Water conservancy project intelligent monitoring and early warning system based on Internet of Things and AI

Abstract

The invention discloses a water conservancy project intelligent monitoring and early warning system based on the Internet of Things and AI. According to the system, firstly, the problem that a traditional monitoring task is blind and disordered is solved by obtaining a potential monitoring demand index of the monitoring task, marking a target task in combination with a threshold value and generating a priority list, resource configuration is optimized, and deployment scientificity and pertinence are improved; secondly, a water conservancy target topology network model is constructed, monitoring links are formed, early warning links are screened, adaptive measures are adopted according to the number grading of the early warning links, full-chain closed-loop management is achieved, and early warning accuracy and response effectiveness are improved; and 3, when an early warning link appears, generating an update priority list, dynamically associating subsequent tasks, quickly responding to chain risks, avoiding omission or lag, enhancing the flexibility and coherence of the system to cope with complex risks, and ensuring the safe operation of the water conservancy project.

Description

Intelligent monitoring and early warning system for water conservancy projects based on IoT and AI Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for water conservancy projects, and more specifically, to an intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI. Background Technology

[0002] Safety monitoring and early warning of water conservancy projects are crucial for ensuring flood control, water supply, and ecological security in river basins. Traditional water conservancy monitoring mainly relies on manual inspections and fixed sensor threshold alarms. The deployment and execution of monitoring tasks are often based on experience or fixed cycles, lacking dynamic quantitative assessment of the inherent risks of the tasks. This makes it difficult to accurately focus monitoring resources on high-risk locations, easily leading to problems such as blind task allocation and unreasonable resource allocation, thus restricting the scientific nature and overall efficiency of monitoring work.

[0003] Existing early warning mechanisms often focus on data exceeding limits at a single monitoring point, failing to fully consider the complex hydrological and hydraulic relationships and risk transmission paths within water conservancy projects and between upstream and downstream areas. This isolated approach to judgment makes it difficult to reveal the complete chain from the source of risk to the weakest link in the project, easily leading to delays or misjudgments in the identification of hidden and interconnected risks. The accuracy of early warnings and the depth of risk disclosure are insufficient, making it difficult to support effective tiered response and precise prevention and control.

[0004] Furthermore, after identifying a risk, traditional monitoring systems typically follow a predetermined analysis process, lacking the ability to dynamically adjust task priorities based on issued early warning information. This static management model struggles to respond quickly to chain reactions that may be triggered by risk transmission and cannot adapt to the complex characteristics of dynamic evolution and multi-point, multi-line connections in water conservancy risks, resulting in weak flexibility and overall coordination effectiveness when dealing with comprehensive and continuous risks. Summary of the Invention

[0005] In view of the shortcomings of existing technologies, the purpose of this invention is to provide an intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI includes:

[0008] The priority task list generation module for water conservancy projects is used to identify all IoT sensing nodes included in the water conservancy project, periodically generate data sensing records for each IoT sensing node, synchronously obtain the potential monitoring demand index for each water conservancy project monitoring task, mark the corresponding water conservancy monitoring target tasks based on the comparison results between the potential monitoring demand index and the potential monitoring demand threshold, and then generate the priority task list for water conservancy projects.

[0009] The water conservancy project task monitoring and early warning module selects the first water conservancy monitoring target task in the priority task list of water conservancy projects, determines the water conservancy target topology network model corresponding to the water conservancy monitoring target task, and then forms multiple water conservancy project monitoring links. It then determines all water conservancy project early warning links. Based on the total number of water conservancy project early warning links, the water conservancy project monitoring tasks are marked as primary early warning monitoring tasks, intermediate early warning monitoring tasks, or key early warning monitoring tasks, and corresponding early warning measures are taken.

[0010] The water conservancy target topology network model determination module is used to determine the water conservancy target topology network model corresponding to the water conservancy monitoring target tasks;

[0011] The early warning update priority task list generation module generates an early warning update priority task list after a water conservancy project early warning link is established. The module then controls the water conservancy project task monitoring and early warning module to perform analysis on each water conservancy monitoring target task in sequence according to the early warning update priority task list.

[0012] Furthermore, the process for obtaining the potential monitoring demand index for water conservancy project monitoring tasks is as follows: Select a water conservancy project monitoring task, obtain the total number of water conservancy project early warning links that appear within a time period q before the current system time, and mark them as DPC; obtain the total number of water conservancy project monitoring links that can be generated by the water conservancy target topology network model corresponding to the water conservancy project monitoring task, and mark them as FMZ; calculate the ratio between DPC and FMZ to obtain the potential monitoring demand index for the water conservancy project monitoring task.

[0013] Furthermore, the process for generating the priority task list for water conservancy projects is as follows: For all water conservancy monitoring target tasks, prioritize them from largest to smallest based on their potential monitoring demand index values, and generate a priority task list for water conservancy projects with a clear order of task execution.

[0014] Furthermore, the data perception record includes the water conservancy project entity ID, the perception data type, the actual data value, and the data perception time.

[0015] Furthermore, the process for forming multiple water conservancy project monitoring links is as follows: obtain the dynamic conductivity of each associated edge in the water conservancy target topology network model; when the dynamic conductivity is higher than the dynamic conductivity threshold, mark the corresponding associated edge as a water conservancy risk edge; and connect multiple water conservancy risk edges sequentially according to their directions to form multiple water conservancy project monitoring links.

[0016] Furthermore, the process for determining the topology network model of the water conservancy target corresponding to the water conservancy monitoring target task is as follows: determine the IoT sensing nodes associated with the water conservancy monitoring target task, take all associated IoT sensing nodes as network nodes, and take the influence associations between network nodes as network edges to form the water conservancy target topology network model.

[0017] Further, the process for obtaining the dynamic transmissibility of the associated edge is as follows: Select an associated edge, clarify the type of the associated edge, mark the IoT sensing nodes at both ends of the associated edge as node a and node b respectively, and determine the dynamic transmissibility of the associated edge based on the type of the associated edge.

[0018] Furthermore, the method for determining the early warning link of a water conservancy project is as follows: Select a water conservancy project monitoring link, obtain the dynamic transmission deviation of each water conservancy risk edge in the monitoring link, calculate the absolute difference of the dynamic transmission deviation of two adjacent water conservancy risk edges in the monitoring link, calculate multiple dynamic transmission deviation distance values, sum and average the multiple dynamic transmission deviation distance values ​​to calculate the comprehensive transmission deviation, obtain the true data value of the source IoT sensing node of the water conservancy project monitoring link, and when the true data value of the source IoT sensing node is outside the reasonable range of data and the comprehensive transmission deviation is lower than the comprehensive transmission deviation threshold, mark the water conservancy project monitoring link as a water conservancy project early warning link.

[0019] Furthermore, the process for obtaining the dynamic transmission deviation of the water conservancy risk edge is as follows: Select a water conservancy risk edge, obtain the dynamic transmission degree of the water conservancy risk edge and the corresponding dynamic transmission degree threshold, calculate the difference between the dynamic transmission degree and the dynamic transmission degree threshold, and calculate the dynamic transmission deviation of the water conservancy risk edge.

[0020] Furthermore, once a water conservancy project early warning link is identified, a list of priority tasks for early warning updates is generated. The process is as follows: All water conservancy project early warning links included in the current water conservancy monitoring target task are determined; the source IoT sensing node for each water conservancy project early warning link is identified; the total number of source IoT sensing nodes is marked as Rw; simultaneously, all water conservancy monitoring target tasks following the current water conservancy monitoring target task in the priority task list are marked as early warning adaptation and adjustment tasks; the early warning update priority index of each early warning adaptation and adjustment task is obtained; and all early warning adaptation and adjustment tasks are prioritized according to their early warning update priority index values ​​from largest to smallest, generating a list of priority tasks for early warning updates with a clear order of task execution.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The system of this invention obtains the potential monitoring demand index of each water conservancy project monitoring task and marks the water conservancy monitoring target tasks by combining the comparison results of potential monitoring demand thresholds, thereby generating a priority task list for water conservancy projects. This breaks the blindness and disorder of task execution in traditional water conservancy monitoring, accurately identifies tasks with high monitoring necessity, prioritizes monitoring resources to core needs, effectively avoids resource waste, and significantly improves the scientificity and pertinence of water conservancy monitoring task deployment. In the monitoring and early warning execution stage, the system constructs a corresponding water conservancy target topology network model, sorts out the correlation between nodes and forms multiple monitoring links, then accurately selects early warning links, and finally conducts graded early warning and takes appropriate measures based on the number of early warning links. This achieves closed-loop management of the entire chain from task focus to risk link tracing and graded response, which can more comprehensively and deeply capture the potential risks of water conservancy projects, make early warning measures more differentiated and adaptable, and significantly improve the accuracy of early warning and the effectiveness of response.

[0023] When a water conservancy project early warning link occurs, the system can promptly generate a list of early warning update priority tasks and control the monitoring and early warning module to analyze subsequent tasks in sequence. Based on the real-time early warning situation, the system can associate subsequent related tasks and quickly respond to potential chain risks. This avoids the problem of risk omission or response delay caused by the rigid order of tasks, greatly improves the system's flexibility and continuity in dealing with complex water conservancy risks, and further ensures the safe operation of water conservancy projects. Attached Figure Description

[0024] Figure 1 is a flowchart of the operation of the system of the present invention;

[0025] Figure 2 is a flowchart of the process for determining the topology network model of water conservancy targets;

[0026] Figure 3 is a flowchart of the process for determining the early warning link of water conservancy projects. Detailed Implementation

[0027] As shown in Figures 1 to 3, the present invention provides an intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI, which includes a water conservancy project priority task list generation module, a water conservancy target topology network model determination module, a water conservancy project task monitoring and early warning module, and an early warning update priority task list generation module.

[0028] The water conservancy project priority task list generation module identifies all IoT sensing nodes included in the water conservancy project, periodically generates data sensing records for each IoT sensing node, and obtains the potential monitoring demand index for each water conservancy project monitoring task (water conservancy project monitoring tasks include, but are not limited to, P1 reservoir flood season overflow risk monitoring task, D2 dam piping hazard monitoring task, and A1 riverbank displacement monitoring task). When the potential monitoring demand index is higher than the potential monitoring demand threshold (the potential monitoring demand threshold is comprehensively set based on historical risk case data, such as statistically analyzing the correspondence between the potential monitoring demand index and the probability of risk occurrence in past water conservancy projects, and setting the lower limit of the potential monitoring demand index interval corresponding to a risk occurrence probability ≥ 85% as the initial threshold), the corresponding water conservancy project monitoring task is marked as a water conservancy monitoring target task, and then a water conservancy project priority task list is generated based on all water conservancy monitoring target tasks.

[0029] The process for obtaining the potential monitoring demand index for water conservancy project monitoring tasks is as follows: Select a water conservancy project monitoring task, obtain the total number of water conservancy project early warning links that appear within a time period q before the current system time, and mark it as DPC. Obtain the total number of water conservancy project monitoring links that can be generated by the water conservancy target topology network model corresponding to the water conservancy project monitoring task, and mark it as FMZ. Calculate the ratio between DPC and FMZ to obtain the potential monitoring demand index for the water conservancy project monitoring task (the purpose is to convert the absolute number of early warnings into relative early warning frequencies, which can avoid the problem of incomparable demand quantification caused by the difference in the total number of theoretical monitoring links for different tasks).

[0030] The process for generating a priority task list for water conservancy projects is as follows: For all water conservancy monitoring target tasks, prioritize them from largest to smallest based on their potential monitoring demand index values ​​to generate a priority task list for water conservancy projects with a clear order of task execution.

[0031] The data sensing record includes the water conservancy project entity ID (each reservoir, dam, and river is an independent water conservancy project entity with different structural characteristics, risk points, and monitoring needs; different water conservancy project entities require the deployment of adapted IoT sensing nodes; the water conservancy project entity ID is the ID that the IoT sensing node is specifically monitoring and sensing, for example: P1 reservoir, P2 reservoir, D1 dam, D2 dam, A1 river, A2 river), sensing data type (sensing data types include water level data, rainfall data, etc.; different water conservancy project entities require different sensing data types; for example, reservoirs require sensing data including water level data, rainfall data, dam stress, etc.; dams require sensing data including water level data, rainfall data, dam soil moisture content, etc.; and rivers require sensing data including water level data, rainfall data, and revetment displacement data), the actual data value (i.e., the specific numerical value of the current actual state of the corresponding type of sensing data), and the data sensing time (i.e., the current time).

[0032] The water conservancy project task monitoring and early warning module selects the first water conservancy monitoring target task in the priority task list of water conservancy projects, determines the corresponding water conservancy target topology network model, and obtains the dynamic conductivity of each associated edge in the water conservancy target topology network model. When the dynamic conductivity is higher than the dynamic conductivity threshold (the dynamic conductivity threshold is different for different types of associated edges; for example: hydrological conduction edge: corresponds to the transmission of natural hydrological elements (such as flow rate → velocity), the risk level is relatively mild, and the threshold can be set to 0.8; engineering control edge: corresponds to the equipment regulation effect (such as water level → gate opening), which directly affects the safety of project operation, and the threshold can be set to 0.75; risk diffusion edge: corresponds to the diffusion of anomalies (such as abnormal humidity → water level). (Anomaly), high risk level requiring rapid response, threshold can be set to 0.7), mark the corresponding associated edge as a water conservancy risk edge, and sequentially connect multiple water conservancy risk edges according to their directions to form multiple water conservancy project monitoring links (Example: Water conservancy risk edge ①: P2 reservoir Doppler flow meter → A1 river ultrasonic velocity sensor (hydrological transmission edge); Water conservancy risk edge ②: A1 river ultrasonic velocity sensor → P1 reservoir water level gauge (hydrological transmission edge); Water conservancy risk edge ③: P1 reservoir rain gauge → P1 reservoir water level gauge (data causal edge); Water conservancy risk edge ④: P1 reservoir water level gauge → P1 reservoir gate opening sensor (engineering control edge); Water conservancy risk edge ⑤: D1 dam soil moisture sensor → P1 reservoir water level gauge ... Level gauge (risk diffusion edge); forming three water conservancy project monitoring links: 1. P2 Doppler flow meter → A1 ultrasonic flow velocity sensor → P1 water level gauge → P1 gate opening sensor; 2. P1 rain gauge → P1 water level gauge → P1 gate opening sensor; 3. D1 soil moisture sensor → P1 water level gauge → P1 gate opening sensor), thereby determining all water conservancy project early warning links, determining the total number of water conservancy project early warning links, when the total number of water conservancy project early warning links is higher than the high number threshold, the corresponding water conservancy project monitoring task is marked as a key early warning monitoring task, when the total number of water conservancy project early warning links is lower than the low number threshold, the corresponding water conservancy project monitoring task is marked as a primary early warning monitoring task, when the water conservancy project When the total number of early warning links falls between the high and low thresholds, the corresponding water conservancy project monitoring tasks are marked as intermediate-level early warning monitoring tasks, and corresponding early warning measures are taken (e.g., for key early warning monitoring tasks, the generation frequency of data sensing records of IoT sensing nodes in all water conservancy project early warning links is increased to the second level, data changes are pushed in real time, and the pre-deployment of emergency resources (such as emergency equipment and materials) is initiated; for intermediate-level early warning monitoring tasks, the generation frequency of data sensing records of IoT sensing nodes in all water conservancy project early warning links is increased to the minute level, and risk contingency plans are prepared; for primary-level early warning monitoring tasks, corresponding responsible positions (such as dike patrol positions) are assigned to patrol at the daily frequency).

[0033] The water conservancy target topology network model determination module is used to determine the water conservancy target topology network model corresponding to the water conservancy monitoring target task. This module determines the IoT sensing nodes associated with the water conservancy monitoring target task (each water conservancy monitoring target task is pre-associated with multiple IoT sensing nodes; for example, the P1 reservoir flood season overflow risk monitoring task not only needs to consider the water level data of P1 reservoir (corresponding to the IoT sensing node of the water level gauge), rainfall data (corresponding to the IoT sensing node of the rain gauge), and gate discharge capacity data (corresponding to the IoT sensing node of the spillway gate opening sensor), but is also affected by the water level data and discharge data of P2 reservoir (upstream reservoir) (corresponding to the IoT sensing node of the Doppler flow meter), the river channel flood discharge capacity data of A1 river channel (inflow channel) (corresponding to the IoT sensing node of the ultrasonic flow velocity sensor), and the sedimentation data (corresponding to the IoT sensing node of the sediment content sensor), and the soil moisture content of the D1 embankment (along the A1 river channel) (corresponding to the IoT sensing node of the soil moisture sensor). Therefore, it is necessary to... The data from these IoT sensing nodes are all correlated. Because the watershed hydrological relationships and upstream and downstream engineering layouts of water conservancy projects are long-term and unchanging, and the influence logic is clear and quantifiable, corresponding IoT sensing nodes can be associated with each water conservancy monitoring target task. All associated IoT sensing nodes are taken as network nodes, and the influence relationships between network nodes are taken as network edges (the definition of network edges is essentially the implementation of objective laws + engineering logic + data causality. Each edge corresponds to a direct and quantifiable actual influence relationship in the water conservancy project. For example, in the topology network of P1 reservoir overflow monitoring, the edge between the IoT sensing node of the P2 reservoir Doppler flow meter and the IoT sensing node of the A1 river ultrasonic flow velocity sensor is based on hydrological and water conservancy laws; the edge between the IoT sensing node of the P1 reservoir water level gauge and the IoT sensing node of the P1 reservoir spillway gate opening sensor is based on engineering function association; the edge between the IoT sensing node of the P1 reservoir rain gauge and the IoT sensing node of the P1 reservoir water level gauge is based on data causality), forming a water conservancy target topology network model.

[0034] The process for obtaining the dynamic transmission degree of associated edges: Select an associated edge and clarify its type (associated edges include three types: hydrological transmission edges, engineering control edges, and risk diffusion edges; the definition of a hydrological transmission edge: nodes a and b are hydrological nodes (such as flow rate, velocity, and water level nodes), and the influence logic is water flow / volume transmission (e.g., a = P2 reservoir Doppler flow meter, b = A1 river channel ultrasonic velocity sensor); the definition of an engineering control edge: a is a status monitoring node, and b is an equipment control node (such as water level → gate, flow rate → control gate). The influencing logic is the transmission of engineering control instructions (e.g., a=P1 reservoir water level gauge, b=P1 gate opening sensor); the definition of the risk diffusion edge: a and b are risk-related nodes (e.g., water content → displacement, sediment content → flood discharge capacity), and the influencing logic is the diffusion of abnormal states (e.g., a=D1 dam soil moisture sensor, b=P1 reservoir water level gauge)). The IoT sensing nodes at both ends of the related edge are marked as node a and node b respectively (node ​​a is the out-degree end, node b is the in-degree end), and the dynamic transmission degree of the related edge is determined based on the type of the related edge.

[0035] The dynamic conductivity is determined differently for different types of associated edges; the dynamic conductivity of hydrological conduction edges is determined as follows: using the formula... The dynamic conductivity of the hydrological conduction edge was calculated. ;in, For hydraulic transmission efficiency, , The change in data for b is the difference between the actual value of b and the initial baseline value. The initial baseline value is the starting data of the corresponding data. For example, if the data type corresponding to b is water level data, then the initial baseline value of b is the starting water level. The preset data change amount for b (i.e., the standard change amount within the expected range set based on the design standards, safety specifications, or historical operating conditions of the water conservancy project for the data type corresponding to b). Let 'a' be the amount of data change. The preset data change for 'a'; The hydraulic connectivity coefficient is set based on the spatial location and engineering layout of a and b, with direct connectivity set at 1.0, indirect connectivity at 0.8, and interception at 0.5. , All are weighting coefficients. Since hydraulic transmission efficiency is a direct quantification of transmission effect, and hydraulic connectivity coefficient is a fundamental prerequisite for transmission, the value of j1 can be 0.6, and the value of j2 can be 0.4.

[0036] The dynamic conductivity of the engineering control edge is determined as follows: using the formula... The dynamic conductivity of the engineering control edges is calculated. ;in, For engineering control efficiency, The ideal value of data b is calculated based on the actual data value of node a according to preset engineering rules (scheduling curve, design standard), and is the ideal value that device b should achieve. To control the accuracy coefficient ( The average deviation rate is the average of the deviation rates of all individual lines. ); , All are weighting coefficients. Since engineering control efficiency is the core objective of control, and the control accuracy coefficient is an auxiliary guarantee for execution, the value of k1 can be 0.7 and the value of k2 can be 0.3.

[0037] The dynamic transmissibility of the risk diffusion edge is determined as follows: using the formula... ; The abnormal coordination rate is calculated by: statistically analyzing all data perception records of a and b over the past n months, obtaining the reasonable range of data corresponding to a, marking the data perception record as the dominant abnormal record when the actual value of a data perception record is outside the reasonable range, obtaining the data perception time of the dominant abnormal record, obtaining all data perception records of b within t time after the data perception time of the dominant abnormal record, and increasing the number of coordinated abnormal records by one when the actual value of at least one data perception record of b within t time is outside the reasonable range of data corresponding to b, and finally calculating the ratio of the total number of dominant abnormal records to the total number of coordinated abnormal records to obtain the abnormal coordination rate, with a value of [0,1]). For threshold proximity (obtaining the true value of the data recorded in the current data perception record for 'a', and simultaneously obtaining the normal baseline value of 'a', which is usually the median value within a reasonable range of data, using the formula...), ); , All are weighting coefficients. Since the abnormal coordination rate is the core criterion for judging risk diffusion, and the threshold proximity is an auxiliary triggering condition for risk diffusion, therefore... The value can be 0.6. The value of can be 0.4.

[0038] The method for determining the early warning link of a water conservancy project is as follows: Select a water conservancy project monitoring link, obtain the dynamic transmission deviation of each water conservancy risk edge in the monitoring link, calculate the absolute difference of the dynamic transmission deviation of two adjacent water conservancy risk edges in the monitoring link, calculate multiple dynamic transmission deviation distance values, sum and average the multiple dynamic transmission deviation distance values ​​to calculate the comprehensive transmission deviation, obtain the real data value of the source IoT sensing node of the water conservancy project monitoring link (example of source IoT sensing node of water conservancy project monitoring link: D1 soil moisture sensor → P1 water level gauge → P1 gate opening sensor, where D1 soil moisture sensor is the source IoT sensing node), when the real data value of the source IoT sensing node is outside the reasonable range of data, and the comprehensive transmission deviation is lower than the comprehensive transmission deviation threshold (the comprehensive transmission deviation threshold is set comprehensively based on the historical risk link data of the water conservancy project), mark the water conservancy project monitoring link as a water conservancy project early warning link.

[0039] The process for obtaining the dynamic transmission deviation of a water conservancy risk edge is as follows: Select a water conservancy risk edge, obtain the dynamic transmission degree of the water conservancy risk edge and the corresponding dynamic transmission degree threshold, calculate the difference between the dynamic transmission degree and the dynamic transmission degree threshold, and calculate the dynamic transmission deviation of the water conservancy risk edge (the purpose of dynamic transmission deviation is to measure the continuity and stability of risk transmission within the link. If the dynamic transmission deviation of each edge within the link is small, it means that the risk can be stably transmitted in the link and is more likely to trigger the risk at the endpoint).

[0040] The early warning update priority task list generation module generates an early warning update priority task list after a water conservancy project early warning link is established. The module then controls the water conservancy project task monitoring and early warning module to perform analysis on each water conservancy monitoring target task in sequence according to the early warning update priority task list.

[0041] Once a water conservancy project early warning link is identified, a list of priority tasks for early warning updates is generated. The process is as follows: Identify all water conservancy project early warning links included in the current water conservancy monitoring target task; identify the source IoT sensing node for each water conservancy project early warning link; mark the total number of source IoT sensing nodes as Rw; simultaneously mark all water conservancy monitoring target tasks after the current water conservancy monitoring target task in the priority task list as early warning adaptation and adjustment tasks; obtain the early warning update priority index for each early warning adaptation and adjustment task; sort all early warning adaptation and adjustment tasks according to their early warning update priority index values ​​from largest to smallest; and generate a list of priority tasks for early warning updates with a clear order of task execution.

[0042] The process for obtaining the early warning update priority index for early warning adaptation and adjustment tasks is as follows: Select an early warning adaptation and adjustment task, obtain the corresponding water conservancy target topology network model (the process is consistent with the determination process of the water conservancy target topology network model corresponding to the water conservancy monitoring target task), when the source IoT sensing node of a water conservancy project early warning link exists in the water conservancy target topology network model, increase the number of early warning associations by one, finally sum the number of early warning associations and mark it as PLS, and obtain the degree centrality of the source IoT sensing node in the water conservancy target topology network model. Calculate the average degree centrality by summing all degree centralities. Through formula The early warning update priority index for this early warning adaptation and adjustment task is calculated. ;in, , All are weighting coefficients. Because it retains the complementarity of the two dimensions while emphasizing the priority given to closely related water conservancy risk response logic, therefore... The value can be 0.3. The value can be 0.7.

[0043] The degree centrality of source IoT sensing nodes in the topological network model of water conservancy targets is obtained as follows: Taking node i as an example, the number of edges directly connected to node i in the network is counted, and this number is the original degree centrality of node i. The original degree centrality is normalized using the formula. The degree centrality of the source IoT sensing nodes in the topological network model of the water conservancy target is calculated; N is the total number of network nodes.

[0044] The above formulas are all dimensionless calculations, and the preset parameters in the formulas should be set by those skilled in the art according to the actual situation.

[0045] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0046] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0047] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A smart monitoring and early warning system for water conservancy projects based on the Internet of Things and AI, characterized in that, include: The priority task list generation module for water conservancy projects is used to identify all IoT sensing nodes included in the water conservancy project, periodically generate data sensing records for each IoT sensing node, synchronously obtain the potential monitoring demand index for each water conservancy project monitoring task, mark the corresponding water conservancy monitoring target tasks based on the comparison results between the potential monitoring demand index and the potential monitoring demand threshold, and then generate the priority task list for water conservancy projects. The water conservancy project task monitoring and early warning module selects the first water conservancy monitoring target task in the priority task list of water conservancy projects, determines the water conservancy target topology network model corresponding to the water conservancy monitoring target task, and then forms multiple water conservancy project monitoring links. It then determines all water conservancy project early warning links. Based on the total number of water conservancy project early warning links, the water conservancy project monitoring tasks are marked as primary early warning monitoring tasks, intermediate early warning monitoring tasks, or key early warning monitoring tasks, and corresponding early warning measures are taken. The water conservancy target topology network model determination module is used to determine the water conservancy target topology network model corresponding to the water conservancy monitoring target tasks; The early warning update priority task list generation module generates an early warning update priority task list after a water conservancy project early warning link is established. The module then controls the water conservancy project task monitoring and early warning module to perform analysis on each water conservancy monitoring target task in sequence according to the early warning update priority task list.

2. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI as described in claim 1, characterized in that, The process for obtaining the potential monitoring demand index for water conservancy project monitoring tasks is as follows: Select a water conservancy project monitoring task, obtain the total number of water conservancy project early warning links that appear within a time period q before the current system time, and mark them as DPC. Obtain the total number of water conservancy project monitoring links that can be generated by the water conservancy target topology network model corresponding to the water conservancy project monitoring task, and mark them as FMZ. Calculate the ratio between DPC and FMZ to obtain the potential monitoring demand index for the water conservancy project monitoring task.

3. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 1, characterized in that, The process for generating a priority task list for water conservancy projects is as follows: For all water conservancy monitoring target tasks, prioritize them from largest to smallest based on their potential monitoring demand index values ​​to generate a priority task list for water conservancy projects with a clear order of task execution.

4. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 1, characterized in that, The data perception record includes the water conservancy project entity ID, the type of perception data, the actual data value, and the time of data perception.

5. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 1, characterized in that, The process of forming multiple water conservancy project monitoring links is as follows: obtain the dynamic conductivity of each associated edge in the topology network model of the water conservancy target; when the dynamic conductivity is higher than the dynamic conductivity threshold, mark the corresponding associated edge as a water conservancy risk edge; and connect multiple water conservancy risk edges in sequence according to their directions to form multiple water conservancy project monitoring links.

6. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 1, characterized in that, The process for determining the topology network model of water conservancy targets corresponding to water conservancy monitoring tasks is as follows: determine the IoT sensing nodes associated with the water conservancy monitoring target task, take all associated IoT sensing nodes as network nodes, and take the influence relationships between network nodes as network edges to form the topology network model of water conservancy targets.

7. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 5, characterized in that, The process for obtaining the dynamic transmissibility of associated edges is as follows: Select an associated edge, clarify the type of associated edge, and mark the IoT sensing nodes at both ends of the associated edge as node a and node b, respectively. Determine the dynamic transmissibility of the associated edge based on the type of associated edge.

8. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 1, characterized in that, The method for determining the early warning link of a water conservancy project is as follows: Select a water conservancy project monitoring link, obtain the dynamic transmission deviation of each water conservancy risk edge in the monitoring link, calculate the absolute difference of the dynamic transmission deviation of two adjacent water conservancy risk edges in the monitoring link, calculate multiple dynamic transmission deviation distance values, sum and average the multiple dynamic transmission deviation distance values ​​to calculate the comprehensive transmission deviation, obtain the true data value of the source IoT sensing node of the water conservancy project monitoring link, when the true data value of the source IoT sensing node is outside the reasonable range of data and the comprehensive transmission deviation is lower than the comprehensive transmission deviation threshold, mark the water conservancy project monitoring link as a water conservancy project early warning link.

9. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 8, characterized in that, The process for obtaining the dynamic transmission deviation of a water conservancy risk edge is as follows: Select a water conservancy risk edge, obtain the dynamic transmission degree of the water conservancy risk edge and the corresponding dynamic transmission degree threshold, calculate the difference between the dynamic transmission degree and the dynamic transmission degree threshold, and obtain the dynamic transmission deviation of the water conservancy risk edge.

10. The intelligent monitoring and early warning system for water conservancy projects based on the Internet of Things and AI according to claim 1, characterized in that, Once a water conservancy project early warning link is identified, a list of priority tasks for early warning updates is generated. The process is as follows: Identify all water conservancy project early warning links included in the current water conservancy monitoring target task; identify the source IoT sensing node for each water conservancy project early warning link; mark the total number of source IoT sensing nodes as Rw; simultaneously mark all water conservancy monitoring target tasks after the current water conservancy monitoring target task in the priority task list as early warning adaptation and adjustment tasks; obtain the early warning update priority index for each early warning adaptation and adjustment task; sort all early warning adaptation and adjustment tasks according to their early warning update priority index values ​​from largest to smallest; and generate a list of priority tasks for early warning updates with a clear order of task execution.

Images