An event-driven method for emergency scenario simulation of sudden events at reservoir dams

Abstract

This invention discloses an event-driven emergency scenario simulation method for reservoir dam emergencies. It constructs an ontology model of the reservoir dam emergency scenario based on event ontology theory; generates a multi-entity Bayesian network model according to the mapping rules between the ontology model and multi-entity Bayesian segments; removes irrelevant branches to construct a specific scenario Bayesian network topology; selects a probability assignment method based on the logical relationships and mechanisms between nodes, initializes probability parameters, and completes the construction of the emergency scenario simulation model; collects current scenario information as evidence and assigns probability values ​​to the root node; uses the driving event as an input variable and determines the probability values ​​of the state variables and output variables of subsequent scenario nodes according to inference rules; and achieves probabilistic simulation and emergency decision assessment of reservoir dam emergency scenarios. This invention enables accurate prediction of reservoir dam emergency scenarios.

Description

Technical Field

[0001] This invention relates to a method for simulating emergency scenarios of reservoir dam emergencies, and more particularly to an event-driven method for simulating emergency scenarios of reservoir dam emergencies, belonging to the field of dam risk management technology in water conservancy engineering. Background Technology

[0002] Reservoirs and dams serve multiple functions, including flood control, power generation, navigation, and irrigation. They are fundamental guarantees for economic and social development and the implementation of major national strategies, and are also important carriers of ecological civilization construction. However, during their operation, reservoirs and dams are subject to unpredictable disaster-causing factors such as inherent defects, mismanagement, extreme weather, and geological disasters. Coupled with insufficient emergency response capabilities in the "prediction-response" emergency management model, dam failures and even dam collapses are a persistent problem, seriously threatening the safety of downstream lives, property, and infrastructure. With global climate change and the increasing frequency of extreme natural disasters, the importance of emergency management for reservoir and dam incidents is further highlighted.

[0003] Emergency management of reservoir dam incidents is the last line of defense for protecting the lives and property of downstream residents and a key aspect of dam risk management. Currently, a relatively comprehensive "one plan, three systems" emergency management system has been initially established, comprising emergency plans, emergency mechanisms, emergency systems, and emergency legislation. This system is based on static emergency plans, selecting high-probability incidents for predictive analysis and planning corresponding emergency management measures. It clarifies the composition of the emergency organization system and the responsibilities and tasks of each unit, develops emergency plans in advance, and activates the emergency plan to respond after an incident occurs, representing a "prediction-response" type of emergency management model. However, based on its management and application, the emergency plan, as a guide for reservoir operators, management units, and the downstream public in responding to incidents, still has many shortcomings. First, emergency plans are stored in text form, which hinders reading and communication among rescue personnel. When an emergency occurs, the command department needs to consult the emergency details based on the level of danger before issuing specific action instructions to the participants, affecting the timeliness of emergency management. Second, reservoir dam emergencies involve complex nonlinear systems involving the elements of nature, the environment, and people. The evolution of the situation and the interaction with emergency measures are significantly uncertain, and emergency plans cannot be quickly and efficiently adjusted to address uncertainties in dynamic scenarios. These shortcomings result in a low level of intelligence in the "prediction-response" emergency management model, poor adaptability of emergency plans, and low efficiency in emergency decision-making, thus restricting the improvement of emergency management capabilities and levels. Reservoir dam emergencies are characterized by uncertainty in evolution, time urgency, and complexity of elements. Emergency departments need to make decisions and judgments based on limited situational information in a very short time, formulate scientific and reasonable emergency plans as quickly as possible, carry out emergency rescue, personnel transfer and rescue, resource allocation, and dynamically adjust emergency measures according to the development of the situation to control the spread of danger and reduce disaster losses to the greatest extent. However, the traditional "prediction-response" emergency management model is difficult to meet the requirements of scientific, dynamic, and effective emergency response. Summary of the Invention

[0004] Purpose of the invention: In view of the problems and shortcomings of the existing technology, the present invention provides an event-driven emergency scenario simulation method for reservoir dam emergencies, so as to realize the quantitative evaluation of the driving effect of disaster-causing factors and emergency measures in emergency management, and provide a reliable decision-making basis for emergency management.

[0005] Technical Solution: An event-driven emergency scenario simulation method for reservoir dam emergencies, comprising the following steps:

[0006] Step 1: Constructing a scenario network structure for sudden events at reservoir dams

[0007] Analysis of case data on reservoir dam emergencies revealed that scenario S consists of four elements: event chain, state elements, causative factors, and emergency activities. A network structure for reservoir dam emergencies was established, consisting of causative factor induction, emergency intervention, state element mapping, and event chain development. The methods used in the analysis of the reservoir dam emergency case data included text extraction and machine learning.

[0008] Step 2: Constructing an ontology model of a reservoir dam emergency scenario

[0009] Based on the network structure of reservoir dam emergency scenarios, a five-element event ontology representation model is used to formally express the knowledge concepts of reservoir dam emergency scenarios. Logical relationships between elements are defined according to the semantic structure of the event ontology, and a data storage framework containing semantic relationships in the reservoir dam emergency scenario domain is constructed, thereby building the reservoir dam emergency scenario ontology model.

[0010] Step 3: Generate a multi-entity Bayesian network model

[0011] Based on the mapping rules of the reservoir dam emergency event scenario ontology model and the multi-entity Bayesian fragment, the mutual mapping of concepts, attributes and relationships in the reservoir dam emergency event scenario ontology model is realized. Key elements in the reservoir dam emergency event scenario ontology model are extracted as network event nodes, and state elements that affect the evolution of the situation are selected as node variable parameters to generate a multi-entity Bayesian network model.

[0012] Step 4: Construct an emergency scenario simulation model

[0013] For multi-entity Bayesian network models, the model structure and scale are dynamically changed according to actual emergency scenarios. Irrelevant branches such as boundary events, conditionally independent events, and irrelevant events are deleted to construct a Bayesian network topology for a specific scenario. The probability assignment method is selected according to the logical relationship and mechanism between nodes, the probability parameters are initialized, and the emergency scenario simulation model is constructed.

[0014] Step 5: Event-driven probability simulation of emergency scenarios

[0015] Collect current scenario information as evidence and assign probability values ​​to the root node. Use the driving event as input variable and determine the probability values ​​of the state variables and output variables of subsequent scenario nodes according to the inference rules to realize the probabilistic deduction of the emergency situation of reservoir dam sudden events.

[0016] Step Six: Emergency Decision Making Based on Scenario Simulation

[0017] Based on scenario simulation results, an emergency decision-making model is constructed, consisting of a decision objective function, boundary conditions, and constraints. Response measures are evaluated based on the probabilities of key scenario nodes and their disastrous consequences during the development of the emergency to determine the optimal emergency management measures.

[0018] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the event-driven emergency scenario simulation method for reservoir dam emergencies as described above.

[0019] A computer-readable storage medium storing a computer program that performs the event-driven emergency scenario simulation method for reservoir dam emergencies as described above.

[0020] Beneficial effects: Compared with the prior art, the event-driven emergency scenario simulation method for reservoir dam emergencies provided by this invention uses modular entity fragments to represent the uncertainty of emergency scenarios, and quantitatively evaluates the driving effects of disaster-causing factors and emergency measures for reservoir dam emergencies based on scenario situation reasoning rules. It solves the problem of difficulty in representing the mutual feedback between scenario situation and response strategies, realizes accurate prediction of reservoir dam emergency scenarios, and determines emergency rescue measures based on scenario simulation results, providing a reliable decision-making basis for emergency management. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the network structure for a reservoir dam emergency scenario in an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram illustrating the representation of a flood inundation event entity in an embodiment of the present invention;

[0023] Figure 3 This is the overtopping dam failure event chain in the scenario ontology model of this invention embodiment;

[0024] Figure 4 This is a schematic diagram of a multi-entity Bayesian network model for sudden rainstorm and flood events in an embodiment of the present invention;

[0025] Figure 5 This is a typical cross-sectional view of the slope in an embodiment of the present invention;

[0026] Figure 6 yes Figure 5 A schematic diagram of the model unit mesh;

[0027] Figure 7 This is a schematic diagram of the emergency scenario simulation model for the PB hydropower station during heavy rain and floods, as described in this embodiment of the invention. Detailed Implementation

[0028] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading the present invention, any modifications of the present invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0029] An event-driven emergency scenario simulation method for reservoir dam emergencies includes the following steps:

[0030] Step 1: Constructing a scenario network structure for sudden events at reservoir dams

[0031] Analysis of reservoir dam emergency incident case data using text extraction or machine learning methods reveals that the emergency incident scenario S consists of four elements: event chain, state elements, disaster-causing factors, and emergency activities. The scenario network structure and logical relationships between elements are shown in [reference needed]. Figure 1 It clarifies the network hierarchy structure consisting of event chain, state element, and driving element, and characterizes the scenario evolution mechanism of reservoir dam emergencies, which is induced by disaster factors, intervened by emergency activities, mapped by state elements, and developed by event chain.

[0032]

[0033] In the formula, S i This is a scene fragment depicting a sudden incident at a reservoir dam. It is the set of event chain elements in scenario segment i, which refers to the scenario network composed of various events in the scenario segment, such as a linear event chain composed of elements such as super-standard flood, gate failure, reservoir water level rise, and flood overflow. It is the set of state elements in scenario segment i. The causal relationship between events is the external representation of the interaction between the state elements of the scenario. That is, the essence of scenario evolution is internal state change, including the rate of change of reservoir water level, reservoir water level elevation, etc. It is the set of emergency activity elements in scenario segment i, which are the influencing factors that play a leading role in the development of the event and the transformation of the scenario, including emergency response plans, emergency rescue measures and evacuation of downstream areas; It is the set of disaster-causing factors in scenario segment i, which refers to the mutating factors in the disaster-prone environment that may exacerbate the risk of dam failure or even lead to dam failure.

[0034] Step 2: Constructing an ontology model of a reservoir dam emergency scenario

[0035] 2.1 Based on the network structure of reservoir dam emergency scenarios, a five-element event ontology representation model is used to formally express the knowledge concept of reservoir dam emergency scenarios.

[0036] e =<A,O,T,P,S>

[0037] In the formula, A represents the action element, O represents the object set, T represents the time when the event occurs, P represents the environment in which the event occurs, and S represents the key attributes and characteristics of the event.

[0038] 2.2 Based on the semantic structure of the event ontology, the logical relationships between elements are defined, and a data storage framework containing the semantic relationships in the domain of reservoir dam emergency events is constructed to build an ontology model of reservoir dam emergency events.

[0039] Step 3: Generate a multi-entity Bayesian network model

[0040] Based on the mapping rules of the reservoir dam emergency event scenario ontology model and the multi-entity Bayesian fragments, the mutual mapping of concepts, attributes, and relationships in the model is realized. Knowledge concepts and state parameters in the reservoir dam emergency event scenario ontology are mapped to entity fragments and nodes within those fragments, and the logical relationships between elements are mapped to the topology of the multi-entity Bayesian network model, thereby generating the multi-entity Bayesian network model.

[0041] The mapping rules are as follows:

[0042] Based on the scenario ontology model structure and multi-entity fragment expressions, mapping rules for the scenario ontology model and multi-entity Bayesian fragments of reservoir dam emergencies are proposed. A conversion channel between the two models is constructed, realizing the mutual mapping of concepts, attributes, and relationships between the models. The mapping rules are as follows: Figure 2 As shown.

[0043] (1) Mapping of situational concepts to multi-entity fragments

[0044] The primary task in mapping the reservoir dam emergency scenario ontology model to the multi-entity Bayesian network model is to establish the association between scenario element concepts and entity fragments. Entity fragments consist of key scenario nodes related to the reservoir dam emergency, while the scenario ontology model is constructed based on scenario element concepts and their relationships; the two models are consistent. Therefore, relevant scenario element concepts defined in the emergency ontology are mapped to entity fragments, and these fragments are named using the scenario concepts from the scenario ontology. First, the top-level event of the reservoir dam emergency is determined, and relevant scenario concepts are obtained from the scenario ontology model. Then, all scenario elements in the reservoir dam emergency scenario ontology model are traversed, and all interrelated scenario element concepts are mapped to entity fragments in the multi-entity Bayesian network.

[0045] (2) Mapping of scenario element state parameters to nodes within entity fragments

[0046] In the scenario ontology model of a reservoir dam emergency, state parameters are used to describe the characteristics, properties, and values ​​of scenario elements. In the multi-entity Bayesian network model, each entity segment has three types of random variable nodes: context nodes, input nodes, and intrinsic nodes, representing the activation conditions, states, and characteristics of the entity segment. Therefore, the set of nodes in a multi-entity Bayesian segment can be generated by mapping the state parameters of the scenario elements in the reservoir dam emergency scenario ontology model. Context nodes serve as the judgment condition for the activation of an entity segment. The preceding events of the entity segment event in the scenario ontology are used as context nodes and represented by Boolean variables to ensure that the entity segment can be activated when the scenario is extrapolated to that node. Intrinsic nodes, as random variables defined in the entity segment, can be directly mapped from the state parameters of the scenario elements corresponding to the entity segment. For ease of calculation, the state parameters are discretized. Input nodes, as the parent nodes of intrinsic nodes, all originate from the intrinsic nodes of the preceding segments. Therefore, the state parameters of the preceding events in the reservoir dam emergency scenario ontology model are mapped to input nodes.

[0047] (3) Mapping of logical relationships between scenario elements to topological structure

[0048] In the ontology of a reservoir dam emergency scenario, the logical relationships between scenario elements include "following relationships," "causal relationships," "exclusion relationships," and "concurrency relationships." "Following relationships" and "causal relationships" can be considered hierarchical relationships, allowing for the identification of parent nodes. These relationships can be constructed using directed edges in a Bayesian network to connect upper and lower nodes, establishing the logical relationships between nodes. "Exclusion relationships" and "concurrency relationships," however, are considered non-hierarchical relationships and have no direct impact on the probabilistic deduction of the emergency scenario. Therefore, the topology network construction only considers hierarchical relationships between scenario elements to simplify the mapping rules of the topology structure. The topology structure is a crucial part of the multi-entity Bayesian network model, comprising directed edges between entity segments and a directed acyclic graph of nodes within entity segments. Based on the logical relationships between scenario elements in the reservoir dam emergency scenario ontology model, the interaction relationships between entity segments in the multi-entity Bayesian network are determined and represented by directed edges. The relationships between state parameters of scenario elements are mapped to the relationships between nodes within entity segments, thereby constructing the topology structure of nodes within entity segments.

[0049] Step 4: Construct a scenario-specific Bayesian network model, i.e., an emergency scenario simulation model.

[0050] The model structure and scale are dynamically changed according to the actual emergency situation. Irrelevant branches such as boundary events, conditionally independent events, and irrelevant events in the multi-entity Bayesian network model are deleted to construct a Bayesian network topology for a specific scenario. The probability assignment method is selected according to the logical relationship and mechanism between nodes, the probability parameters are initialized, and the emergency scenario simulation model is completed. The specific implementation steps are as follows:

[0051] Step 41: Dynamically change the model structure and scale according to the actual emergency situation, and delete irrelevant branches such as boundary events, conditionally independent events and irrelevant events of the multi-entity Bayesian network model to construct a Bayesian network topology for a specific scenario.

[0052] Step 42: Select the probability assignment method based on the logical relationship and mechanism between nodes.

[0053] The assignment methods are:

[0054] (1) Expert experience assignment method based on DS evidence theory

[0055] Experts from different fields were invited to evaluate the probabilities of scenario state nodes based on probability estimation tables from their respective professional perspectives. Since these experts possess different knowledge domains, their evaluations were not consistent. The DS evidence theory is widely used in expert systems, information fusion, and multi-attribute decision-making, and possesses good objectivity and the ability to integrate multi-source information. Therefore, this embodiment uses DS evidence theory to integrate the evaluation opinions of different experts, increasing the objectivity and accuracy of probability assignment.

[0056] (2) Parameter learning method based on sample data

[0057] Sample data analysis methods use historical case data as a sample database to estimate conditional probability distribution parameters. However, the consequences of sudden dam failures are severe, and historical case data is extremely limited, resulting in a small sample data set with some missing data. Therefore, this embodiment selects the Expectation-Maximization (EM) algorithm, which is suitable for samples with missing data. The EM algorithm is an iterative optimization algorithm that estimates missing data variables by alternately performing expectation and maximization steps to construct a complete training sample set. Assume the sample data set of sudden dam failures is X = {X1, X2, ..., X...} n The missing dataset is Y = {Y1, Y2, ..., Y}. n Let θ be the parameter to be estimated, and P(X,Y|θ) be the joint probability distribution of X and Y after θ is determined.

[0058] (3) Structural reliability analysis method

[0059] For scenario nodes in emergencies with clear logical relationships and well-defined mechanisms of action, structural reliability analysis is used to determine node probability parameters. Based on the logical relationships and mechanisms between events, uncertainty influencing factors are identified, and the function functions of the scenario events are established using methods such as finite element analysis and response surface equations.

[0060] Z = g(X1,X2,...,X) n )

[0061] In the formula, Z is the function of a certain scenario event; X1, X2, ..., X n Let Z be a random variable that influences the state of a scenario event. Structural reliability refers to the probability of achieving a predetermined function under specified conditions. The function describes the functional relationship between the structural state and the random variable. When the function Z is less than 0, it indicates that the structure is in a failure state, i.e., P(Z<0) is the failure probability of the structure. Therefore, by solving the function using methods such as the JC method, Monte Carlo method, and the first second-moment method, the scenario node probability parameters can be determined.

[0062] Step 43: Based on the assignment results, initialize the probability parameters and complete the construction of the emergency scenario simulation model.

[0063] Step 5: Event-driven probability simulation of emergency scenarios

[0064] The system collects current scenario information as evidence and assigns probability values ​​to the root node. Using the driving event as input variables, it determines the probability values ​​of the state variables and output variables of subsequent scenario nodes according to inference rules, thus realizing the probabilistic deduction of the emergency situation of a reservoir dam emergency. The specific deduction steps are as follows:

[0065] Step 51: After the initial scenario of the emergency occurs, collect current scenario information as evidence and assign probability values ​​to the root node. Then, predict and infer the emergency situation according to the inference rules to determine the probability values ​​of the state variables and output variables of subsequent scenario nodes. This process is repeated layer by layer until the final scenario node, thereby determining the state probability of each event along the evolution path. In the emergency scenario simulation model, the driving event nodes are input variables. Disaster event nodes are in an uncertain state, while emergency response event nodes, whose variable type is Boolean, are in a non-responsive state.

[0066] Step 52: As the real-world scenario evolves, new catastrophic events or emergency response decisions continuously occur, thus affecting the state probabilities of scenario events. When a driving event occurs, the corresponding event node becomes an evidence node, i.e., a deterministic state. Then, the scenario node probabilities in the model are updated according to the inference rules, thereby driving the dynamic probability deduction of the emergency scenario.

[0067] This step involves the continuous change of parameters and state values ​​in the model as the scenario evolves. When a new disaster-causing event occurs or an emergency decision is made, the state of the nodes in the model changes from an uncertain node to a deterministic node. For example, the original emergency construction of a sub-dam had two states: yes and no. After taking emergency construction measures, the node becomes a deterministic node with a state value of yes and a probability of 100%.

[0068] The rules for situational reasoning are as follows:

[0069] The scenario simulation model uses a Bayesian network to express the relationships between sub-events. Assume that sub-event EN exists in the scenario. i and subevent EN j There is a correlation, let I(i) j1 i j2 ,…,i jn ) is EN j The input variable, S(s) j1 ,s j2 ,…,s jm ) is EN j The state variable, O(o j1 ,o j2 ,…,o jt ) is EN j The output variables. The inference rules of the scenario simulation model are as follows:

[0070] (1) Input variables and state variables

[0071] P(s je )=∑P(s je |i j1 i j2 ,...,i jn )P(i j1 i j2 ,...,i jn (5-1)

[0072] P(s je )=∑P(s je |i j1 i j2 ,...,i jn ,s i1 ,s i2 ,...,s im )P(i ji1 i j2 ,...,i jn ,s i1 ,s i2 ,...,s im (5-2)

[0073] Formula (5-1) indicates that this state variable is only related to EN. j The input variable is related to the state variable, and formula (5-2) indicates that the state variable is related to EN. j Input variables and EN i The state variables are related. Among them, P(s) je ) represents the state variable s je The probability of occurrence; n and m are I j and S i The number of variables.

[0074] (2) State variables and state variables

[0075] P(s je )=∑P(s je |s j1 ,s j2 ,...,s jm )P(s j1 ,s j2 ,...,s jm (5-3)

[0076] Formula (5-3) indicates that this state variable is only related to EN. j Other state variables are relevant. In the formula, P(s) je ) represents the state variable s je The probability of occurrence; m is S j The number of variables.

[0077] (3) Output variables and state variables

[0078] P(o je )=∑P(o je |s j1 ,s j2 ,...,s jm )P(s j1 ,s j2 ,...,s jm (5-4)

[0079] Formula (5-4) indicates that the output variable is related to EN. j State variables are relevant. In the formula, P(o je ) is the output variable o je The probability of occurrence; m is S j The number of variables.

[0080] (4) Relationships between variables in events

[0081]

[0082] Formula (5-5) represents EN i Output variable I j and EN j Input variable O i The intersection is not an empty set. In the formula, t and n are I... j and O i The number of variables.

[0083] Step Six: Construct an emergency decision-making model based on scenario simulation.

[0084] The response measures are evaluated based on the probability of key scenario nodes and the consequences of disasters during the development of an emergency to determine the optimal emergency management measures. The emergency decision-making model based on scenario simulation consists of a decision objective function, boundary conditions, and constraints. The expected loss of life in the downstream risk area is used as the objective function of the emergency decision-making model; the boundary conditions are external factors controlling the decision objective, which are elements of the downstream risk map, including information such as the scope of the downstream risk area, the distribution of at-risk population, and the transportation network; the constraints are key scenario variables affecting the effectiveness of the response, including emergency rescue measures and emergency relief measures.

[0085] A scenario-response emergency decision-making model was constructed using the expected downstream life and economic losses as the objective function:

[0086]

[0087] Constraints:

[0088]

[0089]

[0090] In the formula, Z1 represents the expected downstream life and economic losses; P i L represents the probability of the i-th scenario occurring; i Let x represent the number of lives lost in scenario i; F1 represents the probability calculation function for scenario i; D1 represents the life loss calculation function; x i Let c be the constraint condition, and let c be the random variable representing the scenario element that affects the treatment effect in the i-th scenario; i Let be the boundary condition, representing the flood risk index that affects life in the i-th scenario.

[0091] 6.1 Determine flood risk indicators based on scenario simulations, conduct flood risk analysis, and then estimate the resulting losses.

[0092] 6.2 Input the scenario probability extrapolation results and disaster loss estimation results into the emergency decision-making model to determine emergency rescue measures and generate specific emergency plans.

[0093] Emergency scenario simulation models, on the one hand, enable probabilistic prediction of scenario development trends and real-time monitoring of the impact of driving events on scenario evolution; on the other hand, they quantify and analyze the effectiveness of emergency measures by reasoning about the probabilistic states of scenarios driven by different emergency response measures. By coupling emergency scenario probabilistic simulation with emergency decision-making models that address disaster losses, more scientific and effective emergency measures can be selected, providing a reliable basis for emergency decision-making.

[0094] Additional information needed (existing technology)

[0095] Multi-Entity Bayesian Networks (MEBN)

[0096] Multi-Entity Bayesian Networks (MEBNs) are an extension of ordinary Bayesian Networks. Their core idea is to use modular entity fragments (MFrags) to describe event attributes and relationships. Since the MEBN model expresses knowledge of the event domain based on rule fragments, integrating these fragments using multi-entity rules for logical reasoning results in a clearer and easier-to-update network structure. Essentially, it's a knowledge base representation logic based on rule fragments, with core entity fragments consisting of quintuples.

[0097] F =<C,I,R,G,D>

[0098] In the formula, C is a context node, which is a logical (true or false) random variable representing the enabling condition of an intrinsic node defined in an entity fragment; R is an intrinsic node, which is a random variable defined in the entity fragment; I is an input node, which is an intrinsic node referenced from other entity fragments; G is the topology structure composed of I and R nodes in the fragment; and D is the probability distribution between nodes.

[0099] Figure 2 This diagram illustrates the entity fragment representation of a flood overtopping event. This fragment describes the probabilistic relationship between a rainstorm flood event and a reservoir overtopping event. The rounded rectangle at the top of the fragment is the context node, specifying the activation condition of the fragment; that is, the overtopping event entity fragment is only activated when a flood event occurs. Flood_Type represents the input node I; Overtopping represents the inherent node R; the topology of the events is a directed acyclic graph G, with arrows indicating logical relationships between events, showing that the probability of a flood overtopping event is related to the flood magnitude; the table shows the probability distribution relationship between the events D.

[0100] DS Evidence Theory

[0101] Suppose that in a Bayesian network, there are nodes in a certain scenario (S1, S2, ... S...). m There are m states, and n experts E were invited. n The evaluation was conducted using a selection process (n = 1, 2, 3…). The expert decision results are shown in Table 1.

[0102] Table 1 Expert Evaluation Results

[0103]

[0104] Based on the DS evidence theory, the node state is used as the identification framework, P(A=S) is used as the mass function, and the expert decision opinions are comprehensively fused using the synthesis rule. The fusion formula is as follows:

[0105]

[0106] In the formula, P j (A=S) i The expert's given A=S i The probability of K is the normalization factor, which is used to measure the degree of conflict between different pieces of evidence. The smaller the value, the more contradictory the evidence.

[0107] Example:

[0108] To verify the scientific validity and applicability of the emergency scenario simulation method for reservoir dams, the sudden rainstorm and flood event encountered by the PB hydropower station project was selected as an example for application.

[0109] I. Using machine learning, we extract and identify professional terms related to reservoir dam emergencies from relevant norms, standards and works such as "Guidelines for Constructing Production Safety Accident Scenario", "Construction of Major Emergency Scenario - Theory and Practice", and "Guidelines for the Compilation of Emergency Response Plans for Reservoir Dam Safety Management". Based on the reservoir dam emergency scenario network structure, we complete the construction of the reservoir dam emergency scenario ontology model.

[0110] II. Constructing a multi-entity Bayesian network model for sudden rainstorm and flood events at the PB hydropower station.

[0111] 2.1 Extracting the knowledge network of the rainstorm and flood event chain from the ontology model of sudden events at reservoir dams, see... Figure 3 .

[0112] 2.2 Mapping to generate a multi-entity Bayesian network model for PB rainstorm and flood emergencies.

[0113] Based on the overtopping and dam failure event chain in the scenario ontology model of reservoir dam emergencies, a multi-entity Bayesian network model of the PB hydropower station's rainstorm and flood emergencies is constructed according to the scenario ontology-multi-entity Bayesian fragment mapping rule. For example, extreme rainfall is mapped to rainstorm and flood fragments (scenario element concepts-multi-entity fragments), the causal relationship between reservoir bank landslide events and surge waves is mapped to the parent node relationship between the near-dam reservoir bank landslide fragment and the surge wave fragment in front of the dam, and the logical relationship between the state elements of reservoir bank landslide events and surge wave events is mapped to the landslide volume and sliding surface nodes of the reservoir bank landslide fragment being input nodes for the surge wave height nodes in the surge wave fragment in front of the dam. Figure 4 As shown, the multi-entity Bayesian network model generated by the mapping consists of 11 segments, including gate failure, near-dam bank landslide, spillway structural damage, dam inrush, and dam overtopping.

[0114] III. Constructing a Bayesian Network Model for a Specific Context

[0115] The multi-entity Bayesian network model clarifies the Bayesian network topology of the PB hydropower station's sudden rainstorm and flood event, and provides the causal relationships between nodes within the entity segments. To achieve probabilistic inference of the scenario model, it is necessary to assign values ​​to the node probability parameters, define the element node states and threshold intervals, and determine the probability distribution table, including the root node probability and the conditional probabilities of interactions between nodes. The node probability distribution within the entity segments must be unique to satisfy the consistency constraints of the Bayesian network.

[0116] Based on the analysis method given in step four, define local probability distributions for the node variables in the multireal Bayesian network model.

[0117] The following section uses examples of landslides near the dam and gate failures to illustrate the points.

[0118] A section of landslide near the dam bank

[0119] Landslides and surges near the dam bank caused by slope instability are a common risk in dam overtopping accidents in water conservancy and hydropower projects, and dam failures resulting from such incidents occur frequently. In southwestern my country, many existing reservoir dams are located in high mountains and deep valleys, with complex terrain and geological conditions, steep banks, and numerous potential landslide hazards, making them a significant source of risk threatening dam operation safety. Landslides and surges near the dam bank involve two key stages: slope instability and the subsequent landslide surge.

[0120] (1) Landslide root node near the dam and reservoir bank

[0121] In the numerical analysis of slope stability in rock and soil masses, the finite element method (FEM) is used to analyze the risk of landslides near the dam reservoir bank. A finite element model is constructed based on a typical profile of the tensile deformation body of the right bank slope of the PB hydropower station. The bottom elevation of the model is 600m. Figure 5 and 6 As shown, there are a total of 4071 elements. Horizontal and fixed constraints are applied to the left, right, and bottom boundaries of the model, while there are no constraints at the top.

[0122] For slope reliability, the rock mass shear strength parameters have the most significant impact on slope stability. The rock mass cohesion and internal friction angle are regarded as random variables that follow a normal distribution, and the physical and mechanical parameters of various types of rock masses in tensile fracture deformation bodies are determined.

[0123] Potential failure pathways in the landslide were searched using the strength reduction method. The slip arcs were located at the interface between the loosened deformed mass and the underlying strongly unloaded rock mass, with some slip arcs penetrating the interior of the loosened deformed mass. A stability function for the failure pathways was constructed based on the response surface methodology, and its structure is shown below:

[0124]

[0125]

[0126]

[0127] In the formula, G L (X) is the instability failure function; n is the total number of elements on the slip surface; b i is the length of element i along the slip surface direction; is the normal stress of element i on the slip surface; is the shear stress of element i on the slip surface; f i ′ is the friction coefficient of element i on the slip surface; c i is the cohesion of element i on the slip surface; X is the random variable x 1 , x 2 , ···, x n ; a0, b i and c i are the parameters of the response surface equation.

[0128] According to the distribution characteristics of random parameters, 25 groups of schemes were designed by the orthogonal test method. The undetermined coefficients of the response surface equation were determined through the finite element simulation results and statistical analysis. The multiple correlation coefficient R 2 of the response surface equation reached 0.98, and the fitting effect was good. Based on the slope instability function and the response surface equation, Monte-Carlo simulation was carried out on the most dangerous slip surface. The sampling results showed that the slope instability probability was 3.37×10 -4 . Thus, the state probability of the root node of the landslide of the reservoir bank slope was determined as P(Slope = YES) = 3.37×10 -4 , P(Slope = NO) = 0.999663.

[0129] (2) Node of the wave height in front of the dam

[0130] The wave generated by the landslide was calculated by the Pan Jiazheng algorithm. It was assumed that the river water body was a semi-infinite long water body with parallel sides and a width of B. The cross-section of the reservoir bank within the range L where the landslide body was located was consistent. The calculation formula for the wave height ξ (h″ w ) at the dam site during the time period 0 < t < T was:

[0131]

[0132] In the formula, k r is the reflection coefficient of the wave propagating to the front of the dam; ξ0 is the initial wave height when the sliding body enters the water, which is directly affected by the average water depth of the reservoir , and can be expressed as ξ0(h″ w ) and calculated according to formulas (3-5) to (3-7); x0 is the distance from the farther end of the landslide body to the dam site; L is the width of the landslide body along the reservoir bank; k is the reflection coefficient of the wave propagating to the opposite bank; θ n ​​It is the angle between the nth incident ray of the wave and the normal to the bank slope; n is the number of superimposed waves that produce the maximum wave height in front of the dam, which is determined by the ratio of the landslide duration to the time required for the waves to propagate to the opposite bank, T / Δt, as shown in Table 2, where Δt=B / v c v c Let be the propagation wave speed of the surging waves in the water, calculated according to formula (3-7).

[0133] Table 2. Correspondence between landslide duration and the number of superimposed waves.

[0134]

[0135] ξ0(h″ w )=ξ h (h″ w cos 2 α+ξ v (h″ w sin 2 α (3-5)

[0136]

[0137]

[0138] In the formula, λ is the average thickness of the landslide block; ξ h (h″ w ), ξ v (h″ w The landslide is divided into horizontal and vertical velocity components (v). h v v The surge height caused by ) follows the following formula:

[0139] v h =v·cosα; v v =v·sinα (3-8)

[0140]

[0141]

[0142]

[0143] In the formula, a is the overall slope angle of the landslide; W i a i Let be the weight of the i-th slider and its average slope angle; c and f are the cohesion and friction coefficient of the landslide body; a 0i Let be the slope angle of the line connecting the midpoints of the (i-1)th and i-th sliders; W is the total weight of the landslide.

[0144] The most dangerous sliding surface determined by finite element analysis was taken as the sliding surface. The collapse range was approximately 360m wide, based on the boundary of the tensile deformation body on the geological plan. The riverbed was 80m wide, and the opposite bank slope angle was 40°. Since the landslide body was above the water surface, it was assumed that only frictional force f existed at the contact surface during the slope sliding process, with no pore water pressure. Dynamic strength parameters were used in the calculation. Since the friction coefficient after the landslide started should be the dynamic friction coefficient, the internal friction angle was determined to be 20.96° and the cohesion to be 0.11 MPa, based on the review report of the slope stability study. The landslide body was divided into 18 strips along the horizontal direction, each strip with a width ΔL of 10m. Since the landslide body was above the water surface, it was assumed that only frictional force f existed at the contact surface during the slope sliding process, with no pore water pressure. According to formulas (3-5) to (3-11), the velocity reached its maximum value v at t = 55.73s. max = 5.96 m / s. Taking the normal water level of 850.00 m and the dam crest elevation of 856.00 m as examples, the surge height is calculated to be 5.92 m and 6.14 m respectively.

[0145] Based on the physical fragments, the influencing factors of the surge height in front of the dam include the volume of the landslide mass, the sliding surface, the reservoir water level, and the landslide location. Finite element analysis of slope stability determined the sliding surface, volume of the landslide mass, and location parameters of the tensile fracture. Therefore, the uncertainties surrounding the surge height event in front of the dam are the landslide event and the reservoir water level. Since the reservoir water level is a continuous variable, it is discretized to facilitate the application of Bayesian network inference rules. The reservoir water level is discretized into five state values ​​based on its characteristic values: (790, 850], (850, 853.78], (853.78, 855], (855, 856], and (856, ∞). Since the fractured body is located above the reservoir water level, the two are considered independent events. A functional relationship between the surge height and the reservoir water level is established according to formula (3-4). Therefore, the surge height is also discretized into five state values ​​based on the reservoir water level: 0, (0, 5.92], (5.92, 6.06], (6.06, 6.10], and (6.10, 6.14). Based on the logical relationship between the scenario elements, the probability calculation formula for the surge height state is given as follows:

[0146]

[0147] P(H i )=P(S YES )·P(L i-1 i = 2, 3, 4 (3-13)

[0148] P(H5)=P(S YES )·(P(L4)+P(L5)) (3-14)

[0149] Gate failure segment two

[0150] Discharge capacity is one of the core indicators for ensuring the safe passage of floodwaters through a dam. Under deterministic parameter conditions, the dam discharge can be accurately calculated through hydraulic calculations. However, sudden disasters such as gate failure and spillway structural damage introduce uncertainty into the actual discharge capacity of the dam. Numerous factors affect the normal opening and closing of gates, including gate strength and stiffness, stability, backup power supply, management systems and regulations, and structural aging. Furthermore, the relationships between these influencing factors are complex, making it difficult to accurately assess the gate failure probability through numerical models and logical reasoning. Therefore, an expert experience-based assignment method based on DS evidence theory is used to assess the gate failure probability.

[0151] Based on the survey and statistical results of the gate opening and closing situation during floods, and the assessment basis of gate failure events in relevant literature, a table of probability estimation for individual gate failure is given based on the principles of operability and quantifiability. This table serves as the basis for experts to assess the probability of gate failure. (See Table 3)

[0152] Table 3. Estimation of Failure Probability for a Single Gate

[0153]

[0154] Five senior experts in the field of hydraulic engineering with backgrounds in metal structures and over ten years of experience in reservoir dam safety assessment were invited to evaluate the state probability of individual gates based on the gate failure probability assessment table, using data on the characteristics, design, construction, and operation management of the PB hydropower station project as a reference. The expert assessment results were collected, and the results were evaluated using the evidence theory fusion formula. The final state probability of a single gate failure node was determined to be P(Gate_failure=YES)=0.27×10⁻⁶. -2 , P(Gate_failure=NO)=0.9973.

[0155] IV. Application of Emergency Scenario Probability Simulation for Reservoir Dams

[0156] The states and thresholds of the scenario network nodes are shown in Table 4. Event element nodes are described using Boolean variables; for example, the node state for gate failure is "occurred" or "not occurred." State element nodes are determined based on engineering characteristics and probability distributions; for example, the rainstorm / flood node is determined according to the flood design standards of the PB hydropower station. After determining the node states and probability distributions, based on the assumed scenario information of the PB hydropower station, corresponding entity fragments are selected from the multi-entity Bayesian network model to synthesize a scenario event node chain. Irrelevant events, isolated events, and marginal events are removed, forming a specific scenario Bayesian network model that extends from evidence events to target reasoning events, as shown in Table 4. Figure 7 As shown, this is the scenario simulation model for a sudden rainstorm and flood event at the PB hydropower station.

[0157] Table 4. Network Node Status and Thresholds for Emergency Scenario of Rainstorm and Flood at PB Hydropower Station

[0158]

[0159] In the evolution of a sudden event at a reservoir dam, multiple driving factors change dynamically over time. Scenario simulation is essentially a dynamic recursive process. To ensure the accuracy of the simulation results, the node states of the scenario model are updated in a timely manner based on the driving factors. In actual scenario simulations, uncertainties in disaster-causing factors and emergency measures can lead to different scenarios evolving in different directions. To clarify complex issues and facilitate model construction and verification, the evolution process of the sudden event scenario is simplified, considering only the core event chain and key influencing factors. The simulation process of the initial scenario of a rainstorm and flood at the PB hydropower station and subsequent encounters with different disaster-causing factors is as follows:

[0160] Scenario 1: During the flood season, the PB hydropower station reservoir water level is at the flood control limit of 841.00m, encountering a once-in-10,000-year rainstorm flood. No other disaster-causing factors are found to be abnormal. Using the once-in-10,000-year rainstorm flood as the initial disaster-causing factor, the state probability of the root node of the rainstorm flood in the extrapolation model is updated. In the extrapolation results, the probability of the reservoir water level peak being in the L3(850,853) interval is 97.5%, and the probability of the dam overflowing and breaking is 0.54%. The results indicate that, in the absence of other disaster-causing events, when encountering a once-in-10,000-year rainstorm flood, the safe operation of the hydropower station can be ensured through normal flood control scheduling, without the need to initiate other emergency measures. The PB hydropower station's check flood standard is the most probable flood (PMF), and its flood control capacity exceeds the 10,000-year flood standard. Furthermore, the station has a sound management and maintenance system, its spillway structures and energy dissipation facilities meet regulatory requirements, its gate system is safe and reliable, and the reservoir bank slopes near the dam are reinforced with anchor bolts. Therefore, the probability of disasters such as gate failure, spillway structural damage, and near-dam landslides is low. Based on the current status of the hydropower station, the model simulation results are consistent with the actual risk state of the hydropower station under a 10,000-year torrential flood scenario.

[0161] Scenario 1.1: During the evolution of Scenario 1, the slope safety monitoring system issued an alarm, indicating significant anomalies in slope displacement under the influence of heavy rainfall, and a marked increase in landslide risk. To assess the impact of near-dam bank landslide hazard factors on the overtopping event chain during the flood season, this was input as a deterministic event into the extrapolation model. The extrapolation results show that when the PB hydropower station encounters a once-in-10,000-year rainstorm and flood, the probability of the surge height in front of the dam caused by a near-dam bank landslide being H4(6.06, 6.10] is 78.4%, and the probability of the dam overtopping and failure event is 61.5%. Flood evolution analysis determined that the loss of life under this condition would be 1453 people. Considering the emergency measure of constructing a secondary weir on the dam crest, this was also input as a deterministic event, reducing the probability of the overtopping event to 12.4%, and the loss of life under this condition would be 163 people. Five people were involved. The simulation results show that the successful implementation of the emergency measure of constructing a secondary weir on the dam crest can significantly reduce the probability of dam overtopping and failure caused by near-dam bank landslides. In the event of an overtopping and failure, although constructing the secondary weir on the dam crest will increase the peak flood flow, the significantly reduced probability of failure lowers the expected loss of life from Z1 = 894 to Z2 = 202. The scenario simulation model demonstrates that, under the current scenario, constructing a secondary weir on the dam crest can significantly reduce the risk of overtopping and failure, providing a quantitative basis for emergency decision-making.

[0162] In summary, the scenario simulation results for the PB hydropower station's sudden rainstorm and flood event are largely consistent with the actual engineering situation. The event-driven emergency scenario simulation model can infer the probability changes of scenario node states and accurately predict the development trend of the scenario. The results of the engineering case application are consistent with the actual situation, verifying the scientificity and effectiveness of the method, realizing the quantitative evaluation of the disaster-causing factors and the driving effect of emergency measures, and providing a reliable decision-making basis for emergency management.

[0163] Obviously, those skilled in the art should understand that the steps of the event-driven emergency scenario simulation method for reservoir dam emergencies described in the above embodiments of the present invention can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, and thus stored in a storage device for execution by the computing device. In some cases, the steps shown or described can be executed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Therefore, the embodiments of the present invention are not limited to any particular hardware and software combination.

Claims

1. An event-driven emergency scenario simulation method for reservoir dam emergencies, characterized in that, Includes the following steps: Step 1: Constructing a scenario network structure for sudden events at reservoir dams By analyzing case data of reservoir dam emergencies, we can obtain the scenarios of reservoir dam emergencies. S It consists of four elements: event chain, state element, disaster-causing factor, and emergency activity. The emergency network structure of reservoir dam emergencies is determined as follows: disaster-causing factor induction - emergency activity intervention - state element mapping - event chain development. The methods used for the analysis of reservoir dam emergency case data include text extraction and machine learning. Step 2: Constructing an ontology model of a reservoir dam emergency scenario Based on the network structure of reservoir dam emergency scenarios, a five-element event ontology representation model is used to formally express the knowledge concept of reservoir dam emergency scenarios. The logical relationship between elements is defined according to the semantic structure of the event ontology, and a data storage framework containing the semantic relationship in the domain of reservoir dam emergency scenarios is constructed. Based on this data storage framework, the reservoir dam emergency scenario ontology model is constructed. Step 3: Generate a multi-entity Bayesian network model Based on the mapping rules of the reservoir dam emergency event scenario ontology model and the multi-entity Bayesian fragment, the mutual mapping of concepts, attributes and relationships in the reservoir dam emergency event scenario ontology model is realized. Key elements in the reservoir dam emergency event scenario ontology model are extracted as network event nodes, and state elements that affect the evolution of the situation are selected as node variable parameters to generate a multi-entity Bayesian network model. Step 4: Construct an emergency scenario simulation model For the multi-entity Bayesian network model, the model structure and scale are dynamically changed according to the actual emergency scenarios. Irrelevant branches are deleted to construct the scenario Bayesian network topology. The probability assignment method is selected according to the logical relationship and mechanism between nodes, the probability parameters are initialized, and the emergency scenario simulation model is constructed. Step 5: Event-driven probability simulation of emergency scenarios Collect current situation information as evidence information and complete the probability assignment of the root node. Use the driving event as the input variable of the emergency scenario simulation model, determine the probability values ​​of the state variables and output variables of the subsequent scenario nodes according to the inference rules, and realize the probabilistic simulation of the emergency situation of the reservoir dam sudden event. Step Six: Emergency Decision Making Based on Scenario Simulation Based on the scenario simulation results, an emergency decision-making model consisting of a decision objective function, boundary conditions, and constraints is constructed; the response measures are evaluated based on the probability of key scenario nodes and the disaster consequences during the development of the emergency to determine the optimal emergency management measures. In step three, based on the scenario ontology model structure and multi-entity fragment expressions, mapping rules were proposed between the reservoir dam emergency scenario ontology model and the multi-entity Bayesian fragments. A conversion channel between the two models was constructed, enabling mutual mapping of concepts, attributes, and relationships between the models; including: (1) Mapping of situational concepts to multi-entity fragments; (2) Mapping of scenario element state parameters to nodes within entity segments; (3) The mapping of logical relationships between scenario elements to topological structure; In step six, the emergency decision-making model based on scenario simulation consists of a decision objective function, boundary conditions, and constraints. The expected loss of life of people in the downstream risk area is used as the objective function of the emergency decision-making model. The boundary conditions are external factors that control the decision objective and are elements of the downstream risk map, including the scope of the downstream risk area, the distribution of the at-risk population, and transportation network information. The constraints are key scenario variables that affect the effectiveness of the response, including emergency rescue measures and emergency relief measures. 6.1 Determine flood risk indicators based on scenario simulations, conduct flood risk analysis, and then estimate the resulting losses; 6.2 Input the scenario probability extrapolation results and disaster loss estimation results into the emergency decision-making model to determine emergency rescue measures and generate specific emergency plans.

2. The event-driven emergency scenario simulation method for reservoir dam emergencies according to claim 1, characterized in that, In step one, the scenario network structure for the reservoir dam emergency is as follows: In the formula, S i This is a scenario involving a sudden incident at a reservoir dam. i Fragment; It is a scene segment i The event chain element set refers to the context network composed of various events in a context segment; It is a scene segment i The set of state elements in the middle, the causal relationship between events is the external representation of the interaction between state elements in the situation, that is, the essence of situation evolution is internal state change; It is a scene segment i A collection of elements for emergency response activities in China; It is a scene segment i A set of disaster-causing factors.

3. The event-driven emergency scenario simulation method for reservoir dam emergencies according to claim 1, characterized in that, Step two, constructing an ontology model of a reservoir dam emergency scenario, including the following steps: 2.1 Based on the network structure of reservoir dam emergency scenarios, a five-element event ontology representation model is used to formally express the knowledge concept of reservoir dam emergency scenarios: e=<A,O,T,P,S> In the formula, A As action elements, O For a collection of objects, T The time when the event occurred. P The environment in which the event occurred. S Key attributes and characteristics of an event; 2.2 Define the logical relationships between elements based on the semantic structure of the event ontology, construct a data storage framework that includes the semantic relationships in the domain of reservoir dam emergency events, and construct an ontology model of reservoir dam emergency events based on this data storage framework.

4. The event-driven emergency scenario simulation method for reservoir dam emergencies according to claim 1, characterized in that, In step three, based on the scenario ontology model structure and multi-entity fragment expression, a mapping rule for the scenario ontology model of reservoir dam emergencies and multi-entity Bayesian fragments is proposed, and a conversion channel between the two models is constructed to realize the mutual mapping of concepts, attributes and relationships between the models. (1) Mapping of situational concepts to multi-entity fragments The primary task in mapping the reservoir dam emergency scenario ontology model to the multi-entity Bayesian network model is to establish the association between scenario element concepts and entity fragments. Entity fragments consist of reservoir dam emergency scenario nodes, while the scenario ontology model is constructed based on scenario element concepts and their relationships, and the two models are consistent. Therefore, the relevant scenario element concepts defined in the emergency ontology are mapped to entity fragments, and the entity fragments are named using the scenario concepts in the scenario ontology. First, the top-level event of the reservoir dam emergency is determined, and the relevant scenario concepts are obtained from the scenario ontology model. Then, all scenario elements in the reservoir dam emergency scenario ontology model are traversed, and all scenario element concepts with interrelationships are mapped to entity fragments in the multi-entity Bayesian network. (2) Mapping of scenario element state parameters to nodes within entity fragments In the scenario ontology model of a reservoir dam emergency, state parameters are used to describe the characteristics, properties, and values ​​of scenario elements; in the multi-entity Bayesian network model, each entity segment has three types of random variable nodes: context nodes, input nodes, and intrinsic nodes, to characterize the activation conditions, states, and features of the entity segment. Therefore, the node set in the multi-entity Bayesian fragment can be generated by mapping the state parameters of the scenario elements in the reservoir dam emergency scenario ontology model; the context node serves as the judgment condition for the activation of the entity fragment, taking the preceding events of the entity fragment event in the scenario ontology as the context node and representing them with Boolean variables to ensure that the entity fragment is activated when the scenario is deduced to this node; the inherent node is a random variable defined in the entity fragment, mapping the state parameters of the scenario elements corresponding to the entity fragment to the inherent node, and discretizing the state parameters; the input node, as the parent node of the inherent node, comes from the inherent nodes of the preceding fragment, so the state parameters of the preceding events in the reservoir dam emergency scenario ontology model are mapped to the input node; (3) Mapping of logical relationships between scenario elements to topological structure In the scenario ontology of a reservoir dam emergency, the logical relationships between scenario elements include "following relationship," "causal relationship," "exclusion relationship," and "concurrency relationship." Among them, "following relationship" and "causal relationship" are considered hierarchical relationships. Parent nodes are identified, and directed edges are used in the Bayesian network to connect the hierarchical and superior nodes, thus constructing the node logical relationships. The topology network construction only considers the hierarchical relationships between scenario elements to simplify the mapping rules of the topology structure. The topology structure includes two parts: directed edges between entity segments and directed acyclic graphs of nodes within entity segments. Based on the logical relationships between scenario elements in the reservoir dam emergency scenario ontology model, the interaction relationships between entity segments in the multi-entity Bayesian network are determined and represented by directed edges. The relationships between the state parameters of scenario elements are mapped to the relationships between nodes within entity segments, thus constructing the topology structure of nodes within entity segments.

5. The event-driven emergency scenario simulation method for reservoir dam emergencies according to claim 1, characterized in that, The specific implementation steps for step four are as follows: Step 41: Dynamically change the model structure and scale according to the actual emergency situation, and delete the boundary events, conditionally independent events and irrelevant branches of irrelevant events of the multi-entity Bayesian network model to construct the Bayesian network topology for the specific situation. Step 42: Select a probability assignment method based on the logical relationship and mechanism between nodes; the assignment method includes expert experience assignment method based on DS evidence theory, parameter learning method based on sample data, and structural reliability analysis method; (1) Expert experience assignment method based on DS evidence theory Experts from different fields evaluate the probability of scenario state nodes based on probability estimation tables; (2) Parameter learning method based on sample data Choose an Expectation-Maximization (EM) algorithm suitable for the missing data samples. The Expectation-Maximization (EM) algorithm is an iterative optimization algorithm that estimates the missing data variables by alternately performing the expectation step and the maximization step to construct a complete training sample set. Let the data sample set of reservoir dam emergencies be . The missing dataset is The parameter to be estimated is , for After confirming The joint probability distribution of ; (3) Structural reliability analysis method For scenario nodes in emergencies with clear logical relationships and well-defined mechanisms of action, structural reliability analysis is used to determine node probability parameters. Based on the logical relationships and mechanisms between events, uncertainty influencing factors are identified, and the function functions of the scenario events are established using finite element analysis and response surface equations. In the formula, Z This is a function for a specific scenario event; Let $\mathbf$ be a random variable that influences the state of a situational event; structural reliability refers to the probability of achieving a predetermined function under specified conditions, while the function describes the functional relationship between the structural state and the random variable. Z A value less than 0 indicates that the structure is in a failure state, that is... The failure probability of the structure is given; the function is solved by the JC method, Monte Carlo method or first second moment method to determine the scenario node probability parameters; Step 43: Based on the assignment results, initialize the probability parameters and complete the construction of the emergency scenario simulation model.

6. The event-driven emergency scenario simulation method for reservoir dam emergencies according to claim 1, characterized in that, In step five, current scenario information is collected as evidence information, and the probability of the root node is assigned. The driving event is used as an input variable, and the probability values ​​of the state variables and output variables of subsequent scenario nodes are determined according to the inference rules, thereby realizing the probabilistic deduction of the emergency situation of a reservoir dam emergency. The specific deduction steps are as follows: Step 51: After the initial scenario of the emergency occurs, collect the current scenario information as evidence information and complete the probability assignment of the root node. Then, predict and reason about the emergency situation according to the reasoning rules, determine the probability values ​​of the state variables and output variables of the subsequent scenario nodes, and calculate the reasoning layer by layer until the final scenario node, thereby determining the state probability of each event on the evolution path. Among them, the driving event nodes in the emergency scenario simulation model are the input variables, where the disaster event nodes are in an uncertain state, while the emergency response event nodes with the variable type of Boolean are in a non-responding state. Step 52: As the real scenario evolves, new disaster events or emergency decision responses occur, which in turn affect the state probability of scenario events. When the driving event occurs, the corresponding event node becomes an evidence node, that is, it becomes a deterministic state. Then, the scenario node probability in the model is updated according to the inference rules, thereby driving the dynamic probability deduction of the emergency scenario.

7. The event-driven emergency scenario simulation method for reservoir dam emergencies according to claim 6, characterized in that, Emergency scenario simulation models use Bayesian networks to express the relationships between sub-events; let the sub-events in the scenario be... EN i and subevents EN j There is a relationship between them, let's assume I ( i j1 , i j2 ,…, i jn )for EN j Input variables, S ( s j1 , s j2 ,…, s jm )for EN j State variables, O ( o j1 , o j2 ,…, o jt )for EN j The output variables and the inference rules of the scenario inference model are as follows: (1) Input variables and state variables （5-1） （5-2） Formula (5-1) indicates that the state variable is only related to... EN j The input variables are related, and formula (5-2) indicates that the state variables are related to the input variables. EN j input variables and EN i The state variables are related; among them, State variable S je The probability of occurrence; n , m for I j and S i The number of variables; (2) State variables and state variables （5-3） Formula (5-3) indicates that the state variable is only related to... EN j Other state variables are relevant; where, State variables The probability of occurrence; m for S j The number of variables; (3) Output variables and state variables （5-4） Formula (5-4) represents the output variable and EN j State variables are related; where, For output variables The probability of occurrence; m for S j The number of variables; (4) Relationship between variables between events （5-5） Formula (5-5) means EN i Output variable I j and EN j input variables The intersection is not an empty set; in the formula, t , n for and The number of variables.

8. The event-driven emergency scenario simulation method for reservoir dam emergencies according to claim 1, characterized in that, In step six, A scenario-response emergency decision-making model was constructed using the expected loss of life and economic damage to downstream populations as the objective function: （6-1） Constraints: （6-3） （6-4） In the formula, Z 1 represents the expected loss of life and economic damage to downstream populations; P i For the first i The probability of this scenario occurring; L i For the first i The number of lives lost due to this scenario; F 1 represents the function for calculating the probability of a scenario; D 1 represents the function for calculating life loss; x i As a constraint, it represents the first... i Random variables of situational factors that affect the effectiveness of treatment in various scenarios; c i Let be the boundary condition, representing the th i Flood risk indicators that could affect people's lives in various scenarios.

9. A computer device, characterized in that: The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the event-driven emergency scenario simulation method for reservoir dam emergencies as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that executes the event-driven emergency scenario simulation method for reservoir dam emergencies as described in any one of claims 1-8.

Images