Water conservancy project operation state intelligent diagnosis system based on big data

Abstract

The invention relates to the technical field of hydraulic engineering monitoring, in particular to a hydraulic engineering operation state intelligent diagnosis system based on big data. A hydraulic engineering operation state intelligent diagnosis system based on big data comprises a multi-source data perception and fusion module which constructs a real-time data acquisition network covering basin hydraulic engineering total factors through a plurality of engineering detection devices and an environment perception terminal, acquires engineering environment data, and sends the engineering environment data to a cloud server; engineering environment data are processed through edge computing nodes, and then multi-source heterogeneous data are subjected to fusion governance by using a Hadoop / Spark big data platform; a diagnosis scoring module, a toughness evaluation module, an edge-cloud collaborative computing module and an intelligent decision early warning module. According to the method, a parallel special server is arranged through an innovative hardware architecture, multi-dimensional specialized synchronous calculation is realized, the processing bottleneck of a single server is effectively broken through, and a stable physical basis is constructed for cluster collaborative diagnosis.

Description

Technical Field

[0001] This invention relates to the field of water conservancy project monitoring technology, and in particular to an intelligent diagnostic system for the operational status of water conservancy projects based on big data. Background Technology

[0002] In the current field of water conservancy project monitoring and management, traditional methods mainly rely on periodic manual inspections and offline data analysis. This approach suffers from significant problems with untimely data acquisition. Due to the limitations of the inspection cycle, potential safety hazards such as reservoir leakage and structural deformation are often difficult to detect in a timely manner, increasing the risk of accidents. Simultaneously, offline data analysis is time-consuming, labor-intensive, and inefficient, and is prone to errors in data interpretation or omissions of important information due to human factors, further affecting the accuracy and timeliness of monitoring. Furthermore, traditional assessment methods focus on single indicators or localized areas, lacking a comprehensive consideration of the overall operational status and resilience of water conservancy projects, making it difficult to provide scientifically sound decision support.

[0003] Therefore, there is an urgent need to develop an intelligent diagnostic system for the operational status of water conservancy projects based on big data. This system should overcome the limitations of traditional methods through real-time data acquisition and efficient processing. It should be able to construct a real-time data acquisition network covering all elements of water conservancy projects, and utilize edge computing and big data platforms to achieve the fusion and governance of multi-source heterogeneous data. This would provide more accurate and timely monitoring and evaluation services, offering strong support for the scientific management and decision-making of water conservancy projects. Summary of the Invention

[0004] To overcome the shortcomings of traditional water conservancy project monitoring, such as untimely data acquisition, low processing efficiency, incomplete evaluation, and insufficient decision support, this invention provides an intelligent diagnostic system for the operational status of water conservancy projects based on big data.

[0005] The technical solution is as follows: A smart diagnostic system for the operational status of water conservancy projects based on big data, comprising: The multi-source data sensing and fusion module constructs a real-time data acquisition network covering all elements of water conservancy projects in the basin through various engineering detection equipment and environmental sensing terminals. It collects engineering environmental data, processes the engineering environmental data through edge computing nodes, and uses the Hadoop / Spark big data platform to integrate and manage multi-source heterogeneous data. The diagnostic scoring module constructs a dynamically updated watershed engineering knowledge graph using processed engineering environment data to obtain the hydraulic correlations and functional dependencies of reservoir entities. It then evaluates the clusters of related entities to obtain a first score. By coupling multiple modeling methods and dynamically calibrating the data, it obtains the response of the project under extreme loads and quantitatively evaluates the performance to obtain a second score. Finally, by comparing the impact of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different times, it measures the ecological compliance to obtain a third score. The resilience assessment module calculates the resilience assessment score of the water conservancy project based on the first score, the second score, and the third score, and generates the corresponding collaborative scheduling scheme and resilience weakness report. The edge-cloud collaborative computing module deploys edge computing nodes at the site of important water conservancy projects. It has local data preprocessing and preliminary diagnosis functions, detects anomalies and responds. The diagnostic scoring module and resilience assessment module server run in parallel through the central cloud platform. After bidirectional interaction with the edge nodes through a hybrid redundant communication network, a distributed real-time and centralized global collaborative architecture is formed. The intelligent decision-making and early warning module receives three-dimensional decision information from the central cloud platform, pushes early warning information to managers using AR-HUD and mobile terminal devices, and supports natural language interaction confirmation schemes, feeding back the actual effect to the big data platform to form a closed loop.

[0006] Preferably, the multi-source data sensing and fusion module constructs a real-time data acquisition network covering all elements of the watershed water conservancy project through various engineering detection equipment and environmental sensing terminals, collecting engineering environmental data, including: the engineering detection equipment and its functions include a GNSS displacement monitor providing shortest path parameters for the topological centrality index; a piezometer and stress-strain gauge obtaining the instability critical load value through piezometric data and strain density parameters; and a vibration sensor obtaining the standard load value through the mobile phone's dynamic amplification factor; the environmental sensing terminals and their functions include a water level gauge obtaining the water supply scale index based on recorded time-series data; a rain gauge obtaining the annual average inflow based on recorded data; satellite InSAR technology providing the number of links to other engineering entities for the topological centrality index; a meteorological station obtaining the river condition change index and standard water quality index through an environmental stress model; and a hydrological station obtaining standard biological integrity; the engineering environmental data includes data collected by the engineering detection equipment and data collected by the environmental sensing terminals.

[0007] Preferably, the process of processing engineering environment data through edge computing nodes and integrating multi-source heterogeneous data using the Hadoop / Spark big data platform includes: data cleaning, data fusion, feature extraction, and anomaly detection of the engineering environment data; the data cleaning involves eliminating sensor noise using sliding window filtering and Kalman filtering techniques; the data fusion involves synchronizing multi-sensor data through timestamp alignment and using spatial interpolation methods to complete missing data; the feature extraction involves calculating key engineering safety indicators such as displacement rate and seepage pressure change rate; the anomaly detection involves identifying potential risk events based on threshold models or lightweight machine learning algorithms; the Hadoop / Spark big data platform serves as the technical foundation supporting the integration and governance of multi-source heterogeneous data, where Hadoop provides distributed storage and batch processing computing capabilities, and Spark enables interactive analysis through in-memory computing and elastic distributed datasets.

[0008] Preferably, the diagnostic scoring module constructs a dynamically updated watershed engineering knowledge graph using processed engineering environment data to obtain the hydraulic relationships and functional dependencies of reservoir entities, and evaluates the associated entity clusters to obtain a first score. This includes: the watershed engineering knowledge graph is constructed using natural language processing technology to automatically extract entities and relationships from design documents and scheduling procedure texts, expressing the topological, hydraulic, and functional relationships between projects; the entities include reservoirs, dikes, sluice gates, and inter-basin water transfer projects; the hydraulic relationships are the mutual relationships between these water conservancy project entities (reservoirs, dikes, sluice gates, and inter-basin water transfer projects) based on water flow, with real-time monitoring data continuously updated; the functional dependencies describe the interdependencies of various water conservancy projects within the watershed in terms of flood control, irrigation, and water supply functions; associated entity clusters are classified according to hydraulic relationships and functional dependencies, and a first score is obtained through the evaluation of the associated entity clusters, wherein the scoring algorithm formula is: ; ; ; ; in The associated entity cluster is scored, i.e., the first score; This is the topological centrality index, reflecting the central position of the reservoir in the engineering network diagram; This represents the number of other engineering entities connected to this engineering entity. This represents the number of times the project entity exists on the shortest path between clusters; The functional impact index reflects the degree of damage caused by reservoir failure to its core functional objectives such as flood control, water supply, and ecology. This represents the grade and value of the downstream protected object; This represents the scale of the designed water supply and the importance of the water-supply recipients; This indicates the criticality of the downstream river's ecological baseflow and wetland survival. The scheduling flexibility index reflects the ability of the entire system to compensate for the functional impairment of the project entity by scheduling other projects. The ratio of reservoir capacity to average annual inflow. This represents the level of upstream support capabilities; This is the weighting adjustment coefficient, and , , .

[0009] Preferably, the step of obtaining the engineering response under extreme loads by coupling multiple modeling methods and dynamic data calibration, and quantitatively evaluating performance to obtain a second score, includes: the coupling of multiple modeling methods includes a data-mechanism fusion-based engineering performance degradation model that converts the output of a global climate model into a watershed extreme weather sequence using statistical downscaling, a finite element mechanism model, and an LSTM / Transformer data-driven model; the response under extreme loads is simulated by the data-mechanism fusion-based engineering performance degradation model to simulate various changes in the engineering under these extreme loads; the quantitative performance evaluation is obtained by obtaining a second score based on multiple indices of the reservoir entity under extreme loads, wherein the scoring algorithm formula is: ; ; ; in This is the response score under extreme loads, i.e., the second score; It is a performance stability index, reflecting the degree of deterioration of key performance parameters of the reservoir under extreme loads; The critical load value for instability is obtained through simulation using a digital twin model; The standard load value for design; This is the system redundancy index, reflecting the coverage of backup safety mechanisms when the main functions of the reservoir fail. The recovery potential index reflects the ease with which a reservoir can recover to its basic functional state after being damaged. This represents the estimated repair time, in months. This represents the percentage of estimated repair costs to new construction costs; This is the weighting adjustment coefficient, and , N is the preset value for the data.

[0010] Preferably, the step of measuring ecological compliance and obtaining a third score by comparing the impacts of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different times includes: comparing the impacts of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different times, wherein the impacts include changes in flow amplitude, flood frequency and rise rate indices, changes in water quality and the biological integrity of the surrounding area, and the impacts of reservoir operation on the longitudinal upstream and downstream and lateral connectivity of the river, and obtaining a third score to measure ecological compliance based on the impacts, wherein the scoring algorithm formula is: ; ; in The ecological impact assessment of the reservoir's surrounding area is the third assessment. The River Condition Change Index is a value calculated by comparing the changes in 33 hydrological indicators in five aspects—flow magnitude, occurrence time, frequency, delay, and rate of change—before and after reservoir construction. The water ecological health index reflects the ecological health status of the reservoir and downstream water bodies; Represents a standardized water quality index; Represents a standardized biological integrity index; The connectivity maintenance index measures the impact of a reservoir on the longitudinal upstream and downstream connectivity of a river and the transverse channel-floodplain connectivity. This is the weighting adjustment coefficient, and , N is the preset value for the data.

[0011] Preferably, the resilience assessment module calculates the resilience assessment score of the water conservancy project using the first score, the second score, and the third score, and generates a corresponding collaborative scheduling scheme and a resilience shortcoming report. This includes: the resilience assessment score of the water conservancy project is a comprehensive score of the water conservancy project entity from three dimensions, used to determine the operational status of the project and the degree of change in the project entity; the generation of the collaborative scheduling scheme involves simulating anomaly propagation based on the resilience assessment score in a digital twin environment, where a multi-agent reinforcement learning model finds the optimal collaborative scheduling scheme by continuously optimizing the actions of the agents according to a set global reward function; the resilience shortcoming report is the resilience shortcoming represented by the resilience assessment score, and provides optimization schemes for the relevant resilience shortcoming, wherein the resilience assessment algorithm formula is: ; in The resilience assessment score reflects the overall operational status of an engineering entity; Scoring related entity clusters reflects the degree of system coordination in the project; The response score under extreme loads reflects the safety level of the project. The ecological impact score of the reservoir area reflects the ecological sustainability of the project. This is the weighting adjustment coefficient, and .

[0012] Preferably, the edge-cloud collaborative computing module deploys edge computing nodes at the sites of important water conservancy projects, possessing local data preprocessing and preliminary diagnostic functions, detecting anomalies and responding accordingly. This includes: the important water conservancy project sites encompassing large reservoirs, important dikes, and key water conservancy project nodes—locations of water conservancy projects playing a crucial role within a river basin; the localized data preprocessing involves cleaning the sensor data of the relevant water conservancy project, removing noise and erroneous data, and simultaneously fusing multiple sensor data sets to integrate data of different types and formats into a unified format; the preliminary diagnostic function performs real-time anomaly diagnosis on the preprocessed data locally, quickly identifying any abnormalities occurring in the water conservancy project; and the anomaly detection and response involves reacting through the edge nodes after the preliminary diagnosis detects anomalies, taking timely measures to address emergencies.

[0013] Preferably, the diagnostic scoring module and resilience assessment module servers operate in parallel on the central cloud platform. After bidirectional interaction with edge nodes through a hybrid redundant communication network, a distributed real-time and centralized global collaborative architecture is formed. This includes: the diagnostic scoring module server is located on the central cloud platform and contains scores for reservoir-related entity clusters, response scores under extreme loads, and ecological impact scores around the reservoir; the resilience assessment module is also located on the central cloud platform and calculates the final resilience assessment score of the water conservancy project entity based on the three scores calculated by the diagnostic scoring module, and generates corresponding collaborative scheduling schemes and resilience weakness reports based on the resilience assessment scores; the redundant communication network is a hybrid redundant communication network composed of wired fiber optic networks and wireless 4G or 5G networks.

[0014] Preferably, the intelligent decision-making and early warning module receives three-dimensional decision information output from the central cloud platform, pushes early warning information to management personnel using AR-HUD and mobile terminal devices, and supports natural language interactive confirmation schemes, feeding back the actual effects to the big data platform to form a closed loop. It also includes: the three-dimensional decision information comprising a basin risk chain map, a set of collaborative response schemes, and a climate resilience assessment report; the basin risk chain map visually presents the risk transmission path and impact range caused by the current abnormal event, and identifies the most vulnerable link in the entire basin's water conservancy engineering system; the set of collaborative response schemes provides multi-project collaborative scheduling schemes based on specific circumstances when an abnormal event occurs within the basin; the climate resilience assessment report identifies water conservancy projects with resilience weaknesses and provides suggestions for improvement and optimization based on these weaknesses.

[0015] The beneficial effects of this invention are as follows: This invention utilizes an innovative hardware architecture to set up a parallel dedicated server, enabling multi-dimensional professional synchronous computing, effectively breaking through the processing bottleneck of a single server, and building a solid physical foundation for cluster collaborative diagnosis. Simultaneously, it integrates three types of sensing units to construct an integrated air-space-ground sensing network, supporting security-ecological collaborative monitoring and providing comprehensive data support for ecological load assessment. By clearly defining physical connections and data flow to form an efficient hardware architecture, it can process large-scale watershed data in parallel, quickly identify engineering vulnerabilities and risk transmission relationships, significantly improve watershed safety early warning capabilities and decision-making scientificity, and provide strong technical support for intelligent management of water conservancy projects. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the intelligent diagnostic system for the operation status of water conservancy projects based on big data, as described in this invention. Figure 2 This is a diagram illustrating the formulas for the intelligent diagnostic system for the operational status of water conservancy projects based on big data, as described in this invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0018] A smart diagnostic system for the operational status of water conservancy projects based on big data, such as Figure 1 and Figure 2 As shown, it includes: The multi-source data sensing and fusion module constructs a real-time data acquisition network covering all elements of water conservancy projects in the basin through various engineering detection equipment and environmental sensing terminals. It collects engineering environmental data, processes the engineering environmental data through edge computing nodes, and uses the Hadoop / Spark big data platform to integrate and manage multi-source heterogeneous data. The engineering testing equipment and its functions include a GNSS displacement monitor providing shortest path parameters for the topological centrality index; a piezometer and strain gauge obtaining the instability critical load value through piezometric data and strain density parameters; and a vibration sensor obtaining the standard load value through the mobile phone's dynamic amplification factor. The environmental sensing terminal and its functions include a water level gauge obtaining the water supply scale index based on recorded time-series data; a rain gauge obtaining the annual average inflow based on recorded data; satellite InSAR technology providing the number of links to other engineering entities for the topological centrality index; a meteorological station obtaining the river condition change index and standard water quality index through an environmental stress model; and a hydrological station obtaining standard biological integrity. The engineering environmental data includes data collected by the engineering testing equipment and data collected by the environmental sensing terminal.

[0019] The process involves data cleaning, data fusion, feature extraction, and anomaly detection for engineering environmental data. Data cleaning utilizes sliding window filtering and Kalman filtering techniques to eliminate sensor noise. Data fusion achieves multi-sensor data synchronization through timestamp alignment and uses spatial interpolation to fill in missing data. Feature extraction calculates key engineering safety indicators such as displacement rate and seepage pressure change rate. Anomaly detection identifies potential risk events based on threshold models or lightweight machine learning algorithms. The Hadoop / Spark big data platform serves as the technological foundation supporting the fusion and governance of multi-source heterogeneous data. Hadoop provides distributed storage and batch processing capabilities, while Spark enables interactive analysis through in-memory computing and elastic distributed datasets.

[0020] It should also be noted that the various engineering testing equipment includes GNSS displacement sensors for real-time tracking of spatial position changes of engineering structures, piezometers for detecting internal seepage pressure of earth-rock dam structures, stress-strain gauges for capturing the stress-strain state of concrete or rock masses, and vibration sensors for analyzing the dynamic response characteristics of engineering structures; the environmental sensing terminal includes water level gauges for continuously measuring changes in river or reservoir water levels, rain gauges for real-time recording of precipitation intensity and accumulation, satellite InSAR for acquiring wide-area surface deformation information, meteorological stations for collecting meteorological elements, hydrological stations for detecting hydrological parameters, multi-parameter water quality sensors for detecting water quality, underwater acoustic detectors, and ecological cameras.

[0021] There is a close quantitative correlation between the parameters of the detection instruments and the evaluation models, which together constitute the data foundation for the safety evaluation of water conservancy projects. In the field of engineering structure monitoring, GNSS displacement monitors continuously capture the three-dimensional deformation trajectory of structures and extract key displacement parameters from the topological centrality index through coordinate sequence analysis, thereby quantifying the spatial importance of the project in the watershed network. Embedded seepage pressure sensor arrays accurately measure the pore water pressure distribution inside the dam body. These pressure gradient data directly participate in the calculation of the performance stability index, providing criteria for timely detection of seepage anomalies. Stress-strain sensor networks record the micro-strain response of concrete structures and rock masses. Their strain energy density parameters directly affect the structural stability assessment and indirectly determine the value of the recovery potential index. Vibration acceleration sensors analyze the structural dynamic response spectrum to obtain the dynamic amplification factor under extreme loads such as wind, waves, and earthquakes. These characteristic parameters are directly related to the performance stability score under extreme conditions.

[0022] Environmental sensing systems also play a crucial monitoring role. Water level sensor networks continuously record fluctuations in river and reservoir water levels; this time-series data is transformed into flood control scheduling decisions through hydrological models, directly contributing to the quantification of flood control effectiveness parameters in the functional impact index. Precipitation monitoring stations record the spatiotemporal distribution characteristics of rainfall intensity, providing input parameters for flood forecasting models and directly influencing the comprehensive assessment of engineering flood control capabilities. Satellite InSAR technology acquires large-scale surface deformation fields through interferometry, revealing the correlation between regional tectonic activity and engineering deformation at a macroscopic scale, providing spatial deformation constraints for the topological centrality index. Atmospheric parameters collected by automatic weather stations not only serve meteorological forecasting operations but also indirectly influence the assessment results of engineering adaptability through environmental stress models. Hydrological element sequences acquired by hydrological monitoring stations provide baseline values ​​for calculating the river condition change index ecological indicators, supporting the quantitative evaluation of reservoir ecological impacts.

[0023] The diagnostic scoring module constructs a dynamically updated watershed engineering knowledge graph using processed engineering environment data to obtain the hydraulic correlations and functional dependencies of reservoir entities. It then evaluates the clusters of related entities to obtain a first score. By coupling multiple modeling methods and dynamically calibrating the data, it obtains the response of the project under extreme loads and quantitatively evaluates the performance to obtain a second score. Finally, by comparing the impact of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different times, it measures the ecological compliance to obtain a third score. The watershed engineering knowledge graph is constructed using natural language processing technology to automatically extract entities and relationships from design documents and scheduling procedures, expressing the topological, hydraulic, and functional relationships between projects. The entities include reservoirs, dikes, sluice gates, and inter-basin water transfer projects. The hydraulic relationships are the interrelationships between these water conservancy project entities based on water flow, with real-time monitoring data continuously updated. The functional dependencies describe the interdependencies between various water conservancy projects within the watershed in terms of flood control, irrigation, and water supply. Based on hydraulic relationships and functional dependencies, associated entity clusters are categorized, and a first score is obtained through evaluation of these clusters. The scoring algorithm formula is as follows: ; ; ; ; in The associated entity cluster is scored, i.e., the first score; This is the topological centrality index, reflecting the central position of the reservoir in the engineering network diagram; This represents the number of other engineering entities connected to this engineering entity. This represents the number of times the project entity exists on the shortest path between clusters; The functional impact index reflects the degree of damage caused by reservoir failure to its core functional objectives such as flood control, water supply, and ecology. This represents the grade and value of the downstream protected object; This represents the scale of the designed water supply and the importance of the water-supply recipients; This indicates the criticality of the downstream river's ecological baseflow and wetland survival. The scheduling flexibility index reflects the ability of the entire system to compensate for the functional impairment of the project entity by scheduling other projects. The ratio of reservoir capacity to average annual inflow. This represents the level of upstream support capabilities; This is the weighting adjustment coefficient, and , , .

[0024] The coupled modeling methods include a data-mechanism fusion engineering performance degradation model that converts global climate model output into watershed extreme weather sequences using statistical downscaling, a finite element mechanism model, and an LSTM / Transformer data-driven model; the response under extreme loads is simulated by the data-mechanism fusion engineering performance degradation model to model various changes in the engineering structure under these extreme loads; the quantitative performance evaluation is based on multiple indices of the reservoir entity under extreme loads to obtain a second score, where the scoring algorithm formula is: ; ; ; in This is the response score under extreme loads, i.e., the second score; It is a performance stability index, reflecting the degree of deterioration of key performance parameters of the reservoir under extreme loads; The critical load value for instability is obtained through simulation using a digital twin model; The standard load value for design; This is the system redundancy index, reflecting the coverage of backup safety mechanisms when the main functions of the reservoir fail. The recovery potential index reflects the ease with which a reservoir can recover to its basic functional state after being damaged. This represents the estimated repair time, in months. This represents the percentage of estimated repair costs to new construction costs; This is the weighting adjustment coefficient, and , N is the preset value for the data.

[0025] The comparison of the impacts of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different periods includes changes in flow amplitude, flood frequency and rise rate, water quality and local biological integrity, and the impact of reservoir operation on the longitudinal upstream and downstream and lateral connectivity of the river. A third score measuring ecological compliance is obtained based on these impacts, with the scoring algorithm formula as follows: ; ; in The ecological impact assessment of the reservoir's surrounding area is the third assessment. The River Condition Change Index is a value calculated by comparing the changes in 33 hydrological indicators in five aspects—flow magnitude, occurrence time, frequency, delay, and rate of change—before and after reservoir construction. The water ecological health index reflects the ecological health status of the reservoir and downstream water bodies; Represents a standardized water quality index; Represents a standardized biological integrity index; The connectivity maintenance index measures the impact of a reservoir on the longitudinal upstream and downstream connectivity of a river and the transverse channel-floodplain connectivity. This is the weighting adjustment coefficient, and , N is the preset value for the data.

[0026] It should also be noted that the construction of the watershed engineering knowledge graph utilizes deep learning-based natural language processing technology and a bidirectional attention mechanism to automatically identify water conservancy engineering entities and their multidimensional relationships from unstructured texts of engineering design specifications and scheduling operation procedures. This knowledge system covers basic entities such as reservoirs, dikes, sluice gates, and inter-basin water transfer projects, constructing a semantic network reflecting the hydraulic connections, topological structures, and functional dependencies between projects. Specifically, hydraulic relationships are quantified by constructing a simplified hydrodynamic model based on the Saint-Venant equations and dynamically calibrating parameters using real-time monitoring data; functional dependencies are established by analyzing the synergistic effects of projects in flood control, water supply, and ecology, building a logical framework for functional linkage. Based on these dynamic relationships, the system employs a community detection algorithm to identify clusters of associated entities with strong coupling characteristics, laying a structural foundation for subsequent cluster-level comprehensive evaluation.

[0027] In the evaluation dimension of associated entity clusters, the system integrates complex network theory and hydraulic system engineering methods to construct an evaluation system containing three core indicators. These indicators are calculated using specific formulas, and the final score is derived through integration. Topological centrality index. The eigenvector centrality algorithm not only considers the number of direct connections but also assesses the importance of connection engineering to identify key hubs in the network. It calculates the number of other engineering entities integrated into the connections of the engineering entities. The number of times the project entity exists on the shortest path between clusters. ,Right now ,in Functional Influence Index By constructing a loss function model that comprehensively considers the socio-economic value of the downstream protected area, in this embodiment, a value of 100 is assigned when there is a megacity downstream, 80 when there is an important town downstream, and 40 when there is general farmland downstream. This is denoted as... The population size within the water supply guarantee area is assigned a value of 100 when supplying water to the central city, 70 when supplying water to large irrigation areas, and 50 when supplying water to suburbs. The protection level of ecologically sensitive areas is assigned as follows: 100 for areas that play a decisive role in the ecological base flow of downstream rivers and wetland survival, 80 for areas that play an important role, and 50 for areas that play a general role. The chain reaction caused by the failure of quantitative engineering, namely ,in Scheduling flexibility index By establishing a functional substitution model based on graph neural networks, the feasibility of achieving functional compensation through cluster scheduling under specific engineering failure scenarios is analyzed. The calculation involves the ratio of associated reservoir capacity to average annual inflow. The value is determined by the level of upstream support capabilities: 100 for strong support, 60 for moderate support, and 20 for no support. ,Right now These three dimensions are expressed through formulas. ,in The data is then integrated to form a primary score that reflects the cluster's coordinated operational capabilities.

[0028] For extreme load response assessment, a multi-scale coupled numerical simulation framework was developed. This framework first transforms CMIP6 climate model data into watershed-specific extreme weather sequences using statistical downscaling techniques, then drives an engineering performance evaluation model based on a hybrid of physical mechanisms and data-driven approaches. Specifically, the finite element model accurately simulates the nonlinear response of concrete dams and earth-rock dam materials under extreme loads, while the LSTM-Transformer hybrid architecture uncovers hidden patterns in massive monitoring data, supplementing and correcting time-varying characteristics that are difficult to accurately describe by mechanistic models. Based on the simulation results of this coupled model, the system constructs a second scoring system from three levels: performance stability index. By establishing a time-varying reliability model that considers material degradation, the system failure probability under extreme loads is calculated, as shown in the formula: ,in The instability critical load value is obtained through simulation using a digital twin model. The standard load value is the design load value, and N is a preset data value, which is assigned a value of 100 in this embodiment. System redundancy index. The collaborative support capability of each backup system is assessed through fault tree analysis. For example, if there is an emergency spillway, add 25; if the main and auxiliary spillways are independent, add 25; if the dam body has a multi-layered seepage prevention system, add 25; if the foundation has a cutoff wall and curtain grouting, add 25. Recovery Potential Index Combining structural damage identification results with resource scheduling models, the time period and economic cost required to restore the system to a basic operating state are predicted, as shown in the formula. ,in This represents the estimated repair time, in months. This represents the percentage of estimated repair costs to new construction costs. These three dimensions are expressed through formulas. ,in Then, they are integrated to form a second score.

[0029] In terms of ecological impact assessment, a multi-dimensional evaluation framework covering hydrological conditions, water quality, biological integrity, and connectivity was established. By comparing ecological parameters before and after project construction and at different operational stages, the system quantifies the degree of impact of the project on the watershed ecosystem. River condition change index. By comparing 33 hydrological indicators across five aspects—flow magnitude, occurrence time, frequency, delay, and rate of change—before and after reservoir construction, the degree of anthropogenic alteration of the hydrological cycle is assessed. (Water Ecological Health Index) In this embodiment, a comprehensive water pollution index is adopted, which includes total phosphorus, ammonia nitrogen, permanganate index, and dissolved oxygen. A score ranging from 0 to 100 is calculated using the Nemerow index method, and denoted as [missing information]. The macrobenthic integrity index is calculated by collecting benthic animals or fish, analyzing their number of taxa, proportion of sensitive species, and proportion of pollution-tolerant species, comparing them with the raw data from the reference point, and calculating a score from 0 to 100. It reflects the structural and functional state of the aquatic ecosystem, that is... ,in Connectivity maintenance index By constructing a river corridor resistance model, the hindering effect of engineering operations on longitudinal nutrient transport and lateral undercurrent exchange is quantified. For example, the presence or absence of fishways is scored as 100 for effective and 0 for ineffective; whether the downstream water temperature is stratified is deducted; whether an ecological spillway is retained is added; and whether the dam completely blocks flood overflow is deducted. These three ecological dimensions are then reduced in dimensionality through principal component analysis and combined with expert weighting, using a formula... ,in N is a preset value of 100 for the data, forming the third score.

[0030] Through these three relatively independent yet mutually corroborating evaluation dimensions, a comprehensive assessment system covering engineering coordination, structural safety, and ecological compatibility was constructed. This system not only achieves a comprehensive diagnosis of the operational status of watershed engineering projects, but more importantly, through correlation analysis between quantitative indicators, it reveals the intrinsic connections between the assessment results of different dimensions, providing a scientific basis for precise control and coordinated optimization of watershed engineering projects. This multi-dimensional, multi-scale assessment framework effectively overcomes the limitations of traditional single-indicator assessments, enabling the system to adapt to the assessment needs of complex watershed engineering systems.

[0031] The resilience assessment module calculates the resilience assessment score of the water conservancy project based on the first score, the second score, and the third score, and generates the corresponding collaborative scheduling scheme and resilience weakness report. The resilience assessment score of the water conservancy project is a comprehensive evaluation of the project entity from three dimensions, used to determine the operational status of the project and the degree of change in the project entity. The generation of a collaborative scheduling scheme involves simulating anomaly propagation based on the resilience assessment score in a digital twin environment. A multi-agent reinforcement learning model, according to a set global reward function, continuously optimizes the agents' action choices to find the optimal collaborative scheduling scheme. The resilience bottleneck report identifies resilience bottlenecks based on the resilience assessment score and provides optimization schemes for these bottlenecks. The resilience assessment algorithm formula is: ; in The resilience assessment score reflects the overall operational status of an engineering entity; Scoring related entity clusters reflects the degree of system coordination in the project; The response score under extreme loads reflects the safety level of the project. The ecological impact score of the reservoir area reflects the ecological sustainability of the project. This is the weighting adjustment coefficient, and .

[0032] It should also be noted that the water conservancy project resilience assessment system adopts a multi-dimensional quantitative analysis method, constructing a comprehensive scoring model from three levels: project cluster correlation, extreme load response characteristics, and ecological and environmental impact. This assessment system can not only objectively reflect the degree of conformity between the current operating status of the project entity and the design standards, but also reveal the evolution trend of project performance through time-series comparisons.

[0033] Based on the resilience assessment results, the system initiates a multi-agent collaborative optimization process in a digital twin environment. By constructing a reward function aimed at maximizing the overall safety of the watershed, the reinforcement learning model iteratively optimizes the operational strategies of each engineering agent in the simulated environment. After numerous training cycles, the model can output a collaborative scheduling scheme that balances safety and economy, such as achieving peak shaving through joint scheduling of reservoir groups during flood events.

[0034] The resilience weakness analysis report identifies weak links in the engineering system by comparing the deviations of assessment indicators from safety thresholds. The report uses the analytic hierarchy process (AHP) to determine the weights of each dimension, highlighting key projects with poor cluster coordination, insufficient stability under extreme conditions, or significant negative ecological impacts. For the identified weaknesses, the report proposes actionable improvement suggestions, such as strengthening the structure to improve structural resistance, optimizing scheduling rules to enhance system coordination, and implementing ecological restoration to reduce environmental impact, providing a comprehensive solution for improving the resilience of the engineering system.

[0035] The edge-cloud collaborative computing module deploys edge computing nodes at the site of important water conservancy projects. It has local data preprocessing and preliminary diagnosis functions, detects anomalies and responds. The diagnostic scoring module and resilience assessment module server run in parallel through the central cloud platform. After bidirectional interaction with the edge nodes through a hybrid redundant communication network, a distributed real-time and centralized global collaborative architecture is formed. The important water conservancy project sites include locations of water conservancy projects that play a crucial role in the river basin, such as large reservoirs, important dikes, and key water conservancy project nodes; the localized data preprocessing involves cleaning the sensor data of the relevant water conservancy projects, removing noise and erroneous data, and simultaneously fusing data from multiple sensors to integrate data of different types and formats into a unified format; the preliminary diagnostic function performs real-time anomaly diagnosis on the preprocessed data locally to quickly identify abnormalities in the water conservancy projects; the anomaly detection and response function involves reacting through edge nodes after anomalies are detected in the preliminary diagnosis, and taking timely measures to deal with emergencies.

[0036] The diagnostic scoring module server is located on the central cloud platform. It includes the reservoir-related entity cluster score, the response score under extreme loads, and the ecological impact score around the reservoir. The resilience assessment module is also located on the central cloud platform. It calculates the final resilience assessment score of the water conservancy project entity based on the three scores calculated by the diagnostic scoring module, and generates corresponding collaborative scheduling schemes and resilience deficiency reports based on the resilience assessment scores. The redundant communication network is a hybrid redundant communication network consisting of a wired fiber optic network and a wireless 4G or 5G network.

[0037] It should also be noted that, within the watershed water conservancy engineering system, areas containing strategically important water conservancy facilities are defined as key monitoring nodes. These mainly include large reservoirs that undertake water resource regulation functions, important dikes that ensure flood control safety, and pumping stations and sluice gates that play a crucial scheduling role. These facilities play a vital role in flood storage and drought relief, regional water supply, and flood protection, and therefore become key deployment areas for the system's monitoring network.

[0038] At the data preprocessing level, the system employs a localized data processing strategy to perform preliminary processing on the raw monitoring data collected by various sensors. Since sensors are susceptible to environmental interference and generate noisy data during continuous operation, the raw signals must first be cleaned using filtering algorithms to remove outliers and repair missing data. For heterogeneous data (such as displacement, seepage pressure, and stress) generated by multiple sensors, the system uses a spatiotemporal alignment algorithm to fuse monitoring information with different sampling frequencies and data formats, forming a standardized time-series dataset, providing a high-quality data foundation for subsequent analysis.

[0039] At the edge computing layer, the system possesses real-time anomaly diagnosis capabilities. Based on preprocessed standardized data, embedded diagnostic algorithms continuously evaluate the engineering operation status. When an indicator is detected to exceed a dynamic threshold (such as a sudden change in displacement rate or a sharp increase in seepage flow), the edge node will immediately trigger a local alarm mechanism and initiate emergency control logic. For example, when excessive vibration of the pumping station is detected, the system automatically reduces its frequency and switches to backup equipment, achieving millisecond-level fault isolation.

[0040] In the cloud platform architecture, the diagnostic scoring module generates the basic score data for calculating the final score through a three-party scoring model. First, there is the reservoir-related entity cluster score, which is mainly based on the topological centrality index, functional influence index, and scheduling flexibility index. Then, there is the response score under extreme loads, which is mainly based on the performance stability index, system redundancy index, and recovery potential index. Finally, there is the ecological impact score around the reservoir, which is mainly based on the river situation change index, aquatic ecological health index, and connectivity maintenance index. These three aspects, from the perspectives of systemicity, engineering safety, and eco-friendliness, respectively, make the final score more comprehensive.

[0041] The resilience assessment module assigns weights to the obtained three-party scores to arrive at a final resilience assessment score. This score is not limited to describing the current operating status of the engineering entity; it is derived from a comprehensive analysis of all data. It also reflects whether the engineering entity is suitable for future use and whether it is suitable for continued use in the current cluster. Furthermore, it generates corresponding collaborative scheduling schemes and resilience bottleneck reports based on the resilience assessment score.

[0042] The system's communication network adopts a hybrid fiber optic and 5G network architecture, with a dual-link hot backup design to ensure data transmission reliability. When the primary fiber optic link fails, the 5G backup link automatically switches over within 50ms, ensuring uninterrupted transmission of monitoring data to the cloud platform. This redundant communication mechanism provides robust communication support for engineering monitoring under extreme weather conditions.

[0043] The intelligent decision-making and early warning module receives three-dimensional decision information from the central cloud platform, pushes early warning information to managers using AR-HUD and mobile terminal devices, and supports natural language interaction confirmation schemes, feeding back the actual effect to the big data platform to form a closed loop.

[0044] The three-dimensional decision-making information includes a watershed risk chain map, a set of collaborative response plans, and a climate resilience assessment report. The watershed risk chain map presents the risk transmission path and impact range caused by the current abnormal event in a visual form, and identifies the most vulnerable link in the entire watershed water conservancy system. The set of collaborative response plans provides a multi-project collaborative scheduling scheme based on the specific circumstances when an abnormal event occurs in the watershed. The climate resilience assessment report identifies water conservancy projects with weak resilience and provides suggestions for upgrading and optimization based on these projects.

[0045] It should also be noted that the aforementioned three-dimensional decision-making information system provides comprehensive support for integrated watershed decision-making, specifically comprising three major modules: risk chain mapping, collaborative response plans, and climate resilience assessment. The risk chain mapping dynamically displays the propagation path and impact range of abnormal events within the watershed through visualization, accurately locating weak links in the water conservancy system and providing spatial decision-making basis for risk prevention and control. The collaborative response plan, based on a multi-objective optimization algorithm, generates optimal scheduling strategies according to the anomaly type and engineering status, achieving coordinated operation of watershed projects and optimal resource allocation. The climate resilience assessment employs a forward-looking analysis method to identify projects within the watershed with insufficient capacity to cope with extreme weather and proposes a scientific priority sequence for their modification.

[0046] Although the present invention has been described in detail with reference to the above embodiments, it will be apparent to those skilled in the art that various changes or modifications can be made to the invention without departing from the principles and spirit of the invention as defined by the claims. Therefore, the detailed description of the embodiments in this disclosure is for illustrative purposes only and is not intended to limit the invention; rather, the scope of protection is defined by the content of the claims.

Claims

1. A smart diagnostic system for the operational status of water conservancy projects based on big data, characterized in that: include: The multi-source data sensing and fusion module constructs a real-time data acquisition network covering all elements of water conservancy projects in the basin through various engineering detection equipment and environmental sensing terminals. It collects engineering environmental data, processes the engineering environmental data through edge computing nodes, and uses the Hadoop / Spark big data platform to integrate and manage multi-source heterogeneous data. The diagnostic scoring module constructs a dynamically updated watershed engineering knowledge graph using processed engineering environment data, obtains the hydraulic relationships and functional dependencies of reservoir entities, and evaluates the cluster of related entities to obtain the first score. By coupling multiple modeling methods and dynamically calibrating the data, the response of the project under extreme loads is obtained, and the performance is quantitatively evaluated to obtain a second score; by comparing the impact of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different times, the ecological compliance is measured to obtain a third score. The resilience assessment module calculates the resilience assessment score of the water conservancy project based on the first score, the second score, and the third score, and generates the corresponding collaborative scheduling scheme and resilience weakness report. The edge-cloud collaborative computing module deploys edge computing nodes at the site of important water conservancy projects. It has local data preprocessing and preliminary diagnosis functions, detects anomalies and responds. The diagnostic scoring module and resilience assessment module server run in parallel through the central cloud platform. After bidirectional interaction with the edge nodes through a hybrid redundant communication network, a distributed real-time and centralized global collaborative architecture is formed. The intelligent decision-making and early warning module receives three-dimensional decision information from the central cloud platform, pushes early warning information to managers using AR-HUD and mobile terminal devices, and supports natural language interaction confirmation schemes, feeding back the actual effect to the big data platform to form a closed loop.

2. The intelligent diagnostic system for the operational status of water conservancy projects based on big data as described in claim 1, characterized in that, The multi-source data sensing and fusion module constructs a real-time data acquisition network covering all elements of the watershed water conservancy project through various engineering detection equipment and environmental sensing terminals. It collects engineering environmental data, including: the engineering detection equipment and its functions, such as a GNSS displacement monitor providing shortest path parameters for the topological centrality index; piezometers and strain gauges obtaining instability critical load values ​​through piezometric data and strain density parameters; and vibration sensors obtaining standard load values ​​through the mobile phone's dynamic amplification factor. The environmental sensing terminals and their functions include water level gauges obtaining water supply scale indices based on recorded time-series data; rain gauges obtaining average annual inflow based on recorded data; satellite InSAR technology providing the number of links to other engineering entities for the topological centrality index; meteorological stations obtaining river condition change indices and standard water quality indices through environmental stress models; and hydrological stations obtaining standard biological integrity. The engineering environmental data includes data collected by the engineering detection equipment and data collected by the environmental sensing terminals.

3. The intelligent diagnostic system for the operational status of water conservancy projects based on big data as described in claim 1, characterized in that, The process of processing engineering environment data through edge computing nodes and leveraging the Hadoop / Spark big data platform to integrate and govern multi-source heterogeneous data includes: data cleaning, data fusion, feature extraction, and anomaly detection of the engineering environment data; data cleaning involves using sliding window filtering and Kalman filtering techniques to eliminate sensor noise; data fusion involves synchronizing multi-sensor data through timestamp alignment and using spatial interpolation methods to complete missing data; feature extraction involves calculating key engineering safety indicators such as displacement rate and seepage pressure change rate; anomaly detection involves identifying potential risk events based on threshold models or lightweight machine learning algorithms; the Hadoop / Spark big data platform serves as the technological foundation supporting the integration and governance of multi-source heterogeneous data, where Hadoop provides distributed storage and batch processing computing capabilities, and Spark enables interactive analysis through in-memory computing and elastic distributed datasets.

4. The intelligent diagnostic system for the operation status of water conservancy projects based on big data as described in claim 2, characterized in that, The diagnostic scoring module constructs a dynamically updated watershed engineering knowledge graph using processed engineering environment data to obtain the hydraulic relationships and functional dependencies of reservoir entities. It then evaluates the associated entity clusters to obtain a first score. This includes: the watershed engineering knowledge graph is constructed using natural language processing technology to automatically extract entities and relationships from design documents and scheduling procedures, expressing the topological, hydraulic, and functional relationships between projects; the entities include reservoirs, dikes, sluice gates, and inter-basin water transfer projects; the hydraulic relationships are the mutual relationships between these water conservancy project entities based on water flow, with real-time monitoring data continuously updated; the functional dependencies describe the interdependencies of various water conservancy projects within the watershed in terms of flood control, irrigation, and water supply functions. Based on the hydraulic relationships and functional dependencies, associated entity clusters are classified, and a first score is obtained through the evaluation of these clusters. The scoring algorithm formula is as follows: ； ； ； ； in The associated entity cluster is scored, i.e., the first score; This is the topological centrality index, reflecting the central position of the reservoir in the engineering network diagram; This represents the number of other engineering entities connected to this engineering entity. This represents the number of times the project entity exists on the shortest path between clusters; The functional impact index reflects the degree of damage caused by reservoir failure to its core functional objectives such as flood control, water supply, and ecology. This represents the grade and value of the downstream protected object; This represents the scale of the designed water supply and the importance of the water-supply recipients; This indicates the criticality of the downstream river's ecological baseflow and wetland survival. The scheduling flexibility index reflects the ability of the entire system to compensate for the functional impairment of the project entity by scheduling other projects. The ratio of reservoir capacity to average annual inflow. This represents the level of upstream support capabilities; This is the weighting adjustment coefficient, and , , .

5. The intelligent diagnostic system for the operational status of water conservancy projects based on big data as described in claim 2, characterized in that, The process of obtaining the engineering response under extreme loads by coupling multiple modeling methods and dynamic data calibration, and quantifying performance evaluation to obtain a second score, includes: the coupling of multiple modeling methods includes a data-mechanism fusion-based engineering performance degradation model that converts global climate model output into watershed extreme weather sequences using statistical downscaling, a finite element mechanism model, and an LSTM / Transformer data-driven model; the response under extreme loads is simulated by the data-mechanism fusion-based engineering performance degradation model to simulate various changes in the engineering under these extreme loads; the quantified performance evaluation is based on multiple indices of the reservoir entity under extreme loads to obtain a second score, wherein the scoring algorithm formula is: ； ； ； in This is the response score under extreme loads, i.e., the second score; It is a performance stability index, reflecting the degree of deterioration of key performance parameters of the reservoir under extreme loads; The critical load value for instability is obtained through simulation using a digital twin model; The standard load value for design; This is the system redundancy index, reflecting the coverage of backup safety mechanisms when the main functions of the reservoir fail. The recovery potential index reflects the ease with which a reservoir can recover to its basic functional state after being damaged. This represents the estimated repair time, in months. This represents the percentage of estimated repair costs to new construction costs; This is the weighting adjustment coefficient, and , N is the preset value for the data.

6. The intelligent diagnostic system for the operational status of water conservancy projects based on big data as described in claim 2, characterized in that, The method of measuring ecological compliance by comparing the impacts of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different times to obtain a third score includes: comparing the impacts of reservoir operation on the surrounding aquatic and terrestrial ecosystems at different times, the impacts of flow rate fluctuations, flood frequency and rise rate indices, changes in water quality and local biological integrity, and the impacts of reservoir operation on the longitudinal upstream and downstream and lateral connectivity of the river. A third score for measuring ecological compliance is obtained based on these impacts, and the scoring algorithm formula is as follows: ； ； in The ecological impact assessment of the reservoir's surrounding area is the third assessment. The River Condition Change Index is a value calculated by comparing the changes in 33 hydrological indicators in five aspects—flow magnitude, occurrence time, frequency, delay, and rate of change—before and after reservoir construction. The water ecological health index reflects the ecological health status of the reservoir and downstream water bodies; Represents a standardized water quality index; Represents a standardized biological integrity index; The connectivity maintenance index measures the impact of a reservoir on the longitudinal upstream and downstream connectivity of a river and the transverse channel-floodplain connectivity. This is the weighting adjustment coefficient, and , N is the preset value for the data.

7. The intelligent diagnostic system for the operational status of water conservancy projects based on big data as described in claim 3, characterized in that, The resilience assessment module calculates the resilience assessment score of the water conservancy project based on the first score, second score, and third score, and generates a corresponding collaborative scheduling scheme and a resilience shortcoming report. This includes: the resilience assessment score of the water conservancy project is a comprehensive score of the water conservancy project entity from three dimensions, used to determine the operational status of the project and the degree of change in the project entity; the generation of the collaborative scheduling scheme involves simulating anomaly propagation based on the resilience assessment score in a digital twin environment, and a multi-agent reinforcement learning model finds the optimal collaborative scheduling scheme by continuously optimizing the actions of the agents according to a set global reward function; the resilience shortcoming report is the resilience shortcoming represented by the resilience assessment score, and provides optimization schemes for the relevant resilience shortcoming, wherein the resilience assessment algorithm formula is: ； in The resilience assessment score reflects the overall operational status of an engineering entity; Scoring related entity clusters reflects the degree of system coordination in the project; The response score under extreme loads reflects the safety level of the project. The ecological impact score of the reservoir area reflects the ecological sustainability of the project. This is the weighting adjustment coefficient, and .

8. The intelligent diagnostic system for the operation status of water conservancy projects based on big data according to claim 4, characterized in that, The edge-cloud collaborative computing module deploys edge computing nodes at key water conservancy project sites, possessing local data preprocessing and preliminary diagnostic functions, detecting anomalies and responding accordingly. These key water conservancy project sites include large reservoirs, important dikes, and critical water conservancy project nodes—locations of water conservancy projects playing a crucial role within a river basin. The localized data preprocessing involves cleaning the sensor data from the relevant water conservancy project, removing noise and errors, and simultaneously fusing data from multiple sensors, integrating different types and formats into a unified format. The preliminary diagnostic function performs real-time anomaly diagnosis on the preprocessed data locally, quickly identifying any abnormalities in the water conservancy project. The anomaly detection and response function involves reacting through edge nodes after the preliminary diagnosis detects anomalies, taking timely measures to address emergencies.

9. The intelligent diagnostic system for the operation status of water conservancy projects based on big data according to claim 4, characterized in that, The diagnostic scoring module and resilience assessment module servers operate in parallel on the central cloud platform. After bidirectional interaction with edge nodes through a hybrid redundant communication network, a distributed real-time and centralized global collaborative architecture is formed. This architecture includes: the diagnostic scoring module server, located on the central cloud platform, contains scores for reservoir-related entity clusters, response scores under extreme loads, and ecological impact scores around the reservoir; the resilience assessment module, also located on the central cloud platform, calculates the final resilience assessment score of the water conservancy project entity based on the three scores calculated by the diagnostic scoring module, and generates corresponding collaborative scheduling schemes and resilience weakness reports based on the resilience assessment scores; the redundant communication network is a hybrid redundant communication network consisting of wired fiber optic networks and wireless 4G or 5G networks.

10. The intelligent diagnostic system for the operational status of water conservancy projects based on big data as described in claim 9, characterized in that, The intelligent decision-making and early warning module receives three-dimensional decision information from the central cloud platform, pushes early warning information to managers using AR-HUD and mobile terminal devices, and supports natural language interactive confirmation schemes. It feeds back the actual effects to the big data platform to form a closed loop. The module also includes: the three-dimensional decision information comprising a watershed risk chain map, a set of collaborative response schemes, and a climate resilience assessment report; the watershed risk chain map visually presents the risk transmission path and impact range caused by current abnormal events, and identifies the most vulnerable links in the entire watershed water conservancy system; the set of collaborative response schemes provides multi-project collaborative scheduling plans based on specific circumstances when abnormal events occur within the watershed; and the climate resilience assessment report identifies water conservancy projects with weak resilience and provides suggestions for improvement and optimization based on these projects.

Images