River dynamic risk early warning method and system based on multi-source data

Abstract

The application provides a river dynamic risk early warning method and system based on multi-source data, and relates to the technical field of big data. Firstly, multi-source river monitoring data is engineering deconstructed to generate a river engineering data deconstruction set containing various data engineering representations and associated attributes. A river dynamic risk topology modeling model is established based on the river engineering data deconstruction set to simulate risk transmission. Historical risk data is calibrated by calling a dynamic adaptation process to generate a topology calibration parameter set. Real-time monitoring data is input to perform real-time simulation deduction of risk transmission. Finally, the deduction result is threshold adapted based on the calibration parameter set, and early warning information containing a risk impact area and transmission path coordinates is generated when the threshold is exceeded. The application can accurately and timely warn of river dynamic risks.

Description

Technical Field

[0001] This application relates to the field of big data technology, and more specifically, to a method and system for early warning of dynamic risks in rivers based on multi-source data. Background Technology

[0002] In the field of river management and disaster prevention and mitigation, accurately and timely grasping the dynamic risk status of rivers is crucial for ensuring the safety of people's lives and property along the riverbanks, rationally planning resource utilization, and maintaining ecological balance. Traditional river risk early warning methods mainly rely on data from a single source, such as hydrological data such as water level and flow monitored by hydrological stations, or risk assessment based solely on the physical structural characteristics of the river.

[0003] However, river systems are complex and dynamic, and their risk status is influenced by multiple factors. A single data source cannot fully reflect the true state and potential risks of a river. For example, focusing solely on hydrological data may overlook the impact of changes in the river's physical structure (such as riverbed erosion and channel shifts) on risk; while considering only physical structure data cannot promptly capture the risks brought about by abrupt changes in hydrological conditions (such as floods caused by torrential rains). Furthermore, existing methods are insufficient in simulating risk transmission, failing to accurately describe the transmission process of risk between different regions and elements, thus limiting the accuracy and timeliness of early warning results and failing to meet the needs of modern river management for refined and dynamic risk early warning. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method and system for early warning of river dynamic risks based on multi-source data.

[0005] In conjunction with the first aspect of this application, a method for river dynamic risk early warning based on multi-source data is provided, applied to a river dynamic risk early warning system based on multi-source data, the method comprising: The engineering deconstruction processing of multi-source river monitoring data is performed to generate a set of river engineering data deconstruction. This set of river engineering data deconstruction includes the engineering representation forms and data association attributes of river physical structure data, hydrological movement data, and environmental action data. A dynamic risk topology model for rivers is established based on the deconstruction set of river engineering data. This dynamic risk topology model for rivers completes the engineering simulation of risk transmission through node association, path mapping and parameter coupling. The dynamic adaptation process of the river dynamic risk topology model is invoked to perform topology calibration on historical risk data, generating a topology calibration parameter set, which is directly used to adjust the accuracy of risk transfer simulation. Real-time river engineering monitoring data is input into the river dynamic risk topology model, and real-time simulation of risk transmission is performed to obtain the simulation results of risk transmission. Based on the topology calibration parameter set, threshold adaptation processing is performed on the simulation results of risk transmission. When the simulation results exceed the adaptation threshold range, dynamic risk warning information of the river, including the engineering area affected by the risk and the engineering coordinates of the transmission path, is generated.

[0006] In conjunction with the second aspect of this application, a river dynamic risk early warning system based on multi-source data is provided. The river dynamic risk early warning system based on multi-source data includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the river dynamic risk early warning system based on multi-source data implements the aforementioned river dynamic risk early warning method based on multi-source data.

[0007] In conjunction with a third aspect of this application, a computer-readable storage medium is provided, wherein computer-executable instructions are stored therein, and when the computer-executable instructions are executed, the aforementioned method for river dynamic risk early warning based on multi-source data is implemented.

[0008] Combining any of the above aspects, by performing engineering deconstruction processing on multi-source river monitoring data, an engineering data deconstruction set of rivers is generated, encompassing data on river physical structure, hydrological movement, environmental effects, and other aspects. The data association attributes are clarified, and a dynamic risk topology modeling model of the river is established based on this deconstruction set. Through node association, path mapping, and parameter coupling, an engineering simulation of risk transmission is achieved, which can more realistically reflect the risk propagation mechanism in the river system. A dynamic adaptation process is invoked to calibrate the topology structure of historical risk data, generating a topology calibration parameter set, which can effectively adjust the accuracy of risk transmission simulation and improve the reliability and accuracy of the model. Real-time monitoring data is input into the model for real-time simulation and extrapolation of risk transmission, enabling timely acquisition of dynamic risk changes. Finally, based on the topology calibration parameter set, threshold adaptation processing is performed on the extrapolation results. When the adaptation threshold is exceeded, a dynamic risk warning message containing detailed information about the river is generated, achieving accurate and timely early warning of dynamic river risks. Attached Figure Description

[0009] Figure 1 A flowchart illustrating the river dynamic risk early warning method based on multi-source data provided in this application embodiment. Detailed Implementation

[0010] Figure 1 This paper illustrates a flowchart of a river dynamic risk early warning method based on multi-source data provided in an embodiment of this application, including: Step S110: Perform engineering deconstruction processing on multi-source river monitoring data to generate a river engineering data deconstruction set. This river engineering data deconstruction set includes the engineering representation forms and data association attributes of river physical structure data, hydrological movement data, and environmental action data.

[0011] This embodiment uses a river system in a watershed as an application scenario. This watershed contains different types of river sections, including mountain river sections, plain river sections, and urban river sections. When performing engineering deconstruction processing of multi-source river monitoring data, the specific sources of the multi-source data are first identified. This includes data collected by fixed monitoring stations located at different locations along the river, such as monitoring stations installed in the upper reaches of mountain river sections, the middle reaches of plain river sections, and the lower reaches of urban river sections; data acquired by mobile monitoring equipment, such as data collected by monitoring vessels that regularly patrol the entire watershed; and overall watershed data obtained through remote sensing technology. These data types are diverse, including real-time hydrological data and relevant data about the river's surrounding environment. The engineering deconstruction processing involves integrating and transforming this data from different sources and of different types, transforming it into data with an engineering representation, and identifying the correlations between them. For example, the water flow velocity data collected by monitoring stations in mountain river sections is correlated with the topographic data of the area obtained through remote sensing to analyze the influence of topography on water flow velocity.

[0012] Step S111: Collect multi-source raw river monitoring data output from riverbank monitoring equipment, mobile monitoring platform, and remote sensing monitoring system to generate a multi-source raw river data set. This multi-source raw river data set contains multi-dimensional monitoring records of the river itself and its surrounding environment.

[0013] In the aforementioned watershed scenario, various types of riverbank monitoring equipment are used. In mountainous river sections, water level monitors, flow velocity monitors, and water quality monitors are installed along the banks. These devices continuously collect data at set time intervals: water level monitors collect water level data hourly, flow velocity monitors collect water flow velocity data every ten minutes, and water quality monitors collect water quality data daily. The output data is in a standard monitoring data format, including monitoring time, monitoring location identifier, and specific monitoring values. The mobile monitoring platform consists of a monitoring vessel equipped with multiple sensors, which conducts a monthly patrol of the entire watershed, collecting riverbed topography data, riverbank soil characteristics data, etc., and storing the data in a specific format on the vessel's onboard storage device. The remote sensing monitoring system periodically photographs the watershed via satellite to obtain data such as river area and riverside vegetation cover, in remote sensing image format. When collecting this data, a dedicated data transmission interface is used to transmit data from each monitoring device and system to a data center, adding a unified timestamp and location information to ensure that each data point clearly corresponds to its time and location. The above data is then integrated to generate a multi-source river raw data set, which is stored according to data type, such as water level data, flow velocity data, water quality data, topographic data, vegetation data, etc., for subsequent processing.

[0014] Step S112: Extract relevant data describing the inherent structure of the river from the original dataset of multi-source rivers, and split the data dimensions according to the engineering structure analysis standard to obtain the split structure-related data. The split structure-related data includes cross-sectional dimension description, riverbed material description, and riverbank structure description.

[0015] Extracting data describing the inherent structure of rivers from a multi-source river dataset requires first identifying the data types related to that structure. For example, data related to river cross-sections can be extracted from riverbed topographic data collected by a mobile monitoring platform, information on riverbed material can be extracted from geological survey data, and descriptive data about riverbank structure can be extracted from riverbank engineering construction records. Following engineering structure analysis standards, the extracted data is then dimensionally split. The cross-sectional dimension includes parameters such as the width and depth of the cross-section at different locations along the river. For instance, in a mountainous river section, multiple typical cross-sections are selected, and the width and depth data for each cross-section are extracted. The riverbed material dimension includes information such as the type and particle size distribution of the riverbed soil, such as determining that a section of the riverbed is mainly composed of sand and gravel, and their respective proportions. The riverbank structure dimension involves the slope of the riverbank and the type of slope protection material, such as a section of the riverbank having a specific slope and using masonry as the slope protection material. Through this splitting process, the inherent structure data of the river is decomposed into three distinct dimensions, preparing for subsequent engineering parameter conversion.

[0016] Step S113: Perform engineering parameter transformation on the split structure-related data to convert the natural language description and raw measurement data into a quantifiable representation that can be calculated in engineering design, thereby obtaining river physical structure data.

[0017] For the decomposed structural data, some natural language descriptions, such as "the riverbed soil particles are relatively coarse" and "the riverbank slope is relatively steep," need to be converted into quantitative parameters. Simultaneously, the original measurement data may have inconsistencies in units or formats, which also require processing. For example, "the riverbed soil particles are relatively coarse" needs to be converted into a quantitative parameter of the average diameter of soil particles. By consulting relevant engineering standards, the range of particle diameters corresponding to "relatively coarse" can be determined, and then converted into specific numerical intervals. Similarly, "the riverbank slope is relatively steep" needs to be converted into specific slope values. Based on engineering methods of representing slope, the textual description can be converted into the ratio of vertical height to horizontal distance. For original measurement data, such as cross-sectional width measurements, if some are in meters and others in centimeters, they need to be uniformly converted to meters, retaining a certain number of decimal places. After these conversions, all structural data are transformed into a quantitative representation that can be directly used for calculations in engineering design, forming river physical structure data, which includes multiple quantitative parameters, each with a clear physical meaning and engineering unit.

[0018] Step S1131: Collect the quantitative standards for commonly used river physical parameters in engineering structural design. These quantitative standards for river physical parameters include the quantitative methods and representation forms for cross-sectional dimensions, riverbed materials, and riverbank structures.

[0019] This study collects quantitative standards for commonly used river physical parameters in engineering structural design. These standards are derived from relevant engineering design specifications and industry standards. The quantitative methods for cross-sectional dimensions specify the measurement methods and data representation formats for cross-sectional width and depth. For example, cross-sectional width refers to the horizontal distance from the river surface at a certain cross-section; measurements are taken at multiple locations and the average value is calculated, with the data expressed in meters and rounded to one decimal place. Cross-sectional depth refers to the vertical distance from the riverbed bottom to the water surface, also expressed in meters and rounded to two decimal places. The quantitative methods for riverbed material involve the classification of soil types and particle sizes. Riverbed soils are divided into different types, such as sand, clay, and gravel, each with a specific code. The study also specifies the measurement methods and representation formats for particle size, such as particle diameter expressed in millimeters, with the proportion of particles in different size ranges obtained through sieve analysis. The quantitative methods for riverbank structures include methods for calculating slope and performance parameters of slope protection materials. Slope is expressed as the ratio of vertical height to horizontal distance, accurate to two decimal places. Performance parameters of slope protection materials, such as compressive strength and shear strength, are characterized in specific units and numerical ranges.

[0020] Step S1132: Extract the natural language description content from the split structure-related data, perform semantic parsing on each natural language description, and determine the physical parameter type and feature information corresponding to each natural language description.

[0021] Natural language descriptions are extracted from the decomposed structural data, such as "the riverbed in this section is mainly sandy soil with a small amount of gravel" and "the riverbank uses concrete revetment with a gentle slope." Semantic parsing is performed on each natural language description, and natural language processing techniques are used to identify key information in the description. For "the riverbed in this section is mainly sandy soil with a small amount of gravel," the corresponding physical parameter type is identified as riverbed material, and the feature information includes that the main soil type is sandy soil, the secondary soil type is gravel, and the gravel content. For "the riverbank uses concrete revetment with a gentle slope," the physical parameter type is identified as riverbank structure, and the feature information includes that the revetment material is concrete and the slope characteristic is gentle. Through semantic parsing, the physical parameter type and specific feature information of each natural language description are clarified, laying the foundation for subsequent quantitative conversion.

[0022] Step S1133: Based on the semantic parsing results, and in accordance with the quantification standards for river physical parameters, convert the natural language description into the corresponding quantified parameter expression. The quantified parameter expression includes the parameter name, quantified value, and engineering unit.

[0023] Based on the physical parameter types and feature information obtained from semantic parsing, and in accordance with the collected quantitative standards for river physical parameters, a conversion from natural language descriptions to quantitative parameter expressions is performed. For example, for a semantic parsing result of "the main soil type of the riverbed is sandy soil, with a small amount of gravel," referring to the classification codes of riverbed materials in the quantitative standards, sandy soil corresponds to code A, gravel corresponds to code B, and "small amount" corresponds to a certain range of content. The converted quantitative parameter expression is "Riverbed material code: A (sandy soil), secondary material code: B (gravel), gravel content: [a,b]", where [a,b] is the specific numerical range determined based on "small amount," and the unit is percentage. For "gentle riverbank slope," according to the slope classification in the quantitative standards, "gentle" corresponds to a slope value range of [c,d]. The converted quantitative parameter expression is "Riverbank slope: [c,d]", and the unit is percentage. Each quantitative parameter expression clearly includes the parameter name, quantitative value, and engineering unit, ensuring that the converted parameters have clear engineering significance.

[0024] Step S1134: Extract the original measurement data from the split structural data, perform unified unit processing on the original measurement data, and convert the data of different measurement units into the standard units commonly used in engineering design.

[0025] Raw measurement data is extracted from the disassembled structural data. This data may originate from different monitoring devices and measurement methods, resulting in inconsistent units. For example, cross-sectional width measurements may be in meters or centimeters; cross-sectional depth measurements may be in meters or decimeters. A unit standardization process is performed on this raw measurement data, converting all length units to the standard unit of meters commonly used in engineering design. Specifically, for width data in centimeters, the value is divided by 100 to convert to meters; for depth data in decimeters, the value is divided by 10 to convert to meters. During the conversion process, the appropriate number of decimal places is retained according to the data's precision requirements to ensure accuracy. This unit standardization process ensures that raw measurement data from different sources have a unified measurement standard, facilitating subsequent calculations and analysis.

[0026] Step S1135: According to the accuracy requirements of engineering calculation, retain effective numbers and correct errors in the original measurement data after unit unification to obtain accuracy-optimized measurement data. Match the quantized parameter expression with the accuracy-optimized measurement data. Through data comparison and adjustment, ensure that the language description conversion result of the same physical parameter is consistent with the measurement data, and generate a parameter matching set.

[0027] The original measurement data, after unit standardization, needs to be processed according to the accuracy requirements of engineering calculations, including significant figure retention and error correction. Significant figure retention is determined based on different parameter types and engineering calculation needs; for example, cross-sectional width is retained to one decimal place, and cross-sectional depth to two decimal places. Error correction involves reasonably adjusting the data based on the accuracy of the measuring equipment and the influence of the measurement environment. For example, the measured value of a certain cross-sectional depth may be affected by water flow fluctuations, requiring correction based on historical data and empirical formulas to obtain optimized measurement data. The quantified parameter expression is then matched with the optimized measurement data. For the same physical parameter, such as cross-sectional width, the numerical range in the quantified parameter expression converted from natural language is compared with the optimized measurement data. If discrepancies exist, the conversion process or measurement data is checked for problems and adjustments are made to ensure consistency between the language description conversion result and the measurement data. For example, if the range of cross-sectional width in the quantified parameter expression deviates significantly from the average value of the measurement data, the semantic parsing and conversion process is reviewed again to ensure consistency, ultimately generating a parameter matching set.

[0028] Step S1136: Based on the engineering common sense data of the river's physical structure, perform a rationality analysis on the quantitative parameters in the parameter matching set, remove quantitative parameters that do not conform to engineering logic, perform structured organization on the parameter matching set after the rationality analysis, and group them according to the categories of cross-sectional dimensions, riverbed material, and riverbank structure to generate a structural parameter group set.

[0029] Based on engineering common sense data about river physical structures, such as the width-to-depth ratio of river cross-sections typically falling within a certain range and the distribution of riverbed materials exhibiting certain regularities, a rationality analysis is performed on the quantitative parameters in the parameter matching set. For example, if a cross-section has a width of 'e' and a depth of 'f', its width-to-depth ratio 'e / f' is calculated. If this ratio exceeds the common range for this type of river cross-section, the quantitative parameter is considered to be inconsistent with engineering logic and is discarded. Regarding riverbed material parameters, if a riverbed in a certain area simultaneously exhibits extremely high levels of both sand and clay, this contradicts engineering common sense, as sand and clay typically do not coexist in large quantities, and is therefore also discarded. The parameter matching set that passes the rationality analysis is then structured and grouped according to cross-sectional dimensions, riverbed materials, and riverbank structures. The cross-sectional dimensions group includes parameters such as cross-sectional width and depth; the riverbed material group includes parameters such as soil type code and particle size distribution; and the riverbank structure group includes parameters such as slope and slope protection material code, generating a structural parameter grouping set.

[0030] Step S1137: Add an engineering identifier to the quantized parameters in each structural parameter group set, identifying the engineering application scenario and calculation purpose of the parameter, thus obtaining the identified structural parameters.

[0031] Each quantified parameter in the structural parameter grouping set is labeled with an engineering identifier to clarify its specific application scenario and calculation purpose in engineering. For example, the cross-section width parameter in the cross-section size group is labeled "used for river flow calculation and waterway design"; the cross-section depth parameter is labeled "used for ship navigation assessment and water conservancy project construction". The soil type code in the riverbed material group is labeled "used for riverbed stability analysis and soil erosion prediction"; the particle size distribution parameter is labeled "used for water flow resistance calculation and sediment transport simulation". The slope parameter in the riverbank structure group is labeled "used for riverbank erosion risk assessment and slope stability calculation"; the slope protection material code is labeled "used for riverbank protection engineering design and engineering cost estimation". By adding these engineering identifiers, the application scenario and calculation purpose of each quantified parameter become clearer, resulting in labeled structural parameters.

[0032] Step S1138: Integrate the identified structural parameters, parameter matching sets, and engineering identification descriptions to generate river physical structure data in a quantifiable form that can be calculated in engineering design.

[0033] The river physical structure data is formed by integrating identifiable structural parameters, parameter matching sets, and engineering identifier descriptions. Specifically, the identifiable structural parameters are associated with corresponding data in the parameter matching sets, ensuring that each parameter has its corresponding quantified value, unit, and engineering identifier. The engineering identifier descriptions explain the meaning and application scenarios of each identifier in detail, facilitating reference during subsequent model use. The integrated river physical structure data is stored in a structured data table format, containing fields such as parameter name, quantified value, unit, engineering identifier, and data source, forming a quantitative representation of river physical structure data that can be directly used for calculations in engineering design.

[0034] Step S114: Filter the relevant data describing the water flow motion state in the original data set of multi-source rivers, classify the data types according to the hydrological engineering motion analysis specifications, and obtain the classified motion-related data. The classified motion-related data includes descriptions of water flow velocity, water level changes, and flow fluctuations.

[0035] Data describing the flow motion state was selected from the original dataset of multi-source rivers. This data includes flow velocity data and water level data collected from monitoring stations, as well as flow rate data calculated through flow measurement. Based on the hydrological engineering motion analysis specifications, the data was categorized into the following types: Flow velocity description type includes data on the magnitude and direction of flow velocity at different monitoring points, such as surface flow velocity and flow velocity at a certain depth; Water level change description type includes data on water level changes over time, such as hourly and daily water level data, as well as data on the rise and fall of water levels; Flow rate fluctuation description type includes real-time flow rate values, daily and monthly flow rate changes, and the amplitude of flow rate fluctuations. During the categorization process, appropriate type labels were added to the data, such as "Flow Velocity - Surface," "Water Level - Hourly Change," and "Flow Rate - Daily Fluctuation," to facilitate subsequent processing and analysis.

[0036] Step S115: Implement engineering index mapping on the segmented motion-related data to convert the original monitoring values ​​into standardized index forms that can be called in engineering simulation, thereby obtaining hydrological motion data.

[0037] The segmented motion-related data undergoes engineering index mapping, transforming raw monitoring values ​​into standardized indicators. For water flow velocity data, it is converted into a dimensionless relative velocity index, expressed as the ratio of the real-time water flow velocity at a monitoring point to the historical maximum water flow velocity at that point. For water level change data, it is converted into a water level fluctuation percentage index, calculated as the difference between the current water level and the historical average water level divided by the historical average water level and then multiplied by 100%. For flow fluctuation data, it is converted into a flow fluctuation coefficient index, obtained by dividing the standard deviation of the flow rate by the average value. The values ​​of these standardized indicators range from 0 to 1, facilitating comparison and calculation in engineering simulations. For example, a relative velocity index of 0.6 indicates that the current water flow velocity is 60% of the historical maximum water flow velocity; a water level fluctuation percentage of 10% indicates that the current water level is 10% higher than the historical average water level. Through these transformations, hydrological motion data that can be directly used in engineering simulations is obtained.

[0038] Step S116: Separate the relevant data describing external effects on rivers from the original multi-source river data set, classify the data according to the environmental engineering effect classification standard, and obtain the classified environmentally relevant data. The classified environmentally relevant data includes descriptions of precipitation effects, topographic effects, and human activity effects.

[0039] Data describing external effects on rivers were extracted from the original dataset of multi-source rivers. This data included rainfall data collected by meteorological stations, topographic data, wastewater discharge data from riverside factories, and agricultural irrigation water data. The data was then categorized according to environmental engineering impact classification standards. The rainfall impact category included data on rainfall amount, intensity, and duration; the topographic impact category included data on watershed slope, elevation, and topographic relief; and the human activity impact category included data on industrial wastewater volume, agricultural water consumption, domestic sewage discharge, and water conservancy project operation. During the classification process, the data was divided into these categories, and corresponding labels were added to each category, such as "Rainfall - Daily Rainfall," "Topography - Slope," and "Human Activity - Industrial Wastewater," resulting in categorized environmentally relevant data.

[0040] Step S117: Perform engineering factor transformation on the categorized environmental data to convert the description of external effects into an impact factor form applicable in engineering analysis, thereby obtaining environmental effect data.

[0041] Engineering factor transformations are performed on the categorized environmental data to convert external impact descriptions into influencing factors. Rainfall data in the precipitation impact description is converted into a precipitation impact factor, determined by the ratio of actual rainfall to the region's annual average rainfall. If the actual rainfall is g times the annual average rainfall, the precipitation impact factor is g. Slope data in the topographic impact description is converted into a topographic impact factor; the steeper the slope, the higher the topographic impact factor value. A specific function can be used to convert slope values ​​into factor values ​​between 0 and 1. Industrial wastewater volume in the human activity impact description is converted into a pollution impact factor; the larger the wastewater volume, the higher the pollution impact factor value. Similarly, a conversion function is used to convert wastewater volume into corresponding factor values. These influencing factors are expressed in dimensionless numerical form, facilitating coupling calculations with other parameters in engineering analysis to obtain environmental impact data.

[0042] Step S118: Perform interaction mining operations on river physical structure data, hydrological movement data, and environmental action data, and determine the type of association and degree of influence between different data through data feature comparison and trend correlation analysis.

[0043] This study mines the interaction relationships among river physical structure data, hydrological data, and environmental impact data using association analysis methods from data mining techniques. First, features are extracted from each type of data to obtain feature vectors that characterize the data's features. For example, features such as cross-sectional width, depth, and riverbed material are extracted from river physical structure data; features such as flow velocity, water level fluctuation, and discharge fluctuation are extracted from hydrological data; and features such as precipitation impact factors, topographic impact factors, and pollution impact factors are extracted from environmental impact data. Then, data feature comparison is performed to examine the similarities and differences between different data features and identify potentially correlated feature combinations. Next, trend correlation analysis is conducted to observe whether there is a correlation between the changing trends of different data over time. For example, the relationship between the changing trends of precipitation impact factors and the changing trends of water level fluctuations is analyzed. If both show similar upward or downward trends, a positive correlation exists. The degree of influence is determined by calculating the correlation coefficient, which ranges from -1 to 1; the larger the absolute value, the greater the influence. Through the above analysis, the type of association (e.g., positive correlation, negative correlation) and the specific numerical degree of influence between different data are determined.

[0044] Step S119: Associate and bind the determined association type and degree of influence with river physical structure data, hydrological movement data, and environmental action data to generate a data association attribute table.

[0045] The identified correlation types and their impact levels are then linked and bound to corresponding river physical structure data, hydrological motion data, and environmental impact data. For example, a positive correlation exists between precipitation impact factors and water level fluctuations, with an impact level of h. This correlation information is then bound to the precipitation impact factor parameter in the precipitation impact data and the water level fluctuation parameter in the hydrological motion data. All correlation binding information is then compiled into a data correlation attribute table, which includes columns for the data types of the correlated parties, parameter names, correlation type (positive / negative correlation), and impact level values. For example, a single record might read "Environmental impact data - precipitation impact factor, hydrological motion data - water level fluctuation, positive correlation, h," demonstrating the correlation and impact level between different data sets.

[0046] Step S1110: Integrate river physical structure data, hydrological movement data, environmental action data, and data association attribute tables to generate a set of river engineering data deconstruction.

[0047] By integrating river physical structure data, hydrological motion data, environmental impact data, and data association attribute tables, a set of deconstructed river engineering data is formed. During the integration process, index relationships are established between the various data types, and different types of data are organically linked through the association information in the data association attribute tables. For example, by using the association relationship between river physical structure data and hydrological motion data in the association attribute tables, the impact of a certain physical structure parameter on hydrological motion parameters can be quickly identified.

[0048] Step S120: Establish a river dynamic risk topology model based on the river engineering data deconstruction set. This river dynamic risk topology model completes the engineering simulation of risk transmission through node association, path mapping and parameter coupling.

[0049] In the aforementioned watershed scenario, a dynamic risk topology model for the river is established based on the generated river engineering data deconstruction set. The core idea of ​​this model is to abstract the river system as a topological structure composed of multiple nodes and paths connecting them. The risk transmission process within the river system is simulated through the relationships between nodes, path mapping, and parameter coupling. First, the basic framework and components of the model need to be determined, including node definitions, path construction, and parameter settings. Then, the specific content of the model is filled in based on the data from the river engineering data deconstruction set, ensuring that the model accurately reflects the actual situation of the river system and the characteristics of risk transmission.

[0050] Step S121: Extract the core parameters of the river physical structure data from the river engineering data deconstruction set, and define the river structure location corresponding to the core parameters as a topology node. Each topology node contains structural parameter description and location representation information.

[0051] Core parameters are extracted from the river physical structure data of the river engineering data deconstruction set. These core parameters reflect key characteristics of the river structure, such as the width and depth of important cross-sections, the riverbed material of key river sections, and the slope of important riverbanks. The river structure locations corresponding to these core parameters are defined as topological nodes. For example, the locations of several key river cross-sections within the basin are defined as topological nodes, each with a unique identifier. Each topological node contains a structural parameter description and location representation information. The structural parameter description includes the specific values ​​of the core parameters of the river physical structure at that node, such as a cross-section width of i meters, a depth of j meters, and a riverbed material code of A. The location representation information includes the node's geographical coordinates within the river system, such as latitude and longitude, and its relative position within the river's course, such as upstream node, midstream node, or downstream node. Through these definitions, the key structural locations in the river system are abstracted as nodes in the topological model.

[0052] Step S122: Extract the data association attribute table from the river engineering data deconstruction set, determine the association relationship between topological nodes based on the association type and the degree of influence, and generate a node association list. This node association list records the connection logic between any two topological nodes.

[0053] The data association attribute table is extracted from the deconstructed river engineering data set, recording the association types and influence levels between different data. Based on this information, the association relationships between topological nodes are determined. For example, if the data association attribute table shows a positive correlation between the flow fluctuation of an upstream topological node and the water level change of a downstream topological node, with an influence level of k, then an association relationship is determined between these two topological nodes. A node association list is generated, recording the connection logic between any two topological nodes, including the source node identifier, target node identifier, association type (e.g., positive correlation, negative correlation), and influence level value k. For two topological nodes with a direct association relationship, a record is created in the list, clarifying their connection method and influence, laying the foundation for subsequently constructing risk transmission paths.

[0054] Step S123: Based on the connection logic in the node association list, construct risk transmission paths between topology nodes. Each risk transmission path corresponds to a set of node associations. The path description of the risk transmission path includes the data transmission direction and the impact method.

[0055] Based on the connection logic in the node association list, risk transmission paths are constructed between topological nodes. For each pair of relationships in the node association list, a corresponding risk transmission path is built. For example, for a relationship where the source node is upstream node A, the target node is midstream node B, the association type is positive, and the influence degree is k, a risk transmission path is constructed from node A to node B. The path description includes the data transmission direction and the influence method. The data transmission direction is clearly defined as from the source node to the target node; the influence method describes how changes in the parameters of the source node affect the parameters of the target node. For example, an increase in the flow rate of node A will cause the water level of node B to rise; the influence method is that the flow rate change is transmitted to the downstream node through water flow, thus affecting the water level. Each risk transmission path has a unique identifier, corresponding one-to-one with the relationships in the node association list, forming a complete risk transmission path network.

[0056] Step S124: Extract engineering indicators of hydrological motion data and environmental action data from the river engineering data deconstruction set, and use the engineering indicators of hydrological motion data and environmental action data as attribute parameters of topology nodes to realize the binding of attribute parameters with topology nodes.

[0057] Engineering indicators for hydrological motion data, such as relative velocity, percentage of water level change, and flow fluctuation coefficient, are extracted from the deconstructed set of river engineering data. These indicators, along with engineering indicators for environmental impact data, such as precipitation impact factors, topographic impact factors, and pollution impact factors, are then used as attribute parameters for topological nodes. Each topological node is bound to corresponding attribute parameters based on its location and function within the river system. For example, upstream topological nodes might be bound to attributes such as flow fluctuation coefficient and precipitation impact factor; downstream topological nodes might be bound to attributes such as percentage of water level change and topographic impact factor. By binding attribute parameters to topological nodes, each node possesses rich attribute information. This attribute information plays a crucial role in risk transmission simulation, influencing the intensity and direction of risk transmission.

[0058] Step S125: Based on the attribute parameters of the topology nodes and the degree of influence in the node association list, quantify the transmission strength parameter of each risk transmission path. The transmission strength parameter reflects the degree of contribution of each risk transmission path to risk transmission.

[0059] Based on the attribute parameters of topological nodes and the degree of influence in the node association list, the transmission strength parameter of each risk transmission path is quantified. First, the factors influencing transmission strength are determined, including the attribute parameter values ​​of the source node, the attribute parameter values ​​of the target node, and the degree of influence in the node association list. For example, a larger flow fluctuation coefficient of the source node indicates a stronger risk source, potentially leading to an increased transmission strength parameter; a higher degree of influence also results in a larger transmission strength parameter. By establishing a quantitative model for the transmission strength parameter, these factors are comprehensively considered to calculate the transmission strength parameter for each risk transmission path. The transmission strength parameter ranges from 0 to 1; a larger value indicates a greater contribution of the path to risk transmission, and a higher probability and intensity of risk transmission through that path.

[0060] Step S1251: Extract core indicators related to risk transmission from the attribute parameters of the topological nodes. The node structural stability parameters, hydrological carrying capacity parameters, and environmental interference resistance parameters are the specific contents of these core indicators.

[0061] Core indicators related to risk transmission are extracted from the attribute parameters of topological nodes. These core indicators reflect the key characteristics of nodes in the risk transmission process. Node structural stability parameters measure the river structure's ability to resist risks at the node, such as the stability of the riverbed material and the strength of the riverbank. These parameters comprehensively consider factors such as riverbed soil type, riverbank slope, and slope protection materials. Hydrological carrying capacity parameters indicate the range of hydrological changes the river at the node can withstand, such as maximum flow and highest water level, and are related to attribute parameters such as cross-sectional dimensions and flow velocity. Environmental interference resistance parameters reflect the node's resistance to external environmental effects, such as its ability to withstand changes in precipitation and pollutants, and are related to relevant indicators in the environmental impact data.

[0062] Step S1252: Perform quantitative grading processing on each core indicator. According to the requirements of engineering simulation, divide the numerical range of the core indicators into different level intervals, and each level interval corresponds to a quantitative score.

[0063] Each core indicator undergoes quantitative grading. Based on the needs of engineering simulation and relevant standards, the numerical range of the core indicators is divided into multiple grade intervals. For example, the numerical range of the node structure stability parameter is 0 to 1, which is divided into three grade intervals: high stability, medium stability, and low stability. 0.8 to 1 represents high stability, corresponding to a quantitative score of m; 0.4 to 0.8 represents medium stability, corresponding to a quantitative score of n; and 0 to 0.4 represents low stability, corresponding to a quantitative score of p, where m > n > p. Hydrological carrying capacity parameters and environmental interference resistance parameters are also graded and assigned quantitative scores using a similar method. Through the above processing, continuous core indicator values ​​are converted into discrete quantitative scores, facilitating subsequent calculations and analysis.

[0064] Step S1253: Extract the influence degree description from the node association list, convert the influence degree description into the corresponding influence coefficient, and ensure that the value of the influence coefficient is consistent with the actual association of the influence degree.

[0065] The degree of influence in a node association list is typically described using correlation coefficients, such as 0.7 or -0.5. These influence descriptions are directly converted to their corresponding influence coefficients, with the coefficient values ​​reflecting the actual degree of association. For example, an influence description of 0.7 indicates a strong positive correlation, and the corresponding coefficient is 0.7; an influence description of -0.5 indicates a moderate negative correlation, and the corresponding coefficient is -0.5. The influence coefficient ranges from -1 to 1, with positive numbers representing a positive correlation and negative numbers representing a negative correlation. The absolute value indicates the strength of the influence.

[0066] Step S1254: Set the transmission strength quantification calculation logic. The transmission strength quantification calculation logic takes the quantification score of the core indicator, the influence coefficient and the path length parameter as input, and generates the transmission strength parameter through engineering calculation.

[0067] A logic for quantifying transmission intensity is established, which comprehensively considers the quantified scores of core indicators, influence coefficients, and path length parameters. The path length parameter refers to the actual distance between two topological nodes; the longer the distance, the greater the potential attenuation during risk transmission. The basic idea of ​​the transmission intensity quantification logic is as follows: First, the quantified scores of the core indicators of the source node and the target node are weighted and summed to obtain a comprehensive node score. Then, the comprehensive node score is multiplied by the influence coefficient, and then multiplied by the path length attenuation factor (the longer the path length, the smaller the attenuation factor) to obtain the transmission intensity parameter. For example, the transmission intensity parameter = (quantified score of core indicator of source node × weight 1 + quantified score of core indicator of target node × weight 2) × influence coefficient × path length attenuation factor, where weight 1 and weight 2 are determined according to the importance of the core indicators.

[0068] Step S1255: Substitute the core indicator quantification scores of the two topological nodes corresponding to each risk transmission path into the transmission intensity quantification calculation logic to obtain the node contribution score.

[0069] For each risk transmission path, the core indicator quantification scores of the corresponding source node and target node are obtained. These scores are then substituted into the node comprehensive score calculation part of the transmission intensity quantification calculation logic. For example, the quantification score of the source node's node structural stability is m1, the quantification score of its hydrological carrying capacity is n1, and the quantification score of its environmental resistance is p1; the quantification score of the target node's node structural stability is m2, the quantification score of its hydrological carrying capacity is n2, and the quantification score of its environmental resistance is p2. According to the set weights, such as q1 for node structural stability, q2 for hydrological carrying capacity, and q3 for environmental resistance, the comprehensive score of the source node is calculated as m1×q1+n1×q2+p1×q3, and the comprehensive score of the target node is m2×q1+n2×q2+p2×q3. Then, the comprehensive scores of the source node and the comprehensive scores of the target node are weighted and summed (e.g., each accounting for 50%) to obtain the node contribution score.

[0070] Step S1256: Input the influence coefficient and path length parameters corresponding to the node association list into the transmission intensity quantization calculation logic to obtain the path influence score.

[0071] The impact coefficient for each risk transmission path is retrieved from the node association list, and the path length parameter is measured or calculated. The impact coefficient and path length parameter are then substituted into the corresponding part of the transmission intensity quantification logic to calculate the path impact score. The path length parameter is calculated using the geographical coordinates of the two topological nodes to obtain the actual distance. Then, the distance is converted into a path length attenuation factor according to a preset attenuation function, such as path length attenuation factor = 1 / (1 + path length × s), where s is the attenuation coefficient. Path impact score = impact coefficient × path length attenuation factor.

[0072] Step S1257: Perform collaborative calculations on the node contribution score and path influence score through the transmission intensity quantification calculation logic to generate the initial transmission intensity parameters for each risk transmission path.

[0073] The node contribution score and path influence score are calculated together using the transmission strength quantification logic, for example, by multiplying them to obtain the initial transmission strength parameter. Initial transmission strength parameter = node contribution score × path influence score. Through the above calculation, the impact of both node characteristics and path characteristics on risk transmission strength is comprehensively considered, resulting in a preliminary transmission strength parameter.

[0074] Step S1258: Collect the transmission strength reference standard in the engineering risk simulation, compare the initial transmission strength parameters with the transmission strength reference standard, and determine the direction of parameter adjustment.

[0075] We collected commonly used transmission strength reference standards in engineering risk simulation. These standards specify reasonable ranges for transmission strength parameters of different types of risk transmission paths. We compared the calculated initial transmission strength parameters with the reference standards. If the initial parameters are within the reference standard range, no adjustment is needed; if the initial parameters are higher than the upper limit of the reference standard, the transmission strength parameters need to be reduced; if the initial parameters are lower than the lower limit of the reference standard, the transmission strength parameters need to be increased, thus determining the direction of parameter adjustment.

[0076] Step S1259: Based on the adjustment direction, perform dynamic correction on the initial transmission strength parameters so that the corrected transmission strength parameters meet the actual needs of engineering simulation and the transmission strength reference standard.

[0077] Based on the determined parameter adjustment direction, the initial transfer strength parameters are dynamically corrected. Correction methods can include proportional adjustment, such as reducing the initial parameters by a certain percentage (e.g., t%) when the initial parameters are higher than the upper limit of the reference standard, and increasing them by a certain percentage (e.g., u%) when the initial parameters are lower than the lower limit of the reference standard. Alternatively, a function correction method can be used, calculating the correction value based on the degree of deviation between the initial parameters and the reference standard through a specific functional relationship. After correction, the transfer strength parameters are made to meet the actual needs of engineering simulation and the transfer strength reference standard.

[0078] Step S12510: Perform normalization on the corrected transmission strength parameters to obtain transmission strength parameters that reflect the contribution of the path to risk transmission.

[0079] The corrected transmission strength parameter is normalized, mapping its value to between 0 and 1. The normalization method can be max-min normalization, i.e., transmission strength parameter = (corrected transmission strength parameter - minimum transmission strength parameter across all paths) / (maximum transmission strength parameter across all paths - minimum transmission strength parameter across all paths). This normalization process makes the transmission strength parameters of different risk transmission paths comparable; a larger value indicates a greater contribution of that path to risk transmission.

[0080] Step S126: Perform coupling operations on the attribute parameters, transmission strength parameters, and data association attributes of the topology nodes to generate a node coupling parameter set, which describes the synergistic relationship between node attributes and path transmission.

[0081] A node coupling parameter set is generated by performing coupling operations on the attribute parameters, transmission strength parameters, and data association attributes of topological nodes. First, the coupling relationships between each parameter are determined. For example, the hydrological carrying capacity parameter of a node affects the transmission strength parameter, and the association type and degree of influence in the data association attributes also affect the synergistic effect between nodes. The coupling operation process combines and calculates the above parameters according to certain mathematical relationships, such as multiplying the node's attribute parameters by the transmission strength parameter and then adjusting for the degree of influence in the data association attributes. The generated node coupling parameter set contains multiple parameters, each describing the synergistic relationship between node attributes and path transmission in different aspects, such as the enhancing or weakening effect of node attributes on path transmission strength, and the degree of mutual influence between different nodes through path transmission.

[0082] Step S127: Set up the topology dynamic adjustment instruction receiving process to receive external data update instructions and perform dynamic adaptation operations such as adding or removing topology nodes, optimizing paths, and adjusting parameters.

[0083] A dynamic topology adjustment command receiving process is established, capable of receiving externally sent data update commands in real time. These external data update commands may originate from new monitoring data, engineering modification information, etc., indicating the need to adjust the topology model. Upon receiving a command, the process performs corresponding dynamic adaptation operations based on the command content. If the command requires adding a new topology node, such as adding a node at the location of a newly constructed water conservancy project, the new node is added to the topology model, and its attribute parameters and relationships are set. If the command requires deleting a no longer important node, the node and its related paths and parameters are removed from the model. If the command involves path optimization, such as changes in risk transmission paths due to river regulation, the connection relationships and transmission strength parameters of the corresponding paths are adjusted. If the command requires parameter adjustment, such as updating node attribute parameters or path transmission strength parameters, the corresponding parameters in the model are modified. Through this dynamic adjustment, the topology model can adapt to changes in the river system.

[0084] Step S128: Based on the topology nodes, node association list, risk transmission path, node coupling parameter set and dynamic adjustment instruction receiving process, construct the initial topology modeling structure, which has the basic functions of risk transmission simulation.

[0085] Based on the determined topology nodes, node association list, risk transmission paths, node coupling parameter sets, and dynamic adjustment command receiving process, an initial topology modeling structure is constructed. Topology nodes are arranged in the model according to their geographical location and association relationships to form a node network. Connections are established between nodes according to the node association list and risk transmission paths, and transmission strength parameters are labeled. The node coupling parameter set is associated with nodes and paths to achieve the synergistic effect of node attributes and path transmission. The dynamic adjustment command receiving process is integrated into the model, enabling it to receive and process external commands. The initial topology modeling structure possesses the basic functions of risk transmission simulation, capable of simulating the process of risk transmission from one node to other nodes through paths, and calculating the intensity changes of risk during the transmission process.

[0086] Step S129: Perform an engineering adaptation operation on the initial topology modeling structure, incorporating the structural mechanics principles and hydrological motion-related technical points in river engineering analysis, so that the initial topology modeling structure meets the accuracy requirements of engineering simulation.

[0087] An engineering adaptation process is performed on the initial topology model, integrating structural mechanics principles and hydrological motion-related techniques from river engineering analysis into the model. Structural mechanics principles are used to analyze the structural stability at topological nodes, such as calculating riverbank slope stability and analyzing the erosion resistance of the riverbed. By introducing relevant mechanical calculation formulas and parameters, the model can more accurately simulate the structural response of nodes under risk. Hydrological motion-related techniques include hydrodynamic equations and sediment transport laws. These techniques are applied to the simulation of risk transmission paths, considering the impact of factors such as flow velocity and flow rate on risk transmission, such as the diffusion rate and range of hazardous substances in the flow. Through this engineering adaptation, the simulation accuracy of the initial topology model is improved, meeting the requirements of engineering simulation.

[0088] Step S1210: Integrate the optimized topological nodes, relationships, risk transmission paths, node coupling parameter sets, and dynamic adjustment instruction receiving process to generate a river dynamic risk topology model that completes risk transmission through node association, path mapping, and parameter coupling.

[0089] The optimized and engineered topology nodes, relationships, risk transmission paths, node coupling parameter sets, and dynamic adjustment command receiving process are integrated to form the final river dynamic risk topology model. During integration, consistency among the components is ensured, node attribute parameters match their relationships and risk transmission paths, and the node coupling parameter set accurately describes the synergistic effect between nodes and paths. The dynamic adjustment command receiving process can correctly receive and process external commands, enabling dynamic model updates. The final river dynamic risk topology model can perform engineered simulations of risk transmission processes in river systems through node associations, path mapping, and parameter coupling.

[0090] Step S130: Call the dynamic adaptation process of the river dynamic risk topology model to perform topology calibration operation on historical risk data and generate a topology calibration parameter set, which is directly used to adjust the accuracy of risk transmission simulation.

[0091] In the aforementioned watershed scenario, to improve the simulation accuracy of the river dynamic risk topology model, it is necessary to invoke the model's dynamic adaptation process to perform topology calibration on historical risk data. Historical risk data contains detailed information about past river risk events. By analyzing and utilizing this data, the model's topology parameters are adjusted, enabling the model to more accurately simulate the actual risk transmission process. The generated topology calibration parameter set will be used for subsequent accuracy adjustments in risk transmission simulations.

[0092] Step S131: Collect complete engineering monitoring data and risk development records recorded during the occurrence of historical river dynamic risks, and generate a historical river dynamic risk engineering dataset. This historical river dynamic risk engineering dataset contains multi-dimensional engineering data before and after the occurrence of risks and details of risk evolution.

[0093] Data related to the occurrence of historical river dynamic risks is collected, including complete engineering monitoring data and risk development records before, during, and after the occurrence of risks. Engineering monitoring data comes from the monitoring equipment at the time, such as water level monitoring data, flow velocity monitoring data, and water quality monitoring data. Risk development records include information such as the type of risk event (e.g., flood, water pollution), occurrence time, impact range, and development trend. This data is then organized and integrated to generate a historical river dynamic risk engineering dataset, which is stored according to risk events. Each risk event corresponds to a set of multi-dimensional engineering data and risk evolution details.

[0094] Step S132: Perform data format standardization processing on the historical river dynamic risk engineering dataset to obtain a standardized historical engineering dataset. Extract key node parameters, path transmission data and risk impact range data when the risk occurs from the standardized historical engineering dataset to generate a historical risk core dataset.

[0095] Data format standardization was performed on the historical river dynamic risk engineering dataset to unify the data format, units, and precision. For example, monitoring data in different formats were converted into a unified tabular format, all length units were standardized to meters, and time units were standardized to hours. This standardization process yielded a standardized historical engineering dataset. From this dataset, key node parameters at the time of risk occurrence were extracted, such as water level, flow velocity, and water quality parameters at each topological node at the moment of risk occurrence; path transmission data, such as the intensity and time of risk transmission along each path; and risk impact range data, such as the length and area of ​​the river affected by the risk. These extracted data were then integrated to generate a core historical risk dataset, which focuses on key information at the time of risk occurrence and is used for comparison with model simulation results.

[0096] Step S133: Input the historical risk core dataset into the dynamic adaptation process of the river dynamic risk topology model, start the topology simulation operation, and obtain the historical risk topology simulation results. The historical risk topology simulation results include the simulated node association status, path transmission strength, and risk diffusion trend.

[0097] The core historical risk dataset is input into the dynamic adaptation process of the river dynamic risk topology model. The dynamic adaptation process sets the initial and boundary conditions of the model based on the input data. Then, topology simulation is initiated. The model simulates the occurrence and development of historical risk events based on its own topology, node relationships, and transmission strength parameters. After the simulation, the historical risk topology simulation results are obtained, including the simulated node relationship status (e.g., which nodes have risk transmission relationships), path transmission strength (the transmission strength values ​​of each path during the simulation), and risk diffusion trends (the diffusion of risk in time and space, such as the change in the risk's impact range over time).

[0098] Step S134: Extract the actual risk development records from the historical risk core dataset, dynamically compare them with the historical risk topology simulation results, and determine the deviation data between the simulation results and the actual situation. The deviation data includes node parameter deviation, path transmission deviation, and range prediction deviation.

[0099] Actual risk development records are extracted from the historical risk core dataset, including actual node parameter changes, actual risk transmission paths and intensities, and actual risk impact ranges. These actual records are dynamically compared with the results of historical risk topology simulations, point-by-point across time and space. For example, comparing the simulated water level at a specific node at a particular time with the actual monitored water level, the difference is calculated to obtain the node parameter deviation; comparing the simulated transmission intensity of a certain path with the transmission intensity inferred from actual data, the path transmission deviation is obtained; comparing the simulated risk impact range with the actual impact range, the range prediction deviation is obtained. These deviation data are then organized and recorded to form a deviation dataset.

[0100] Step S135: Substitute the deviation data into the preset calibration calculation process, set the calibration target based on the accuracy requirements of the engineering simulation, and generate the node calibration coefficient, path calibration coefficient and range calibration coefficient by reverse derivation of the data deviation.

[0101] The deviation data is substituted into a pre-defined calibration calculation process. The purpose of this process is to adjust model parameters based on the deviation data to reduce the discrepancy between simulation results and actual conditions. First, calibration targets are set based on the accuracy requirements of engineering simulations, such as the allowable ranges for node parameter deviations, path transmission deviations, and range prediction deviations. Then, by reverse-engineering the data deviations, the causes of the deviations are analyzed to determine the model parameters that need adjustment. For node parameter deviations, node calibration coefficients are derived to adjust the attribute parameters of nodes in the model; for path transmission deviations, path calibration coefficients are derived to adjust the transmission strength parameters of the paths; and for range prediction deviations, range calibration coefficients are derived to adjust the calculation parameters for the risk diffusion range. The generation of calibration coefficients requires comprehensive consideration of the magnitude and direction of the deviations, as well as the sensitivity of the model parameters to the deviations.

[0102] Step S1351: Analyze the accuracy requirements of the engineering simulation, determine the allowable deviation range of node parameters, path transmission and range prediction in the risk transmission simulation, and use the allowable deviation range as the core basis for calibration targets.

[0103] The accuracy requirements for engineering simulations are typically determined based on the importance of the river project, the severity of risk events, and relevant industry standards. For example, for important urban river sections, the allowable deviation range for node parameters (such as water level) is relatively small; for general rural river sections, the allowable deviation range can be appropriately relaxed. The allowable deviation ranges for node parameters, path transmission, and extent prediction in risk transmission simulations are determined, such as ±v for node parameters, ±w for path transmission, and ±x for extent prediction. These allowable deviation ranges serve as the core basis for calibration objectives, ensuring that the deviation between the adjusted model simulation results and the actual situation is controlled within the allowable range during the calibration process.

[0104] Step S1352: Set the basic execution logic of the calibration operation process. The basic execution logic corresponds to the operation steps of node deviation calibration, path deviation calibration and range deviation calibration.

[0105] The basic execution logic of the calibration operation is defined, which consists of three parallel operation steps: node deviation calibration, path deviation calibration, and range deviation calibration. The node deviation calibration step compares the node parameter deviation with the allowable deviation range. If the deviation exceeds the range, a node calibration coefficient is calculated based on the magnitude and direction of the deviation, and the node's attribute parameters are adjusted. The path deviation calibration step analyzes the causes of path transmission deviations, such as unreasonable transmission strength parameter settings, and then calculates the path calibration coefficient to correct the transmission strength parameters. The range deviation calibration step adjusts relevant parameters in the risk diffusion model, such as the diffusion coefficient, for range prediction deviations, generating range calibration coefficients. Each step has clearly defined inputs (deviation data, allowable deviation range) and outputs (calibration coefficients).

[0106] Step S1353: Input the node parameter deviation in the deviation data into the node deviation calibration step of the calibration operation process. This step is based on the calibration target and calculates the adjustment ratio of the node parameters through an engineering algorithm to generate the initial node calibration coefficient.

[0107] The node parameter deviations in the deviation data are input into the node deviation calibration step of the calibration operation process. This step first determines whether the node parameter deviation is within the allowable deviation range. If it is within the range, the initial node calibration coefficient is 1 (no adjustment is needed); if it exceeds the range, the adjustment ratio of the node parameters is calculated based on the calibration target using an engineering algorithm. The engineering algorithm can be a proportional control algorithm, such as adjustment ratio = (allowable deviation upper limit - node parameter deviation) / node parameter deviation. The initial node calibration coefficient is generated based on the adjustment ratio, and the initial node calibration coefficient = 1 + adjustment ratio. For example, if the node parameter deviation is y, the allowable deviation upper limit is z, and y > z, then the adjustment ratio = (zy) / y, and the initial node calibration coefficient = 1 + (zy) / y.

[0108] Step S1354: Input the path transfer deviation in the deviation data into the path deviation calibration step of the calibration calculation process. Based on the calibration target and combined with the engineering characteristics of path transfer, this step calculates the adjustment ratio of the path transfer parameters and generates the initial path calibration coefficient.

[0109] The path transmission deviation from the deviation data is input into the path deviation calibration step of the calibration process. Based on the calibration target and the engineering characteristics of path transmission (such as path length and node attributes), the adjustment ratio of the path transmission parameters is calculated. For example, if the path transmission deviation is caused by an excessively high transmission strength parameter and the path length is long, a larger adjustment ratio may be required. Through an engineering algorithm, such as a weighted calculation based on deviation magnitude and path characteristics, the adjustment ratio is obtained, generating the initial path calibration coefficient: Initial path calibration coefficient = 1 + Path transmission parameter adjustment ratio.

[0110] Step S1355: Input the range prediction deviation in the deviation data into the range deviation calibration step of the calibration calculation process. This step refers to the calibration target and, based on the engineering technical points of risk diffusion, calculates the adjustment ratio of the range prediction parameter and generates the initial range calibration coefficient.

[0111] Input the range prediction deviation from the deviation data into the range deviation calibration step of the calibration calculation process. Referring to the calibration target and based on key engineering techniques for risk diffusion, such as the diffusion coefficient and the influence of water flow velocity on diffusion, calculate the adjustment ratio of the range prediction parameters. For example, if the range prediction deviation is due to an unreasonable diffusion coefficient setting, adjust the value of the diffusion coefficient according to the magnitude and direction of the deviation, calculate the adjustment ratio, and generate the initial range calibration coefficient: Initial range calibration coefficient = 1 + Range prediction parameter adjustment ratio.

[0112] Step S1356: Extract the parameter coupling relationship in the river dynamic risk topology model, and perform correlation verification on the initial node calibration coefficient, initial path calibration coefficient and initial range calibration coefficient to ensure that the calibration coefficients conform to the parameter coupling logic.

[0113] The parameter coupling relationships in the river dynamic risk topology model are extracted. These relationships describe the mutual influence between node parameters, path transmission parameters, and range prediction parameters. For example, the hydrological carrying capacity parameter of a node affects the path transmission intensity parameter, which in turn affects the risk diffusion range parameter. Correlation checks are performed on the initial node calibration coefficients, initial path calibration coefficients, and initial range calibration coefficients to check whether the adjusted calibration coefficients conform to these parameter coupling relationships. For example, if the node calibration coefficient increases (indicating an enhancement of the node attribute parameter), according to the parameter coupling relationship, the path transmission calibration coefficient should also change accordingly. If the initial path calibration coefficient does not change as expected, it indicates a correlation problem, and the calibration coefficients need to be adjusted.

[0114] Step S1357: Based on the correlation verification results, adjust the values ​​of the initial calibration coefficients, and set the calibration coefficient optimization target operation logic. The calibration coefficient optimization target operation logic aims to minimize the deviation between the simulation results and the actual historical data, and performs optimization operation on the adjusted calibration coefficients.

[0115] Based on the correlation verification results, the initial calibration coefficients are adjusted to ensure they conform to parameter coupling logic. Then, a calibration coefficient optimization logic is established, aiming to minimize the deviation between the simulation results and actual historical data. Using an optimization algorithm (such as gradient descent), the adjusted calibration coefficients are continuously optimized, gradually reducing the deviation between the model simulation results and actual historical data until it reaches its minimum.

[0116] Step S1358: Through multiple rounds of iterative calculations, the target calculation logic for optimizing the calibration coefficients reaches a convergent state, resulting in the optimized node calibration coefficients, path calibration coefficients, and range calibration coefficients.

[0117] The calibration coefficients are optimized through multiple rounds of iterative calculations. In each iteration, the current calibration coefficients are applied to the model for simulation, and the deviation between the simulation results and actual historical data is obtained. The calibration coefficients are then adjusted based on the deviation. This process is repeated until the calibration coefficient optimization logic reaches a convergent state, meaning the deviation no longer decreases significantly with increasing iterations. At this point, the optimized node calibration coefficients, path calibration coefficients, and range calibration coefficients are obtained.

[0118] Step S1359: Integrate the optimized node calibration coefficients, path calibration coefficients, range calibration coefficients, and association rules in the calibration process to generate node calibration coefficients, path calibration coefficients, and range calibration coefficients.

[0119] The optimized node calibration coefficients, path calibration coefficients, range calibration coefficients, and the association rules determined during the calibration process are integrated to form the final node calibration coefficients, path calibration coefficients, and range calibration coefficients. The association rules describe the interrelationships between different calibration coefficients, such as how the path calibration coefficient should be adjusted when the node calibration coefficient changes. The integrated calibration coefficients will be used to adjust the model's topology parameters, improving the model's simulation accuracy.

[0120] Step S136: Perform correlation integration on the node calibration coefficient, path calibration coefficient and range calibration coefficient, and determine the synergistic effect rules among the calibration coefficients based on the parameter coupling relationship in the river dynamic risk topology model.

[0121] A correlational integration was performed on the node calibration coefficients, path calibration coefficients, and range calibration coefficients. Based on the parameter coupling relationships in the river dynamic risk topology model, the mutual influence among the calibration coefficients was analyzed. For example, changes in the node calibration coefficients may lead to changes in the path calibration coefficients, which in turn affect the range calibration coefficients. Through correlational integration, the synergistic effect rules among the calibration coefficients were determined, clarifying how other related calibration coefficients should be adjusted accordingly when one calibration coefficient is adjusted, to ensure the overall consistency and rationality of the model parameters.

[0122] Step S137: Based on the synergistic effect rule, bind various calibration coefficients to the topological parameters of the river dynamic risk topology model to generate a preliminary calibration parameter set, which contains the correspondence between structural parameters and calibration coefficients.

[0123] Based on the synergistic effect rule, node calibration coefficients are bound to topological node attribute parameters in the model, path calibration coefficients are bound to transmission strength parameters of risk transmission paths, and range calibration coefficients are bound to calculation parameters of risk diffusion range. A preliminary calibration parameter set is generated, recording the correspondence between each topological parameter and its corresponding calibration coefficient; for example, the water level parameter of node A corresponds to node calibration coefficient 'a', and the transmission strength parameter of path B corresponds to path calibration coefficient 'b', etc. Through these bindings, the topological parameters can be dynamically adjusted based on the calibration coefficients during model runtime.

[0124] Step S138: Perform dynamic optimization processing on the preliminary calibration parameter set through multiple rounds of historical data simulation verification, adjust the values ​​of calibration coefficients and synergistic rules, so that the deviation between the topology simulation results and the actual historical risk records meets the preset requirements.

[0125] The initial calibration parameter set is dynamically optimized using multiple rounds of historical data simulation verification. In each round of verification, the initial calibration parameter set is applied to the river dynamic risk topology model to simulate historical risk events. The simulation results are compared with actual historical risk records, and the deviation is calculated. If the deviation does not meet the preset requirements, the values ​​of the calibration coefficients and the synergistic effect rules are adjusted according to the deviation, and then the next round of simulation verification is performed. This process is repeated until the deviation between the topology simulation results and the actual historical risk records meets the preset requirements, resulting in the optimized calibration parameter set.

[0126] Step S139: Integrate the optimized calibration coefficients, synergy rules, and structural parameter correspondences to generate a topology calibration parameter set.

[0127] The optimized calibration coefficients, synergy rules, and structural parameter correspondences are integrated to form a topology calibration parameter set. This set includes all calibration coefficients used to adjust the model's topology parameters, as well as the synergy rules between these coefficients and the correspondences between structural parameters and calibration coefficients. This topology calibration parameter set will be directly used to adjust the accuracy of risk transmission simulations, improving the model's accuracy in simulating real-world risk events.

[0128] Step S140: Input the real-time river engineering monitoring data into the river dynamic risk topology model, perform real-time simulation of risk transmission, and obtain the simulation results of risk transmission.

[0129] In the aforementioned watershed scenario, real-time river engineering monitoring data is collected in real time by monitoring equipment deployed along the riverbanks, including data on water level, flow velocity, water quality, and rainfall. This real-time data is processed according to the format requirements of the river engineering data deconstruction set and then input into the river dynamic risk topology model. The model adjusts its parameters using a topology calibration parameter set, and then performs real-time risk transmission simulation, simulating the entire process from the occurrence to the transmission of potential risks in the current river system. The final result of the risk transmission simulation includes information such as the intensity of the risk at each node, the transmission path, and the scope of impact.

[0130] Step S141: Activate the engineering data acquisition function of the multi-dimensional real-time monitoring equipment to continuously collect real-time data on river physical structure, hydrological movement, and environmental effects, and generate a real-time river monitoring data stream.

[0131] The engineering-engineered data acquisition function of multi-dimensional real-time monitoring equipment deployed within the watershed is activated. This equipment includes water level gauges, current meters, water quality sensors, rain gauges, and weather stations. Water level gauges and current meters are installed at different cross-sections of the river to collect real-time data on water level and flow velocity. Water quality sensors monitor parameters such as pollutant concentration and dissolved oxygen in the water. Rain gauges and weather stations collect environmental data such as rainfall, temperature, and wind speed. The above equipment continuously collects data at a set sampling frequency (e.g., once per minute), generating a real-time river monitoring data stream. Each data record in the data stream includes information such as the collection time, equipment identification, monitoring parameter name, and value, and is transmitted to the data processing center in real time via a data transmission network.

[0132] Step S142: Perform engineering transformation processing on each data point in the real-time river monitoring data stream. According to the representation standard of the river engineering data deconstruction set, the original real-time data is transformed into engineering real-time data to obtain the real-time engineering monitoring dataset.

[0133] Each data point in the real-time river monitoring data stream undergoes engineering transformation. Referring to the representation standards of the river engineering data deconstruction set, the raw real-time data is converted into a unified engineering format. For example, water level data is converted into a relative water level index (the ratio of the current water level to the historical average water level), flow velocity data into a relative velocity index, and rainfall data into precipitation impact factors. Simultaneously, unit standardization and outlier handling are performed, removing obviously unreasonable data (such as values ​​outside the normal range). After transformation, a real-time engineering monitoring dataset is obtained, containing real-time data with a structure consistent with the river engineering data deconstruction set, facilitating input into the river dynamic risk topology model for processing.

[0134] Step S143: Extract the core parameters from the real-time engineering monitoring dataset, and assign the core parameters to the corresponding topology nodes according to the topology node definition of the river dynamic risk topology model, so as to realize the binding of real-time parameters and topology nodes.

[0135] Core parameters are extracted from real-time engineering monitoring datasets. These parameters are essential for the model to simulate risk transmission, such as percentage water level fluctuation, flow fluctuation coefficient, and precipitation impact factor. Following the topology node definition of the river dynamic risk topology model, the core parameters are assigned to corresponding topology nodes based on their acquisition location and type. For example, the percentage water level fluctuation parameter for a specific cross-section is assigned to the topology node corresponding to that cross-section, and the precipitation impact factor for a specific region is assigned to the relevant topology nodes within that region. Through this assignment, real-time parameters are bound to topology nodes, ensuring that each topology node has the latest real-time attribute parameters.

[0136] Step S144: Call the parameter coupling operation logic in the river dynamic risk topology model, perform coupling operation on the bound real-time node parameters and the node coupling parameter set to generate real-time node coupling data. This real-time node coupling data reflects the collaborative state of nodes and paths under real-time parameters.

[0137] The parameter coupling operation logic in the river dynamic risk topology model is invoked. This logic calculates the synergistic effect between nodes and paths based on the real-time attribute parameters of nodes and the coupling relationships in the node coupling parameter set. For example, multiplying the node's water level fluctuation percentage parameter by the water level-discharge coupling coefficient in the node coupling parameter set yields the node's influence on the transmission intensity of related paths. By performing similar coupling operations on all relevant parameters, real-time node coupling data is generated, which details the synergistic state between each topological node and its connected paths under the current real-time parameters, such as the degree to which a node enhances or weakens the transmission intensity of the path.

[0138] Step S145: Based on real-time node coupling data, start the path transmission simulation function of the river dynamic risk topology model to simulate the transmission process of risk in topology nodes and transmission paths under real-time parameters, and obtain the real-time transmission simulation sequence.

[0139] Based on real-time node coupling data, the path transmission simulation function of the river dynamic risk topology model is initiated. This function simulates the process of risk transmission between topological nodes through paths, based on the collaborative state of nodes and paths in the real-time node coupling data and the transmission strength parameters of the paths. During the simulation, risk starts from the initial node and is transmitted to other nodes sequentially through each path according to the magnitude of the transmission strength parameter. Upon receiving the risk, each node determines whether to continue transmission and the transmission strength based on its own attribute parameters and node coupling data. The simulation results are output in the form of a real-time transmission simulation sequence, which includes information such as the risk intensity of each topological node and the transmission status of each path at different time points.

[0140] Step S1451: Extract the real-time coupling parameter value of each topology node from the real-time node coupling data. The real-time coupling parameter value reflects the state of the node under real-time parameters and the strength of its association with other nodes.

[0141] The real-time coupling parameter values ​​of each topology node are extracted from the real-time node coupling data. These parameter values ​​are obtained through parameter coupling operations and reflect the node's state under the current real-time parameters and the strength of its association with other nodes. For example, the real-time coupling parameter values ​​of a topology node include the node's own risk-bearing capacity, the risk acceptance coefficient of upstream nodes, and the risk transmission coefficient of downstream nodes. The magnitude of these parameter values ​​directly affects the risk transmission process between nodes. For example, the larger the transmission coefficient, the greater the possibility and intensity of the node transmitting risk to downstream nodes.

[0142] Step S1452: Based on the real-time coupling parameter values ​​of the topology nodes, determine the initial risk triggering state of each topology node. The triggering state reflects the specific conditions under which the node can initiate risk transmission.

[0143] Based on the real-time coupling parameter values ​​of the topology nodes, the initial risk triggering state of each node is determined. The initial risk triggering state is divided into two types: triggered and non-triggered. When the real-time coupling parameter value of a node meets certain conditions, the node is triggered and has the ability to initiate risk transmission. For example, when the percentage change in water level of a node exceeds a set threshold and the precipitation impact factor reaches a certain value, the initial risk triggering state of that node is triggered. The triggering state reflects the specific conditions under which a node can initiate risk transmission; only nodes in the triggered state will transmit risk to other connected nodes.

[0144] Step S1453: Extract the risk transmission paths and transmission intensity parameters from the river dynamic risk topology model, and construct a path network for real-time transmission simulation. This path network contains all executable risk transmission paths and their corresponding intensity information.

[0145] Risk transmission paths and transmission intensity parameters are extracted from the river dynamic risk topology model. These paths are constructed based on node relationships, and the transmission intensity parameters are calibrated using a topology calibration parameter set. A path network for real-time transmission simulation is constructed, with topology nodes as vertices and risk transmission paths as edges, each edge labeled with its corresponding transmission intensity parameter. The path network contains all executable risk transmission paths and their corresponding intensity information.

[0146] Step S1454: In accordance with the association order of the topology nodes, the transmission function of the topology nodes with the initial risk trigger state is activated in sequence, so that the risk is transmitted from the initial node to the associated nodes.

[0147] Following the order of the topology nodes' relationships (e.g., from upstream to downstream, or starting from the risk source node), the initial risk triggering status of each topology node is checked sequentially. For topology nodes with a triggering status, their propagation function is activated, transmitting the risk to associated nodes along the path in the path network. During propagation, the risk intensity is attenuated or amplified according to the propagation intensity parameter of the path. For example, a propagation intensity parameter of 0.6 means that when the risk is propagated through this path, its intensity is 60% of the initial intensity.

[0148] Step S1455: During the risk transmission process, the transmission intensity change data of each path is collected in real time. Combined with the dynamic update of real-time node coupling data, the path transmission intensity parameters are adjusted to keep the transmission process consistent with the real-time status.

[0149] During risk transmission, real-time data on transmission intensity changes for each path is collected, including how transmission intensity changes over time. Simultaneously, real-time node coupling data may be dynamically updated with new monitoring data input, affecting the collaborative state between nodes and paths. Based on the updated transmission intensity data and real-time node coupling data, path transmission intensity parameters are dynamically adjusted to ensure the risk transmission process aligns with the real-time state of the river system. For example, if the real-time coupling parameter value of a node increases, it may be necessary to correspondingly increase the transmission intensity parameters of paths connected to that node.

[0150] Step S1456: Record the time, intensity, and state changes of each topology node receiving risk transmission, and generate a node transmission record. This node transmission record describes in detail the changes of the node during risk transmission.

[0151] Record the time, intensity, and state changes of each topology node that receives risk transmission. Reception time refers to the moment a node begins to receive risk; transmission intensity refers to the numerical value of the received risk; node state changes include changes in the node's risk-bearing capacity and trigger state. Organize this information into node transmission records, with each record corresponding to the changes of a node during the risk transmission process, detailing the entire process from receiving risk to processing risk and then transmitting risk.

[0152] Step S1457: Based on node transmission records and path transmission intensity change data, generate a transmission process data sequence for each risk transmission path. This transmission process data sequence contains transmission status information at different time points.

[0153] Based on node transmission records and path transmission intensity change data, a transmission process data sequence is generated for each risk transmission path. This transmission process data sequence is time-axis-based, recording the transmission status information of the path at different points in time, such as transmission intensity, whether risk is being transmitted, the starting node, and the target node. Through the transmission process data sequence, the dynamic changes of each path during the risk transmission process can be understood.

[0154] Step S1458: Perform time synchronization processing on the data sequence of the transmission process of all paths to obtain a time-synchronized transmission sequence. Extract the key transmission events in the time-synchronized transmission sequence, including the specific content of risk transmission initiation, transmission intensity peak, and transmission path switching as key transmission events.

[0155] Time synchronization processing is performed on the transmission process data sequences of all paths, unifying the transmission status information of different paths onto the same timeline to obtain a time-synchronized transmission sequence. Key transmission events are extracted from this sequence; these events represent crucial nodes in the risk transmission process. The risk transmission initiation event is the moment when the risk begins to transmit along a certain path; the peak transmission intensity event is the moment when the transmission intensity of a path reaches its maximum value; and the transmission path switching event is the moment when the risk switches from one path to another. Extracting these key transmission events helps in analyzing the patterns and characteristics of risk transmission.

[0156] Step S1459: Integrate the time synchronization transmission sequence, node transmission records, and key transmission event information to obtain the real-time transmission simulation sequence.

[0157] By integrating time-synchronized transmission sequences, node transmission records, and key transmission event information, a real-time transmission simulation sequence is formed. This real-time transmission simulation sequence comprehensively reflects the real-time process of risk transmission through topological nodes and paths in the river system, including multi-dimensional information such as time, nodes, paths, and transmission intensity.

[0158] Step S146: Extract the node calibration coefficients and path calibration coefficients from the topology calibration parameter set, perform accuracy calibration processing on the real-time transmission simulation sequence, correct the deviations in the simulation process, and obtain the calibrated transmission simulation sequence.

[0159] Node calibration coefficients and path calibration coefficients are extracted from the topology calibration parameter set and applied to the real-time transmission simulation sequence. For node-related simulation data (such as node risk intensity), node calibration coefficients are used for correction; for path-related simulation data (such as path transmission intensity), path calibration coefficients are used for correction. Through this accuracy calibration process, deviations caused by model parameter settings or initial conditions during the simulation are corrected, resulting in a calibrated transmission simulation sequence and improving the accuracy of the simulation results.

[0160] Step S147: Perform time-dimensional continuous processing on the calibrated transmission simulation sequence to convert the discrete simulation data into a continuous risk transmission trend curve, which describes the changes in risk transmission over time.

[0161] The calibrated simulated sequence is then processed to be continuous in the time dimension. Since the simulated data is usually collected at discrete time points, it needs to be converted into a continuous function curve using methods such as interpolation. For example, linear interpolation or spline interpolation methods can be used to generate a continuous risk transmission trend curve based on discrete time points and corresponding risk intensity values. This risk transmission trend curve, with time on the horizontal axis and risk intensity on the vertical axis, intuitively describes the changes in risk transmission at different time points, including characteristics such as risk increase, decrease, and peak.

[0162] Step S148: Based on the risk transmission trend curve, extract the risk transmission status data of key time nodes. The risk transmission status data of key time nodes includes node risk intensity, path transmission efficiency, and risk coverage.

[0163] Based on the risk transmission trend curve, key time nodes are identified, such as the time when the risk intensity reaches its peak, the time when the risk begins to spread, and the time when the risk's impact range is greatest. At these key time nodes, risk transmission status data is extracted, including node risk intensity (the risk intensity value of each topological node at that time), path transmission efficiency (the risk transmission efficiency of each path at that time, such as the amount of risk transmitted per unit time), and risk coverage (the range of river areas affected by the risk at that time). This data can help analyze the state and characteristics of the risk at critical moments.

[0164] Step S149: Integrate the risk transmission status data at key time points with the topological structure information of the river dynamic risk topology model to generate a complete data record of the risk transmission process.

[0165] This method integrates risk transmission status data at key time points with the topological information of the river dynamic risk topology model. The topological information includes the geographical location of nodes and the connectivity of pathways. Through this integration, the risk transmission status data is mapped to specific nodes and pathways, generating a complete data record of the risk transmission process. This complete data record contains detailed information on when, where, with what intensity, and along what pathway the risk was transmitted.

[0166] Step S1410: Integrate the risk transmission trend curve, risk transmission status data at key time nodes, and complete data records to obtain the risk transmission simulation results.

[0167] By integrating risk transmission trend curves, risk transmission status data at key time points, and complete data records of the risk transmission process, a risk transmission simulation result is formed. This simulation result comprehensively reflects the transmission process and characteristics of risk in the river system, including the temporal trend of risk changes, spatial distribution, and the status of key nodes.

[0168] Step S150: Perform threshold adaptation processing on the risk transmission simulation results based on the topology calibration parameter set. When the simulation results exceed the adaptation threshold range, generate dynamic river risk early warning information containing the engineering area affected by the risk and the engineering coordinates of the transmission path.

[0169] In the aforementioned watershed scenario, threshold adaptation processing is performed on the risk transmission simulation results based on the topology calibration parameter set. First, an adaptation threshold range is set according to the calibration coefficients in the topology calibration parameter set and the accuracy requirements of the engineering simulation. This range includes node risk thresholds, path transmission thresholds, and range coverage thresholds. Then, the risk transmission simulation results are compared with the adaptation threshold range. If the node risk intensity, path transmission efficiency, or risk coverage in the simulation results exceeds the corresponding threshold range, dynamic river risk early warning information is generated. The early warning information includes the engineering area affected by the risk (such as a specific river section or riverside area) and the engineering coordinates of the transmission path (such as the coordinates of the path's start and end points, and the coordinates of key turning points), so that relevant departments can take countermeasures.

[0170] For example, in step S151: extract node calibration coefficients, path calibration coefficients and range calibration coefficients from the topology calibration parameter set, and set threshold calculation logic in combination with the accuracy requirements of engineering simulation. This threshold calculation logic can generate an adaptive threshold corresponding to the risk transmission simulation results.

[0171] Node calibration coefficients, path calibration coefficients, and range calibration coefficients are extracted from the topology calibration parameter set. These coefficients reflect the calibration status of the model parameters. A threshold calculation logic is established based on the accuracy requirements of the engineering simulation, such as the maximum allowable node risk intensity, maximum path transmission efficiency, and maximum risk coverage. This threshold calculation logic calculates an appropriate threshold corresponding to the risk transmission simulation results based on the calibration coefficients and accuracy requirements. For example, the node risk threshold = baseline node risk value × node calibration coefficient × accuracy adjustment coefficient, where the baseline node risk value is determined based on historical data and engineering standards, and the accuracy adjustment coefficient is set according to the accuracy requirements of the engineering simulation.

[0172] Step S152: Input the risk transmission trend curve and the risk transmission status data of key time nodes from the risk transmission simulation results into the threshold calculation logic to obtain the initial adaptation threshold range. The initial adaptation threshold range includes the node risk threshold, the path transmission threshold, and the range coverage threshold.

[0173] The risk transmission trend curve and risk transmission status data at key time nodes from the risk transmission simulation results are input into the threshold calculation logic. The risk transmission trend curve provides information on how risk changes over time, while the risk transmission status data at key time nodes provides information on the risk status at a specific moment. Based on this data and the established calculation rules, the threshold calculation logic calculates an initial adaptive threshold range. This initial adaptive threshold range includes the node risk threshold (the maximum allowable risk intensity for each topology node), the path transmission threshold (the maximum allowable transmission efficiency for each path), and the range coverage threshold (the maximum allowable risk coverage area).

[0174] Step S153: Extract the data association attribute table from the river engineering data deconstruction set, and perform association adjustment on the initial adaptation threshold range to ensure that the threshold range matches the association type and degree of influence between different data.

[0175] A data association attribute table is extracted from the deconstructed set of river engineering data. This table records the association types and influence levels between different data points. An association adjustment is performed on the initial fitting threshold range, adjusting the threshold range based on the information in the data association attribute table. For example, if the risk intensity of a node has a strong positive correlation with the flow fluctuation of an upstream node, and the influence is high, then the risk threshold of that node should be adjusted according to the flow threshold of the upstream node to match the threshold range of the associated data. Through this association adjustment, the fitting threshold range becomes more reasonable and accurately reflects the mutual influence between different data points.

[0176] Step S154: Based on the topological parameters of the river dynamic risk topology model, perform spatial adaptation processing on the adjusted adaptation threshold range, and associate the threshold range with the engineering coordinates of the topological nodes to obtain the spatialized adaptation threshold.

[0177] Based on the topological parameters of the river dynamic risk topology model, such as the geographical coordinates of topological nodes and the spatial distribution of paths, spatial adaptation processing is performed on the adjusted adaptation threshold range. The threshold range is associated with the engineering coordinates of the topological nodes; for example, a node risk threshold corresponding to its geographical location is assigned to each topological node, and a path transmission threshold corresponding to its spatial location is assigned to each path. Through spatial adaptation processing, a spatialized adaptation threshold is obtained, giving the threshold range a clear spatial attribute and accurately reflecting the risk threshold requirements of different geographical locations.

[0178] Step S155: Compare the risk coverage data in the risk transmission simulation results with the range coverage threshold in the spatialization adaptation threshold to determine whether the risk coverage exceeds the threshold limit.

[0179] The risk coverage data (such as the river length, area, and coordinate range of the specific region affected by the risk) in the risk transmission simulation results are compared with the range coverage threshold in the spatialization adaptation threshold. It is then checked whether the risk coverage exceeds the area defined by the range coverage threshold. If the coordinate range of the risk coverage exceeds the coordinate range of the range coverage threshold, or if the affected area / length exceeds the threshold value, then the risk coverage is determined to exceed the threshold limit.

[0180] Step S156: Compare the node risk intensity data in the risk transmission simulation results with the node risk threshold in the spatialization adaptation threshold to determine whether the node risk intensity exceeds the threshold standard.

[0181] The risk intensity data of each topological node in the risk transmission simulation results are compared with the node risk threshold of the corresponding node in the spatialization adaptation threshold. Each node's risk intensity is checked individually to see if it exceeds its corresponding node risk threshold. If a node's risk intensity value is greater than its node risk threshold, then the node's risk intensity is determined to exceed the threshold standard.

[0182] Step S157: Compare the path transmission efficiency data in the risk transmission simulation results with the path transmission threshold in the spatialization adaptation threshold to check whether the path transmission efficiency exceeds the threshold range.

[0183] The transmission efficiency data of each path in the risk transmission simulation results are compared with the corresponding path transmission threshold in the spatialization adaptation threshold. The path transmission threshold is usually a range. The path transmission efficiency is checked to see if it is within this range. If the path transmission efficiency value is less than the lower limit of the threshold range or greater than the upper limit, it is determined that the path transmission efficiency exceeds the threshold range.

[0184] Step S158: When any one of the risk coverage, node risk intensity, or path transmission efficiency exceeds the corresponding threshold, extract the key time node, topology node, and transmission path information corresponding to the excess item.

[0185] When any of the risk coverage, node risk intensity, or path transmission efficiency exceeds the corresponding threshold, the system automatically extracts relevant information for the excess item. For the case where the risk coverage exceeds the threshold, the system extracts the key time point at which the risk coverage begins to exceed the threshold, the involved topology nodes, and related transmission path information. For the case where the node risk intensity exceeds the threshold, the system extracts the key time point at which the node's risk intensity reaches the threshold and the transmission path information related to that node. For the case where the path transmission efficiency exceeds the threshold, the system extracts the key time point at which the path transmission efficiency exceeds the threshold, and the starting and target topology nodes of the path.

[0186] Step S159: Based on the extracted key time nodes, topological nodes and transmission path information, and combined with the engineering coordinate data in the river dynamic risk topology modeling model, determine the engineering coordinate sequence of the engineering area boundary and transmission path of the risk impact.

[0187] Based on the extracted key time nodes, topological nodes, and transmission path information, and combined with engineering coordinate data (such as the latitude and longitude coordinates of topological nodes and spatial coordinates of the path) from the river dynamic risk topology modeling model, the boundary of the engineering area affected by the risk is determined. The boundary of the engineering area is determined by connecting the coordinate points of the affected topological nodes or by the coordinate range of the risk coverage area. Simultaneously, the engineering coordinate sequence of the transmission path is determined, which includes key coordinate points on the transmission path, such as the starting point coordinates, ending point coordinates, and turning point coordinates, forming a complete path coordinate chain.

[0188] Step S1510: Integrate the engineering area boundary affected by the risk, the engineering coordinate sequence of the transmission path, and the specific data exceeding the threshold, and generate dynamic river risk early warning information containing the engineering area affected by the risk and the engineering coordinates of the transmission path in accordance with the requirements of the engineering early warning information format.

[0189] The system integrates the engineering area boundaries affected by the risk (such as the set of boundary coordinates and area names), the engineering coordinate sequence of the transmission path (such as the latitude and longitude of key coordinate points along the path), and specific data exceeding the threshold (such as the magnitude of the exceeded risk and the extent of the exceeded risk). Following the requirements for engineering-based early warning information formats, this information is organized into a standardized text format, including the warning level, risk type, affected area, transmission path, and threshold exceedance details, to generate dynamic river risk early warning information. This information is then used to promptly notify relevant departments and personnel to take preventative and response measures.

[0190] In some embodiments, the river dynamic risk early warning system based on multi-source data used to perform the above methods can be any electronic device with data computing, processing, and storage functions. This river dynamic risk early warning system based on multi-source data can be used to implement the text processing methods or text processing model processing methods provided in the above embodiments.

[0191] Typically, a river dynamic risk early warning system based on multi-source data includes a processor and a memory. The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory is used to store a computer program configured to be executed by one or more processors to implement the above-described text processing method or text processing model.

[0192] This application provides a computer program product, which includes computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are executed by a processor, the processor will execute the river dynamic risk early warning method based on multi-source data provided in this application.

[0193] This application provides a computer-readable storage medium storing computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are executed by a processor, the processor will execute the river dynamic risk early warning method based on multi-source data provided in this application.

Claims

1. A method for dynamic risk early warning of rivers based on multi-source data, characterized in that, The method includes: The engineering deconstruction processing of multi-source river monitoring data is performed to generate a set of river engineering data deconstruction. This set of river engineering data deconstruction includes the engineering representation forms and data association attributes of river physical structure data, hydrological movement data, and environmental action data. A dynamic risk topology model for rivers is established based on the deconstruction set of river engineering data. This dynamic risk topology model for rivers completes the engineering simulation of risk transmission through node association, path mapping and parameter coupling. The dynamic adaptation process of the river dynamic risk topology model is invoked to perform topology calibration on historical risk data, generating a topology calibration parameter set, which is directly used to adjust the accuracy of risk transfer simulation. Real-time river engineering monitoring data is input into the river dynamic risk topology model, and real-time simulation of risk transmission is performed to obtain the simulation results of risk transmission. Based on the topology calibration parameter set, threshold adaptation processing is performed on the simulation results of risk transmission. When the simulation results exceed the adaptation threshold range, dynamic risk warning information of the river, including the engineering area affected by the risk and the engineering coordinates of the transmission path, is generated.

2. The river dynamic risk early warning method based on multi-source data according to claim 1, characterized in that, The engineering deconstruction processing of multi-source river monitoring data generates a set of river engineering data deconstructions, including: Collect raw monitoring data of multiple sources of rivers from riverside monitoring equipment, mobile monitoring platforms and remote sensing monitoring systems to generate a raw data set of multiple sources of rivers. This raw data set of multiple sources of rivers contains multi-dimensional monitoring records of the river itself and its surrounding environment. Extract relevant data describing the inherent structure of rivers from the original dataset of multi-source rivers, and split the data dimensions according to the engineering structure analysis standard to obtain the split structure-related data, which includes cross-sectional dimension description, riverbed material description, and riverbank structure description. The decomposed structural data is transformed using engineering parameters to convert natural language descriptions and raw measurement data into a quantifiable representation that can be computed in engineering design, thus obtaining river physical structure data. The relevant data describing the water flow motion state in the original dataset of multi-source rivers are screened, and the data types are divided according to the hydrological engineering motion analysis specifications to obtain the divided motion-related data, which includes descriptions of water flow velocity, water level changes, and flow fluctuations. The segmented motion-related data are mapped using engineering indicators, converting the original monitoring values ​​into standardized indicator forms that can be used in engineering simulations, thus obtaining hydrological motion data. The relevant data describing external effects on rivers are separated from the original dataset of multi-source rivers. The data are then classified according to the classification standard of environmental engineering effects to obtain classified environmental data. The classified environmental data includes descriptions of precipitation effects, topographic effects, and human activity effects. The classified environmental data are subjected to engineering factor transformation to convert the description of external effects into an impact factor form that can be applied in engineering analysis, thus obtaining environmental effect data. Interaction mining operations are performed on river physical structure data, hydrological motion data, and environmental action data. The types of associations and the degree of influence between different data are determined by data feature comparison and trend correlation analysis. The determined association types and degree of influence are associated and bound with river physical structure data, hydrological movement data, and environmental action data to generate a data association attribute table; By integrating river physical structure data, hydrological movement data, environmental impact data, and data association attribute tables, a set of river engineering data deconstruction is generated.

3. The river dynamic risk early warning method based on multi-source data according to claim 1, characterized in that, The method for establishing a dynamic risk topology model for rivers based on the deconstruction set of river engineering data includes: The core parameters of the river physical structure data are extracted from the river engineering data deconstruction set. The river structure location corresponding to the core parameters is defined as a topological node. Each topological node contains structural parameter description and location representation information. Extract the data association attribute table from the river engineering data deconstruction set, determine the association relationship between topological nodes based on the association type and the degree of influence, and generate a node association list. This node association list records the connection logic between any two topological nodes. Based on the connection logic in the node association list, risk transmission paths between topology nodes are constructed. Each risk transmission path corresponds to a set of node associations. The path description of the risk transmission path includes the data transmission direction and the impact method. Engineering indicators of hydrological motion data and environmental action data are extracted from the river engineering data deconstruction set. These engineering indicators are then used as attribute parameters of topology nodes to achieve the binding of attribute parameters with topology nodes. Based on the attribute parameters of topological nodes and the degree of influence in the node association list, the transmission strength parameter of each risk transmission path is quantified. The transmission strength parameter reflects the degree of contribution of each risk transmission path to risk transmission. Perform coupling operations on the attribute parameters, transmission strength parameters, and data association attributes of the topology nodes to generate a node coupling parameter set, which describes the synergistic relationship between node attributes and path transmission. Set up a dynamic topology adjustment instruction receiving process to receive external data update instructions and perform dynamic adaptation operations such as adding or removing topology nodes, optimizing paths, and adjusting parameters. Based on topology nodes, node association lists, risk transmission paths, node coupling parameter sets, and dynamic adjustment instruction receiving processes, an initial topology modeling structure is constructed, which has the basic functions of risk transmission simulation. An engineering adaptation operation is performed on the initial topology modeling structure, incorporating the structural mechanics principles and hydrological motion-related technical points in river engineering analysis, so that the initial topology modeling structure meets the accuracy requirements of engineering simulation. By integrating and optimizing the topological nodes, relationships, risk transmission paths, node coupling parameter sets, and dynamic adjustment instruction receiving processes, a dynamic risk topology model for rivers is generated, which simulates risk transmission through node association, path mapping, and parameter coupling.

4. The river dynamic risk early warning method based on multi-source data according to claim 1, characterized in that, The dynamic adaptation process of calling the river dynamic risk topology model performs topology calibration on historical risk data to generate a topology calibration parameter set, including: Collect complete engineering monitoring data and risk development records recorded during the occurrence of historical river dynamic risks, and generate a historical river dynamic risk engineering dataset. This historical river dynamic risk engineering dataset contains multi-dimensional engineering data before and after the occurrence of risks and details of risk evolution. The historical river dynamic risk engineering dataset is standardized to obtain a standardized historical engineering dataset. Key node parameters, path transmission data and risk impact range data at the time of risk occurrence are extracted from the standardized historical engineering dataset to generate a core historical risk dataset. Input the core historical risk dataset into the dynamic adaptation process of the river dynamic risk topology model, start the topology simulation operation, and obtain the historical risk topology simulation results. The historical risk topology simulation results include the simulated node association status, path transmission strength and risk diffusion trend. Extract actual risk development records from the core historical risk dataset and dynamically compare them with the historical risk topology simulation results to determine the deviation data between the simulation results and the actual situation. The deviation data includes node parameter deviation, path transmission deviation, and range prediction deviation. Substitute the deviation data into the preset calibration calculation process, set the calibration target based on the accuracy requirements of engineering simulation, and generate node calibration coefficients, path calibration coefficients and range calibration coefficients by reverse derivation of data deviations. The node calibration coefficient, path calibration coefficient, and range calibration coefficient are integrated and correlated. Based on the parameter coupling relationship in the river dynamic risk topology model, the synergistic effect rules among the calibration coefficients are determined. Based on the synergistic effect rule, various calibration coefficients are bound to the topological parameters of the river dynamic risk topology model to generate a preliminary calibration parameter set, which contains the correspondence between structural parameters and calibration coefficients. The initial calibration parameter set was dynamically optimized through multiple rounds of historical data simulation verification, adjusting the values ​​of calibration coefficients and synergistic rules to ensure that the deviation between the topology simulation results and the actual historical risk records met the preset requirements. By integrating and optimizing the calibration coefficients, synergy rules, and structural parameter correspondences, a topology calibration parameter set is generated.

5. The river dynamic risk early warning method based on multi-source data according to claim 1, characterized in that, The process involves inputting real-time river engineering monitoring data into a river dynamic risk topology model, performing real-time risk transmission simulation, and obtaining the risk transmission simulation results, including: Activate the engineering data acquisition function of the multi-dimensional real-time monitoring equipment to continuously collect real-time data on river physical structure, hydrological movement, and environmental effects, and generate a real-time river monitoring data stream. Each data point in the real-time river monitoring data stream is subjected to engineering transformation processing. According to the representation standard of the river engineering data deconstruction set, the original real-time data is converted into engineering real-time data to obtain the real-time engineering monitoring dataset. Extract the core parameters from the real-time engineering monitoring dataset, and assign the core parameters to the corresponding topology nodes according to the topology node definition of the river dynamic risk topology model, so as to achieve the binding of real-time parameters with topology nodes; The parameter coupling operation logic in the river dynamic risk topology model is called to perform coupling operation on the bound real-time node parameters and the node coupling parameter set to generate real-time node coupling data. This real-time node coupling data reflects the collaborative state of nodes and paths under real-time parameters. Based on real-time node coupling data, the path transmission simulation function of the river dynamic risk topology model is activated to simulate the transmission process of risk in topology nodes and transmission paths under real-time parameters, and obtain the real-time transmission simulation sequence. Extract the node calibration coefficients and path calibration coefficients from the topology calibration parameter set, perform accuracy calibration processing on the real-time transmission simulation sequence, correct the deviations in the simulation process, and obtain the calibrated transmission simulation sequence; The time dimension of the calibrated transmission simulation sequence is continuously processed to transform the discrete simulation data into a continuous risk transmission trend curve, which describes how the risk changes over time. Based on the risk transmission trend curve, risk transmission status data at key time nodes are extracted. The risk transmission status data at key time nodes includes node risk intensity, path transmission efficiency, and risk coverage. By associating and integrating the risk transmission status data at key time points with the topological structure information of the river dynamic risk topology model, a complete data record of the risk transmission process is generated. By integrating the risk transmission trend curve, risk transmission status data at key time points, and complete data records, the risk transmission simulation results are obtained.

6. The river dynamic risk early warning method based on multi-source data according to claim 2, characterized in that, The process involves performing engineering parameter transformation on the split structure-related data, converting natural language descriptions and raw measurement data into a quantifiable representation that can be computed in engineering design, to obtain river physical structure data, including: This paper collects quantitative standards for commonly used river physical parameters in engineering structural design. These standards include quantitative methods and representation forms for cross-sectional dimensions, riverbed materials, and riverbank structures. Extract the natural language description content from the structure-related data after splitting, perform semantic parsing on each natural language description, and determine the physical parameter type and feature information corresponding to each natural language description; Based on the semantic parsing results, and in accordance with the quantitative standards for river physical parameters, the natural language description is converted into the corresponding quantitative parameter expression, which includes the parameter name, quantitative value, and engineering unit. Extract the original measurement data from the structural data after splitting, perform unified unit processing on the original measurement data, and convert data from different measurement units into standard units commonly used in engineering design; According to the accuracy requirements of engineering calculations, the original measurement data after unit unification is retained effectively and the error is corrected to obtain accuracy-optimized measurement data. The quantitative parameter expression is associated and matched with the accuracy-optimized measurement data. Through data comparison and adjustment, the language description conversion result of the same physical parameter is kept consistent with the measurement data, and a parameter matching set is generated. Based on engineering common sense data of river physical structure, a rationality analysis is performed on the quantitative parameters in the parameter matching set, and quantitative parameters that do not conform to engineering logic are eliminated. The parameter matching set after rationality analysis is then structured and grouped according to the categories of cross-sectional size, riverbed material, and riverbank structure to generate a structural parameter group set. Add an engineering identifier to the quantized parameter in each structural parameter group set to identify the engineering application scenario and calculation purpose of the parameter, thus obtaining the identified structural parameters. By integrating identifiable structural parameters, parameter matching sets, and engineering identifier descriptions, river physical structure data in a quantifiable form that can be calculated in engineering design is generated.

7. The river dynamic risk early warning method based on multi-source data according to claim 3, characterized in that, The transmission strength parameters for each risk transmission path, based on the attribute parameters of the topology nodes and the influence of the node association list, include: Core indicators related to risk transmission are extracted from the attribute parameters of topological nodes. The specific contents of these core indicators are node structural stability parameters, hydrological carrying capacity parameters, and environmental interference resistance parameters. For each core indicator, a quantitative grading process is performed. According to the needs of engineering simulation, the numerical range of the core indicators is divided into different level intervals, and each level interval corresponds to a quantitative score. Extract the influence description from the node association list, convert the influence description into the corresponding influence coefficient, and ensure that the value of the influence coefficient is consistent with the actual association of influence. The logic for quantifying transmission strength is set up. This logic takes the quantified score of the core indicator, the influence coefficient and the path length parameter as input, and generates the transmission strength parameter through engineering calculation. Substitute the core indicator quantified scores of the two topological nodes corresponding to each risk transmission path into the transmission intensity quantification calculation logic to obtain the node contribution score; Input the influence coefficient and path length parameters corresponding to the node association list into the transmission intensity quantification calculation logic to obtain the path influence score; By performing collaborative calculations on the node contribution score and path impact score through the transmission intensity quantification calculation logic, the initial transmission intensity parameters for each risk transmission path are generated. Collect transmission strength reference standards in engineering risk simulation, compare the initial transmission strength parameters with the transmission strength reference standards, and determine the direction of parameter adjustment; Based on the adjustment direction, the initial transmission strength parameters are dynamically corrected so that the corrected transmission strength parameters meet the actual needs of engineering simulation and the transmission strength reference standard. The modified transmission strength parameters are normalized to obtain transmission strength parameters that reflect the contribution of the path to risk transmission.

8. The river dynamic risk early warning method based on multi-source data according to claim 4, characterized in that, The process of substituting deviation data into a preset calibration calculation flow, setting calibration targets based on the accuracy requirements of engineering simulation, and generating node calibration coefficients, path calibration coefficients, and range calibration coefficients through reverse derivation of data deviations includes: The accuracy requirements of engineering simulation are analyzed, and the allowable deviation ranges of node parameters, path transmission and range prediction in risk transmission simulation are determined. The allowable deviation range is used as the core basis for calibration targets. Define the basic execution logic of the calibration operation process, which corresponds to the operation steps of node deviation calibration, path deviation calibration and range deviation calibration respectively; The node parameter deviation in the deviation data is input into the node deviation calibration step of the calibration operation process. This step, based on the calibration target, calculates the adjustment ratio of the node parameters through an engineering algorithm to generate the initial node calibration coefficient. The path transfer deviation in the deviation data is input into the path deviation calibration step of the calibration calculation process. This step calculates the adjustment ratio of the path transfer parameters based on the calibration target and the engineering characteristics of path transfer, and generates the initial path calibration coefficient. The range prediction deviation in the deviation data is input into the range deviation calibration step of the calibration operation process. This step, with reference to the calibration target and based on the key engineering techniques of risk diffusion, calculates the adjustment ratio of the range prediction parameters and generates the initial range calibration coefficient. Extract the parameter coupling relationships in the river dynamic risk topology model, and perform correlation verification on the initial node calibration coefficient, initial path calibration coefficient, and initial range calibration coefficient to ensure that the calibration coefficients conform to the parameter coupling logic; Based on the correlation verification results, the values ​​of the initial calibration coefficients are adjusted, and the calibration coefficient optimization target operation logic is set. This calibration coefficient optimization target operation logic aims to minimize the deviation between the simulation results and the actual historical data, and performs optimization operation on the adjusted calibration coefficients. Through multiple rounds of iterative calculations, the calibration coefficient optimization target calculation logic reaches a convergent state, resulting in the optimized node calibration coefficient, path calibration coefficient, and range calibration coefficient. By integrating and optimizing the node calibration coefficients, path calibration coefficients, range calibration coefficients, and association rules during the calibration process, node calibration coefficients, path calibration coefficients, and range calibration coefficients are generated.

9. The river dynamic risk early warning method based on multi-source data according to claim 5, characterized in that, The method, based on real-time node coupling data, initiates the path transmission simulation function of the river dynamic risk topology model to simulate the transmission process of risk in topology nodes and transmission paths under real-time parameters, obtaining a real-time transmission simulation sequence, including: Extract the real-time coupling parameter value of each topology node from the real-time node coupling data. This real-time coupling parameter value reflects the state of the node under real-time parameters and the strength of its association with other nodes. Based on the real-time coupling parameter values ​​of the topology nodes, the initial risk triggering state of each topology node is determined. The triggering state reflects the specific conditions under which the node has the ability to initiate risk transmission. Extract risk transmission paths and transmission intensity parameters from the river dynamic risk topology model, and construct a path network for real-time transmission simulation. This path network contains all executable risk transmission paths and their corresponding intensity information. According to the association order of the topology nodes, the transmission function of the topology nodes with the initial risk triggering state is activated in sequence, so that the risk is transmitted from the initial node to the associated nodes. During the risk transmission process, the transmission intensity change data of each path is collected in real time. Combined with the dynamic update of real-time node coupling data, the path transmission intensity parameters are adjusted to keep the transmission process consistent with the real-time status. Record the time, intensity, and state changes of each topology node receiving risk transmission, and generate a node transmission record. This node transmission record describes in detail the changes of the node during risk transmission. Based on node transmission records and path transmission intensity change data, a transmission process data sequence is generated for each risk transmission path. This transmission process data sequence contains transmission status information at different time points. Perform time synchronization processing on the data sequence of the transmission process of all paths to obtain a time-synchronized transmission sequence. Extract the key transmission events in the time-synchronized transmission sequence, such as risk transmission initiation, transmission intensity peak, and transmission path switching, as the specific content of the key transmission events. By integrating the time synchronization transmission sequence, node transmission records, and key transmission event information, a real-time transmission simulation sequence is obtained.

10. A river dynamic risk early warning system based on multi-source data, characterized in that, The method includes a processor and a computer-readable storage medium storing machine-executable instructions, which, when executed by a computer, implement the river dynamic risk early warning method based on multi-source data as described in any one of claims 1-9.

Images