Estuary coast embankment risk early warning evaluation system and method under composite disaster-inducing factors

Abstract

The invention discloses an estuary coast embankment risk early warning evaluation system and method under composite disaster-inducing factors, and relates to the technical field of hydraulic engineering safety monitoring and risk evaluation. Comprising a multi-source data acquisition module, a data preprocessing module, a dynamic index system construction module, an adaptive weight determination module, a risk assessment module, a refined early warning grade division module and a visual decision support module. Real-time sensing, quantitative evaluation and graded early warning of the dike safety state are achieved, the method can be directly applied to coastal city disaster prevention and reduction, dike project full-life-cycle operation and maintenance, major storm surge disaster emergency response and other scenes, the technical bottlenecks of single disaster-inducing factor, static weight and insufficient early warning precision in a traditional evaluation method are effectively broken through, and the method has a wide application prospect. And a precise technical support is provided for the safety protection of the estuary coast embankment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water conservancy safety monitoring and risk assessment, in particular to a river mouth and coastal embankment risk early warning assessment system and method under composite disaster-causing factors. BACKGROUND

[0002] The river mouth and coastal area is a special geographical unit of land and sea interaction, which is not only the core area of global economic and social development, but also the high-incidence area of natural disasters. According to statistics, more than 70% of storm surge disasters in the world are concentrated in the river mouth and coastal area, and the direct economic loss caused by embankment collapse is more than 10 billion yuan per year. Embankment engineering has become a key barrier to resist marine dynamic and hydrological disasters, and its safety assessment technology has been the research focus in the field of water conservancy for a long time.

[0003] The existing embankment risk assessment technology has the following significant limitations, which is difficult to meet the safety protection needs under the composite disaster situation: 1. Single disaster-causing factor: Traditional methods focus on single water level or wave action, completely ignoring the coupling effect of wave-flow-tide-wind speed, resulting in that the evaluation results cannot reflect the complex action mechanism of actual disasters; 2. Static evaluation model: The existing system generally uses a fixed weight model, which cannot dynamically adjust the importance of indicators according to the disaster situation. For example, the weight of wave force during a storm surge should be significantly higher than that in normal working conditions, but the traditional analytic hierarchy process (AHP) cannot realize real-time weight correction; 3. Insufficient early warning accuracy: The grading standards of key indicators such as over-warning water level amplitude and over-warning duration are too simplified in conventional evaluation methods, and there is a lack of quantitative correlation with embankment structure response; 4. Weak data fusion capability: It fails to effectively integrate real-time data from sensor networks, historical hazard records, remote sensing images and other multi-source heterogeneous data, resulting in that the evaluation results are one-sided and cannot fully reflect the safety state of the embankment.

[0004] Therefore, it is of important practical significance to develop an evaluation system that can comprehensively consider the dynamic coupling effect of composite disaster-causing factors, fuse multi-source data, and realize accurate early warning, in order to improve the disaster prevention and mitigation capability of river mouth and coastal embankments. SUMMARY

[0005] The embodiments of the present application provide a river mouth and coastal embankment risk early warning assessment system and method under composite disaster-causing factors, which realizes real-time perception, quantitative evaluation and graded early warning of the safety state of the embankment through multi-source heterogeneous data fusion technology and adaptive weight optimization algorithm. It can be directly applied to disaster prevention and mitigation in coastal cities, embankment engineering full life cycle operation and maintenance, and emergency response to major storm surge disasters, effectively breaking through the technical bottlenecks of "single disaster-causing factor, static weight and insufficient early warning accuracy" in traditional evaluation methods, and providing accurate technical support for the safety protection of river mouth and coastal embankments.

[0006] The estuary coast embankment risk early warning evaluation system under the composite disaster factor comprises: A multi-source data acquisition module adapted to acquire multi-source heterogeneous data related to embankment risk; A data preprocessing module adapted to preprocess the multi-source heterogeneous data to generate a standardized index data set; A dynamic index system construction module adapted to establish a hierarchical index system covering disaster factors, structural characteristics and environmental impact based on the preprocessed data; An adaptive weight determination module adapted to dynamically adjust the weights of each index in combination with historical disaster data and real-time context; A risk assessment module adapted to obtain an embankment risk comprehensive index through coupling analysis and fuzzy comprehensive calculation based on the indexes and weights; A refined early warning level division module adapted to divide the risk into multiple levels and specify threshold values and disposal measures according to the risk comprehensive index and embankment response characteristics; A visual decision support module adapted to visually display and push the data, evaluation results and early warning information to the decision-making end.

[0007] Further, the multi-source heterogeneous data includes real-time monitoring data, historical data and remote sensing data, wherein the real-time monitoring data is adapted to reflect the current stress and environmental action of the embankment; the historical data is adapted to reflect the design, danger and maintenance process of the embankment; and the remote sensing data is adapted to reflect the structural integrity of the embankment appearance.

[0008] Further, the multi-source data acquisition module comprises a real-time monitoring submodule, a historical data submodule and a remote sensing data submodule, which are respectively used to collect real-time monitoring data of sensors deployed along the embankment, call database information to obtain historical data, and extract image features to obtain remote sensing data.

[0009] Further, the data preprocessing process of the data preprocessing module on the multi-source heterogeneous data includes outlier rejection, missing value filling, and quantitative index normalization and qualitative index quantification, which realizes the cleaning and unification of multi-source monitoring data, and finally forms a standardized index data set that can be directly used for risk assessment, wherein the data preprocessing module adopts an improved criterion combined with a sliding window algorithm to identify outliers, and uses a time series prediction model to fill in missing values.

[0010] Further, the data preprocessing module adopts a normalization method based on disaster threshold for quantitative indexes, and adopts a fuzzy comprehensive evaluation method for quantification of qualitative indexes.

[0011] Further, the dynamic index system construction module is used for establishing a dynamic evaluation index system based on a target layer-criterion layer-index layer, and integrates disaster-causing factors, dike structures and environmental impact indexes to support dike safety risk assessment and early warning.

[0012] Further, the criterion layer includes a disaster-causing factor subsystem, a dike structure subsystem and an environmental impact subsystem, wherein the disaster-causing factor subsystem is used for representing the direct disaster pressure of external natural action such as wave, water flow, tide level and wind speed on the safety of the dike, the dike structure subsystem is used for reflecting the influence of internal structure characteristics such as dike top elevation, slope stability, permeability and apparent damage on the disaster resistance of the dike, and the environmental impact subsystem is used for comprehensively considering the adjusting effect of meteorological, hydrological and surrounding environmental changes on the overall risk level of the dike.

[0013] Further, the self-adaptive weight determination module calculates objective weights by improving an entropy weight method and dynamically corrects them in combination with a disaster situation intensity index.

[0014] Further, the fine early warning grade division module divides the risk grade into five levels, and sets corresponding monitoring, patrol and rescue measures for different grades.

[0015] The method for early warning and evaluation of the risk of the estuary coast dike under the composite disaster-causing factor comprises the following steps: Step 1, system initialization: setting parameters such as safety threshold, danger threshold, coupling coefficient and aging coefficient, and establishing a database connection; Step 2, data acquisition: the multi-source data acquisition module acquires real-time data every 10 minutes, and synchronously updates historical data and remote sensing data; Step 3, data preprocessing: the data preprocessing module performs outlier processing and standardization on the newly acquired data to generate a standardized index data set; Step 4, index quantization and weight calculation: the dynamic index system construction module outputs 17 index quantization values, and the self-adaptive weight determination module calculates real-time dynamic weights; Step 5, risk assessment: the risk assessment module sequentially calculates subsystem indexes and a safety comprehensive index; Step 6, early warning grade determination: the fine early warning grade division module determines the early warning grade according to the safety comprehensive index; Step 7, result display and pushing: the visual decision support module updates the display content, and if the early warning grade is orange or red, automatically sends an alarm information to the relevant person in charge; Step 8, cyclic monitoring: returning to step 2 to realize continuous dynamic evaluation.

[0016] The above technical solutions provided by the embodiments of the present application have at least the following beneficial effects: 1. Full coverage of composite disaster-causing factors: For the first time, wave-current-tide-typhoon-seepage multi-factor coupling is incorporated into the evaluation model, considering the structure characteristics of the dike and environmental impact, and the evaluation results are more in line with the actual disaster scenario.

[0017] 2. Dynamic weight self-adaptation: The subjective and objective weights are fused by group analytic hierarchy process and improved entropy weight method, and real-time adjustment is realized by combining disaster situation identification, solving the limitations of traditional static weight.

[0018] 3. High warning accuracy: A 5-level refined warning level is constructed, and the dike aging coefficient and coupling coefficient are introduced, so that the matching degree of the warning result and the actual safety state of the dike is improved.

[0019] 4. Deep fusion of multi-source data: Real-time sensor data, historical hazard data and remote sensing data are integrated to eliminate the influence of data heterogeneity, and the reliability of the evaluation results is significantly improved.

[0020] 5. Strong decision support capability: Through the visualization platform and intelligent warning push, an integrated solution of "monitoring-evaluation-disposal" is provided for management personnel, and the emergency response time is shortened.

[0021] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the written description, claims, and accompanying drawings.

[0022] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0023] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings: Figure 1 The structure schematic diagram of the river estuary coastal dike risk warning evaluation system under the composite disaster-causing factor disclosed by the embodiment of the present application; Figure 2 The process schematic diagram of the river estuary coastal dike risk warning evaluation system under the composite disaster-causing factor disclosed by the embodiment of the present application; Figure 3 The dynamic change diagram of each index of the disaster-causing factor subsystem disclosed by the embodiment of the present application; Figure 4 The time sequence change and warning level division schematic diagram of the dike comprehensive risk index disclosed by the embodiment of the present application; Figure 5 The dike comprehensive risk index distribution schematic diagram disclosed by the embodiment of the present application.

[0024] Figure label: 1. Multi-source data acquisition module; 11. Real-time monitoring sub-module; 12. Historical data sub-module; 13. Remote sensing data sub-module; 2. Data preprocessing module; 3. Dynamic indicator system construction module; 31. Target layer; 32. Criterion layer; 33. Indicator layer; 4. Adaptive weight determination module; 5. Risk assessment module; 6. Refined early warning level classification module; 7. Visual decision support module. Detailed Implementation

[0025] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0026] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a risk early warning and assessment system and method for estuary and coastal dikes under multiple disaster-causing factors, focusing on solving the following four core technical problems: 1. How to construct a coupled evaluation model that includes multiple factors such as waves, water flow, wind speed, and tides, and comprehensively incorporate key parameters such as the stability of the dike structure, seepage characteristics, and maintenance history; 2. How to achieve dynamic adaptive adjustment of the weights of evaluation indicators and adapt to different disaster scenarios through proactive scenario identification; 3. How to establish a refined early warning classification standard to improve the matching degree between early warning results and the actual safety status of dikes; 4. How to efficiently integrate multi-source monitoring data and eliminate the impact of data heterogeneity on the reliability of assessment results.

[0027] To achieve the above objectives, the present invention adopts the following technical solution: The risk early warning and assessment system for estuary and coastal dikes under multiple disaster-causing factors proposed in this invention, such as... Figure 1 As shown, the system includes a multi-source data acquisition module 1, a data preprocessing module 2, a dynamic indicator system construction module 3, an adaptive weight determination module 4, a risk assessment module 5, a refined early warning level classification module 6, and a visualization decision support module 7. The functions of each module are as follows: 1. Multi-source data acquisition module 1, suitable for acquiring multi-source heterogeneous data related to dike risks. The multi-source heterogeneous data includes: real-time monitoring data, historical data and remote sensing data. Among them, real-time monitoring data is suitable for reflecting the current stress and environmental effects on the dike; historical data is suitable for reflecting the dike design, danger and maintenance process; and remote sensing data is suitable for reflecting the integrity of the dike's appearance and structure.

[0028] This module achieves full-dimensional data coverage through a three-dimensional acquisition system of "real-time monitoring + historical data + remote sensing data". The multi-source data acquisition module 1 includes a real-time monitoring submodule 11, a historical data submodule 12, and a remote sensing data submodule 13, which are respectively used to collect real-time monitoring data from sensors deployed along the dike, retrieve historical data from the database, and extract remote sensing data from image features. Real-time monitoring submodule 11: consists of a high-precision sensor array deployed along the dike, with specific parameters as follows: (1) Wave sensor (sampling frequency 50Hz): collects effective wave height, wave period, and wave direction; (2) Hydrological sensors: synchronously measure flow velocity (accuracy ±0.01m / s) and tide level (accuracy ±0.005m); (3) Meteorological sensors: monitor wind speed (accuracy ±0.1m / s) and wind direction; (4) Structural monitoring sensors: piezometers and displacement gauges to collect stress and strain data inside the dike.

[0029] Historical Data Submodule 12: Retrieves three types of core data through the database interface: (1) Design parameters of the dike (top elevation, slope ratio, material properties, etc.); (2) Historical records of dangerous situations (data on the combination of disaster-causing factors and the breach process of dike breaches since 1950). (3) Hazard removal and reinforcement inspection files (engineering measures, construction parameters and effect evaluation reports of each maintenance).

[0030] Remote sensing data submodule 13: Receives UAV inspection images (resolution 0.1m) and satellite remote sensing data, and automatically extracts information on apparent defects such as cracks in the top of the dike, damage to the slope, and collapse of the dike body through image recognition algorithms.

[0031] 2. Data preprocessing module 2 is suitable for establishing a hierarchical indicator system covering disaster-causing factors, structural characteristics, and environmental impacts based on preprocessed data. The preprocessing process for multi-source heterogeneous data in data preprocessing module 2 includes: outlier removal, missing value imputation, and normalization of quantitative indicators and quantification of qualitative indicators, achieving the cleaning and unification of multi-source monitoring data, and ultimately forming a standardized indicator dataset that can be directly used for risk assessment. Specifically, data preprocessing module 2 employs an improved... The criteria combine a sliding window algorithm to identify outliers and use a time series prediction model to fill in missing values.

[0032] Data preprocessing module 2 uses a normalization method based on disaster thresholds for quantitative indicators and a fuzzy comprehensive evaluation method for qualitative indicators.

[0033] To address the heterogeneity and noise issues of multi-source data, a two-step processing scheme is adopted: Outlier handling: Adopting improved The criteria (combined with the sliding window algorithm) identify abnormal data caused by sensor malfunctions and environmental interference; for missing values, an LSTM-based time series prediction model is used to fill in the gaps.

[0034] Data standardization: Quantitative indicators: An improved normalization method is adopted, and a disaster threshold is introduced as a benchmark value. The formula is as follows:

[0035] in, As the indicator safety threshold, As the indicator of the danger threshold, These are the original index values. This is the standardized value.

[0036] Qualitative indicators, such as the apparent condition of the levee, are converted into quantitative values ​​of 0-100 points using the fuzzy comprehensive evaluation method. The specific classifications are as follows: intact (100 points, no warning), slightly damaged (80 points, no warning), moderately damaged (60 points, depending on the disaster-causing factor data to determine whether a low-risk warning is triggered), severely damaged (30 points, triggering a medium-risk warning), and on the verge of collapse (0 points, triggering an extremely high-risk warning).

[0037] 3. Dynamic indicator system construction module 3 is suitable for establishing a hierarchical indicator system covering disaster-causing factors, structural characteristics and environmental impacts based on preprocessed data. Dynamic indicator system construction module 3 is used to establish a dynamic evaluation indicator system based on target layer 31-criteria layer 32-indicator layer 33, and to uniformly incorporate disaster-causing factors, dike structure and environmental impact indicators to support dike safety risk assessment and early warning.

[0038] Module 3 of the dynamic indicator system construction module constructs a three-layer evaluation indicator system consisting of "target layer 31 - criterion layer 32 - indicator layer 33", which includes 17 core indicators. The time series performance of each indicator under typical working conditions is as follows: Figure 3 As shown: Target Layer 31: Comprehensive Safety Index of Dikes (value range 0-1 points, the higher the score, the higher the risk level). Criterion Layer 32: Disaster-causing factor subsystem, dike structure subsystem, environmental impact subsystem; Indicator Layer 33: Disaster-causing factor subsystem (10 items): significant wave height, wave period, current velocity, tide level, wind speed, wave run-up, overtopping volume, magnitude of exceeding warning level, duration of exceeding warning level, storm surge increase; details are as follows: Valid wave height (early warning correlation: when the predicted value is ≥2.5m, the risk value contributed by this indicator increases, which drives down the composite index and may trigger an early warning). Wave period (early warning correlation: coupling effect with significant wave height, increasing disaster risk and affecting early warning level); Flow velocity (early warning correlation: when the predicted value is ≥2.0m / s, the score of this indicator decreases, exacerbating the overall risk and potentially triggering an early warning). Tide level (early warning correlation: when the predicted value is ≥3.5m (safe threshold), this indicator begins to affect the early warning judgment; when it is ≥4.8m (danger threshold), it significantly promotes the upgrading of the early warning level). Wind speed (early warning correlation: when the predicted value is ≥10.8m / s, relevant early warning monitoring will be activated; when it is ≥24.5m / s, the risk of disaster is greatly increased and a high-level early warning is likely to be triggered). Wave rise (early warning correlation: when the predicted value is ≥ 90% of the seawall crest elevation, it indicates a risk of wave overtopping and may trigger an early warning). Overwater surge (early warning correlation: predicted value ≥ 0.5m) 3 (At / s・m, a wave-related warning is triggered). The extent of exceeding the warning level (warning correlation: when the predicted value is ≥0.5m, the warning level is upgraded by one level); Duration of exceeding the warning level (early warning correlation: when the predicted duration of exceeding the warning level is ≥24 hours, the risk of damage to the dikes accumulates, prompting the upgrade of the warning level). Storm surge (warning correlation: when the predicted surge value is ≥1.0m, determine whether a high-risk warning is triggered based on the tide level data); Dike Structure Subsystem (7 items): Dike crest elevation (early warning association: when the predicted value of tide level + storm surge is close to or exceeds the dike crest elevation, a wave breach warning is triggered), slope ratio (early warning association: when the slope ratio does not meet the design requirements and encounters predicted high flow velocity and high waves, a slope collapse warning is triggered), permeability coefficient (early warning association: when the predicted value of permeability coefficient is ≥ 1.5 times the design value, a seepage risk is indicated, and an early warning may be triggered), dike body crack length (early warning association: when the predicted crack length is ≥ 5m, a structural risk warning is triggered in combination with disaster-causing factor data), slope integrity (early warning association: when the predicted slope damage area is ≥ 10%, a slope protection warning is triggered), reinforcement times (early warning association: when the number of reinforcement times is low and the predicted value of disaster-causing factors is high, the warning level is upgraded), dike patrol time (early warning association: when the patrol interval exceeds 144 hours and the disaster-causing factor is predicted to be risky, a patrol warning is triggered).

[0039] 4. Adaptive weight determination module 4 is suitable for dynamically adjusting the weights of each indicator by combining historical disaster data and real-time scenarios. The adaptive weight determination module 4 calculates objective weights by improving the entropy weight method and dynamically corrects them by combining the disaster scenario intensity index.

[0040] By integrating subjective and objective weighting methods with a dynamic adjustment mechanism, real-time optimization of weights can be achieved. Basic weight calculation: Subjective weights: Group hierarchical analysis (G-AHP) was used to invite 10 experts in the field of water conservancy (including 5 professor-level senior engineers and 5 front-line engineering experts) to construct a judgment matrix and determine the initial weights; Objective weighting: An improved entropy weighting method is adopted, and historical risk data is introduced as a penalty factor to assign higher entropy weights to key indicators that have historically led to levee breaches (such as the magnitude of water level exceeding the warning level and the amount of waves that have surged). For example, the objective weight of the magnitude of water level exceeding the warning level is 0.20, and the objective weight of the amount of waves that have surged is 0.15. This enhances the correlation between the weights and the actual disasters, and makes these key early warning indicators account for a higher proportion in the comprehensive assessment.

[0041] Dynamic adjustment mechanism: Based on the disaster scenario identification results, the basic weights are adjusted in real time through a fuzzy inference system. The adjustment formula is as follows:

[0042] in, for Time of the first The dynamic weight of each indicator Based on the weights, This is the weighting adjustment factor (values ​​range from 0.2 to 0.5). for Moment-based situation intensity index (value 0~1, for extreme scenarios such as storm surge and typhoon) Approaching 1).

[0043] 5. Risk assessment module 5 is suitable for obtaining the comprehensive risk index of dikes through coupled analysis and fuzzy comprehensive calculation based on indicators and weights.

[0044] A two-step combined model of "subsystem evaluation - comprehensive evaluation" is adopted: Subsystem Assessment: The disaster-causing factor subsystem adopts a weighted summation method with coupling coefficient correction, considering the nonlinear interaction effect of waves and water flow, as shown in the following formula:

[0045] in, As a disaster-causing factor subsystem index, For the first Item and the The coupling coefficient of the indicators (e.g., the coupling coefficient between waves and water level is taken as 0.3).

[0046] Where S is the index of the dike structure subsystem; Comprehensive assessment: The comprehensive safety index of the dike is calculated based on the variable fuzzy set theory, and the dike aging coefficient is introduced. (Values ​​range from 0.8 to 1, depending on the age of the dike and the degree of material aging), the formula is as follows:

[0047] SI stands for Risk Composite Index.

[0048] 6. The refined early warning level classification module 6 is suitable for classifying risks into multiple levels and specifying thresholds and response measures based on the comprehensive risk index and levee response characteristics. The refined early warning level classification module 6 divides the risk level into five levels and sets corresponding monitoring, patrol and emergency response measures for different levels.

[0049] Based on the structural response characteristics of the dikes, the early warning levels are divided into five levels, with clearly defined thresholds and corresponding response requirements, such as... Figure 4 and 5 As shown: Safety (below 0.6 points, no warning): All predicted values ​​of indicators are within the safe range (e.g., effective wave height <1.5m, flow velocity <1.0m / s, tide level <2.0m, wind speed <8.0m / s), the dike shows no deformation or abnormal seepage, and no special treatment is required. Low risk (0.6-0.7 points, yellow-green warning): The predicted values ​​of individual disaster-causing factors are close to the warning value (such as wind speed prediction of 10.8-17.1m / s (6-8 level wind), tide level prediction of 2.5-3.3m), the dike structure is intact, and the monitoring frequency needs to be increased (such as shortening the real-time monitoring data collection interval from 10 minutes to 5 minutes). Medium risk (0.7-0.8 points, yellow alert): Multiple indicators are predicted to exceed warning levels (e.g., water level exceeding warning level by 0.3-0.5m, wave volume by 0.3-0.5m). 3 / s・m、wind speed 17.2-24.4m / s (8-9 level wind),the embankment has slight deformation, and the patrol plan needs to be activated (if patrol personnel are increased, patrol once every 2 hours). High risk (0.8-0.9 points, orange alert): Key indicators are predicted to reach dangerous thresholds (such as seepage exceeding the design allowable value by 1.2 times, embankment crack length of 5-10m, tide level of 4.0-4.5m, wind speed of 24.5-32.6m / s (10-11 level wind)), obvious cracks or abnormal seepage appear in the embankment, and it is necessary to organize personnel evacuation and prepare emergency materials (such as 10,000 sandbags and 5,000㎡ of geotextile). Extremely high risk (0.9-1 points, red alert): The dike is on the verge of collapse, and large-scale piping or slope collapse is predicted (e.g., piping diameter predicted ≥0.5m, slope displacement predicted ≥10mm / d), and the predicted values ​​of disaster-causing factors are extremely high (e.g., effective wave height ≥3.5m, tide level ≥5.0m, wind speed ≥32.7m / s (level 12 wind)). Emergency rescue plans must be activated immediately (e.g., mobilizing rescue teams and activating backup flood control facilities).

[0050] 7. Visualized Decision Support Module 7 is suitable for intuitively displaying and pushing data, evaluation results, and early warning information to the decision-making end.

[0051] Develop a visualization platform to achieve full-process visualization of "data-assessment-early warning": Real-time data monitoring: Dynamically displays the monitoring values ​​and change curves of each sensor (including predicted data curves), and supports clicking on sensors to view detailed parameters (e.g., clicking on the tide sensor can view the real-time tide level, the predicted tide level for the next 24 hours, and the corresponding warning level). Assessment results are presented using a risk map, with red areas indicating high-risk sections (corresponding to warning levels of orange and red) and blue areas indicating low-risk sections (corresponding to warning levels of safe and yellow-green). The key disaster-causing factors prediction data for each risk section are also marked (e.g., the red area is marked with "tide level prediction 4.2m, wind speed prediction 28m / s"). Early warning information push: Tiered early warning information is pushed via SMS and APP, along with specific handling suggestions (e.g., orange warning push: "The tide level is expected to reach 4.4m and the wind speed is 29m / s in the next 12 hours. It is recommended to immediately organize the evacuation of personnel and allocate sandbags, geotextiles and other emergency supplies to the XX section of the road"). Historical trend analysis: Provides the risk index change curve and forecast trend for the past 30 days (e.g., predicting that the risk index will rise from 0.65 to 0.82 in the next 3 days, and the warning level will rise from low risk to high risk), and supports the export of assessment reports (including warning data, judgment basis and handling suggestions).

[0052] like Figure 2 As shown, the risk early warning assessment method for estuary and coastal dikes under combined disaster-causing factors includes the following steps: Step 1, System Initialization: Set parameters such as safety threshold, danger threshold, coupling coefficient, and aging coefficient for each indicator, and establish a database connection; Step 2, Data Acquisition: Multi-source data acquisition module 1 acquires real-time data every 10 minutes, and synchronously updates historical data and remote sensing data; Step 3, Data Preprocessing: Data preprocessing module 2 performs outlier handling and standardization on the newly collected data to generate a standardized indicator dataset; Step 4, Indicator Quantification and Weight Calculation: The dynamic indicator system construction module 3 outputs quantified values ​​of 17 indicators, and the adaptive weight determination module 4 calculates the real-time dynamic weights. Step 5, Risk Assessment: Risk assessment module 5 calculates the subsystem index and the comprehensive safety index in sequence; Step 6, Warning Level Determination: The refined warning level classification module 6 determines the warning level based on the comprehensive safety index; Step 7, Results Display and Push: The Visual Decision Support Module 7 updates the displayed content. If the warning level is orange or red, an alarm message is automatically sent to the relevant person in charge. Step 8, cyclic monitoring: Return to step 2 to achieve continuous dynamic evaluation.

[0053] Example (a) Implementation Environment The system described in this invention is deployed on a river estuary dike in southeastern coastal my country (15 km long, built in 1995, and reinforced in 2010). Sensor deployment: One set of wave sensor, hydrological sensor and meteorological sensor is deployed every 1km along the dike, and two additional sets of structural monitoring sensors are deployed at key sections of the dike (such as bends and structures penetrating the dike). Data storage: Real-time monitoring data is stored in a database (retained for 10 years), and remote sensing images and historical hazard data are stored in the cloud; Computing platform: The evaluation algorithm is run on a cloud computing server (8 cores and 16GB of memory), with a response time of ≤1 second.

[0054] (II) Implementation Steps Data Acquisition and Preprocessing: Data acquisition content: Wave sensor acquires significant wave height (range 0.0~6.0m) and average wave period (range 0.5~20s); Hydrological sensor acquires water flow velocity (range 0.0~3.5m / s) and tide level (range -10.0~10.0m). Outlier handling: Adopting improved The criteria identify and remove abnormal data caused by wind speed sensor malfunctions (such as instantaneous wind speeds exceeding the reasonable range, such as 70 m / s); for missing seepage pressure data, an LSTM neural network model is used to complete the data. Data standardization: Water level data is standardized by using 2.0m as the safety threshold and 4.5m as the danger threshold; "2cm crack on the top of the dike" identified by drone inspection is defined as moderate damage.

[0055] Weighting and Risk Assessment: Basic weights: The subjective weight of "above-warning level" is determined by G-AHP with a value of 0.15, and its objective weight is determined by the improved entropy weight method with a value of 0.20. The basic weight is (0.15×0.4+0.20×0.6)=0.18. Dynamic Adjustment: Situation Intensity Index during Typhoons =0.9, weight adjustment coefficient =0.3, dynamic weight ; Comprehensive Index Calculation: Disaster-Causing Factor Subsystem Index Structural subsystem index aging coefficient Comprehensive safety index point.

[0056] Warning Level Determination and Response: Warning level: 0.68 points indicates "low risk"; Response measures: The system pushes early warning information through the APP, prompting managers to strengthen dike patrols and focus on monitoring seepage in sections where water levels exceed warning levels.

[0057] (III) Implementation Results During the year of operation of this system, it successfully issued early warnings for three moderate-risk events caused by typhoons, with an accuracy rate of 92%. Compared with traditional assessment methods, the emergency response time was shortened by 40%, effectively preventing the escalation of dike risks.

[0058] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to the specific order or hierarchy described.

[0059] In the detailed description above, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features in a single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, with each claim representing a separate preferred embodiment of the invention.

[0060] Those skilled in the art will also understand that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments herein can be implemented as electronic hardware, computer software, or a combination thereof. To clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the described functionality in alternative ways for each specific application; however, such implementation decisions should not be construed as departing from the scope of this disclosure.

[0061] The steps of the methods or algorithms described in conjunction with the embodiments herein can be directly embodied in hardware, software modules executed by a processor, or a combination thereof. The software modules can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in a user terminal. Alternatively, the processor and storage medium can exist as discrete components in the user terminal.

[0062] For software implementation, the techniques described in this application can be implemented using modules (e.g., procedures, functions, etc.) that perform the functions described in this application. This software code can be stored in memory units and executed by a processor. The memory units can be implemented within the processor or outside the processor; in the latter case, they are communicatively coupled to the processor via various means, as is well known in the art.

[0063] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that the various embodiments can be further combined and arranged. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

Claims

1. A risk early warning and assessment system for estuaries and coastal dikes under multiple disaster-causing factors, characterized in that: include: The multi-source data acquisition module is suitable for acquiring multi-source heterogeneous data related to dike risks. The data preprocessing module is suitable for preprocessing multi-source heterogeneous data to generate standardized indicator datasets; The dynamic indicator system construction module is suitable for establishing a hierarchical indicator system covering disaster-causing factors, structural characteristics, and environmental impacts based on preprocessed data. The adaptive weight determination module is suitable for dynamically adjusting the weights of each indicator by combining historical disaster data and real-time scenarios. The risk assessment module is suitable for obtaining a comprehensive risk index for dikes through coupled analysis and fuzzy comprehensive calculation based on indicators and weights. The refined early warning level classification module is suitable for classifying risks into multiple levels and specifying thresholds and response measures based on the comprehensive risk index and levee response characteristics; The visualization decision support module is suitable for intuitively displaying and pushing data, evaluation results, and early warning information to the decision-making end.

2. The system as described in claim 1, characterized in that, Multi-source heterogeneous data includes: real-time monitoring data, historical data, and remote sensing data. Real-time monitoring data is suitable for reflecting the current stress and environmental effects on the dike; historical data is suitable for reflecting the design, danger, and maintenance process of the dike; and remote sensing data is suitable for reflecting the structural integrity of the dike.

3. The system as described in claim 2, characterized in that, The multi-source data acquisition module includes a real-time monitoring submodule, a historical data submodule, and a remote sensing data submodule, which are used to collect real-time monitoring data from sensors deployed along the dike, retrieve historical data from database information, and extract remote sensing data from image features, respectively.

4. The system as described in claim 1, characterized in that, The data preprocessing module performs preprocessing on multi-source heterogeneous data, including outlier removal, missing value imputation, and normalization of quantitative indicators and quantification of qualitative indicators. This process cleanses and unifies multi-source monitoring data, ultimately forming a standardized indicator dataset that can be directly used for risk assessment. The data preprocessing module employs an improved... The criteria combine a sliding window algorithm to identify outliers and use a time series prediction model to fill in missing values.

5. The system as described in claim 4, characterized in that, The data preprocessing module uses a normalization method based on disaster thresholds for quantitative indicators and a fuzzy comprehensive evaluation method for qualitative indicators.

6. The system as described in claim 1, characterized in that, The dynamic indicator system construction module is used to establish a dynamic evaluation indicator system based on the target layer, criterion layer, and indicator layer, which integrates disaster-causing factors, dike structure, and environmental impact indicators to support dike safety risk assessment and early warning.

7. The system as described in claim 6, characterized in that, The criteria layer includes a disaster-causing factor subsystem, a dike structure subsystem, and an environmental impact subsystem. The disaster-causing factor subsystem is used to characterize the direct disaster pressure on dike safety from external natural forces such as waves, water flow, tide level, and wind speed. The dike structure subsystem is used to reflect the impact of internal structural characteristics such as dike crest elevation, slope stability, permeability, and apparent damage on the dike's disaster resistance capacity. The environmental impact subsystem is used to comprehensively consider the moderating effect of meteorological, hydrological, and surrounding environmental changes on the overall risk level of the dike.

8. The system as described in claim 1, characterized in that, The adaptive weight determination module calculates objective weights by improving the entropy weight method and dynamically corrects them by combining the disaster situation intensity index.

9. The system as described in claim 1, characterized in that, The refined early warning level classification module divides the risk level into five levels and sets corresponding monitoring, inspection and emergency response measures for different levels.

10. A method for risk early warning and assessment of estuary and coastal dikes under multiple disaster-causing factors, using the system described in any one of claims 1 to 9, characterized in that, Includes the following steps: Step 1, System Initialization: Set parameters such as safety threshold, danger threshold, coupling coefficient, and aging coefficient for each indicator, and establish a database connection; Step 2, Data Acquisition: The multi-source data acquisition module collects real-time data every 10 minutes and updates historical data and remote sensing data synchronously. Step 3, Data Preprocessing: The data preprocessing module performs outlier handling and standardization on the newly collected data to generate a standardized indicator dataset; Step 4, Indicator Quantification and Weight Calculation: The dynamic indicator system construction module outputs quantified values ​​for 17 indicators, and the adaptive weight determination module calculates real-time dynamic weights. Step 5, Risk Assessment: The risk assessment module calculates the subsystem index and the comprehensive safety index sequentially; Step 6, Warning Level Determination: The refined warning level classification module determines the warning level based on the comprehensive safety index; Step 7, Results Display and Push: The visualization decision support module updates the displayed content. If the warning level is orange or red, an alarm message is automatically sent to the relevant person in charge. Step 8, cyclic monitoring: Return to step 2 to achieve continuous dynamic evaluation.

Images