Method, system, device and program for dynamic assessment of risk of chain disaster of dangerous rock on bank

Abstract

The application discloses a kind of bank dangerous rock chain generation disaster risk dynamic evaluation method, system, equipment and procedure, it is related to rock mass monitoring technical field, the method includes: obtaining multi-source dynamic monitoring data;Based on the multi-source dynamic monitoring data, construct bank dangerous rock disaster chain dynamic directed graph model;Obtain key environmental driving factor and its influence path;Respectively calculate the dynamic influence weight of the key environmental driving factor to each directed edge in the bank dangerous rock disaster chain dynamic directed graph model;Based on the dynamic influence weight, dynamically weight directed edge, and based on the directed edge intensity after dynamic weighting, calculate comprehensive risk value in real time.Based on the processing scheme of the present application, the quantitative analysis of the whole process of chain generation disaster dynamic risk is realized, the uncertainty of model initialization is fundamentally overcome, and the reliability of long-term application and the continuous improvement of evaluation accuracy are ensured.

Description

Technical Field

[0001] This invention relates to the field of rock mass monitoring technology, and in particular to a method, system, equipment and program for dynamic assessment of the risk of chain-like disasters caused by unstable rock masses on reservoir banks. Background Technology

[0002] The reservoir bank area has a complex geological structure, and under the coupled effects of periodic water level fluctuations, rainfall, weathering, and other internal and external dynamics, numerous unstable rock masses often develop. The instability of these unstable rock masses is not an isolated event and can easily trigger chain-reaction disasters. These disasters are characterized by their suddenness, chain-like evolution, wide affected area, and great destructive power, posing a serious threat to navigation, infrastructure, the lives and property of residents along the riverbank, and the reservoir's operational management. Therefore, accurate and timely assessment of the risk of chain-reaction disasters caused by unstable rock masses along the reservoir bank is a crucial prerequisite for achieving scientific early warning and proactive prevention and control.

[0003] Currently, the relevant risk assessment technologies mainly have the following limitations and drawbacks: First, the assessment focuses on a single object, neglecting the chain reaction and systemic nature of disasters.

[0004] Existing technologies mostly focus on the stability assessment of the unstable rock mass itself, using limit equilibrium methods, numerical simulations, or qualitative-semi-quantitative index systems for static analysis. These methods treat collapse as the terminal event, completely severing its attribute as the initiating link of the disaster chain. They fail to conduct correlation analysis and quantitative assessment of subsequent derivative disasters such as surges and secondary landslides that may be triggered. This leads to one-sided risk assessment conclusions that cannot reflect the real and expanded impact range that the disaster may cause, thus seriously underestimating the overall risk.

[0005] Second, the assessment process is static and rigid, and cannot respond to dynamic changes in the environment.

[0006] The stability of reservoir bank boulders and the development of cascading hazards are highly dependent on time-varying environmental driving factors such as reservoir water level and rainfall. However, most existing methods treat these factors as static background parameters or simple threshold conditions, such as performing stability calculations only at specific water levels. This results in assessments lagging behind environmental changes and failing to provide timely risk warnings during critical risk periods such as rapid decline in reservoir water level or continuous heavy rainfall.

[0007] Third, the methods of risk quantification are one-sided and fail to characterize the chain accumulation and amplification effects of risk.

[0008] There is also a risk prediction method that considers disaster chains, which uses a list or simple probability multiplication to predict risks. The risk value is usually expressed as an isolated probability or a vague level, making it difficult for decision-makers to accurately grasp the real-time level of risk, the main contributing path, and the evolution trend.

[0009] In summary, while the aforementioned methods for predicting dangerous rockfall risks can provide safety warnings to a certain extent, several shortcomings have been found in practical application, preventing them from achieving optimal results. These shortcomings can be summarized as follows: Therefore, it is evident that the existing methods for predicting the risk of unstable rockfalls still have inconveniences and shortcomings in use, and urgently need further improvement. How to create a new dynamic assessment method for the risk of chain-like disasters caused by unstable rockfalls along reservoir banks has become a pressing goal for the industry. Summary of the Invention

[0010] In view of this, the present disclosure provides a method for dynamic assessment of the risk of reservoir bank rock chain disasters, which at least partially solves the problems existing in the prior art.

[0011] In a first aspect, this disclosure provides a method for dynamic assessment of the risk of reservoir bank rock chain disasters, the method comprising the following steps: Acquire multi-source dynamic monitoring data; wherein, the multi-source dynamic monitoring data includes: three-dimensional displacement time series data of the unstable rock mass, acoustic emission signal of the unstable rock mass, hydrological data, meteorological data, and width variation data of the main control fracture of the unstable rock mass; A dynamic directed graph model of the reservoir bank dangerous rock disaster chain is constructed based on the aforementioned multi-source dynamic monitoring data; Obtain key environmental driving factors and their impact paths; calculate the dynamic impact weight of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank dangerous rock disaster chain; The directed edges are dynamically weighted based on the dynamic influence weights, and the comprehensive risk value is calculated based on the real-time strength of the dynamically weighted directed edges.

[0012] According to a specific implementation of this disclosure, the step of constructing a dynamic directed graph model of the reservoir bank rockfall disaster chain based on the multi-source dynamic monitoring data includes: Define disaster chain nodes; wherein, the disaster chain nodes include the cumulative state of rock mass damage, the initiation and movement state of rock mass collapse, the impact and splash state of the collapse body, the propagation and rise state of surge waves, the state of bank erosion / structure impact, and the state of secondary landslide / river channel siltation. Based on the disaster mechanism, directed edges between nodes are defined according to the causal relationship and development sequence between nodes, and the basic transfer intensity of each directed edge is quantified; The multi-source dynamic monitoring data is mapped to the state values ​​of the nodes.

[0013] According to a specific implementation of an embodiment of this disclosure, obtaining key environmental driving factors and their influence paths includes: Key environmental driving factors were identified based on the disaster-causing mechanism of the chain reaction of unstable rock formations on the target reservoir bank. These key environmental driving factors include: reservoir water level fluctuations, rainfall, and fissure water pressure. The influence pathways of the reservoir water level fluctuations include: affecting stability by changing the seepage pressure, buoyancy force, and physical and mechanical properties of the rock mass; The impact pathways of the rainfall include: increasing the self-weight of the soil and rock mass, reducing the shear strength of the structural surface, and generating dynamic water pressure that affects stability; The influence path of the fissure water pressure includes: directly affecting the effective stress on the fissure surface.

[0014] According to a specific implementation of this disclosure, the step of calculating the dynamic influence weights of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank rockfall disaster chain includes: The dynamic influence weight of directed edges is calculated based on the following formula. : ; in, For directed edges exist The dynamic impact weight at any given moment; For the edge For driving factors Sensitivity coefficient; As driving factor exist The normalized impact at any given time; The net effect of each environmental factor; This is the normalized influence function of key environmental driving factors.

[0015] According to a specific implementation of this disclosure, the normalized influence function of the key environmental driving factors includes: reservoir water level fluctuation influence function, rainfall influence function, and fissure water pressure influence function. The influence function of reservoir water level fluctuation is calculated based on the following method. : ; in, This refers to the 24-hour fluctuation of the water level. It is a symbolic function; For reference amplitude; It is the hyperbolic tangent function; This is the current water level; The critical water level for reservoir bank stability; This is the highest water level in history. It is a step function; Weighted by the rate of water level change; For the absolute height of the influence weight, satisfying ,and ; The rainfall impact function is calculated using the following method. : ; in, For the past Hourly cumulative rainfall; This serves as a reference threshold for cumulative rainfall. Real-time rainfall intensity; This serves as a reference threshold for rainfall intensity. Weighting based on cumulative rainfall impact; The weighting of rainfall intensity must satisfy the following conditions. ,and ; The influence function of fracture water pressure is calculated based on the following formula: ; ; in, This is a function representing the influence of crack propagation on stability. This represents the current crack aperture. This represents the initial aperture of the crack. This is a function representing the effect of internal water pressure on stability. This refers to the pore water pressure within the fracture. This is a reference value for pore water pressure. This is the proportionality coefficient.

[0016] According to a specific implementation of an embodiment of this disclosure, the step of dynamically weighting the directed edges based on the dynamic influence weight includes: The real-time strength of directed edges is calculated based on the following formula. : ; in, For directed edges At any moment Real-time intensity; For directed edges Basic transfer strength; For directed edges exist The dynamic influence weight matrix at any given time; The starting node is calculated based on the following formula. Risk status value : ; in, The deformation damage index; It is an activity index; These are the weighting coefficients. ; The deformation damage index is calculated based on the following formula. ; ; in, This represents the real-time crack width. The critical crack width; For key point displacement rates; This is the critical displacement rate.

[0017] The activity index is calculated based on the following formula. : ; in, Real-time acoustic emission rate; The background sound emission rate.

[0018] According to a specific implementation of this disclosure, the step of calculating the comprehensive risk value based on the real-time strength of dynamically weighted directed edges includes: The overall risk value is calculated based on the following formula:

[0019] in, for The comprehensive dynamic risk value at any given moment; End node Risk contribution weight, satisfying ; For nodes exist Risk status value at any given time; Nodes are calculated based on the following formula. exist Risk status value at any given time:

[0020] in, For nodes exist Risk status value at any given time; Predecessor node exist Risk status value at any given time; For directed edges At any moment Real-time intensity; For nodes The severity coefficient of the consequences.

[0021] Secondly, this disclosure provides a dynamic risk assessment system for reservoir bank rockfall chain disasters, the system comprising: The data acquisition module is configured to acquire multi-source dynamic monitoring data; wherein, the multi-source dynamic monitoring data includes: three-dimensional displacement time series data of the unstable rock mass, acoustic emission signal of the unstable rock mass, hydrological data, meteorological data, and width variation data of the main control fracture of the unstable rock mass; The directed graph module is configured to construct a dynamic directed graph model of the reservoir bank rockfall disaster chain based on the multi-source dynamic monitoring data; The dynamic influence weight module is configured to obtain key environmental driving factors and their influence paths; and to calculate the dynamic influence weight of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank dangerous rock disaster chain. The comprehensive risk module is configured to dynamically weight directed edges based on the dynamic influence weights, and calculate the comprehensive risk value based on the real-time strength of the dynamically weighted directed edges.

[0022] Thirdly, embodiments of this disclosure also provide an electronic device, the electronic device comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor. When the instructions are executed by the at least one processor, they enable the at least one processor to perform the dynamic assessment method for reservoir bank rock chain disaster risk as described in the first aspect or any implementation thereof.

[0023] Fourthly, embodiments of this disclosure also provide a non-transitory computer-readable storage medium storing computer instructions that, when executed by at least one processor, cause the at least one processor to perform the dynamic assessment method for reservoir bank rock chain disaster risk in the first aspect or any implementation thereof.

[0024] Fifthly, this disclosure also provides a computer program product, which includes a computing program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions that, when executed by a computer, cause the computer to perform the dynamic assessment method for reservoir bank rock chain disaster risk in the first aspect or any implementation thereof.

[0025] The dynamic assessment method for the risk of chain-like disasters caused by unstable rockfalls on the reservoir bank in this embodiment of the present disclosure achieves a fundamental transformation in the risk assessment of unstable rockfalls on the reservoir bank from static, single-point, and lagging assessment to dynamic, chain-like, and forward-looking assessment, which is specifically reflected in the following four aspects: 1. It has achieved the quantification and visualization of the dynamic risks of the entire process of chain-related disasters.

[0026] This invention transforms physical processes into a computable risk propagation network through a dynamically weighted disaster chain directed graph model, quantifying the propagation and amplification of risks on the chain in real time, making the originally ambiguous chain-generated risks computable, traceable, and predictable.

[0027] 2. It enables real-time and precise response of risk assessment to environmental driving factors.

[0028] By constructing a dynamic environmental impact weight matrix, real-time monitoring data is transformed into differentiated modulation coefficients for each link in the disaster chain. This enables the assessment results to sensitively reflect dynamic stresses such as sudden drops in water levels and continuous rainfall, achieving a precise depiction of the real-time changes in risk levels of the same unstable rock mass under different hydrological and meteorological conditions, greatly improving the timeliness and accuracy of early warnings.

[0029] 3. An adaptive learning closed loop of monitoring-evaluation-feedback was constructed, making the model more and more accurate with use.

[0030] The invention’s unique feedback iteration mechanism uses actual disaster data or new monitoring data to automatically correct basic parameters and sensitivity coefficients through Bayesian updates, parameter optimization, and other methods. This enables the system to improve itself and adapt to slow changes in geological conditions, fundamentally overcoming the uncertainty of model initialization and ensuring the reliability of long-term applications and the continuous improvement of evaluation accuracy.

[0031] 4. A structured and deep integration of multi-source heterogeneous monitoring data and risk assessment models has been achieved.

[0032] This invention defines a starting node state function to uniformly transform heterogeneous data streams such as cracks, microseismic events, InSAR, water levels, and rainfall into the initial state and dynamic weights driving the risk propagation model. This structured fusion mechanism ensures that the model is always driven by the latest and most comprehensive data, providing a solid data foundation for dynamic assessment and maximizing the utilization of monitoring data value. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of a dynamic risk assessment method for reservoir bank rock chain disasters provided in an embodiment of the present disclosure; Figure 2 A flowchart illustrating a method for dynamic assessment of reservoir bank rock chain disaster risk, provided in this embodiment of the disclosure; Figure 3 This is a schematic diagram of the structure of a dynamic risk assessment system for reservoir bank rock chain disasters provided in an embodiment of this disclosure; Figure 4 A schematic diagram of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0034] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0035] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0036] It should be noted that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. Furthermore, this device and / or this method can be implemented using other structures and / or functionalities besides one or more of the aspects set forth herein.

[0037] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0038] This invention provides a method for dynamic assessment of the risk of chain-related disasters caused by unstable rockfalls along reservoir banks. It constructs a closed-loop assessment framework encompassing dynamic data perception, disaster chain modeling, dynamic risk calculation, and model feedback updates. First, real-time data on the unstable rock mass and its environmental drivers are collected using a multi-source sensor network. Second, the chain-related process of rockfall collapse, impact surge, and secondary disasters is abstracted into a dynamic directed graph model. Nodes in the graph represent disaster states, and directed edges (or simply edges) represent disaster development paths and are associated with basic transfer intensity. Then, real-time monitoring data is quantified into a dynamic influence weight matrix to correct the disaster development intensity of each path. Based on the weighted disaster chain graph, a comprehensive dynamic risk value is calculated using a risk propagation algorithm. Finally, actual disaster information is used to optimize the model parameters. This invention, by integrating time-series data and the physical chain-related process, achieves a quantitative and dynamic assessment of the risk of chain-related disasters along reservoir banks, significantly improving the timeliness and accuracy of risk identification and providing an effective tool for precise prevention and control of geological disasters in reservoir areas.

[0039] Figure 1 This is a schematic diagram of the dynamic assessment method for reservoir bank rock chain disaster risk provided in this embodiment of the disclosure.

[0040] Figure 2 To and Figure 1 The flowchart of the corresponding dynamic assessment method for reservoir bank rock chain disaster risk.

[0041] like Figure 1 As shown, in step S110, multi-source dynamic monitoring data is acquired; wherein, the multi-source dynamic monitoring data includes: three-dimensional displacement time series data of the unstable rock mass, acoustic emission signal of the unstable rock mass, hydrological data, meteorological data, and width variation data of the main control fracture of the unstable rock mass.

[0042] An integrated monitoring network is deployed in the target reservoir shore area to collect multi-source dynamic monitoring data periodically (e.g., every hour) or triggered (e.g., after an earthquake), and the data is unified to the same timestamp and data format to provide a real-time, heterogeneous data foundation for dynamic assessment.

[0043] More specifically, the multi-source dynamic monitoring data acquisition and preprocessing includes the following steps: 1. Collect multi-source dynamic monitoring data, including: three-dimensional displacement time series data of unstable rock mass, acoustic emission signal of unstable rock mass, hydrological data, meteorological data, and width variation data of main control fractures of unstable rock mass.

[0044] Based on the geological structure, potential failure modes, and surrounding environment of the target unstable rock mass, the monitoring scheme is optimized. Sensors are deployed at key locations of the unstable rock mass (such as controlling fissures and potential collapse blocks) and threatened areas (such as riverbanks and waterways that may be affected by swells).

[0045] The three-dimensional displacement time series data of the unstable rock mass is acquired periodically (e.g., weekly) or triggered (e.g., after a sudden change in displacement rate). Acquisition methods include installing high-precision GNSS receivers on the surface of the unstable rock mass and at stable reference points, using a three-dimensional laser scanner or UAV photogrammetry, etc., to acquire the overall point cloud model of the unstable rock mass. This invention does not impose any limitations on these methods.

[0046] The invention continuously monitors acoustic emission signals generated by internal fractures in rock masses. Acquisition methods include drilling around key fractures inside the unstable rock mass and deploying a three-component microseismic sensor array, etc. However, this invention does not impose any limitations on these methods.

[0047] Hydrological data of the waters near the unstable rock mass can be obtained through methods such as installing water level gauges; however, this invention does not impose any limitations on these methods.

[0048] Obtain meteorological data for the assessment area, including real-time rainfall intensity and 24-hour cumulative rainfall.

[0049] Data on the width variation of the fractures can be obtained through methods such as installing fracture gauges on both sides of the main control fracture in the unstable rock mass, and no restrictions are imposed here.

[0050] 2. Data Preprocessing The collected multi-source dynamic monitoring data is preprocessed, including data cleaning, noise filtering, coordinate unification, and normalization of physical quantities with different dimensions, converting them into dimensionless relative indices.

[0051] The preprocessed multi-source dynamic monitoring data is then resampled and aligned according to the same time base (such as the hour) to generate a time series data table in a unified format.

[0052] This step is the data foundation for all subsequent dynamic evaluation work, ensuring the real-time, synchronous, and accurate nature of the input data.

[0053] More specifically, we now proceed to step S120.

[0054] In step S120, a dynamic directed graph model of the reservoir bank dangerous rock disaster chain is constructed based on the multi-source dynamic monitoring data.

[0055] In this embodiment of the invention, the construction of a dynamic directed graph model of the reservoir bank rockfall disaster chain based on the multi-source dynamic monitoring data includes: defining disaster chain nodes; wherein, the disaster chain nodes include the rock mass damage accumulation state, rock mass collapse initiation and movement state, collapse impact and splash state, surge propagation and rise state, bank slope erosion / structure impact state, and secondary landslide / river channel siltation state; according to the disaster mechanism, the directed edges between nodes are defined according to the causal relationship and development sequence between nodes, and the basic transfer intensity of each directed edge is quantified; the multi-source dynamic monitoring data is mapped to the state values ​​of the nodes.

[0056] More specifically, constructing a dynamic directed graph model of the reservoir bank rockfall hazard chain includes the following steps: 1. Define disaster chain nodes Based on the analysis of the geological structure, historical disaster cases and disaster-causing mechanisms of the target reservoir bank, the continuous chain disaster process is discretized into a series of key and distinguishable disaster physical states. Each state is defined as a graph node, and the node represents a stage in the disaster evolution process.

[0057] For example: Graph Node This indicates the cumulative damage state of the unstable rock mass, characterizing the initiation and propagation of cracks within the rock mass, macroscopically manifested as the continuous opening of fissures, active acoustic emission, and accelerated local deformation. This is the starting point of the disaster chain, and its state is directly driven by the monitoring data from step S110.

[0058] Graph Node This indicates the initiation and movement state of a rock mass collapse, signifying the breakthrough of the main controlling structural plane, the collapse of the rock mass detaching from the parent rock, and the initiation of movement along the slope. Its triggering and... This is directly related to instability.

[0059] Graph Node This indicates the impact and splash state of the collapsed body, representing the impact of the collapsed body on the reservoir water at a certain speed and volume, converting kinetic energy into water potential energy, and generating an initial impact surge. The initial height of the surge is a key intensity indicator of this node.

[0060] Graph Node It represents the propagation and rise of the swell, indicating the initial swell propagating far from the water surface and rising and scouring when it reaches the opposite bank or upstream and downstream slopes. Its intensity is related to the initial height of the swell, the propagation distance, underwater topography, etc.

[0061] Graph Node It indicates the state of bank erosion / structure impact, characterizing the scouring and soaking effect of surge rise on the bank slope, or the direct impact on structures such as ports and ships, which may lead to new damage.

[0062] Graph Node Indicates the state of secondary landslides / river channel siltation, characterizing Erosion or impact can induce slope instability, leading to new landslides, or large-scale collapses / landslides that block local river channels, which is a possible end of a disaster chain.

[0063] 2. Define directed edges between nodes Based on the disaster mechanism, determine the causal relationships and development sequence between nodes using directed edges. From the pioneer node Point to the successor node ,in, For disaster from state Development to a certain state One possible path.

[0064] For example: directed edges Indicates the accumulation of damage To the start of the collapse This represents the process of quantitative change leading to qualitative change in damage; directed edge Indicates the start of the collapse To impact and splash This represents the conversion of the kinetic energy of the collapsed body into the wave energy of the water; directed edge Indicates from impact splash To the spread of the swell , representing the propagation process of waves; directed edge Indicates propagation from the swell. Impact on the bank slope This indicates the effect of waves on the disaster-bearing body; oriented edge Indicates impact from bank slope To secondary landslide This indicates the triggering of secondary disasters, etc. Depending on the terrain, there may be other pathways, which will not be elaborated here.

[0065] 3. Quantify the basic transfer intensity

[0066] For each directed edge Assign a base transfer strength ( This indicates that a given disaster is in a state under standard or average environmental conditions. Its development to a certain state The conditional probability or relative strength.

[0067] Basic transfer strength Determined in the following ways: (1) Computational method based on physical model For disaster development stages with clear physical mechanisms and describable by mathematical formulas, physical models are used to determine the basic transfer intensity.

[0068] For example, for (Collapse → Impact and splash) Calculate the initial surge height based on the energy conservation formula. :

[0069] in, This is a comprehensive empirical coefficient, which is related to the shape of the landslide body, the angle of entry into the water, and the depth of the water body; The volume of the collapsed body (m) 3 ); Acceleration due to gravity (m / s²) 2 ); The vertical drop (m) from the center of gravity of the collapse to the water surface; The empirical index is preferably 2 / 3; The empirical index is preferably 1 / 2.

[0070] The physical quantities obtained through physical models (such as...) The base transfer strength is obtained by normalizing the data to the [0,1] interval. : Method 1 (Proportion Method): .in, It is the maximum wave height that the assessment area may generate under the maximum credible collapse scenario, determined through extreme case simulation or historical maximum record.

[0071] Method 2 (Probability Mapping Method): If there is a known relationship between the physical quantity and the probability of occurrence, a mapping can be established. For example, assuming wave height... Exceeding the preset threshold If it is considered to constitute a disaster at that time, then Can be defined as ,pass Calculation of probability distribution.

[0072] (2) Analysis method based on historical statistics For edges with numerous historical cases and obvious statistical patterns, the basic transition strength is obtained based on the statistical analysis of historical case data. .

[0073] Historical cases from the assessment area and areas with similar geological conditions were collected to form a sample set. Each case included: whether an event occurred... arrive The transfer, and when (or before) the transfer occurs. Quantitative characteristics of the state (e.g., crack width, displacement rate, etc.).

[0074] Historical statistical analysis method :

[0075] in, In a state The number of cases; for China's development into The number of cases.

[0076] Considering state characteristics Acquisition: Establishing models through logistic regression, survival analysis, etc. State characteristic variables X (such as fracture aperture, acoustic emission frequency) and the probability of transition The relationship. Taken as standard state Conditional probability under (historical averages of each feature variable): .

[0077] Logistic Regression Example: Using historical data to fit coefficients , , Then, the standard state value , Substitute in, and you will get .

[0078] (3) Delphi method For processes with complex mechanisms and scarce data, the Delphi method is used to determine the relative intensity.

[0079] First, an expert group (including experts in geology, engineering geology, hydrology, geotechnical engineering and other fields) was formed.

[0080] Then, we provide experts with disaster chain models and geological data, and invite them to discuss each chain separately. of The value (or range) is evaluated independently in multiple rounds. After each round, expert opinions are summarized and anonymously fed back until the opinions converge.

[0081] Using the Analytic Hierarchy Process (AHP), experts perform pairwise comparisons of the edges, construct a judgment matrix, calculate eigenvectors to obtain the relative weights of each edge, and then scale the weights using known anchoring cases. value.

[0082] Finally, the average of the Delphi method results was taken; a consistency test was performed on the AHP results. ).

[0083] For the key edge Choose at least two of the above methods to estimate independently. Compare the results. If the difference is significant (a difference exceeding 0.2 is considered significant when data is insufficient), the reasons need to be analyzed (such as inaccurate physical model parameters, statistical sample bias, differences in expert perception, etc.), and the final initial values ​​should be determined through discussion in a meeting.

[0084] 4. Define the quantization function for the node state. For certain nodes (especially the starting node) Define the state quantization function. This maps multi-source dynamic monitoring data to the state values ​​of the node. .For example, (Damage Accumulation) Status Value It can be defined as:

[0085] in, This represents the current crack aperture. This represents the initial aperture of the crack. The current acoustic emission rate; The background acoustic emission rate; The weight for the crack aperture; Let be the weight of the acoustic emission rate, satisfying α+β=1.

[0086] The higher the value, the more stable the system is. The higher the risk level of the situation.

[0087] 5. Initial model validation and calibration Using historical disaster cases or physical simulation results, the initial directed graph model (including nodes, edges, and...) is constructed. Validate the model. Check if it can reproduce known disaster chain processes, and adjust any unreasonable connections or... The value is adjusted until the model output basically matches the historical data.

[0088] This step is crucial for abstracting from physical processes to computational models, and it serves the following purposes: 1) It structures and discretizes the complex, continuous chain-like disaster physical processes into a network composed of nodes and edges, which is the foundation for computer logic operations and risk propagation; 2) Through The inherent and relatively stable development trends of each link in the disaster chain were quantified, providing a benchmark anchor for the subsequent introduction of dynamic environmental impacts; 3) The defined node state function established a bridge between the original monitoring data and the internal state variables of the model, enabling the data to drive the model.

[0089] Next, proceed to step S130.

[0090] In step S130, key environmental driving factors and their influence paths are obtained; the dynamic influence weights of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank rockfall disaster chain are calculated respectively.

[0091] In this embodiment of the invention, obtaining key environmental driving factors and their influence paths includes: obtaining key environmental driving factors based on the disaster-causing mechanism of the target reservoir bank unstable rock chain disaster; the key environmental driving factors include: reservoir water level fluctuation, rainfall, and fissure water pressure; wherein, the influence path of the reservoir water level fluctuation includes: affecting stability by changing the bank slope seepage pressure, buoyancy force, and physical and mechanical properties of the rock mass; the influence path of the rainfall includes: increasing the self-weight of the rock and soil mass, reducing the shear strength of the structural surface, and generating dynamic water pressure, thus affecting stability; the influence path of the fissure water pressure includes: directly affecting the effective stress of the fissure surface.

[0092] In this embodiment of the invention, the step of calculating the dynamic influence weights of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank rockfall disaster chain includes: The dynamic influence weight of directed edges is calculated based on the following formula. : ; in, For directed edges exist The dynamic impact weight at any given moment; For the edge For driving factors The sensitivity coefficient; As driving factor exist The normalized impact at any given time; The net effect of each environmental factor; This is the normalized influence function of key environmental driving factors.

[0093] In this embodiment of the invention, the normalized influence function of the key environmental driving factors includes: reservoir water level fluctuation influence function, rainfall influence function, and fissure water pressure influence function; The influence function of reservoir water level fluctuation is calculated using the following method. : ; in, This refers to the 24-hour fluctuation of the water level. It is a symbolic function; For reference amplitude; It is the hyperbolic tangent function; This is the current water level; This is the critical water level for reservoir bank stability; This is the highest water level in history. It is a step function; Weighted by the rate of water level change; For the absolute height of the influence weight, satisfying ,and ; The rainfall impact function is calculated using the following method. : ; in, For the past Hourly cumulative rainfall; This serves as a reference threshold for cumulative rainfall. Real-time rainfall intensity; This serves as a reference threshold for rainfall intensity. Weighting based on cumulative rainfall impact; The weighting of rainfall intensity must satisfy the following conditions. ,and ; The influence function of fracture water pressure is calculated based on the following formula: ; ; in, This is a function representing the influence of crack propagation on stability. This represents the current crack aperture. This represents the initial aperture of the crack. This is a function representing the effect of internal water pressure on stability. This refers to the pore water pressure within the fracture. This is a reference value for pore water pressure. This is the proportionality coefficient.

[0094] More specifically, calculating the dynamic impact weight matrix of environmental driving factors on the disaster chain includes the following steps: 1. Identify key environmental drivers and their impact pathways This study analyzes the formation mechanism of the target reservoir bank unstable rock chain disaster, identifies the key time-varying environmental factors that significantly influence the disaster's development process, and clarifies which specific links (directed edges) in the directed graph of the disaster chain each factor primarily affects. ).

[0095] For example, typical factor-path mappings mainly include: Reservoir water level fluctuations affect stability by altering bank slope seepage pressure, buoyancy, and the physical and mechanical properties of the rock mass (such as softening), primarily acting on... (Promoting the collapse of unstable rock masses) and (Promotes secondary landslides). Its characteristic parameter is the current water level. 24-hour water level fluctuation .

[0096] Rainfall affects stability by increasing the self-weight of the soil and rock mass, reducing the shear strength of structural surfaces, and generating hydrodynamic pressure. Its main effects are... and Its characterization parameter is the cumulative rainfall in the preceding period. Real-time rainfall intensity .

[0097] Fracture water pressure: directly affects the effective stress on the fracture surface, mainly acting on... Its characterization parameter is the fracture aperture. It can be reflected indirectly, or by specifically deploying pore water pressure gauges to monitor fracture water pressure. .

[0098] Establish a mapping table: explicitly record each driving factor in matrix or list form. And the directed edges it affects The correspondence.

[0099] 2. Define the normalized influence function of each key environmental driving factor.

[0100] For each environmental driving factor Design a normalized influence function. Its function is to transform the raw monitoring data into a dimensionless scalar of relative influence, with values ​​ranging from [-1,1] to [0,1].

[0101] The following is a detailed description through examples: The influence function of reservoir water level fluctuation is calculated based on the following formula. (acts on) , wait):

[0102] in, This refers to the 24-hour fluctuation of the water level. For symbolic functions, exist Returns -1 at that time. The value returns +1, as a rapid drop in water level is usually the most detrimental to stability, hence this item is negative; For reference amplitude (m), for example, 2.0m / day. The hyperbolic tangent function maps the input to the (0,1) saturation interval; Current water level (m); The critical water level for bank stability (m) is the water level above which the bank stability is significantly reduced due to soaking and softening. This is the highest water level in history (m); For the step function (Heaviside function), when The value is 1 when the time is right, and 0 otherwise. Weighted by the rate of water level change; For the absolute height of the influence weight, satisfying The relative importance of the rate of water level change and the absolute height are measured respectively, and The effect is more significant due to the rate of water level change.

[0103] The critical water level for reservoir bank stability was obtained using the following method: 1. Based on exploration data (drilling, geophysical exploration), establish a geological model of the slope's stratigraphic structure, sliding surface (or potential sliding surface) morphology, and soil and rock parameters (unit weight, shear strength index values).

[0104] 2. Hydrogeological model coupling: Determine the dynamic relationship between the groundwater level and the reservoir water level (recharge and discharge conditions), and clarify parameters such as permeability coefficient and specific yield.

[0105] 3. Load and operating condition settings: Consider self-weight, reservoir water pressure, seepage force (dynamic water pressure), seismic force, etc., and use the reservoir water level as a variable.

[0106] 4. Iterative Calculation: In software (such as GeoStudio-SLOPE / W, Lizheng, PLAXIS), set a series of water level elevations to perform stability calculations and plot the safety factor. The curve showing the relationship between the curve and the reservoir water level, and the curve with The water level corresponding to the intersection of (or standard values) is... .

[0107] It combines two adverse impact modes: rapid water level drop (negative contribution) and continuous high water level immersion (positive contribution). The more negative the value, the more unfavorable the water level conditions are to stability.

[0108] Calculate the rainfall impact function based on the following formula. (acts on) , wait):

[0109] in, For the past Hourly (T is usually taken as 24 or 48) cumulative rainfall (mm); The cumulative rainfall reference threshold (mm), such as 50mm, is related to soil and rock saturation and can be obtained based on statistical analysis of historical disaster data and rainfall data, or through simulation based on a hydrological-mechanical coupling physical model. Real-time rainfall intensity (mm / h); The reference threshold for rainfall intensity (mm / h), such as 20 mm / h, is related to runoff and hydrodynamic pressure. It can be obtained based on statistical analysis of historical disaster data and high-resolution rainfall data, or based on physical models and numerical simulations. Weighting based on cumulative rainfall impact; The weighting of rainfall intensity must satisfy the following conditions. , (Cumulative rainfall has a more lasting impact on stability).

[0110] Combining the cumulative softening effect of previous rainfall with the infiltration and scouring effect of instantaneous rainfall, when The larger the value, the more unfavorable the rainfall conditions.

[0111] The fissure water pressure influence function (acting on) is calculated based on the following formula. ):

[0112]

[0113] in, The impact of fracture development on stability; The current crack aperture (mm); The initial or background aperture of the crack (mm); The effect of internal water pressure on stability; The pore water pressure within the fracture (kPa); Reference value for pore water pressure (kPa); This is a proportionality coefficient, used to control the magnitude of the influence; preferably, it is 0.5.

[0114] Methods for obtaining pore water pressure reference values ​​include: using seepage theory and unsaturated soil mechanics to simulate the distribution and evolution of pore water pressure under conditions of water level fluctuation and rainfall infiltration, and calculating the corresponding stability safety factor.

[0115] 1. Establish an unsteady seepage model: Based on the survey data, use software (such as Geo-SEEP / W, HYDRUS, TOUGH2) to establish a two-dimensional or three-dimensional seepage model and determine the hydraulic parameters of the soil and rock mass (permeability coefficient, soil-water characteristic curve).

[0116] 2. Set boundary conditions: Apply time-varying boundary conditions such as reservoir water level fluctuations and rainfall infiltration.

[0117] 3. Coupled stability analysis: The calculated pore water pressure field is imported into the limit equilibrium method or finite element strength reduction method model.

[0118] 4. Through parameter sensitivity analysis or iterative calculation, identify the factors affecting the slope safety factor. The pore water pressure at a critical point on the potential slip surface (typically the slip zone exit or middle) when the pressure drops to a critical value (e.g., 1.0). This value is the theoretical pore water pressure. .

[0119] 3. Determine each directed edge Sensitivity coefficient matrix to environmental factors Quantify the development path of each disaster chain (directed edge) The sensitivity of each pair to various environmental driving factors. driving factors Assign a sensitivity system number. It is a real number whose absolute value indicates the strength of sensitivity, and its sign indicates the direction of influence (positive indicates that the factor enhances the transfer strength of this side, and negative indicates inhibition).

[0120] The sensitivity coefficient was determined using the following method: Physical model inversion: Using a water-rock coupled numerical model, the rate of change of the stability coefficient of the unstable rock mass under different water levels and rainfall conditions is simulated, thereby calculating the probability of collapse. Influence coefficient , .

[0121] Statistical analysis: Based on historical monitoring data and disaster event records, logistic regression or survival analysis is performed to establish the statistical relationship between the probability of disaster occurrence (or activation of a certain link) and various environmental factors. The regression coefficient is then used. The estimated value.

[0122] Expert experience verification: This is achieved through the Delphi method or the Analytic Hierarchy Process (AHP), where domain experts verify (…). , The relative importance of each pair is compared pairwise to obtain the coefficient matrix. .

[0123] 4. Calculate the dynamic influence weight of each directed edge.

[0124] For each directed edge in the disaster chain directed graph ,exist At any given moment, all environmental driving factors are considered. Calculate its dynamic impact weight. :

[0125] in, For the edge For driving factors The sensitivity coefficient, if Does not affect the edge Then its corresponding ; As driving factor exist The normalized impact at any given time. This represents the net effect of each environmental factor. If the net effect is positive, then... This indicates that the current environment has promoted / exacerbated the disaster along its path. → The development of a path amplifies its activity or potential for migration; if the net effect is negative, then... This indicates that the current environment has suppressed / slowed down the development of this path. It must be a non-negative value, with a lower limit of 0.

[0126] 5. Assemble the dynamic influence weight matrix All directed edges exist Dynamic influence weights calculated at each time step According to the adjacency relationship of the graph model, assemble into a square array ,in This represents the total number of nodes in the disaster chain. (Row index of the matrix) Represents the starting node, column index Represents the target node. When from... arrive There are directed edges Matrix elements ,otherwise . It is a time The changing matrix encapsulates the dynamic modulation information of all current environmental conditions on each development path of the entire disaster chain network.

[0127] This step serves as a bridge connecting real-time data and static disaster chain models. It is the core algorithm module for achieving dynamic assessment and has the following functions: 1) It transforms monitoring data with clear physical meaning but diverse formats into... 1) Transform the function into a unified, comparable influence quantity; 2) Through the sensitivity coefficient matrix It accurately characterizes the differentiated impacts of different environmental factors on different disaster stages, reflecting the physical mechanism and refinement of the system; 3) The final output dynamic influence weight matrix This will be used as a time-varying parameter input into the comprehensive risk calculation in the next stage, so that the risk assessment results can reflect dynamic environmental stresses such as reservoir water level fluctuations and rainfall processes in real time.

[0128] Next, proceed to step S140.

[0129] In step S140, the directed edges are dynamically weighted based on the dynamic influence weights, and the comprehensive risk value is calculated based on the real-time strength of the dynamically weighted directed edges.

[0130] In this embodiment of the invention, the step of dynamically weighting the directed edges based on the dynamic influence weight includes: The real-time strength of directed edges is calculated based on the following formula. : ; in, For directed edges At any moment Real-time intensity; For directed edges Basic transfer strength; For directed edges exist The dynamic influence weight matrix at any given time; The starting node is calculated based on the following formula. Risk status value : ; in, For state quantization function; The deformation damage index; It is an activity index; These are the weighting coefficients. ; The deformation damage index is calculated based on the following formula. ; ; in, This represents the real-time crack width. The critical crack width; For key point displacement rates; This is the critical displacement rate.

[0131] The activity index is calculated based on the following formula. : ; in, Real-time acoustic emission rate; The background sound emission rate.

[0132] In this embodiment of the invention, the calculation of the comprehensive risk value based on the real-time strength of the dynamically weighted directed edges includes: The overall risk value is calculated based on the following formula:

[0133] in, for The comprehensive dynamic risk value at any given moment; End node Risk contribution weight, satisfying ; For nodes exist Risk status value at any given time; Nodes are calculated based on the following formula. exist Risk status value at any given time:

[0134] in, For state quantization function; For nodes exist Risk status value at any given time; Predecessor node exist Risk status value at any given time; For directed edges At any moment Real-time intensity; For nodes The severity coefficient of the consequences.

[0135] More specifically, the calculation of the comprehensive risk value based on the dynamic weighted disaster chain diagram includes the following steps: 1. Calculate the real-time strength of directed edges based on the following formula. :

[0136] in, For directed edges The basic transition intensity represents the rate at which a disaster transitions from a state of equilibrium under standard environmental conditions. Transition to state The inherent strength or conditional probability; For directed edges exist The dynamic influence weight matrix at any given time; 2. Calculate the starting node based on the following formula. Risk status value :

[0137] in, For state quantization function; The deformation damage index; It is an activity index; Weighting coefficients ( The relative importance of deformation and activity indices is used to balance the data, and is determined based on historical data analysis or by experts.

[0138] The deformation damage index is calculated based on the following formula. ;

[0139] in, This represents the real-time crack width. The critical crack width; For key point displacement rates; This is the critical displacement rate.

[0140] The critical fracture width can be determined using the discrete element method (such as UDEC, 3DEC) or the extended finite element method to establish rock mass models containing fractures of different widths. Simulations of the stress intensity factor at the fracture tip under gravity and seismic forces, along with the fracture width, determine the critical width leading to rapid fracture propagation or penetration. Furthermore, simulations of the deformation and failure processes of fractures of different widths under water pressure are conducted to analyze the fracture width corresponding to the critical hydraulic fracturing condition leading to rock instability.

[0141] The critical displacement rate can be obtained based on creep theory and model inversion. This method uses a soil rheological model to predict and inversely calculate the critical rate, and specifically includes the following steps: 1. Using the constant-rate creep stage data obtained from monitoring, the parameters of the rheological model are inverted.

[0142] 2. Substitute the parameters obtained from the inversion into the model and perform extrapolation prediction. Solve the model equations to calculate the displacement rate corresponding to the theoretical acceleration creep initiation point, which is the critical displacement rate based on the model.

[0143] 3. By fitting the empirical material failure prediction formula proposed by Voigt to the monitoring data, the failure time can be extrapolated and the characteristic values ​​near the critical rate can be obtained.

[0144] The activity index is calculated based on the following formula. :

[0145] in, Real-time acoustic emission rate; The background sound emission rate.

[0146] It directly reflects the immediate risk level of the unstable rock mass itself at the current moment.

[0147] 3. Perform risk propagation and aggregation calculations along the dynamically weighted disaster chain diagram. by As initial values, based on the topological order of the constructed disaster chain directed graph (from... (to the end node), the risk value is along the weighted edge Propagate forward and aggregate the risk state values ​​for each node. For non-starting nodes ( Its risk state value is the sum of the risk contributions of all predecessor nodes that may propagate to this node.

[0148] The calculation formula is:

[0149] in, For nodes exist The risk status value at any given moment indicates the extent to which the disaster has developed. The level of potential risk inherent in a given state; Predecessor node exist Risk status value at any given time; For directed edges Real-time intensity; For nodes The severity coefficient (a constant greater than 0) is used to quantify the severity of a disaster once it develops to a certain extent. The status, which is the relative loss or severity of the consequences that may be caused, is a comprehensive assessment of multiple dimensions such as casualties, economic losses, social impact, and ecological damage. It is usually obtained through expert scoring, analytic hierarchy process (AHP), or historical loss statistics normalization.

[0150] from Begin by sequentially calculating all its successor nodes. of And then with these nodes As input, their successor nodes are calculated until all end nodes have been calculated; this is a forward propagation process.

[0151] 4. Calculate the comprehensive dynamic risk value

[0152] The risk state values ​​of all terminal nodes (i.e., nodes with no outgoing edges) in the directed graph of the disaster chain are... By performing a weighted summation, a characterization of the entire unstable rock mass can be obtained. Quantitative indicators of the comprehensive risk level of disasters in real time :

[0153] in, for The overall dynamic risk value at any given moment. The higher this value, the higher the overall risk of a chain reaction disaster occurring at that moment and causing serious consequences. End node Risk contribution weight, satisfying This reflects the relative importance of different end-point disaster patterns in the overall risk assessment and can be set according to the distribution of regional disaster-bearing bodies, historical disaster situation, or management objectives.

[0154] Integrating from the initial state of the unstable rock mass ( ), through all possible disaster chain paths ( The spread and amplification of ) ultimately leads to different disaster consequences ( The risk of unstable rock masses is reflected in the overall probability and severity of chain disasters and adverse consequences caused by unstable rock masses under the current dynamic environmental conditions.

[0155] 5. Risk Level Classification and Early Warning Based on the calculation The value is mapped to a preset risk level to achieve intuitive expression and early warning of risk.

[0156] Thresholds for classifying each risk level were determined through methods such as historical disaster case analysis, numerical simulation, and expert consultation. The risk is divided into 5 levels: low risk ( Level 1), Medium-risk ( Level 1), higher risk ( Level 1), High-risk ( Level 1), extremely high risk ( class).

[0157] when At that time, it was determined to be Level (low risk), displayed in blue, routine monitoring.

[0158] when At that time, it was determined to be Level 1 (Medium Risk), displayed in green, requiring close monitoring.

[0159] when At that time, it was determined to be Level 1 (Higher Risk): Displayed in yellow, warning issued, and monitoring encrypted.

[0160] when At that time, it was determined to be Level 1 (High Risk): Display orange, issue an alert, conduct on-site inspections, and prepare for emergency response.

[0161] when Determined as Level 1 (Extremely High Risk): Displayed in red, an alert is issued, an emergency response is activated, and evacuation is organized if necessary.

[0162] This step is the final stage of risk quantitative calculation and output, and has the following functions: 1) By organically combining the static basic attributes of the disaster chain with the dynamic environmental impact, a completely time-varying risk propagation network is formed; 2) Through the graph propagation algorithm, the quantitative simulation of the accumulation, amplification and aggregation of risks along the disaster chain is realized, which is consistent with the physical nature of the evolution of chain-generated disaster risks; 3) The final output R(t) is a comprehensive, dynamic and comparable risk quantification index, which provides a scientific and objective decision-making basis for risk comparison between different time points and different unstable rock masses, as well as for threshold-based graded early warning.

[0163] Next, proceed to step S150.

[0164] In step S150, the model feedback update and evaluation iteration are performed.

[0165] More specifically, the model feedback update and evaluation iteration includes the following steps: 1. Define feedback trigger events and data capture Establish a trigger mechanism for model feedback updates. The system will automatically initiate the feedback update process and lock the relevant time window when any of the following conditions are met. All monitoring data, intermediate calculation results, and final early warning records within the system: Actual disaster event occurrence: A disaster event related to the assessment objective is monitored or verified to have occurred, regardless of its size (e.g., localized collapse, significant surge, abrupt change in bank deformation). Record the event type and time of occurrence. Actual scale and scope of impact.

[0166] There is a significant difference between model predictions and actual conditions: when the model predicts a risk level... When the level remains high but there are no actual anomalies over a long period, or when the model has a low risk level but an unannounced emergency occurs.

[0167] Add or upgrade monitoring data sources: When new, more accurate sensors (such as distributed fiber optics, InSAR) are deployed and a stable data stream is obtained.

[0168] Periodic calibration cycle reached: Set fixed time intervals (such as quarterly or annually) for systematic review and calibration.

[0169] 2. Inversion and parameter correction based on actual events Using information from actual disaster events, the key parameters in the model are inverted and optimized, with a focus on correcting the base transfer intensity. and environmental factor sensitivity coefficient .

[0170] right The corrections include: when an actual change from state... arrive Disasters (such as collapses, which trigger...) Then, use the time when the event occurs. Based on the data before and after, reassess the likelihood of this path shifting:

[0171] in, Before the correction Estimated value; For the revised version Estimated value; For use in initial estimation The equivalent number of historical samples (which can be regarded as the confidence level of prior information); This represents the evidence of transfer observed in the current event. It is 1 if the event occurs, 0 otherwise, and 0 if the model predicts a high risk even though the event has not occurred.

[0172] Dynamically adjust based on actual events. To make it closer to reality, when the event occurs... A moderate increase, or frequent model warnings even if no event occurs. It may be overestimated and needs to be reduced appropriately.

[0173] right Correction: Analyze environmental driving factors when the event occurs. The contribution of the model to the risk predicted by the model. If the deviation from the actual value is large, the sensitivity coefficient can be fine-tuned through optimization algorithms (such as the least squares method). This makes the model in Make use of the revised version at all times Calculated and ultimate risk It can more accurately match the actual situation of an event (such as whether it occurs or not, or the scale of its occurrence).

[0174] Data assimilation: Assimilate the observed data of the actual event (such as the volume of landslides and the height of surge waves) into the model as true values.

[0175] 3. Model structure expansion and update When a new disaster chain link emerges that is not covered by the initial model, the directed graph model constructed in step S120 is structurally updated.

[0176] Specifically, the following steps are included: Adding new nodes: Identifying new disaster states For example, the initial model only considered swells, but actual observations revealed that the collapse triggered disturbance of sediment deposits at the reservoir bottom, necessitating the addition of nodes. (Resuspension of sediment at the bottom of the reservoir).

[0177] Adding or modifying directed edges: Establishing connections between new nodes and existing nodes. or Determine its basic transfer strength. The initial value can be estimated based on limited data of new events or mechanistic analogies.

[0178] Determine the dynamic weight parameters of the new edge: for the newly added edge Assign its sensitivity coefficients to various environmental driving factors. The initial value can be set based on the physical mechanism assumptions and optimized in subsequent iterations.

[0179] Record version changes to the model structure to ensure the historical traceability of evaluation results.

[0180] 4. Optimization of state calculation function based on new data source When new monitoring methods are introduced (such as planar deformation revealed by high-precision InSAR and three-dimensional acoustic emission localization), the risk status of the starting node is optimized for calculation. function .

[0181] Specifically, this includes the following steps: processing new monitoring data (such as InSAR deformation rate fields) ) is introduced as an input variable. In the calculation function:

[0182] in, , , For the new weighting coefficients, satisfying New coefficients are determined by training with historical data (such as regression analysis). It has a stronger correlation with the development of subsequent disasters.

[0183] 5. Systematic recalibration and uncertainty assessment After periodic calibration cycles or after accumulating a sufficient number of event samples, the model is systematically recalibrated and uncertainty analyzed.

[0184] Specifically, the following steps are included: Global parameter optimization: Using the monitoring data sequences and corresponding observation states (whether safe or not) for all time steps within a time window (e.g., the past year) as the training set, the model predicts risk. Using the overall fit with the actual state (e.g., the area under the ROC curve, AUC) as the objective function, optimization algorithms (such as genetic algorithms and particle swarm optimization) are employed to optimize a set of core parameters in the model (including...). , , , as well as Global optimization is performed using the weight coefficients in the function.

[0185] Uncertainty quantification: Based on parameter optimization results and event samples, evaluate key parameters (such as...) The posterior probability distribution of is given, and its confidence interval is provided (preferred). ).

[0186] Warning threshold adjustment: Based on the performance of the optimized model on the new validation set, the risk level threshold is reassessed and adjusted to control the false positive rate and false negative rate at an acceptable level.

[0187] After the above feedback updates, the model is more sensitive to the response of the unstable rock mass under conditions of rapid water level decline, and to the initial state. The judgment is more accurate, and the overall prediction performance is improved.

[0188] This step is crucial for ensuring the assessment system's continued effectiveness, self-improvement, and adaptive evolution, and has the following functions: 1) Overcoming the uncertainty of model initialization; 2) Adapting to the time-varying nature of geological conditions; 3) Improving the accuracy and reliability of early warnings; 4) Achieving progressive model complexity: Through structural updates, the model can describe newly discovered disaster phenomena, and its assessment capabilities are continuously enhanced.

[0189] The proposed method for dynamic assessment of reservoir bank rock chain-related disaster risks enables quantitative and dynamic assessment of these risks, overcoming the limitations of traditional static assessment methods. By integrating multi-source time-series data with physical chain-related processes, it significantly improves the timeliness and accuracy of risk identification, providing reliable technical support for real-time early warning, scientific prevention and control, and emergency decision-making in reservoir geological disaster areas.

[0190] Figure 3 The present invention illustrates a dynamic risk assessment system 300 for reservoir bank rock chain disasters, which includes a data acquisition module 310, a directed graph module 320, a dynamic influence weight module 330, and a comprehensive risk module 340.

[0191] The data acquisition module 310 is used to acquire multi-source dynamic monitoring data; wherein, the multi-source dynamic monitoring data includes: three-dimensional displacement time series data of the unstable rock mass, acoustic emission signal of the unstable rock mass, hydrological data, meteorological data, and width variation data of the main control fracture of the unstable rock mass; The directed graph module 320 is used to construct a dynamic directed graph model of the reservoir bank dangerous rock disaster chain based on the multi-source dynamic monitoring data; The dynamic influence weight module 330 is used to obtain key environmental driving factors and their influence paths; and to calculate the dynamic influence weight of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank dangerous rock disaster chain. The comprehensive risk module 340 is used to dynamically weight the directed edges based on the dynamic influence weights, and calculate the comprehensive risk value based on the real-time strength of the dynamically weighted directed edges.

[0192] See Figure 4 This disclosure also provides an electronic device 40, which includes: At least one processor; and, The memory is communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform the dynamic risk assessment method for reservoir bank rock chain disasters in the foregoing method embodiments.

[0193] This disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the dynamic risk assessment method for reservoir bank rock chain disasters in the aforementioned method embodiments.

[0194] This disclosure also provides a computer program product, which includes a computing program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions that, when executed by a computer, cause the computer to perform the dynamic risk assessment method for reservoir bank rock chain disasters in the aforementioned method embodiments.

[0195] The following is for reference. Figure 4 The diagram illustrates a structural schematic of an electronic device 40 suitable for implementing embodiments of the present disclosure. The electronic devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0196] like Figure 4 As shown, electronic device 40 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 402 or a program loaded from storage device 408 into random access memory (RAM) 403. The RAM 403 also stores various programs and data required for the operation of electronic device 40. The processing unit 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0197] Typically, the following devices can be connected to I / O interface 405: input devices 406, including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 407, including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 408, including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows electronic device 40 to communicate wirelessly or wiredly with other devices to exchange data. Although electronic device 40 with various devices is shown in the figure, it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0198] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 409, or installed from a storage device 408, or installed from a ROM 402. When the computer program is executed by the processing device 401, it performs the functions defined in the methods of embodiments of this disclosure.

[0199] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0200] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0201] The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire at least two Internet Protocol (IP) addresses; send a node evaluation request including the at least two IP addresses to a node evaluation device, wherein the node evaluation device selects an IP address from the at least two IP addresses and returns it; and receive the IP address returned by the node evaluation device; wherein the acquired IP address indicates an edge node in a content delivery network.

[0202] Alternatively, the aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: receive a node evaluation request including at least two Internet Protocol (IP) addresses; select an IP address from the at least two IP addresses; and return the selected IP address; wherein the received IP address indicates an edge node in the content delivery network.

[0203] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0204] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0205] The units described in the embodiments of this disclosure can be implemented in software or in hardware. The name of a unit does not necessarily limit the unit itself; for example, the first acquisition unit can also be described as "a unit that acquires at least two Internet Protocol addresses".

[0206] It should be understood that the various parts of this disclosure can be implemented in hardware, software, firmware, or a combination thereof.

[0207] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A dynamic assessment method for the risk of reservoir bank rock chain disasters, characterized in that, The method includes the following steps: Acquire multi-source dynamic monitoring data; wherein, the multi-source dynamic monitoring data includes: three-dimensional displacement time series data of the unstable rock mass, acoustic emission signal of the unstable rock mass, hydrological data, meteorological data, and width variation data of the main control fracture of the unstable rock mass; Based on the multi-source dynamic monitoring data, a dynamic directed graph model of the reservoir bank dangerous rock disaster chain is constructed; wherein, the graph nodes are the stages in the evolution process of the physical state of the disaster; and the directed edges are the paths from the pioneer nodes to the successor nodes. Obtain key environmental driving factors and their impact paths; calculate the dynamic impact weight of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank dangerous rock disaster chain; The directed edges are dynamically weighted based on the dynamic influence weights, and the comprehensive risk value is calculated based on the real-time strength of the dynamically weighted directed edges. The calculation of the dynamic influence weights of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank rockfall disaster chain includes: The dynamic influence weight of directed edges is calculated based on the following formula. : ； in, For directed edges exist The dynamic impact weight at any given moment; For the edge For driving factors The sensitivity coefficient; As driving factor exist The normalized impact at any given time; The net effect of each environmental factor; This is the normalized influence function of key environmental driving factors; The construction of a dynamic directed graph model of the reservoir bank rockfall disaster chain based on the multi-source dynamic monitoring data includes: Define disaster chain nodes; wherein, the disaster chain nodes include the cumulative state of rock mass damage, the initiation and movement state of rock mass collapse, the impact and splash state of the collapse body, the propagation and rise state of surge waves, the state of bank erosion / structure impact, and the state of secondary landslide / river channel siltation. Based on the disaster mechanism, directed edges between nodes are defined according to the causal relationship and development sequence between nodes, and the basic transfer intensity of each directed edge is quantified; The multi-source dynamic monitoring data is mapped to the state values ​​of the nodes.

2. The dynamic assessment method for the risk of chain-like disasters caused by unstable rock formations on reservoir banks according to claim 1, characterized in that, The acquisition of key environmental driving factors and their impact pathways includes: Key environmental driving factors were identified based on the disaster-causing mechanism of the chain reaction of unstable rock formations on the target reservoir bank. These key environmental driving factors include: reservoir water level fluctuations, rainfall, and fissure water pressure. The influence pathways of the reservoir water level fluctuations include: affecting stability by changing the seepage pressure, buoyancy force, and physical and mechanical properties of the rock mass; The impact pathways of the rainfall include: increasing the self-weight of the soil and rock mass, reducing the shear strength of the structural surface, and generating dynamic water pressure that affects stability; The influence path of the fissure water pressure includes: directly affecting the effective stress on the fissure surface.

3. The dynamic assessment method for the risk of reservoir bank rock chain disasters according to claim 1, characterized in that, The normalized influence functions of the key environmental driving factors include: reservoir water level fluctuation influence function, rainfall influence function, and fissure water pressure influence function; The influence function of reservoir water level fluctuation is calculated using the following method. : ； in, This refers to the 24-hour fluctuation of the water level. It is a symbolic function; For reference amplitude; It is the hyperbolic tangent function; This is the current water level; This is the critical water level for reservoir bank stability; This is the highest water level in history. It is a step function; Weighted by the rate of water level change; For the absolute height of the influence weight, satisfying ,and ; The rainfall impact function is calculated using the following method. : ； in, For the past Hourly cumulative rainfall; This serves as a reference threshold for cumulative rainfall. Real-time rainfall intensity; This serves as a reference threshold for rainfall intensity. Weighting based on cumulative rainfall impact; The weighting of rainfall intensity must satisfy the following conditions. ,and ; The influence function of fracture water pressure is calculated based on the following formula: ； ； in, This is a function representing the influence of crack propagation on stability. This represents the current crack aperture. This represents the initial aperture of the crack. This is a function representing the effect of internal water pressure on stability. This refers to the pore water pressure within the fracture. This is a reference value for pore water pressure. This is the proportionality coefficient.

4. The dynamic assessment method for the risk of reservoir bank rock chain disasters according to claim 1, characterized in that, The dynamic weighting of directed edges based on the dynamic influence weights includes: The real-time strength of directed edges is calculated based on the following formula. : ； in, For directed edges At any moment Real-time intensity; For directed edges Basic transfer strength; For directed edges exist The dynamic influence weight matrix at any given time; The starting node is calculated based on the following formula. Risk status value : ； in, For state quantization function; The deformation damage index; It is an activity index; These are the weighting coefficients. ; The deformation damage index is calculated based on the following formula. ; ； in, This represents the real-time crack width. The critical crack width; For key point displacement rates; The critical displacement rate; The activity index is calculated based on the following formula. : ； in, Real-time acoustic emission rate; The background sound emission rate.

5. The dynamic assessment method for the risk of reservoir bank rock chain disasters according to claim 1, characterized in that, The calculation of the comprehensive risk value based on the real-time strength of dynamically weighted directed edges includes: The overall risk value is calculated based on the following formula: ； in, for The comprehensive dynamic risk value at any given moment; End node Risk contribution weight, satisfying ; For nodes exist Risk status value at any given time; Nodes are calculated based on the following formula. exist Risk status value at any given time: ； in, For nodes exist Risk status value at any given time; Predecessor node exist Risk status value at any given time; For directed edges At any moment Real-time intensity; For nodes The severity coefficient of the consequences.

6. A dynamic assessment system for the risk of reservoir bank rock chain disasters, characterized in that, The system includes: The data acquisition module is configured to acquire multi-source dynamic monitoring data; wherein, the multi-source dynamic monitoring data includes: three-dimensional displacement time series data of the unstable rock mass, acoustic emission signal of the unstable rock mass, hydrological data, meteorological data, and width variation data of the main control fracture of the unstable rock mass; The directed graph module is configured to construct a dynamic directed graph model of the reservoir bank rockfall disaster chain based on the multi-source dynamic monitoring data; wherein, the graph nodes are the stages in the evolution of the physical state of the disaster; and the directed edges are the paths from the pioneer nodes to the successor nodes. The dynamic influence weight module is configured to obtain key environmental driving factors and their influence paths; and to calculate the dynamic influence weight of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank dangerous rock disaster chain. The comprehensive risk module is configured to dynamically weight directed edges based on the dynamic influence weights, and calculate the comprehensive risk value based on the real-time strength of the dynamically weighted directed edges. The calculation of the dynamic influence weights of the key environmental driving factors on each directed edge in the dynamic directed graph model of the reservoir bank rockfall disaster chain includes: The dynamic influence weight of directed edges is calculated based on the following formula. : ； in, For directed edges exist The dynamic impact weight at any given moment; For the edge For driving factors The sensitivity coefficient; As driving factor exist The normalized impact at any given time; The net effect of each environmental factor; This is the normalized influence function of key environmental driving factors; The construction of a dynamic directed graph model of the reservoir bank rockfall disaster chain based on the multi-source dynamic monitoring data includes: Define disaster chain nodes; wherein, the disaster chain nodes include the cumulative state of rock mass damage, the initiation and movement state of rock mass collapse, the impact and splash state of the collapse body, the propagation and rise state of surge waves, the state of bank erosion / structure impact, and the state of secondary landslide / river channel siltation. Based on the disaster mechanism, directed edges between nodes are defined according to the causal relationship and development sequence between nodes, and the basic transfer intensity of each directed edge is quantified; The multi-source dynamic monitoring data is mapped to the state values ​​of the nodes.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, cause the at least one processor to perform the dynamic assessment method for the risk of reservoir bank rock chain disasters as described in any one of claims 1 to 5.

8. A computer program product, characterized in that, The computer program product includes a computing program stored on a non-transitory computer-readable storage medium, the computer program including program instructions that, when executed by a computer, cause the computer to perform the dynamic assessment method for the risk of reservoir bank rock chain disasters as described in any one of claims 1 to 5.

Images