compare the water receding status level index with the preset status level standard, identify the water receding status pattern between the water receding status level index and the preset status level standard, and obtain the alarm command and early warning log tracking link of the irrigation area.

[0011] In a preferred embodiment, when the multi-source data acquisition module for the drainage outlet transmits the flow rate information, water level information, and conductivity information of the irrigation area's drainage outlet to the edge processing node to obtain the multi-source detection data of the irrigation area, it is specifically used for:

[0012] Obtain flow rate, water level, and conductivity information from the irrigation district's drainage outlets;

[0013] The flow rate information, water level information, and conductivity information are timestamped to obtain synchronous monitoring data of the irrigation area;

[0014] The format integrates the synchronous monitoring data to obtain the water status information of the irrigation area;

[0015] The water body status information is packaged into transmission data packets using a wireless communication protocol, and a secure communication connection is established with the edge processing node.

[0016] The transmission data packet is sent to the edge processing node to obtain multi-source detection data of the irrigation area.

[0017] In a preferred embodiment, when the detection data anomaly removal module removes abnormal data points from the multi-source detection data based on real-time characteristics and distribution patterns to obtain filtered data for the irrigation area, it is specifically used for:

[0018] Initialize the communication protocol between the edge processing node and the monitoring node in the irrigation area, and establish the sensing network topology of the irrigation area;

[0019] Based on the perception network topology, the state characteristics of the monitoring nodes in the irrigation area are obtained in real time to obtain the node operation status map of the irrigation area;

[0020] Based on the node operation status map, the dynamic change characteristics of the sensing network topology are analyzed to obtain the adaptive networking strategy of the irrigation area;

[0021] According to the adaptive networking strategy, abnormal fluctuation data in the multi-source detection data are classified and filtered to obtain the primary filtered data of the irrigation area.

[0022] Spatiotemporal interpolation is performed on the missing time periods in the primary filtered data to obtain the optimized monitoring data of the irrigation area;

[0023] The integrity of the optimized monitoring data is verified through a consistency check mechanism to obtain the filtered data of the irrigation area.

[0024] In a preferred embodiment, when the filtered data compression strategy module executes a data compression strategy for the irrigation district based on the spatiotemporal correlation of the filtered data and the laws governing the correlation characteristics of the filtered data, to obtain compressed data for the irrigation district, it is specifically used for:

[0025] Identify the spatiotemporal correlation rules between different monitoring points in the filtered data to obtain the correlation rule table of the irrigation area;

[0026] Based on the association rule table, a data coding scheme for the irrigation area is formulated, and the coding mechanism of the irrigation area is obtained;

[0027] Based on the encoding mechanism, feature extraction is performed on the filtered data to obtain the data feature set of the irrigation area;

[0028] According to a preset compression ratio, the data feature set is subjected to volume optimization to obtain the compressed data stream of the irrigation area;

[0029] The integrity of key features in the compressed data stream is verified to obtain the compressed data of the irrigation area.

[0030] In a preferred embodiment, when the compressed data feature analysis module performs piecewise regression analysis of abnormal segments and trend features in the compressed data based on threshold boundary conditions and trend change patterns in the compressed data, and obtains the irrigation district's trend parameters and drainage dynamic characteristic parameters, it is specifically used for:

[0031] Based on preset threshold boundary conditions, data segments in the compressed data that exceed the boundary range are marked to obtain the abnormal data segment identifiers of the irrigation area;

[0032] The abnormal data segments are analyzed for trend trends, and the trend change nodes of the abnormal data segments are identified to obtain the trend turning point set of the irrigation area.

[0033] By integrating the dynamic characteristic parameters of the trend inflection point set, the water receding trend parameters of the irrigation area are obtained;

[0034] Based on the set of trend turning points, the compressed data is divided into continuous segments to obtain the data segment sequence of the irrigation area;

[0035] By fitting a trend line that shows the changing patterns in the data segment sequence, a set of trend line segments for the irrigation area is obtained;

[0036] By comprehensively analyzing the slope and interrelationships of the trend line segments, the changing trend parameters of the irrigation area are obtained.

[0037] In a preferred embodiment, when the compressed data feature parsing module performs the integration of the dynamic feature parameters of the trend inflection point set to obtain the drainage trend parameters of the irrigation area, it is specifically used for:

[0038] Based on the set of trend turning points, the slope value and duration of the trend line segments are extracted to obtain the line segment slope sequence and duration sequence of the irrigation area;

[0039] Based on the slope sequence and duration sequence, a comprehensive index of the receding water trend parameters is calculated. The formula for calculating the comprehensive index of the receding water trend is as follows:

[0040] ;

[0041] In the formula, The comprehensive index of the receding trend, The first line segment in the slope sequence The slope of each trend segment, The duration sequence of the first The duration of each trend segment. The median of the slope sequence. The number of trend segments concentrated at the trend reversal points. These are preset weighting factors.

[0042] In a preferred embodiment, when the compressed data feature parsing module performs trend analysis on the abnormal data segments and identifies the trend change nodes of the abnormal data segments to obtain the trend inflection point set of the irrigation area, it is specifically used for:

[0043] Extract the sequence of continuous data points from the abnormal data segment to obtain the time series dataset of the irrigation area;

[0044] Change rate detection is performed on the time series dataset, and abnormal data points of change rate in the time series data are identified to obtain the potential change point set of the irrigation area;

[0045] Remove the pseudo-change points from the potential change point set to obtain the change node set of the irrigation area;

[0046] Based on the set of changing nodes, the compressed data is divided into segments to obtain the set of trend turning points of the irrigation area.

[0047] In a preferred embodiment, when the drainage status level analysis module performs collaborative analysis of the meteorological environment data and irrigation operation data of the irrigation district with the drainage dynamic characteristic parameters to obtain the drainage status level index of the irrigation district, it is specifically used for:

[0048] The meteorological environment data, irrigation operation data and drainage dynamic characteristic parameters of the irrigation area are integrated from multiple sources to obtain the multi-source dataset of the irrigation area.

[0049] The multi-source datasets are classified and organized according to three dimensions: environmental characteristics, operational status, and dynamic changes, to establish a status assessment framework for the irrigation district.

[0050] Based on the aforementioned state assessment framework, the dynamic characteristic parameters of the drainage are evaluated to obtain the drainage state level index of the irrigation area.

[0051] In a preferred embodiment, when the drainage status early warning generation module compares the drainage status level index with a preset status level standard, identifies the drainage status pattern between the drainage status index and the preset status standard, and obtains the alarm command and early warning log tracking link for the irrigation district, it is specifically used for:

[0052] The water receding status level index is compared item by item with the preset status level standard to obtain the level difference identifier of the irrigation area;

[0053] Analyze the distribution characteristics of the grade difference indicators to identify the abnormal state pattern types of the irrigation area;

[0054] Based on the historical early warning rule base of the irrigation district, the early warning rules are matched for the abnormal state pattern types to obtain the alarm information of the irrigation district;

[0055] The alarm information is distributed to the management terminal of the irrigation area to complete the transmission of early warning information;

[0056] The generation time, warning level, and distribution path of the alarm command are recorded synchronously to obtain the warning log tracking link of the irrigation area.

[0057] In a preferred embodiment, when the drainage status early warning generation module performs early warning rule matching on the abnormal status pattern type according to the historical early warning rule library of the irrigation district to obtain the alarm information of the irrigation district, it is specifically used for:

[0058] The typical abnormal pattern features in the historical early warning rule base of the irrigation area are retrieved to obtain the rule feature set of the irrigation area;

[0059] The similarity between the abnormal state pattern type and the rule feature set is compared to obtain the matching degree evaluation result of the irrigation area;

[0060] Based on the matching degree evaluation results, the early warning rules for the irrigation area are filtered to obtain the candidate rule set for the irrigation area;

[0061] By combining the real-time environmental factors of the irrigation area, the applicability of the candidate rule set is verified to obtain the early warning rules for the irrigation area;

[0062] Based on the aforementioned early warning rules, the early warning level and response measures for the irrigation district are configured, and the alarm information for the irrigation district is obtained.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] 1. This invention significantly improves the information processing quality and efficiency of irrigation district drainage monitoring through multi-module collaborative operation. The multi-source data acquisition module at the drainage outlet performs timestamp alignment and format integration of flow, water level, and conductivity information, and then transmits it to the edge processing node through a secure communication connection to ensure the synchronization and integrity of multi-source detection data. The detection data anomaly removal module formulates an adaptive networking strategy based on the dynamic changes of the sensing network topology, performs spatiotemporal interpolation and consistency verification after hierarchical screening of abnormal fluctuation data, and effectively improves the reliability of filtered data. The filtered data compression module constructs an encoding scheme based on spatiotemporal correlation rules, and optimizes data volume under the premise of verifying the integrity of key features, which greatly improves the efficiency of data transmission and storage.

[0065] 2. This invention can accurately improve the effectiveness of irrigation district drainage status assessment and early warning. The compressed data feature analysis module marks abnormal data segments by setting preset thresholds, identifies trend turning points and fits trend lines, and accurately extracts change trend parameters and drainage dynamic feature parameters. The drainage status level analysis module performs multi-dimensional collaborative analysis of meteorological environmental data, irrigation operation data and drainage dynamic feature parameters, making the generated drainage status level indicators more consistent with the actual drainage situation. The drainage status early warning generation module combines historical early warning rule base with real-time environmental factors to perform rule matching, ensuring the accuracy of alarm commands and the adaptability of response measures. At the same time, it generates early warning log tracking links, providing traceable decision-making basis for irrigation district drainage management and comprehensively improving the practical value of the monitoring system. Attached Figure Description

[0066] Figure 1A system architecture diagram of an irrigation district drainage monitoring system based on the Internet of Things is provided in an embodiment of the present invention;

[0067] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0068] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0070] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0071] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0072] In practice, the server-side equipment deployed in an IoT-based irrigation drainage monitoring system may consist of one or more devices. This IoT-based irrigation drainage monitoring system can be implemented as a business instance, a virtual machine, or hardware devices. For example, it can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, it can be understood as software deployed on a cloud node, providing an IoT-based irrigation drainage monitoring system to various user terminals. Alternatively, it can be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage each user terminal. Or, it can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide an IoT-based irrigation drainage monitoring system to various user terminals.

[0073] In terms of implementation, the IoT-based irrigation drainage monitoring system and the user terminal are mutually compatible. That is, if the IoT-based irrigation drainage monitoring system is implemented as an application installed on a cloud service platform, the user terminal is a client that establishes a communication connection with the application; or if the IoT-based irrigation drainage monitoring system is implemented as a website, the user terminal is implemented as a webpage; or if the IoT-based irrigation drainage monitoring system is implemented as a cloud service platform, the user terminal is implemented as a mini-program in an instant messaging application.

[0074] like Figure 1 The figure shown is a system architecture diagram of an irrigation area drainage monitoring system based on the Internet of Things provided in an embodiment of the present invention.

[0075] The IoT-based irrigation area drainage monitoring system 100 described in this invention can be installed on a cloud server. In terms of implementation, it can function as one or more service devices, or as an application installed in the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed into a website. Depending on the functions implemented, the IoT-based irrigation area drainage monitoring system 100 may include a multi-source data acquisition module 101 for drainage inlets, a data anomaly removal module 102, a data compression strategy module 103, a compressed data feature analysis module 104, a drainage status level analysis module 105, and a drainage status early warning generation module 106. The modules described in this invention can also be called units, referring to a series of computer program segments that can be executed by an electronic device's processor and perform a fixed function, stored in the electronic device's memory.

[0076] In this embodiment of the invention, in an IoT-based irrigation drainage monitoring system, each of the above-mentioned modules can be implemented independently and can call other modules. Here, "calling" can be understood as one module connecting to multiple modules of another type and providing corresponding services to those connected modules. The IoT-based irrigation drainage monitoring system provided by this embodiment of the invention allows for adjustments to the applicability of the system architecture without modifying the program code. This is achieved by adding modules and directly calling them, enabling cluster-based horizontal expansion and flexibly expanding the IoT-based irrigation drainage monitoring system. In practical applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in a cloud server.

[0077] The following describes, with reference to specific embodiments, the various components and specific workflow of an IoT-based irrigation district drainage monitoring system:

[0078] The multi-source data acquisition module 101 at the drainage outlet is used to transmit the flow rate information, water level information, and conductivity information of the drainage outlet in the irrigation area to the edge processing node to obtain the multi-source detection data of the irrigation area.

[0079] In this embodiment of the invention, when the multi-source data acquisition module for the drainage outlet transmits the flow rate information, water level information, and conductivity information of the irrigation area's drainage outlet to the edge processing node to obtain the multi-source detection data of the irrigation area, it is specifically used for:

[0080] Obtain flow rate, water level, and conductivity information from the irrigation district's drainage outlets;

[0081] The flow rate information, water level information, and conductivity information are timestamped to obtain synchronous monitoring data of the irrigation area;

[0082] The format integrates the synchronous monitoring data to obtain the water status information of the irrigation area;

[0083] The water body status information is packaged into transmission data packets using a wireless communication protocol, and a secure communication connection is established with the edge processing node.

[0084] The transmission data packet is sent to the edge processing node to obtain multi-source detection data of the irrigation area.

[0085] A flow sensor is installed at the central axis of the water flow channel at the irrigation area's drainage outlet. This sensor detects the volume or velocity of the water flowing through that location in real time and converts the detected physical quantity into digital flow information through an internal signal conversion circuit. A water level sensor is fixed at a pre-set mounting hole on the side wall of the drainage outlet. The sensor probe is in direct contact with the water body. It calculates the distance by sensing the pressure exerted by the water body on the probe or by receiving ultrasonic reflection signals, and then converts these physical signals into digital water level information. A conductivity sensor is immersed in the water body at the drainage outlet. This sensor detects the conductivity of ions in the water body and converts the electrical signal corresponding to the conductivity into digital conductivity information. This completes the acquisition of flow, water level, and conductivity information at the irrigation area's drainage outlet.

[0086] Using the standard time provided by the high-precision clock module built into the edge processing node as the timestamp reference, whenever the flow sensor, water level sensor, or conductivity sensor collects a piece of information, it immediately requests the current standard timestamp from the edge processing node through the real-time communication link between the sensor and the edge processing node. Then, the obtained standard timestamp is bound to the corresponding flow information, water level information, or conductivity information one by one to ensure that each piece of information has a completely consistent time identifier. After all the collected flow information, water level information, and conductivity information have completed the timestamp binding operation, the synchronous monitoring data of the irrigation area is obtained.

[0087] A unified water status information format is pre-defined in the system. This format includes four fixed fields: an information type field to clearly indicate the type of data, a timestamp field to fill in the bound time identifier, a data value field to fill in the specific detection data, and a sensor number field to mark the unique identifier of the sensor collecting the data. Each piece of information in the synchronous monitoring data is filled in the fields according to this preset format. For example, for a flow rate information, the flow rate needs to be marked in the information type field, the bound time identifier needs to be filled in the timestamp field, the specific flow rate data needs to be filled in the data value field, and the unique identifier of the corresponding flow sensor needs to be marked in the sensor number field. After all the synchronous monitoring data is filled in the fields in this way, the water status information of the irrigation area is obtained.

[0088] Low-power wide-area network (LPWAN) wireless communication protocol is selected as the data transmission protocol. Following the data packet structure specified in this protocol, the integrated water state information is used as the core data payload of the data packet. A header containing the data acquisition terminal device address, the edge processing node device address, and the total data packet length is added before the core data payload. A cyclic redundancy check (CRC) field for verifying data integrity is added after the core data payload, thus completing the packaging of the water state information into the transmission data packet. When establishing a secure communication connection, a pre-shared key encryption method is used. The data acquisition terminal and the edge processing node pre-store the same key. When the data acquisition terminal sends a connection request to the edge processing node, it encrypts its own device address using the pre-shared key and sends it along with the data. After receiving the connection request, the edge processing node decrypts the encrypted device address using the pre-shared key, obtains the device address, and verifies whether the device address falls within the system's authorized scope. If the verification is successful, a secure communication connection is established with the data acquisition terminal.

[0089] The data acquisition terminal sends the packaged data packets to the edge processing node via the established low-power wide-area network (LPWAN) secure communication link, according to the transmission frequency and power specified in the protocol. Upon receiving the data packets, the edge processing node first reads the cyclic redundancy check (CR) field at the end of the data packet. Based on the header information and core data payload, it recalculates the CR value and compares it with the CR field in the data packet. If they match, the data packet is considered complete and untampered. Subsequently, the edge processing node removes the header information and CR field according to the LWAN protocol's unpacking rules, extracting the water state information. Since this water state information already includes timestamp-aligned and format-integrated flow, water level, and conductivity information, the edge processing node summarizes and organizes this information to obtain the multi-source monitoring data for the irrigation area.

[0090] The beneficial effects are that the multi-source data acquisition module at the drainage outlet ensures that the flow rate, water level, and conductivity information collected from the irrigation area's drainage outlet are highly synchronized in time by sequentially executing the steps of information acquisition, timestamp alignment, format integration, data packaging and secure connection establishment, and data transmission. This avoids monitoring errors caused by time deviations in different types of data. Furthermore, the unified format integration gives the data a standardized structure, facilitating direct processing in subsequent stages. The wireless communication protocol packaging and secure communication connection effectively prevent data loss, tampering, or leakage during transmission. Ultimately, the multi-source monitoring data obtained after transmission to the edge processing node possesses integrity, accuracy, and security, providing high-quality data support for subsequent abnormal data removal, data compression, and feature analysis within the entire irrigation area drainage monitoring system. This ensures that subsequent modules can perform efficient processing based on reliable data, thereby improving the operational stability and data processing effectiveness of the entire monitoring system.

[0091] The detection data anomaly removal module 102 is used to remove abnormal data points in the multi-source detection data according to the real-time characteristics and distribution patterns in the multi-source detection data, so as to obtain the filtered data of the irrigation area.

[0092] In this embodiment of the invention, when the detection data anomaly removal module removes abnormal data points from the multi-source detection data based on real-time characteristics and distribution patterns to obtain filtered data for the irrigation area, it is specifically used for:

[0093] Initialize the communication protocol between the edge processing node and the monitoring node in the irrigation area, and establish the sensing network topology of the irrigation area;

[0094] Based on the perception network topology, the state characteristics of the monitoring nodes in the irrigation area are obtained in real time to obtain the node operation status map of the irrigation area;

[0095] Based on the node operation status map, the dynamic change characteristics of the sensing network topology are analyzed to obtain the adaptive networking strategy of the irrigation area;

[0096] According to the adaptive networking strategy, abnormal fluctuation data in the multi-source detection data are classified and filtered to obtain the primary filtered data of the irrigation area.

[0097] Spatiotemporal interpolation is performed on the missing time periods in the primary filtered data to obtain the optimized monitoring data of the irrigation area;

[0098] The integrity of the optimized monitoring data is verified through a consistency check mechanism to obtain the filtered data of the irrigation area.

[0099] By pre-setting fixed parameters for communication between edge processing nodes and irrigation district monitoring nodes, including communication frequency, data transmission format, and device authentication method, the edge processing node sends a protocol initialization command to all monitoring nodes. After receiving the command, the monitoring nodes automatically configure their own communication parameters to match the settings of the edge processing node. After the parameter matching is completed, the edge processing node collects the actual installation location, unique device identifier, and current communication connection status of each monitoring node. Based on this information, a network structure model is constructed with the edge processing node as the core and each monitoring node distributed according to its actual location and labeled with its communication status. This initializes the communication protocol and establishes the sensing network topology of the irrigation district.

[0100] The communication links of each monitoring node are determined from the topology of the sensing network. The edge processing node sends status query commands to each monitoring node through these links at fixed time intervals. After receiving the commands, the monitoring node collects its own power supply voltage, data acquisition interval, communication signal strength with the edge processing node, and internal hardware operating temperature in real time. It packages this status information in a preset format and feeds it back to the edge processing node through the original communication link. After receiving the feedback information from all monitoring nodes, the edge processing node classifies and organizes the status characteristics of each monitoring node according to the device identifier, and presents the device identifier of each monitoring node and its corresponding power supply voltage, data acquisition frequency, communication signal strength, and hardware temperature in the form of a chart, thus obtaining the node operation status map of the irrigation area.

[0101] By comparing the node operation status maps generated at different time points, the changes in the status of each monitoring node are identified, such as whether the communication signal strength continues to decline, whether the power supply voltage is lower than the normal operating threshold, and whether the hardware temperature exceeds the safe range. The impact of these abnormal monitoring nodes on the sensing network topology is analyzed, such as whether abnormal nodes cause data acquisition interruption in specific areas or interfere with the communication stability of adjacent nodes. Targeted adjustment plans are formulated according to the degree of impact. For example, for nodes with weak communication signals, the communication frequency with edge processing nodes is adjusted or adjacent normal nodes are designated as relays; for nodes with low power supply voltage, low power warnings are sent and their data acquisition frequency is reduced to reduce power consumption; for nodes with high hardware temperature, high-load acquisition tasks are suspended until the temperature drops. These adjustment plans are integrated into a dynamically executable rule system to obtain the adaptive networking strategy for the irrigation area.

[0102] Based on the abnormal fluctuation classification standard clearly defined in the adaptive networking strategy, the standard classifies data fluctuations into different levels. For example, a fluctuation exceeding 50% of the normal range and corresponding monitoring node status is classified as Level 1 abnormality; exceeding 30% but node status is normal is Level 2 abnormality; and exceeding 10% and lasting for more than a preset duration is Level 3 abnormality. The edge processing node calls multi-source detection data, classifies it by flow rate, water level, and conductivity, compares it with their respective normal ranges, and marks each data point with an abnormality level based on the node status judgment results. First, Level 1 abnormal data is removed. For Level 2 abnormal data, it is further verified whether it is temporary interference. If it is confirmed to be abnormal, it is removed. For Level 3 abnormal data, it is removed if the duration exceeds the preset limit. The filtered data is summarized and organized to obtain the primary filtered data of the irrigation area.

[0103] First, identify missing time periods in the primary filtered data that do not contain the corresponding data type. Then, obtain the normal data adjacent to the missing time periods. For example, if data for a certain time period in the morning is missing, extract the last normal data before that time period and the first normal data after that time period. At the same time, collect the synchronous data of the two monitoring nodes that are closest to the node to which the missing data belongs and are in normal status within the missing time period. Using a spatiotemporal interpolation method, first estimate the theoretical data of each time period within the missing time period according to the time dimension, and then combine it with the spatial dimension, set weights according to the distance between the adjacent nodes and the target node, and perform weighted calculation on the theoretical data and the synchronous data of the adjacent nodes. Use the calculation results as supplementary data for the missing time period to fill the missing position of the primary filtered data to obtain the optimized monitoring data of the irrigation area.

[0104] A consistency verification mechanism is established, and verification rules are clarified. These include ensuring that data changes at adjacent time periods of the same monitoring node conform to physical laws and that different types of data have logical relationships. Edge processing nodes review and optimize monitoring data one by one according to the rules, checking whether the changes in water level at adjacent time periods are compliant, whether the trends of flow rate and water level changes are consistent, and whether the fluctuations in conductivity are within the normal range. If a data point does not conform to the rules, the status of the corresponding node and the data of adjacent nodes at the same time are retrieved again for verification. If a problem is confirmed, the data is removed and re-interpolated and supplemented. Once all data conforms to the verification rules, and it is confirmed that there are no logical contradictions, no omissions, and that the data conforms to physical laws, the data is deemed complete, and the filtered data of the irrigation area is obtained.

[0105] The beneficial effects of the anomaly removal module are as follows: It sequentially executes the following steps: initializing the communication protocol between edge processing nodes and monitoring nodes, establishing the sensing network topology, acquiring node operational status maps based on the topology, analyzing network dynamic changes based on the maps and generating adaptive networking strategies, filtering abnormal fluctuation data from multi-source detection data according to the strategy, performing spatiotemporal interpolation on missing periods of primary filtered data, and verifying and optimizing the integrity of monitoring data through consistency checks. First, it ensures stable communication between nodes with a unified communication protocol and a clear network structure, laying a reliable foundation for subsequent data processing. Then, by monitoring node operational status and network dynamic changes in real time, the system can proactively adapt to changes in the network environment and avoid... This system avoids data acquisition deviations caused by node anomalies or network structure adjustments; then, it accurately and hierarchically removes abnormal data according to an adaptive strategy to reduce interference from invalid data in subsequent processing; it supplements missing data periods through spatiotemporal interpolation to fill data gaps and improve data continuity; finally, it performs consistency verification to ensure that the data conforms to physical laws and logical relationships, eliminating data contradictions and omissions. The resulting filtered data is accurate, complete, and reliable, providing high-quality data input for subsequent modules such as filtered data compression and compressed data feature analysis. This effectively improves the efficiency and accuracy of subsequent data processing in the entire irrigation district drainage monitoring system, further enhancing the overall stability and practical value of the system.

[0106] The filtered data compression strategy module 103 is used to construct a data compression strategy for the irrigation area based on the spatiotemporal correlation in the filtered data and the laws of correlation characteristics in the filtered data as a theoretical basis, so as to obtain compressed data of the irrigation area.

[0107] In this embodiment of the invention, when the filtered data compression strategy module executes a data compression strategy for the irrigation area based on the spatiotemporal correlation of the filtered data and the laws governing the correlation characteristics of the filtered data, to obtain compressed data for the irrigation area, it is specifically used for:

[0108] Identify the spatiotemporal correlation rules between different monitoring points in the filtered data to obtain the correlation rule table of the irrigation area;

[0109] Based on the association rule table, a data coding scheme for the irrigation area is formulated, and the coding mechanism of the irrigation area is obtained;

[0110] Based on the encoding mechanism, feature extraction is performed on the filtered data to obtain the data feature set of the irrigation area;

[0111] According to a preset compression ratio, the data feature set is subjected to volume optimization to obtain the compressed data stream of the irrigation area;

[0112] The integrity of key features in the compressed data stream is verified to obtain the compressed data of the irrigation area.

[0113] The flow rate, water level, and conductivity information of all monitoring points are extracted from the filtered data. The data changes of each monitoring point at different time periods are sorted out in chronological order. At the same time, the monitoring point areas are divided according to spatial location. The data change trends of monitoring points in adjacent areas or areas with the same function are compared within the same time period. For example, it is observed whether the water level of different drainage outlet monitoring points in the same irrigation area shows a synchronous rising or falling trend during the same irrigation period, whether the flow rate has a stable proportional relationship due to location differences, and whether the conductivity remains similar due to water connectivity. These stable temporal and spatial correlations are recorded one by one and organized into a table containing monitoring point identifiers, correlation types, and correlation rule descriptions to obtain the correlation rule table of the irrigation area.

[0114] Based on the correlation patterns of data from different monitoring points recorded in the association rule table, for monitoring point data with synchronous change patterns, coding rules are formulated. One core monitoring point is used as the baseline data, and other related point data are recorded as differences from the baseline data. For example, the core point records the complete water level value, while related points only record the difference between the core point's water level and the core point's water level. For flow data with fixed proportional relationships, coding rules are formulated that only the baseline point's flow value and proportional coefficient are recorded, and the flow of related points is derived from the baseline value and proportional coefficient. Simultaneously, data type identifiers, difference value recording formats, and proportional coefficient storage methods are clearly defined. These coding rules are integrated into a unified and executable rule system, resulting in the irrigation district's coding mechanism.

[0115] According to the data type identifiers specified in the coding mechanism, the flow rate information, water level information, and conductivity information in the filtered data are first classified and divided to ensure that each type of data corresponds to a unique type identifier. Then, for each type of data, the information that reflects the core characteristics of the data is extracted by referring to the association relationship in the association rule table. For example, for water level data, the water level change sequence of the core monitoring point and the water level difference sequence between the associated point and the core point are extracted; for flow rate data, the flow rate value sequence of the benchmark point and the corresponding proportional coefficient are extracted. These extracted core feature information are organized according to the format specified in the coding mechanism to ensure that each feature information contains the corresponding monitoring point identifier, data type identifier, and feature content, forming a structured feature set, thus obtaining the data feature set of the irrigation area.

[0116] Referring to the ratio between the final data volume and the original data volume required by the preset compression ratio, the content of the data feature set is optimized. For recurring feature patterns in the data feature set, a simplified identifier is used to replace the recurring feature pattern, and the content of the feature pattern corresponding to the identifier and the number of occurrences are recorded at the end of the data to avoid storing the same features repeatedly. At the same time, redundant identifiers in the data feature set that do not affect data reconstruction are removed, and only necessary key identifier information is retained. By reducing the storage volume of the data feature set in the above way, until the data volume meets the requirements of the preset compression ratio, the compressed data stream of the irrigation area is obtained.

[0117] First, clarify the scope of key features in the compressed data stream, including the complete data change trends of core monitoring points, the correlation patterns between monitoring points, and the completeness of data type identification. Then, reverse the compressed data stream according to the rules in the encoding mechanism to obtain the restored monitoring data. Compare the key features in the restored monitoring data with the corresponding key features in the original filtered data item by item to check whether the data change trends of the restored core points are consistent with the original, whether the difference values ​​or proportions of the associated points are consistent with the correlation patterns in the original data, and whether the data type identification is complete and error-free. If all key features can be accurately restored and completely match the key features in the original data, the key features in the compressed data stream are determined to be complete, and the compressed data stream at this time is the compressed data of the irrigation area.

[0118] The beneficial effects are as follows: the data compression strategy module sequentially executes the following steps: identifying the spatiotemporal correlation rules between different monitoring points in the filtered data to form a correlation rule table; formulating a data encoding scheme based on the correlation rule table to establish an encoding mechanism; extracting features from the filtered data according to the encoding mechanism to obtain a data feature set; optimizing the volume of the data feature set according to a preset compression ratio to generate a compressed data stream; and verifying the integrity of key features of the compressed data stream to obtain compressed data. First, the encoding mechanism is built based on the spatiotemporal correlation and association characteristics of the filtered data to ensure that the data encoding has clear targeting and logical rationality, avoiding information confusion or omission of core information caused by unfounded encoding. Then, through feature extraction, it accurately focuses on the monitoring of water discharge in the filtered data. Valuable core information is proactively filtered out, removing irrelevant and redundant content to provide efficient and accurate processing targets for subsequent data compression. Then, data volume is optimized according to a preset compression ratio, significantly reducing the space required for data storage and bandwidth consumption during transmission, improving data flow efficiency within the system, and reducing hardware resource waste. Finally, through key feature integrity verification, it is strictly ensured that no core information affecting the analysis of drainage status is lost during compression. This ensures that the final compressed data maintains both conciseness and reliable monitoring and analysis value, providing high-quality, low-redundancy data input for the subsequent compressed data feature analysis module, further supporting the efficiency of the entire irrigation district drainage monitoring system's data processing and the accuracy of the analysis results.

[0119] The compressed data feature analysis module 104 is used to perform segmented regression analysis of abnormal segments and trend features in the compressed data according to the threshold boundary conditions and trend change rules in the compressed data, so as to obtain the change trend parameters and water receding dynamic feature parameters of the irrigation area.

[0120] In this embodiment of the invention, when the compressed data feature analysis module performs piecewise regression analysis of abnormal segments and trend features in the compressed data based on threshold boundary conditions and trend change patterns in the compressed data, and obtains the irrigation area's trend parameters and drainage dynamic characteristic parameters, it is specifically used for:

[0121] Based on preset threshold boundary conditions, data segments in the compressed data that exceed the boundary range are marked to obtain the abnormal data segment identifiers of the irrigation area;

[0122] The abnormal data segments are analyzed for trend trends, and the trend change nodes of the abnormal data segments are identified to obtain the trend turning point set of the irrigation area.

[0123] By integrating the dynamic characteristic parameters of the trend inflection point set, the water receding trend parameters of the irrigation area are obtained;

[0124] Based on the set of trend turning points, the compressed data is divided into continuous segments to obtain the data segment sequence of the irrigation area;

[0125] By fitting a trend line that shows the changing patterns in the data segment sequence, a set of trend line segments for the irrigation area is obtained;

[0126] By comprehensively analyzing the slope and interrelationships of the trend line segments, the changing trend parameters of the irrigation area are obtained.

[0127] When the compressed data feature parsing module integrates the dynamic feature parameters of the trend inflection point set to obtain the drainage trend parameters of the irrigation area, it is specifically used for:

[0128] Based on the set of trend turning points, the slope value and duration of the trend line segments are extracted to obtain the line segment slope sequence and duration sequence of the irrigation area;

[0129] Based on the slope sequence and duration sequence, a comprehensive index of the receding water trend parameters is calculated. The formula for calculating the comprehensive index of the receding water trend is as follows:

[0130] ;

[0131] In the formula, The comprehensive index of the receding trend, The first line segment in the slope sequence The slope of each trend segment, The duration sequence of the first The duration of each trend segment. The median of the slope sequence. The number of trend segments concentrated at the trend reversal points. These are preset weighting factors.

[0132] When the compressed data feature parsing module performs trend analysis on the abnormal data segments and identifies the trend change nodes of the abnormal data segments to obtain the trend inflection point set of the irrigation area, it is specifically used for:

[0133] Extract the sequence of continuous data points from the abnormal data segment to obtain the time series dataset of the irrigation area;

[0134] Change rate detection is performed on the time series dataset, and abnormal data points of change rate in the time series data are identified to obtain the potential change point set of the irrigation area;

[0135] Remove the pseudo-change points from the potential change point set to obtain the change node set of the irrigation area;

[0136] Based on the set of changing nodes, the compressed data is divided into segments to obtain the set of trend turning points of the irrigation area.

[0137] The preset threshold boundary conditions are the normal value ranges for the flow, water level, and conductivity information contained in the compressed data, as pre-defined by the system. For example, the upper and lower limits of normal flow fluctuations, the highest and lowest safe operating values ​​of water levels, and the normal range of conductivity. Each piece of flow, water level, and conductivity data in the compressed data is compared with the corresponding preset threshold boundary one by one. If a certain type of data continuously exceeds its corresponding normal value range within a certain time period, the time period and the corresponding data category are marked to form identification information containing abnormal time periods and abnormal data types, thus obtaining the abnormal data segment identification of the irrigation area.

[0138] The marked abnormal data segments are arranged in chronological order, and the direction of data change over time is analyzed segment by segment. For example, in an abnormal water level segment, is the data constantly rising, falling, or fluctuating around a certain value over time? During the analysis, the focus is on the specific location where the direction of data change changes. For example, if the data in an abnormal segment first rises to a certain value and then begins to fall, the time point corresponding to that value is a trend change node. All such trend change nodes identified in all abnormal data segments are organized in chronological order, and the time, data value, and abnormal segment to which each node belongs are recorded to obtain the set of trend turning points of the irrigation area.

[0139] Key information corresponding to each trend inflection point is extracted from the trend inflection point set, including the time span between two adjacent trend inflection points, the maximum change in data between these two inflection points, and dynamic characteristic parameters such as the average rate of data change. These parameters are then categorized and organized according to the abnormal data type. For example, for trend inflection points in sections with abnormal water levels, the magnitude of water level changes and the corresponding time intervals between adjacent inflection points are compiled. These parameters are used to determine the trend of water level changes during the receding process. For instance, if the water level drop between adjacent inflection points is large and the time interval is short, it indicates that the receding speed is fast during this stage. These parameters that can reflect the overall trend of receding water are summarized to obtain the receding water change trend parameters of the irrigation area.

[0140] Using each trend inflection point in the trend inflection point set as a boundary marker, the entire compressed data is segmented. Specifically, the first continuous segment starts from the beginning of the compressed data and ends at the first trend inflection point; the second continuous segment starts from the first trend inflection point and ends at the second trend inflection point; and so on, until the end of the compressed data. The data in each continuous segment maintains a consistent trend. These sequentially divided continuous segments are then organized into an ordered set to obtain the data segment sequence of the irrigation district.

[0141] For each consecutive segment in the data segment sequence, collect the time information and data value information corresponding to all data within that segment. Use the time information as the horizontal axis and the data value as the vertical axis to mark the positions of all data points within that segment in the coordinate system. Then, based on the distribution of these data points, draw a line that best fits the distribution trend of all data points. For example, if the data points in a certain segment gradually increase over time and are relatively concentrated, draw a straight line from the lower left to the upper right to make the line as close as possible to each data point. After completing this fitting operation for each consecutive segment, collect all the fitted lines in segment order to obtain the trend line segment set of the irrigation area.

[0142] The slope of each trend line segment in the trend line segment set is calculated one by one. The sign of the slope determines whether the data trend represented by the line segment is upward or downward, and the steepness of the slope indicates the speed of data change. For example, a positive and relatively steep slope indicates that the data is rising rapidly, while a negative and relatively gentle slope indicates that the data is falling slowly. At the same time, the relationship between two adjacent trend line segments is analyzed. For example, if the slope of the first line segment is positive and the slope of the second line segment is negative, it indicates that the data trend has changed from upward to downward. If the absolute value of the slope of the first line segment is greater than that of the second line segment, it indicates that the speed of data change has slowed down. The slope characteristics of all line segments and the relationship characteristics of adjacent line segments are comprehensively organized to form information that can comprehensively reflect the overall changes in the compressed data, and thus obtain the trend parameters of the irrigation area.

[0143] Extract the time information and corresponding data values ​​of all trend inflection points from the trend inflection point set. Determine that two consecutive adjacent trend inflection points constitute a trend segment. For each trend segment, subtract the data value of the previous trend inflection point from the data value of the subsequent trend inflection point to obtain the data change of the segment. Simultaneously, subtract the time of the previous trend inflection point from the time of the subsequent trend inflection point to obtain the time change of the segment. Divide the data change by the time change to obtain the slope value of the trend segment. Arrange the slope values ​​of all trend segments in chronological order on the time axis to obtain the segment slope sequence of the irrigation area. At the same time, take the time change corresponding to each trend segment as the duration of the segment. Arrange the durations of all trend segments in the same chronological order as the slope values ​​to obtain the duration sequence of the irrigation area.

[0144] First, arrange all slope values ​​in the slope sequence in ascending order and find the slope value in the middle of the sequence, taking it as the median. Then, calculate the first part of the value by multiplying each slope value in the slope sequence by the corresponding duration in the duration sequence. Sum all products to obtain a total product. Simultaneously, sum all durations in the duration sequence to obtain a total duration. Divide the sum of products by the total duration to obtain the first part of the value. Next, calculate the second part of the value by subtracting the median of each slope value in the slope sequence. Take the absolute value of each subtraction to obtain the absolute difference between each slope value and the median. Sum all absolute differences to obtain a total absolute difference. Divide the total absolute difference by the number of trend segments to obtain the average absolute difference. Multiply the average absolute difference by a preset weighting factor to obtain the second part of the value. Finally, add the first part of the value and the second part of the value to obtain the comprehensive index of the receding trend parameter.

[0145] The slope of each trend segment in the trend slope sequence is obtained by extracting two consecutive trend turning points from the set of trend turning points. The data value corresponding to the second trend turning point is subtracted from the data value corresponding to the first trend turning point to obtain the data change between the two turning points. Then, the time of the second trend turning point is subtracted from the time of the first trend turning point to obtain the time change between the two turning points. The data change is divided by the time change to calculate the slope of the trend segment formed by the adjacent turning points. The slopes of all trend segments are arranged in chronological order on the time axis to form the trend slope sequence, and each slope comes from this sequence.

[0146] The duration of each trend segment in the duration sequence is obtained by extracting two consecutive trend turning points from the set of trend turning points, subtracting the time of the previous trend turning point from the time of the latter trend turning point, and the time difference is the duration of the trend segment formed by these two turning points. The durations of all trend segments are arranged in order of their corresponding slopes to form the duration sequence, and each duration comes from this sequence.

[0147] The median of the slope sequence is obtained by arranging all the slope values ​​in the line segment slope sequence in ascending order. If the number of slopes in the sequence is odd, the slope value in the middle position after arrangement is taken as the median; if the number of slopes in the sequence is even, the average of the two middle slope values ​​after arrangement is taken as the median. This median is directly obtained from the sorted line segment slope sequence.

[0148] The number of trend segments concentrated at trend reversal points is the total number of slope values ​​contained in the statistical segment slope sequence. Since each slope value corresponds to a trend segment composed of adjacent trend reversal points, this number is equal to the total number of slope values ​​in the segment slope sequence, which is obtained by counting the slope values ​​in the segment slope sequence.

[0149] The preset weighting factor is a fixed value set in advance by the system before deployment, based on the characteristics of historical drainage monitoring data of the irrigation area, including the common range of slope fluctuations during historical drainage processes and the magnitude of the impact of different fluctuation degrees on the judgment of drainage trends. It is used to adjust the weight of the deviation between the slope and the median on the final comprehensive index. This value comes from the system's pre-configuration.

[0150] The result calculated by the formula is the comprehensive index of the receding trend. The first part of the calculation process is as follows: multiply each slope in the slope sequence of the line segments by the corresponding duration in the duration sequence to obtain the product of each slope and the corresponding duration, and add all the products to get the sum of the products; at the same time, add all the durations in the duration sequence to get the sum of the durations; divide the sum of the products by the sum of the durations. This part of the result reflects the average slope of all trend line segments after considering the duration weight, which can reflect the overall trend direction and average rate of change of the receding process.

[0151] The second part of the formula is calculated as follows: Subtract the median of each slope in the slope sequence from the median of the slope sequence to obtain the difference between each slope and the median; take the absolute value of each difference to avoid cancellation of positive and negative values; add all the absolute values ​​to obtain the sum of absolute differences; divide the sum of absolute differences by the number of trend segments to obtain the average absolute difference; and then multiply the average absolute difference by a preset weighting factor. This part of the result reflects the degree of deviation of the slope of each trend segment from the median of the overall slope, and can adjust the interference of individual abnormal slopes on the comprehensive index, so that the index is more in line with the actual situation of the receding trend.

[0152] Adding the calculation results from the first part of the formula to the second part yields a comprehensive index of drainage trend. This index comprehensively considers both the overall trend of drainage and the fluctuations in slope at each stage, providing a quantitative description of the drainage trend. As a core parameter of the drainage trend in the irrigation district, this index provides precise quantitative evidence for the subsequent drainage status level analysis module, helping to accurately determine the overall situation and characteristics of drainage in the irrigation district and ensuring the accuracy of the drainage status level assessment.

[0153] The time range and data type of the abnormal data are determined from the abnormal data segment identifier. The compressed data values ​​corresponding to each moment in the time range are extracted one by one in the order from early to late. At the same time, the specific time information corresponding to each data value is recorded. The extracted time and data values ​​are arranged in chronological order to form an ordered set containing time series and corresponding data series. This set is the time series dataset of the irrigation area.

[0154] For consecutive adjacent data points in the time series dataset, the change in value between the preceding and following data points is calculated by subtracting the value of the preceding data point from the value of the following data point. The time interval between the preceding and following data points is then calculated by subtracting the time interval from the time interval of the following data point. The rate of change between these two adjacent data points is calculated by dividing the change in value by the time interval. This process is repeated to calculate the rate of change between all adjacent data points in the time series dataset. Then, each rate of change is compared with the overall trend of multiple adjacent rates of change before and after it. If a rate of change differs significantly from the rates before and after it and exceeds the allowable range of normal data fluctuations, the data point following that rate of change is marked as an anomalous data point. All marked anomalous data points are then aggregated to obtain the potential change point set for the irrigation district.

[0155] For each data point in the potential change point set, observe the direction and magnitude of change of the consecutive data points preceding and following that data point. If the direction of change of the data points before and after that data point remains consistent, and the magnitude of change does not show a significant trend change, but only the rate of change of that data point itself is abnormal, then that data point is determined to be a pseudo-change point and is removed from the potential change point set. After removing all pseudo-change points from the potential change point set, the remaining data points are those that can truly reflect the trend of data change. These points are then organized to form the change node set of the irrigation district.

[0156] All data points in the set of change nodes are arranged in chronological order. Each arranged change node is used as a dividing point to divide the entire compressed data into segments. Each change node corresponds to the end of one trend segment and the beginning of another trend segment in the compressed data. These change nodes with clear trend boundary significance are marked, and the time, data value, and position of each node in the compressed data are recorded. These marked nodes are then summarized to form a set, which is the trend turning point set of the irrigation area.

[0157] The beneficial effects are that the compressed data feature analysis module, through multi-step collaborative execution, first accurately marks abnormal segments in the compressed data that exceed the normal range based on preset threshold boundary conditions, thus identifying key objects for subsequent targeted analysis; then, it extracts continuous data points from the abnormal segments to form a time series dataset, and combines change rate detection to identify potential change points and eliminate spurious change points, effectively eliminating data fluctuation interference and ensuring that a set of trend turning points that truly reflects trend reversals is obtained; subsequently, based on the set of turning points, the slope and duration of trend segments are extracted, and dynamic feature parameters are quantified and integrated by calculating the comprehensive index of the receding trend, giving the receding trend parameters a clear quantitative basis and reliability; at the same time, the compressed data is divided into continuous segments by the set of turning points, the trend lines of each segment are fitted, and the slope and interrelationships of the trend segments are comprehensively analyzed to fully capture the changing patterns of the data at different stages, allowing the trend parameters to fully present the overall changing characteristics of the compressed data. The entire process is progressive, achieving precise location of abnormal sections and accurate identification of trend reversals. Through quantitative analysis and pattern extraction, it ensures that the obtained drainage trend parameters are both accurate and comprehensive, providing high-quality feature data support for subsequent drainage status level analysis and effectively improving the effectiveness and reliability of data feature analysis in irrigation area drainage monitoring.

[0158] The drainage status level analysis module 105 is used to perform collaborative analysis of the meteorological environment data and irrigation operation data of the irrigation area with the drainage dynamic characteristic parameters to obtain the drainage status level index of the irrigation area.

[0159] In this embodiment of the invention, when the drainage status level analysis module performs collaborative analysis of the meteorological environment data and irrigation operation data of the irrigation district with the drainage dynamic characteristic parameters to obtain the drainage status level index of the irrigation district, it is specifically used for:

[0160] The meteorological environment data, irrigation operation data and drainage dynamic characteristic parameters of the irrigation area are integrated from multiple sources to obtain the multi-source dataset of the irrigation area.

[0161] The multi-source datasets are classified and organized according to three dimensions: environmental characteristics, operational status, and dynamic changes, to establish a status assessment framework for the irrigation district.

[0162] Based on the aforementioned state assessment framework, the dynamic characteristic parameters of the drainage are evaluated to obtain the drainage state level index of the irrigation area.

[0163] Meteorological environmental data, such as precipitation, temperature, and wind speed, are acquired from meteorological monitoring equipment deployed in the irrigation district. This data is collected in real time by the meteorological monitoring equipment and transmitted to the system via communication links. Irrigation operation data, such as irrigation duration, irrigation area, and single irrigation water volume, are acquired from the irrigation control system of the irrigation district. This data is recorded and stored synchronously by the irrigation control system during irrigation operations. Receding water dynamic characteristic parameters are obtained after processing by the compressed data feature parsing module, including information such as receding water change trend parameters and receding water trend comprehensive index. These three types of data are aligned along the same time dimension to ensure a one-to-one correspondence between meteorological environmental data, irrigation operation data, and receding water dynamic characteristic parameters within the same time period. Then, the data format is standardized, and all data are organized into a structured dataset containing time stamps, data types, and data content, resulting in a multi-source dataset for the irrigation district.

[0164] Data items related to meteorology were selected from multi-source datasets and categorized into the environmental characteristics dimension. This dimension specifically includes precipitation, temperature, and wind speed data, reflecting the impact of the current natural environmental conditions in the irrigation area on drainage. Data items related to irrigation operations were selected and categorized into the operational status dimension. This dimension specifically includes irrigation duration, irrigation area, and single irrigation volume data, reflecting the impact of irrigation activities on drainage. Data items related to changes in drainage itself were selected and categorized into the dynamic change dimension. This dimension specifically includes drainage change trend parameters and a comprehensive drainage trend index, directly reflecting the real-time changes in drainage. These three dimensions and their constituent data items are presented in a structured format, clarifying the data source and meaning of each dimension, thus establishing a status assessment framework for the irrigation area.

[0165] In the state assessment framework, data from the environmental characteristics dimension is analyzed first. If precipitation data shows recent rainfall, it is determined that the natural environment may lead to an increase in drainage volume. If the temperature is high, it may accelerate water evaporation and affect drainage. Next, data from the operational status dimension is analyzed. If the irrigation duration is long and the single irrigation volume is large, it is determined that irrigation activities may increase drainage volume. Combining the analysis results of these two dimensions, the dynamic characteristics parameters of drainage in the dynamic change dimension are evaluated. For example, if the comprehensive drainage trend index shows a rapid drainage rate, combined with the absence of heavy rainfall in the environmental characteristics and the small irrigation volume in the operational status, it is determined that the drainage rate is within the normal range and corresponds to a lower level. If the drainage rate is rapid and there is heavy rainfall in the environmental characteristics or a large irrigation volume in the operational status, it is determined that there may be an anomaly and corresponds to a higher level. Through such a comprehensive assessment, the specific level of the drainage status reflected by the dynamic characteristics parameters of drainage is determined, and the drainage status level index of the irrigation area is obtained.

[0166] The beneficial effects are as follows: The drainage status level analysis module integrates multi-source data, including irrigation district meteorological environment data, irrigation operation data, and drainage dynamic characteristic parameters. It aligns monitoring data from different sources and of different types according to the time dimension and unifies the format, forming a complete and closely related multi-source dataset. This avoids the one-sidedness of analysis caused by scattered data and inconsistent formats. Furthermore, the multi-source dataset is categorized and organized according to three dimensions: environmental characteristics, operational status, and dynamic changes, establishing a status assessment framework. This clearly defines the attributes and functions of various data types and clarifies the impact logic of different dimensions of data on the drainage status, providing a structured and systematic analytical basis for subsequent assessments. Finally, when assessing the drainage dynamic characteristic parameters according to this framework, it can comprehensively judge whether the drainage status reflected by the dynamic characteristic parameters is reasonable by combining environmental conditions and actual irrigation operations. This avoids assessment bias caused by relying solely on a single drainage data point, ensuring that the final drainage status level index of the irrigation district is both comprehensive and accurate. This provides a reliable level basis for the subsequent drainage status early warning generation module, effectively improving the scientific rigor and practical value of irrigation district drainage status assessment.

[0167] The water receding status early warning generation module 106 is used to compare the water receding status level index with the preset status level standard, identify the water receding status pattern between the water receding status level index and the preset status level standard, and obtain the alarm command and early warning log tracking link of the irrigation area.

[0168] In this embodiment of the invention, when the drainage status early warning generation module compares the drainage status level index with the preset status level standard, identifies the drainage status pattern between the drainage status index and the preset status level standard, and obtains the alarm command and early warning log tracking link of the irrigation area, it is specifically used for:

[0169] The water receding status level index is compared item by item with the preset status level standard to obtain the level difference identifier of the irrigation area;

[0170] Analyze the distribution characteristics of the grade difference indicators to identify the abnormal state pattern types of the irrigation area;

[0171] Based on the historical early warning rule base of the irrigation district, the early warning rules are matched for the abnormal state pattern types to obtain the alarm information of the irrigation district;

[0172] The alarm information is distributed to the management terminal of the irrigation area to complete the transmission of early warning information;

[0173] The generation time, warning level, and distribution path of the alarm command are recorded synchronously to obtain the warning log tracking link of the irrigation area.

[0174] When the drainage status early warning generation module performs early warning rule matching on the abnormal status pattern type based on the historical early warning rule library of the irrigation district to obtain the alarm information of the irrigation district, it is specifically used for:

[0175] The typical abnormal pattern features in the historical early warning rule base of the irrigation area are retrieved to obtain the rule feature set of the irrigation area;

[0176] The similarity between the abnormal state pattern type and the rule feature set is compared to obtain the matching degree evaluation result of the irrigation area;

[0177] Based on the matching degree evaluation results, the early warning rules for the irrigation area are filtered to obtain the candidate rule set for the irrigation area;

[0178] By combining the real-time environmental factors of the irrigation area, the applicability of the candidate rule set is verified to obtain the early warning rules for the irrigation area;

[0179] Based on the aforementioned early warning rules, the early warning level and response measures for the irrigation district are configured, and the alarm information for the irrigation district is obtained.

[0180] The preset status level standard is set in advance by the system according to the safety management needs of irrigation district drainage. It includes the drainage status level range corresponding to different safety levels, such as the specific ranges corresponding to the safety level, warning level, and danger level of each dimension, including drainage rate, drainage volume, and water conductivity. Each dimension level included in the drainage status level index is compared one by one with the corresponding dimension level range in the preset status level standard. If the drainage status level index of a certain dimension exceeds or falls below the safety level range of the corresponding dimension in the preset standard, the difference in that dimension is clearly marked, such as "drainage rate level exceeds the preset safety level range" or "conductivity level is lower than the preset safety level range". All the difference marks of the dimensions are organized into a structured set of identifiers to obtain the level difference identifier of the irrigation district.

[0181] The system extracts information on the number of dimensions involved in the differences, the specific degree of difference in each dimension, and the duration of the differences from the grade difference identifiers. It then analyzes the distribution characteristics of this information, such as determining whether the difference exists only in a single dimension or multiple dimensions, whether the degree of difference slightly exceeds the standard or severely exceeds the standard, and whether the difference is transient or persistent. Based on these distribution characteristics, the system classifies and identifies the differences. If the differences involve multiple dimensions and are persistent, and the degree of difference in each dimension reaches or exceeds the warning level, it is identified as a "multi-dimensional persistent warning abnormal pattern." If the differences exist only in a single dimension and are transient with a slight degree of difference, it is identified as a "single-dimensional instantaneous slight abnormal pattern." Through this classification and identification, the abnormal state pattern type of the irrigation area is obtained.

[0182] The historical early warning rule base is a pre-stored collection of various abnormal state patterns that have occurred in the irrigation district in the past, along with corresponding early warning handling rules. Each rule clearly records the early warning level corresponding to the abnormal state pattern type, the response measures to be triggered, and the information notification scope. The currently identified abnormal state pattern type is compared one by one with all abnormal state pattern types stored in the historical early warning rule base. Historical pattern records that completely match the current abnormal state pattern type are found, and the corresponding early warning level, response measures, and notification scope are extracted from these records. This information is then integrated to form structured information containing "early warning level - abnormal content - response measures - notification scope," thus obtaining the irrigation district's alarm information.

[0183] The system has a built-in dedicated communication transmission module that pre-stores the fixed communication addresses of all irrigation district management terminals. Upon receiving an alarm message, the communication transmission module encapsulates the alarm message into a standard transmission packet according to a preset information format and sends it to all preset irrigation district management terminals through a stable communication link. After transmission, the communication module receives a reception confirmation signal from each management terminal to ensure that each management terminal successfully receives the alarm message and completes the early warning information transmission.

[0184] At the moment an alarm command is generated, the system automatically invokes its built-in time recording function to record the current time. Simultaneously, it extracts the warning level information from the generated alarm information. During alarm information distribution, the system records the complete information transmission path in real time, including the process of information being transferred from the system's core processing module to the communication transmission module, the specific terminal identifiers sent by the communication transmission module to each management terminal, and communication node information during transmission. The recorded alarm command generation time, warning level, and distribution path information are organized chronologically into a structured log set containing the corresponding relationships for each record. This set clearly traces the entire process of alarm command generation and transmission, providing a clear link for tracking the warning logs of the irrigation district.

[0185] The historical early warning rule base pre-stores various typical abnormal state patterns that have occurred in the irrigation district in the past, along with their corresponding feature information. This feature information includes the number of dimensions involved in the anomaly, the duration range of the anomaly in each dimension, and the severity level of the anomaly. The system opens the storage file of the historical early warning rule base through a preset access path, extracts the complete feature information corresponding to each typical abnormal pattern in the base, and organizes this feature information according to the structure of "abnormal pattern name - feature item - feature content" to form a structured set containing the features of all typical abnormal patterns, thus obtaining the rule feature set of the irrigation district.

[0186] First, key features are extracted from the currently identified abnormal state pattern types. These features are consistent with the features of typical abnormal patterns in the rule feature set, including the number of dimensions involved in the anomaly, the duration of the anomaly in each dimension, and the severity level of the anomaly. Each extracted feature is compared one by one with the features corresponding to each typical abnormal pattern in the rule feature set. If the content of a feature is completely identical, it is determined to be a complete match; if the content partially overlaps, it is determined to be a partial match. Based on the matching results of all feature items, the overall similarity between the current abnormal state pattern type and each typical abnormal pattern is evaluated. For example, if all feature items are completely matched, it is evaluated as a high match; if most feature items are partially matched, it is evaluated as a medium match; and if a few feature items are matched, it is evaluated as a low match. These evaluation results are compiled into a structured document to obtain the matching degree evaluation result of the irrigation area.

[0187] The system pre-sets screening criteria, retaining only the early warning rules corresponding to typical anomaly patterns marked as high or medium matching in the matching evaluation results, excluding early warning rules corresponding to low matching. Based on the names of typical anomaly patterns corresponding to high and medium matching recorded in the matching evaluation results, the system searches and extracts complete early warning rules associated with these pattern names from the historical early warning rule database. These early warning rules include the corresponding early warning level, the response measures to be triggered, and the scope of information notification recipients. All extracted early warning rules are arranged in descending order of matching degree, ensuring that early warning rules with high matching degrees are at the top, forming a set containing multiple qualified early warning rules, thus obtaining the candidate rule set for the irrigation district.

[0188] Real-time environmental factors include current precipitation, wind speed, crop growth stage, and soil moisture in the irrigation area. This information is acquired in real-time by the system from meteorological monitoring equipment and sensors in the irrigation area. Each early warning rule in the candidate rule set is extracted one by one, and the historical environmental factors corresponding to the rule's typical abnormal pattern are compared with the current real-time environmental factors to determine whether the real-time environmental factors affect the rule's applicability. For example, if a candidate rule corresponds to a historical environment of no precipitation, and the current real-time environment is continuous heavy precipitation, it is necessary to determine whether the heavy precipitation causes a difference in the essential cause between the current abnormal state pattern and the historical pattern. If the essential cause is the same, the rule is still applicable; if the essential cause is different, the rule is not applicable. All applicable early warning rules are retained, and inapplicable early warning rules are removed. The final single or integrated early warning rule is the early warning rule for the irrigation area.

[0189] The system directly extracts the corresponding warning level information and response measures from the established warning rules. Following the system's preset alarm information format, it fills in the corresponding fields with the warning level, abnormal state mode type, response measures, and the effective time of the warning rule, ensuring that the information in each field is accurate, complete, and without omissions or errors. The completed information is then validated to ensure it meets the display requirements of the management terminal, resulting in a structured document containing all key warning information, thus obtaining the irrigation district's alarm information.

[0190] The beneficial effects are as follows: The flood discharge status early warning generation module accurately locates differences in each dimension and forms level difference identifiers by comparing the flood discharge status level indicators with preset status level standards item by item, providing a clear target for anomaly analysis. Furthermore, by analyzing the distribution characteristics of the level difference identifiers, scattered differences are integrated into anomaly status pattern types with clear attributes, avoiding one-sided anomaly judgments. In the rule matching stage, typical anomaly pattern features from the historical early warning rule library are first retrieved to construct a rule feature set. Similarity comparison and matching degree filtering are used to obtain a set of candidate rules that fit the current anomaly. Then, the applicability of the rules is verified by combining real-time environmental factors in the irrigation area, ensuring that the early warning rules are based on historical experience and fit the current reality, effectively avoiding misjudgments caused by rigidly applying rules. The early warning level and response measures configured based on the verified early warning rules make the alarm information both targeted and practical. The alarm information is promptly distributed to the management terminal to ensure the timeliness of early warning transmission. The alarm log tracking link, which records the alarm command generation time, early warning level, and distribution path, provides a complete basis for subsequent anomaly tracing, responsibility verification, and rule optimization. The entire process, from differential location, anomaly identification, rule matching to early warning transmission and traceability, forms a closed-loop guarantee, significantly improving the accuracy, timeliness, and traceability of irrigation area drainage warnings, and providing strong support for the safe management and control of irrigation area drainage.

[0191] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0192] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0193] Finally, it should be noted that 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An Internet of Things-based irrigation district drainage monitoring system, characterized in that, The system includes a multi-source data acquisition module for drainage outlets, a module for removing abnormal detection data, a module for filtering and compressing data, a module for analyzing compressed data features, a module for analyzing drainage status levels, and a module for generating early warnings for drainage status, wherein: The multi-source data acquisition module for the drainage outlet is used to transmit the flow rate information, water level information, and conductivity information of the drainage outlet in the irrigation area to the edge processing node to obtain the multi-source detection data of the irrigation area. The abnormal data removal module is used to remove abnormal data points from the multi-source detection data based on the real-time characteristics and distribution patterns in the multi-source detection data, thereby obtaining the filtered data of the irrigation area. The filtered data compression strategy module is used to construct a data compression strategy for the irrigation area based on the spatiotemporal correlation in the filtered data and the laws of correlation characteristics in the filtered data, so as to obtain compressed data of the irrigation area. The compressed data feature analysis module is used to perform piecewise regression analysis on the abnormal segments and trend features in the compressed data according to the threshold boundary conditions and trend change rules in the compressed data, so as to obtain the change trend parameters and water receding dynamic feature parameters of the irrigation area. The drainage status level analysis module is used to collaboratively analyze the meteorological environment data and irrigation operation data of the irrigation area with the drainage dynamic characteristic parameters to obtain the drainage status level index of the irrigation area. The water receding status early warning generation module is used to compare the water receding status level index with the preset status level standard, identify the water receding status pattern between the water receding status level index and the preset status level standard, and obtain the alarm command and early warning log tracking link of the irrigation area.

2. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 1, characterized in that, When the multi-source data acquisition module for the drainage outlet transmits the flow rate, water level, and conductivity information of the irrigation area's drainage outlet to the edge processing node to obtain the multi-source detection data of the irrigation area, it is specifically used for: Obtain flow rate, water level, and conductivity information from the irrigation district's drainage outlets; The flow rate information, water level information, and conductivity information are timestamped to obtain synchronous monitoring data of the irrigation area; The format integrates the synchronous monitoring data to obtain the water status information of the irrigation area; The water body status information is packaged into transmission data packets using a wireless communication protocol, and a secure communication connection is established with the edge processing node. The transmission data packet is sent to the edge processing node to obtain multi-source detection data of the irrigation area.

3. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 1, characterized in that, When the anomaly removal module removes abnormal data points from the multi-source detection data based on real-time characteristics and distribution patterns to obtain filtered data for the irrigation area, it is specifically used for: Initialize the communication protocol between the edge processing node and the monitoring node in the irrigation area, and establish the sensing network topology of the irrigation area; Based on the perception network topology, the state characteristics of the monitoring nodes in the irrigation area are obtained in real time to obtain the node operation status map of the irrigation area; Based on the node operation status map, the dynamic change characteristics of the sensing network topology are analyzed to obtain the adaptive networking strategy of the irrigation area; According to the adaptive networking strategy, abnormal fluctuation data in the multi-source detection data are classified and filtered to obtain the primary filtered data of the irrigation area. Spatiotemporal interpolation is performed on the missing time periods in the primary filtered data to obtain the optimized monitoring data of the irrigation area; The integrity of the optimized monitoring data is verified through a consistency check mechanism to obtain the filtered data of the irrigation area.

4. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 1, characterized in that, When the filtered data compression strategy module executes a data compression strategy for the irrigation district based on the spatiotemporal correlation of the filtered data and the laws governing the correlation characteristics of the filtered data, to obtain compressed data for the irrigation district, it is specifically used for: Identify the spatiotemporal correlation rules between different monitoring points in the filtered data to obtain the correlation rule table of the irrigation area; Based on the association rule table, a data coding scheme for the irrigation area is formulated, and the coding mechanism of the irrigation area is obtained; Based on the encoding mechanism, feature extraction is performed on the filtered data to obtain the data feature set of the irrigation area; According to a preset compression ratio, the data feature set is subjected to volume optimization to obtain the compressed data stream of the irrigation area; The integrity of key features in the compressed data stream is verified to obtain the compressed data of the irrigation area.

5. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 1, characterized in that, When the compressed data feature analysis module performs piecewise regression analysis of abnormal segments and trend features in the compressed data based on threshold boundary conditions and trend change patterns in the compressed data, and obtains the changing trend parameters and drainage dynamic characteristic parameters of the irrigation area, it is specifically used for: Based on preset threshold boundary conditions, data segments in the compressed data that exceed the boundary range are marked to obtain the abnormal data segment identifiers of the irrigation area; The abnormal data segments are analyzed for trend trends, and the trend change nodes of the abnormal data segments are identified to obtain the trend turning point set of the irrigation area. By integrating the dynamic characteristic parameters of the trend inflection point set, the water receding trend parameters of the irrigation area are obtained; Based on the set of trend turning points, the compressed data is divided into continuous segments to obtain the data segment sequence of the irrigation area; By fitting a trend line that shows the changing patterns in the data segment sequence, a set of trend line segments for the irrigation area is obtained; By comprehensively analyzing the slope and interrelationships of the trend line segments, the changing trend parameters of the irrigation area are obtained.

6. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 5, characterized in that, When the compressed data feature parsing module integrates the dynamic feature parameters of the trend inflection point set to obtain the drainage trend parameters of the irrigation area, it is specifically used for: Based on the set of trend turning points, the slope value and duration of the trend line segments are extracted to obtain the line segment slope sequence and duration sequence of the irrigation area; Based on the slope sequence and duration sequence, a comprehensive index of the receding water trend parameters is calculated. The formula for calculating the comprehensive index of the receding water trend is as follows: ； In the formula, The comprehensive index of the receding trend, The first line segment in the slope sequence The slope of each trend segment, The duration sequence of the first The duration of each trend segment. The median of the slope sequence. The number of trend segments concentrated at the trend reversal points. These are the preset weighting factors.

7. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 5, characterized in that, When the compressed data feature parsing module performs trend analysis on the abnormal data segments and identifies the trend change nodes of the abnormal data segments to obtain the trend inflection point set of the irrigation area, it is specifically used for: Extract the sequence of continuous data points from the abnormal data segment to obtain the time series dataset of the irrigation area; Change rate detection is performed on the time series dataset, and abnormal data points of change rate in the time series data are identified to obtain the potential change point set of the irrigation area; Remove the pseudo-change points from the potential change point set to obtain the change node set of the irrigation area; Based on the set of changing nodes, the compressed data is divided into segments to obtain the set of trend turning points of the irrigation area.

8. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 1, characterized in that, When the drainage status level analysis module performs collaborative analysis of the meteorological environment data, irrigation operation data, and drainage dynamic characteristic parameters of the irrigation area to obtain the drainage status level index of the irrigation area, it is specifically used for: The meteorological environment data, irrigation operation data and drainage dynamic characteristic parameters of the irrigation area are integrated from multiple sources to obtain the multi-source dataset of the irrigation area. The multi-source datasets are classified and organized according to three dimensions: environmental characteristics, operational status, and dynamic changes, to establish a status assessment framework for the irrigation district. Based on the aforementioned state assessment framework, the dynamic characteristic parameters of the drainage are evaluated to obtain the drainage state level index of the irrigation area.

9. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 1, characterized in that, When the flood discharge status early warning generation module compares the flood discharge status level index with the preset status level standard, identifies the flood discharge status pattern between the index and the standard, and obtains the alarm command and early warning log tracking link for the irrigation area, it is specifically used for: The water receding status level index is compared item by item with the preset status level standard to obtain the level difference identifier of the irrigation area; Analyze the distribution characteristics of the grade difference indicators to identify the abnormal state pattern types of the irrigation area; Based on the historical early warning rule base of the irrigation district, the early warning rules are matched for the abnormal state pattern types to obtain the alarm information of the irrigation district; The alarm information is distributed to the management terminal of the irrigation area to complete the transmission of early warning information; The generation time, warning level, and distribution path of the alarm command are recorded synchronously to obtain the warning log tracking link of the irrigation area.

10. The irrigation district drainage monitoring system based on the Internet of Things as described in claim 9, characterized in that, When the drainage status early warning generation module performs early warning rule matching on the abnormal status pattern type based on the historical early warning rule library of the irrigation district to obtain the alarm information of the irrigation district, it is specifically used for: The typical abnormal pattern features in the historical early warning rule base of the irrigation area are retrieved to obtain the rule feature set of the irrigation area; The similarity between the abnormal state pattern type and the rule feature set is compared to obtain the matching degree evaluation result of the irrigation area; Based on the matching degree evaluation results, the early warning rules for the irrigation area are filtered to obtain the candidate rule set for the irrigation area; By combining the real-time environmental factors of the irrigation area, the applicability of the candidate rule set is verified to obtain the early warning rules for the irrigation area; Based on the aforementioned early warning rules, the early warning level and response measures for the irrigation district are configured, and the alarm information for the irrigation district is obtained.

Images