A Verification Method for Dam Failure Path Analysis of Hydropower Projects Driven by Multi-Source Data

Abstract

This invention discloses a multi-source data-driven method for verifying the path analysis of hydropower dam failures, belonging to the field of hydropower dam failure path analysis. The method includes entity extraction and relation extraction based on geological, meteorological, biological, and engineering datasets to construct a hydropower dam failure knowledge graph; supplementing and correcting erroneous and missing relations in the knowledge graph using path ranking and tensor decomposition algorithms; verifying the entity relations in the corrected knowledge graph to obtain a verified hydropower dam failure knowledge graph; and determining all paths from each hazard-causing factor to the dam failure node using a breadth-first search algorithm based on the verified knowledge graph, and visualizing each path to complete the path analysis of the hydropower dam failure event node. This invention solves the problem of low reliability of assessment results due to excessive reliance on human experience in existing methods.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of dam breach path analysis of hydropower hubs, and particularly relates to a multi-source data driven dam breach path analysis and verification method for hydropower hubs. BACKGROUND

[0002] Accurate identification of disaster-causing factors and impact paths of dam breach of hydropower hubs is the key to ensuring the safe operation of hydropower hubs. The existing analysis methods have many shortcomings: traditional methods are mostly based on single-discipline knowledge, and it is difficult to comprehensively consider the comprehensive influence of geological, meteorological, engineering and human factors; the integration ability of multi-source heterogeneous data is weak, and the potential relationship behind the data cannot be fully tapped; in terms of threshold selection and impact path analysis, it often relies on artificial experience, lacks objectivity and accuracy, and the reliability of the risk assessment results is low. Therefore, a more scientific and comprehensive method is needed to analyze the disaster-causing factors and impact paths of dam breach of hydropower hubs. SUMMARY

[0003] In view of the above shortcomings in the prior art, the multi-source data driven dam breach path analysis and verification method for hydropower hubs provided by the application solves the problem of low reliability of evaluation results caused by excessive reliance on artificial experience in the prior art.

[0004] In order to achieve the above-mentioned application purposes, the technical scheme adopted by the application is as follows: a multi-source data driven dam breach path analysis and verification method for hydropower hubs, comprising:

[0005] obtaining multi-source data related to dam breach of hydropower hubs;

[0006] dividing the multi-source data into a geological data set, a meteorological data set, a biological data set and an engineering data set;

[0007] performing entity extraction and relationship extraction based on the geological data set, the meteorological data set, the biological data set and the engineering data set to obtain a final entity set and relationship information between entities, and converting the final entity set and the relationship information between entities into graph data to construct a dam breach knowledge graph of hydropower hubs;

[0008] respectively using a path ranking algorithm and a tensor decomposition algorithm to supplement and correct the incorrect relationships and missing relationships in the dam breach knowledge graph of hydropower hubs, to obtain a corrected dam breach knowledge graph of hydropower hubs;

[0009] verifying the entity relationships in the corrected dam breach knowledge graph of hydropower hubs to obtain a verified dam breach knowledge graph of hydropower hubs;

[0010] According to the verified dam breach knowledge graph of hydropower hubs, all paths from each disaster-causing factor to the dam breach node are determined by using a breadth-first search algorithm, and each path is visualized to complete the node path analysis of the dam breach event of hydropower hubs.

[0011] The present application has the advantages that: the present application breaks through the limitations of traditional water and electricity hub breach analysis methods, evaluates the water and electricity hub breach risk from the perspective of multi-source data integration and comprehensive analysis, greatly expands the breadth and depth of analysis. Abandoning the traditional mode of relying only on single-discipline knowledge, comprehensively covering geological, meteorological, biological and human (engineering) and other factors, it has a more comprehensive analysis perspective. By constructing a knowledge graph to mine the potential relationship of data, it is no longer limited by the problem of multi-source heterogeneous data integration, can more systematically grasp the breach risk, and is suitable for various complex water and electricity hub scenes; the dam breach path quantum topology verification model DPRQTVM can deeply understand the entity-relation interaction with the help of quantum entanglement coding of quantum encoder, explicitly model the interaction between entities and relations in the path, more accurately capture semantic relationships, and deepen the understanding of the path. The space-time resolver can effectively process space-time information, decompose and process the path in time and space dimensions, fully mine the value of space-time information, better adapt to the characteristics of different types of paths, especially suitable for dam breach paths involving space-time changes. The topology verification network combines graph neural networks and attention mechanisms to verify the topology structure of the path, which can effectively identify unreasonable path structures, improve the robustness of the verification results, and the hybrid decision module makes decisions based on credibility, further enhancing the reliability of the verification results. In addition, the system architecture design has good adaptability and expandability, which can not only handle paths of different lengths and types, but also due to the independence and modular design of each module, it is easy to apply to knowledge graph verification work in other fields.

[0012] Further, the entity extraction and relation extraction are performed, specifically:

[0013] The entity types related to dam breach are determined, including: disaster-causing factor entity, dam breach event entity, space-time entity, attribute entity, and engineering part entity;

[0014] Based on each entity type, a plurality of regular expressions of different complexity levels are designed, and based on each regular expression, the geological data set, the meteorological data set, the biological data set, and the engineering data set are subjected to entity extraction and relation extraction, to obtain the entity set extracted from different data and the relation information between entities in each entity set;

[0015] Based on each entity set and the relation information between entities in each entity set, the information of the same entity is integrated to obtain the final entity set and the relation information between entities.

[0016] The beneficial effects of the above further scheme are: by determining the disaster-causing factors related to dam collapse, dam collapse events and other entity types, the information extraction is more targeted; different complexity regular expressions are designed to flexibly process data from different sources and improve the comprehensiveness of information extraction; the same entity information is integrated to form a complete final entity set and relationship information, providing a comprehensive and accurate data basis for subsequent in-depth analysis and decision support of dam collapse information.

[0017] Further, the verified dam breach knowledge graph of the hydropower junction is specifically:

[0018] A dam breach path quantum topology verification model for verifying entity relationships is constructed.

[0019] The relationship of the corrected dam breach knowledge graph of the hydropower junction is verified by using the dam breach path quantum topology verification model, and the verified dam breach knowledge graph of the hydropower junction is obtained.

[0020] The beneficial effects of the above further scheme are: by constructing the dam breach path quantum topology verification model, the entity relationships in the corrected dam breach knowledge graph of the hydropower junction can be verified from the perspective of quantum topology, and the characteristics of quantum mechanics and topology structure analysis are used to provide a novel and accurate method for relationship verification of the knowledge graph. This verification method can more deeply and comprehensively check the accuracy and completeness of the relationships in the knowledge graph, which helps to find potential errors or inconsistencies, thereby improving the quality and reliability of the dam breach knowledge graph of the hydropower junction, and providing more accurate and effective knowledge support for safety evaluation, risk prediction and other aspects of the hydropower junction.

[0021] Further, the dam breach path quantum topology verification model includes a path extraction module, a quantum encoder, a space-time resolver, a dynamic gate fusion module, a topology verification network, a credibility evaluation module and a hybrid decision module.

[0022] The path extraction module is configured to extract a dam breach path from the corrected dam breach knowledge graph of the hydropower junction.

[0023] The quantum encoder is configured to convert the dam breach path into a quantum state and encode the path through quantum entanglement, and output the quantum encoded path.

[0024] The space-time resolver is configured to decompose the quantum encoded path into a time convolution branch and a space relationship branch, and obtain time features and space features, respectively.

[0025] The dynamic gate fusion module is configured to fuse the time features and the space features, and output the space-time fusion features.

[0026] The topology verification network is used to construct a topology graph based on spatiotemporal fusion features, calculate the topology vulnerability index of the topology graph, and process and verify the topology graph through graph neural networks and attention mechanisms to output the credibility assessment results of the dam failure path.

[0027] The credibility assessment module is used to quantify the credibility assessment results of the topological vulnerability index and the dam failure path, and obtain the quantification results.

[0028] The hybrid decision-making module is used to make decisions based on the quantification results and output verification results.

[0029] The beneficial effects of the above-mentioned further scheme are as follows: The quantum topology verification model of dam failure path consists of multiple modules such as path extraction and quantum encoding. Each module works in concert. First, the dam failure path is extracted and the information processing efficiency is improved by quantum encoder with quantum state and entanglement encoding. Then, the spatiotemporal features are accurately extracted and fused by spatiotemporal decomposer and dynamic gating fusion module. Next, the topology verification network combines graph neural network and attention mechanism to deeply analyze the topology structure. Finally, the credibility assessment module and hybrid decision module perform quantitative assessment and decision-making, thereby comprehensively and accurately verifying the dam failure path in the hydropower hub failure knowledge graph and ensuring the quality and reliability of the knowledge graph.

[0030] Furthermore, the expression for the quantum-encoded path is:

[0031]

[0032]

[0033]

[0034]

[0035] in, The path after quantum encoding; for The path states of the layered quantum circuit are connected to each layer through tensor products; It is an exponential function with the natural constant as its base; The imaginary unit; For the first Hermitian matrix dynamically generated by layered quantum circuits; To correct the linear unit; For the first Trainable qubit parameters of layered quantum circuits; For the first Normalized dam-break path of layered quantum circuits; It is the tensor product; It is the hyperbolic tangent function; trainable qubit parameters for the 1st normalized breach path for the 1st trainable qubit parameters for the 1st normalized breach path for the 1st layer index diagonal phase matrix diagonal element transpose trainable qubit parameters for the 1st entangling gate parameters learnable by the 1st classical feature projection by ReLU activation for the 1st classical feature projection by tanh activation for the 1st classical feature projection by tanh activation for the 1st path position encoding unitary matrix for the 1st breach path for the 1st path position encoding unitary matrix for the 1st breach path for the 1st breach path for the 1st L2 norm multilayer perceptron diagonal matrix constructor

[0036] The above further scheme has the beneficial effect that the quantum encoding path expression realizes efficient information coding by means of quantum states, tensor products, etc., flexibly adapts to data by means of numerous trainable parameters, extracts nonlinear features by means of activation functions and multilayer perceptrons, provides support for topology construction in combination with special matrix operations, can effectively improve the breach path information processing capability, optimize the encoding effect, capture complex relationships, and help topology analysis.

[0037] Further, the expression for decomposing the quantum encoded path into a time convolution branch and a spatial relationship branch is:

[0038]

[0039]

[0040] wherein, spatiotemporal feature long short-term memory network, processing time sequence dependency relationship real part of quantum path encoding, reflecting causal strength quantum encoded path outer product operation graph attention network, processing spatial correlation between nodes imaginary part of quantum path encoding, reflecting phase relationship Sigmoid function, compressing weights to the interval (0, 1); trainable vector; taking the mean of temporal features, capturing global temporal patterns; temporal features, inputting the real part of the quantum state; taking the max of spatial features, highlighting key regions; spatial features, inputting the imaginary part of the quantum state; dynamic coupling weight; normalized cross-correlation function; transpose; L2 norm.

[0041] The above further scheme has the beneficial effects that: the expression effectively decomposes and extracts the spatiotemporal features of the quantum-encoded path by fusing long short-term memory networks, graph attention networks, and the like, and adaptively adjusts the weights and measures the correlation using special functions and operations, can capture the temporal sequence dependence and spatial position correlation in the dam breach process, and comprehensively and deeply presents the spatiotemporal dynamic characteristics of the dam breach path, providing key information support for the hydropower hub dam breach knowledge graph verification and dam breach risk analysis.

[0042] Further, the expression of the spatiotemporal fusion feature is:

[0043]

[0044] wherein, spatiotemporal fusion feature; Sigmoid function, compressing weights to the interval (0, 1); temporal feature; spatial feature; trainable weight; and the splicing result of and; weight calculated by ; smooth ReLU function, ensuring the output to be positive; trainable weight; outer product operation; Hadamard product; nonlinear coupling coefficient calculated by .

[0045] ​The beneficial effects of the further scheme are that the comprehensive feature expression can adaptively fuse time and space features through trainable weights, special activation functions and operations, the Sigmoid function and the softplus function ensure reasonable weight value range and positive output, and operations such as outer product and Hadamard product deeply mine the correlation between features, comprehensively integrate the space-time information of the dam-break path, provide more accurate and rich feature expression for subsequent topological verification, and improve the accuracy of understanding and analysis of the complex dam-break process.

[0046] Further, the expression of the topological vulnerability index and the reliability evaluation result of the dam-break path is:

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055] wherein, is a topological vulnerability index; is a topological graph; is a diameter of a connected branch ; is a branch; is a total number of vertices; is a first-order Betti number; is a vertex set; is an edge set; is a feature of an th topological graph vertex; is a space-time fusion feature corresponding to an th topological graph vertex; is a topological graph edge between an th topological graph vertex and an th topological graph vertex; is an indicator function, and is 1 when the condition is met, otherwise is 0; is a space-time fusion feature corresponding to an th topological graph vertex; is an L2 norm; is a distance threshold value; is the median of the upper triangle of the distance matrix, excluding the diagonal part; is the credibility evaluation result of the dam-break path; is the normalization factor; is the total number of local subgraphs; is the Betti number corresponding to the m-th local subgraph; is the Betti number corresponding to the m-th local subgraph; is the multi-head attention mechanism; is the feature representation of the m-th local subgraph; is the feature representation of the m-th local subgraph; is the feature representation of the global graph; is the natural constant; is the Betti number calculation function of the homology group matrix; is the homology group matrix corresponding to the m-th local subgraph. is the homology group matrix corresponding to the m-th local subgraph.

[0056] The beneficial effects of the above further scheme are: the above expression generates topological edges accurately by constructing a topological graph, combining connectivity, ring structure and other characteristic indicators, and using distance thresholds, and comprehensively integrates local and global features by means of a multi-head attention mechanism to obtain a dam-break path credibility evaluation result. It can systematically quantify and evaluate the stability of the topological structure and the reliability of the dam-break path from the perspective of graph theory, providing a scientific basis for the risk analysis of the dam-break of the hydroelectric hub, and improving the understanding and control ability of the complex situation of dam-break.

[0057] Further, the expression of the quantification result is:

[0058]

[0059] wherein, is the quantification result; is the function type; is the topological vulnerability index; is the topological graph; is the hyperbolic tangent function; is the credibility evaluation result of the dam-break path.

[0060] The beneficial effects of the above further scheme are: the quantification result expression fuses the topological vulnerability index and the dam-break path credibility evaluation result, processes and normalizes them using the S-shaped function and the hyperbolic tangent function, and comprehensively quantifies the topological structure stability and the dam-break path reliability information, providing a simple and unified numerical result for dam-break risk assessment, which is convenient for intuitive comparison and analysis of the risk degree of different dam-break scenarios, and provides more explicit data support for safety decision-making of the hydroelectric hub.

[0061] Further, the expression of the verification result is:

[0062]

[0063] wherein, is a verification result; is a path valid; is a lower limit value of quantization result for judging path valid; is a path partially valid; is an upper limit value of quantization result for judging path invalid; is a path invalid; is a quantization result.

[0064] The above further scheme has the beneficial effects that: the verification result expression presents the complex dam-break path verification situation in three clear classifications of “valid”, “partially valid” and “invalid” through setting different threshold intervals of the quantization result, so that the verification result is intuitive and easy to understand, and it is convenient for relevant personnel to quickly judge the reliability of the dam-break path of the hydropower hub, to provide clear and direct basis for subsequent decision-making (such as engineering reinforcement, risk warning, etc.), and to improve the decision-making efficiency and accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 is a flow chart of the method of the present application.

[0066] Figure 2 is a system architecture diagram of the dam-break path quantum topology verification model DPRQTVM in the embodiment of the present application.

[0067] Figure 3 is a knowledge graph path diagram of dam-break caused by meteorological factors in the embodiment of the present application.

[0068] Figure 4 is a knowledge graph path diagram of dam-break caused by geological factors in the embodiment of the present application.

[0069] Figure 5 is a knowledge graph path diagram of dam-break caused by biological factors in the embodiment of the present application.

[0070] Figure 6 is a knowledge graph path diagram of dam-break caused by human (engineering) factors in the embodiment of the present application. DETAILED DESCRIPTION

[0071] The specific embodiments of the present application are described below to facilitate those skilled in the art to understand the present application, but it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, it is obvious that various changes are within the spirit and scope of the present application defined and determined by the appended claims, and all the inventions utilizing the concept of the present application are within the scope of protection.

[0072] As Figure 1As shown, in one embodiment of the present application, a multi-source data-driven hydropower hub dam-break path analysis verification method comprises:

[0073] Obtaining multi-source data related to hydropower hub dam-break;

[0074] Dividing the multi-source data into a geological data set, a meteorological data set, a biological data set, and an engineering data set;

[0075] Performing entity extraction and relationship extraction based on the geological data set, the meteorological data set, the biological data set, and the engineering data set to obtain a final entity set and relationship information between entities, and converting the final entity set and the relationship information between entities into graph data to construct a hydropower hub dam-break knowledge graph;

[0076] Respectively using a path ranking algorithm and a tensor decomposition algorithm to supplement and correct the erroneous relationships and missing relationships in the hydropower hub dam-break knowledge graph to obtain a corrected hydropower hub dam-break knowledge graph;

[0077] Verifying the entity relationships in the corrected hydropower hub dam-break knowledge graph to obtain a verified hydropower hub dam-break knowledge graph;

[0078] According to the verified hydropower hub dam-break knowledge graph, determining all paths from each disaster-causing factor to the dam-break node through a breadth-first search algorithm, and visualizing each path to complete the hydropower hub dam-break event node path analysis.

[0079] In this embodiment, the network crawler technology is used to obtain multi-source data related to hydropower hub dam-break from various professional websites, academic databases, government report websites, etc. At the same time, literature management software is used to collect various academic literature, engineering books, etc. After that, the collected data is classified into four categories: geological data, meteorological data, biological data, and human (engineering) data through artificial classification.

[0080] Data acquisition: Scrapy is used to write a simple crawler program to analyze the webpage structure of professional hydropower engineering websites and locate HTML elements containing hydropower hub related data. Historical meteorological data of the area where the hydropower hub is located, including rainfall, temperature, etc., is obtained from the water conservancy department website. Geological structure, earthquake activity, etc. data is obtained from the geological survey agency website. At the same time, using literature management software, connecting academic databases (CNKI, Web of Science, Google Scholar), searching related academic literature through keywords such as "hydropower hub dam-break", "disaster-causing factor", "hydropower hub engineering safety", "dam-break mechanism", "geological disaster and hydropower hub", "meteorological factors affecting hydropower hub", "human factors leading to hydropower hub risk", "hydropower hub operation and management failure", "dam structure damage reason", "flood impact on hydropower hub", "hydropower hub response under earthquake action", etc. Import the software for management.

[0081] Data Classification: The collected data is classified. Volcanic activity, soil and rock properties, geological structure, and dam foundation stability are classified as geological data; rainfall, temperature changes, extreme weather events, floods, and other meteorological variables are classified as meteorological data; data on the impact of plant activity, animal activity, microbial activity, algal influence, and insect erosion on the project are classified as biological data; and data on design defects, construction problems, management oversights, insufficient maintenance, operational errors, third-party sabotage, lack of monitoring, inadequate emergency response, economic factors, and policy factors are classified as human (engineering) data.

[0082] The entity extraction and relation extraction are performed as follows:

[0083] Identify the entity types related to dam failure, including: hazard-causing entity, dam failure event entity, spatiotemporal entity, attribute entity, and engineering component entity;

[0084] Based on various entity types, several regular expressions of different complexities were designed, and entity extraction and relation extraction were performed on geological datasets, meteorological datasets, biological datasets and engineering datasets based on each regular expression to obtain entity sets extracted from different data and relation information between entities in each entity set.

[0085] Based on the entity sets and the relationship information between entities within each entity set, the information of the same entity is integrated to obtain the final entity set and the relationship information between entities.

[0086] In this embodiment, based on the data collected and classified in the previous step, the entity types related to hydropower dam failure are first determined, including disaster-causing entity, dam failure event entity, spatiotemporal entity, attribute entity, and engineering component entity. Then, a set of regular expressions is used to extract various entities and relationships from the preprocessed data. The extracted entities and relationships are then stored in the form of a graph to construct a hydropower dam failure knowledge graph. Finally, a knowledge graph completion algorithm is used to modify, predict, and supplement erroneous and missing relationships in the graph.

[0087] Construct a set of regular expressions for multi-source data:

[0088] By designing regular expression patterns of varying complexity, we can handle large amounts of complex text structures and semantic relationships to achieve entity name extraction. Table 1 shows examples of regular expressions.

[0089] Table 1

[0090]

[0091] ([Disaster-causing factor].+?): Matches disaster-causing factors such as geology, meteorology, biology, and man-made engineering (e.g., "rainstorm" or "abnormal permeability coefficient").

[0092] ([Dam breach event].+?): Matches dam breach events (e.g., "overtopping" "piping failure").

[0093] ([Time point / period].+?): Matches time descriptions (e.g., "July 2023" "lasts 72 hours").

[0094] ([Facility].+?): Matches attribute components (e.g., "dam foundation" "spillway").

[0095] Entity extraction and relationship extraction:

[0096] According to the regular expressions, extract various entities (disaster-causing factors, dam breach events, time and space, attributes) related to dam breaches and the influence relationships between entities from geological, meteorological, biological, and human (engineering) data texts. After extracting the entities, determine whether the entities from different sources are the same entity based on entity names, attributes, etc. If the names of two entities are exactly the same, they are determined to be the same entity. For the same entity, integrate the related information in different texts. For entities with similar names but differences, or using the TF-IDF (Term Frequency-Inverse Document Frequency) combined with cosine similarity method to calculate the similarity between entities. First, convert the text containing the entity into a vector form, calculate the weight of each word through TF-IDF, and reflect its importance in the text. Then, use the cosine similarity to calculate the cosine value of the angle between two vectors. The closer the value is to 1, the higher the similarity between the two entities. Set a threshold (0.8), when the similarity is higher than the threshold, further manually review and determine whether it is the same entity; if the similarity is lower than the threshold, it is determined as different entities.

[0097] Knowledge graph fusion and optimization:

[0098] Identify errors and omissions in the relationships between entities (disaster-causing factors, dam breach events, time and space, attributes).

[0099] Prioritize paths with fewer than or equal to three entities and relationships, as these simple paths are more likely to have missing or incorrect information. Additionally, manually review each relationship path in combination with professional knowledge, actual data, and logical rules to identify all problematic paths. For incorrect relationship paths, use the Path Ranking Algorithm (PRA), and for missing relationship paths, use the RESCAL model for tensor decomposition. Combine both algorithms to modify and supplement the paths.

[0100] Determine the optimal weight between the PRA and RESCAL algorithms.

[0101] There are two kinds of weight distribution criteria for path ranking algorithm (PRA) and tensor decomposition algorithm (RESCAL). One is based on the characteristics of the dataset, and the other is based on experimental verification. For the first one, the dataset with rich path features and the dataset with complex relationships and implicit semantics are considered. For the second one, the FB15k-237 dataset is taken as an example, and the training set, validation set and test set are divided. Nine kinds of weight combinations are set, and the mean reciprocal rank (MRR) and Hits@K are used as evaluation indicators. The PRA and RESCAL models are trained respectively, and the performance of each combination is evaluated on the validation set. The optimal combination is selected to determine the weight on the test set. After the weight is determined, the PRA and RESCAL models are used to process the remaining error relationships and missing relationships in the knowledge graph. First, the PRA is used to correct the marked error relationships one by one in the knowledge graph. According to the calculated path feature similarity, the alternative relationship is found and the knowledge graph is updated. Then, the RESCAL model is used to scan the knowledge graph, predict and supplement the potential missing relationships, and add the predicted relationships to the knowledge graph to perfect the knowledge graph.

[0102] The weight selection method is:

[0103] ① Weight distribution based on dataset characteristics:

[0104] Dataset with rich path features: If there are many obvious and traceable paths between entities in the knowledge graph, PRA can play a more important role. Because PRA can use path search to mine the potential relationships between entities, it can be given a higher weight of 0.7, and RESCAL is given a weight of 0.3.

[0105] Dataset with complex relationships and implicit semantics: When the relationships in the knowledge graph are complex and have many implicit semantics, it is difficult to describe them through simple paths. At this time, the advantage of RESCAL, which can learn the potential representation of entities and relationships, is obvious. At this time, RESCAL is given a weight of 0.7, and PRA is given a weight of 0.3.

[0106] ② Weight distribution based on experimental verification:

[0107] Prepare the knowledge graph dataset and divide it into training set, validation set and test set.

[0108] Set different weight combinations, (0.1, 0.9), (0.2, 0.8), (0.3, 0.7), (0.4, 0.6), (0.5, 0.5), (0.6, 0.4), (0.7, 0.3), (0.8, 0.2), (0.9, 0.1).

[0109] ③ Determine the weight:

[0110] The dataset selected is the common knowledge graph dataset FB15k-237 containing 14,541 entities, 237 relationships, and more than 175,000 triples. The mean reciprocal rank (MRR) and Hits@K are used as evaluation indicators, where MRR is the average of the reciprocal ranks of all query triples, and Hits@K represents the proportion of query triples whose correct answers are ranked in the top K positions in the prediction list. In terms of experimental procedures, the dataset is first divided into training set, validation set and test set according to the ratio of 8:1:1; then the PRA and RESCAL models are trained respectively; then different weight combinations are tried, and the performance of each combination is evaluated on the validation set; finally, the weight combination with the best performance on the validation set is selected, and the final evaluation is carried out on the test set to determine the weight, and the weight example is shown in Table 2.

[0111] Table 2

[0112]

[0113] From the experimental results, it can be seen that when the PRA weight is 0.4 and the RESCAL weight is 0.6, the evaluation indicators of the model on the validation set and the test set all reach the optimal. This indicates that in this dataset and task scenario, the weight combination can make the knowledge graph completion performance of the fusion model best.

[0114] The verified water and electricity hub breach knowledge graph is specifically:

[0115] A dam breach path quantum topology verification model for verifying entity relationships is constructed.

[0116] The relationship verification of the modified water and electricity hub breach knowledge graph is carried out by using the dam breach path quantum topology verification model, and the verified water and electricity hub breach knowledge graph is obtained.

[0117] As shown in Figure 2 The dam breach path quantum topology verification model includes a path extraction module, a quantum encoder, a space-time decomposer, a dynamic gate fusion module, a topology verification network, a credibility evaluation module and a hybrid decision module.

[0118] The path extraction module is used to extract the dam breach path from the modified water and electricity hub breach knowledge graph.

[0119] The quantum encoder is used to convert the dam breach path into a quantum state, and encode the path through quantum entanglement, and output the quantum encoded path.

[0120] The space-time decomposer is used to decompose the quantum encoded path into a time convolution branch and a space relationship branch, and obtain time features and space features respectively.

[0121] The dynamic gating fusion module is configured to fuse the time feature and the space feature, and output a spatio-temporal fusion feature;

[0122] The topology verification network is configured to construct a topology graph based on the spatio-temporal fusion feature, calculate a topology vulnerability index of the topology graph, and process and verify the topology graph through a graph neural network and an attention mechanism, and output a credibility evaluation result of the dam-break path;

[0123] The credibility evaluation module is configured to quantitatively evaluate the topology vulnerability index and the credibility evaluation result of the dam-break path, and obtain a quantitative result.

[0124] The mixed decision module is configured to make a decision according to the quantitative result, and output a verification result.

[0125] In this embodiment, according to the constructed knowledge graph, a dam-break path quantum topology verification model DPRQTVM is introduced to carry out deep analysis and verify the entity and the relationship information between entities in the text.

[0126] The dam-break path quantum topology verification model DPRQTVM is used to verify the above knowledge graph. By introducing quantum coding, space-time decomposition and topology verification technologies, the accuracy, robustness and adaptability of the knowledge graph verification are improved, and different types of paths can be processed more effectively, especially in the verification of key fields such as dam-break paths.

[0127] The expression of the quantum coded path is:

[0128]

[0129]

[0130]

[0131]

[0132] wherein, is the quantum coded path; is the path state of the quantum circuit of the i-th layer, which is connected through a tensor product; is an exponential function with a natural constant as the base; is an imaginary unit; is the i-th layer quantum circuit dynamically generated Hermitian Hermite matrix; is a rectified linear unit; is the trainable quantum bit parameter of the i-th layer quantum circuit; is the trainable quantum bit parameter of the i-th layer quantum circuit; is the trainable quantum bit parameter of the i-th layer quantum circuit; is the trainable quantum bit parameter of the i-th layer quantum circuit; normalized collapsed path of the layer quantum circuit is a tensor product is a hyperbolic tangent function is the i-th trainable qubit parameters of the i-th layer quantum circuit is the i-th normalized collapsed path of the i-th layer quantum circuit is the layer index is a diagonal phase matrix is a diagonal element is a transpose is the i-th entangling gate parameters of the i-th layer quantum circuit is the i-th classical feature projection of the i-th layer quantum circuit through ReLU activation is the i-th classical feature projection of the i-th layer quantum circuit through tanh activation is the i-th path position encoding unitary matrix of the i-th layer quantum circuit is the i-th collapsed path of the i-th layer quantum circuit is an L2 norm is a multi-layer perceptron is a diagonal matrix constructor

[0133] The expression of decomposing the quantum encoded path into a temporal convolution branch and a spatial relation branch is:

[0134]

[0135]

[0136] wherein, is a spatio-temporal feature is a long short-term memory network, processing temporal dependency is a real part of quantum path encoding, reflecting causal strength is a quantum encoded path is an outer product operation is a graph attention network, processing spatial correlation between nodes is an imaginary part of quantum path encoding, reflecting phase relation is a Sigmoid function, compressing weights into (0, 1) interval is a trainable vector is a mean over time features, capturing global temporal pattern is a time feature, inputting quantum state real part To take the maximum value of spatial features, highlight the key area; For spatial features, input the imaginary part of the quantum state; For dynamic coupling weight; For normalized cross-correlation function; For transpose; For L2 norm.

[0137] The expression of the spatio-temporal fusion feature is:

[0138]

[0139] Wherein, The spatio-temporal fusion feature; Sigmoid function, compress the weight to the interval (0, 1); Temporal feature; Spatial feature; Trainable weight; For And The splicing result; The weight calculated by ; Smooth ReLU function, ensure the output is positive; Trainable weight; Outer product operation; Hadama product; Nonlinear coupling coefficient calculated by .

[0140] The expression of the topological vulnerability index and the credibility evaluation result of the dam collapse path is:

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149] Wherein, Topological vulnerability index; Topological graph; For connected components The diameter; For branches; The total number of vertices; It is a first-order Betti number; For vertex set; It is an edge set; For the first Features of the vertices of a topological graph; For the first Spatiotemporal fusion features corresponding to vertices of a topological graph; For the first The vertex of the topological graph and the first A topological graph with vertices and edges; This is an indicator function; its value is 1 if the condition is met, and 0 otherwise. For the first Spatiotemporal fusion features corresponding to vertices of a topological graph; It is an L2 norm; Distance threshold; To take the median of the upper triangular portion of the distance matrix, excluding the diagonal; The results of the credibility assessment of the dam failure path; Normalization factor; This represents the total number of local subgraphs. For the first The number of Betti numbers corresponding to each local subgraph; This is a multi-head attention mechanism; For the first Feature representation of a local subgraph; A feature representation of the global graph; It is a natural constant; The Betti number calculation function is used for the homology group matrix; For the first The homology group matrix corresponding to each local subgraph.

[0150] The expression for the quantization result is:

[0151]

[0152] in, To quantify the results; It is a function type; It serves as a topological vulnerability indicator. It is a topological graph; It is the hyperbolic tangent function; The results represent the credibility assessment of the dam failure path.

[0153] The expression for the verification result is:

[0154]

[0155] wherein, is a verification result; is a path valid; is a lower limit value of a quantized result for judging a path valid; is a path partially valid; is an upper limit value of a quantized result for judging a path invalid; is a path invalid; is a quantized result.

[0156] In this embodiment, the verification model operation steps are as follows:

[0157] The input path is input into the quantum encoder to perform quantum encoding to obtain a quantum-encoded path; the quantum-encoded path is input into the space-time decomposer to obtain time features and space features through the time convolution branch and the space relationship branch, respectively; the time features and the space features are input into the dynamic gate fusion module to perform fusion to obtain comprehensive features; the comprehensive features are input into the topology verification network to construct a topology graph and perform processing and verification to obtain a credibility evaluation result of the path; the credibility evaluation result is quantitatively evaluated by the credibility evaluation module; according to the evaluated credibility, a decision is made by the hybrid decision module, and a verification result is output.

[0158] Specific examples are as follows:

[0159] Case 1. Verification of a rainfall-induced dam breach path

[0160] Input path: extreme rainfall → water level exceeding limit → overtopping → dam weakening → dam breach;

[0161] Quantum encoding process:

[0162]

[0163] Running output:

[0164] Quantum encoding time consumption: 12.3 ms;

[0165] Space-time feature decomposition:

[0166] Time component: [0.82, 0.76, 0.91] (L2-norm = 1.32);

[0167] Space component: [0.68, 0.73, 0.85] (L2-norm = 1.24);

[0168] Dynamic coupling weight: = 0.79;

[0169] Topology verification:

[0170] Betti number: = 3, = 1;

[0171] Persistent homology score: 0.87;

[0172] Final confidence: 0.83 → quantum decision VALID.

[0173] Case 2. Earthquake-induced dam breach path validation

[0174] Input path: high earthquake intensity → liquefaction → landslide → impact → dam breach;

[0175] Temporal-spatial decomposition process:

[0176] ;

[0177] Run output:

[0178] Quantum encoding time consumption: 9.7 ms;

[0179] Temporal-spatial feature decomposition:

[0180] Temporal component: [0.91, 0.88, 0.95] (L2-norm = 1.58);

[0181] Spatial component: [0.72, 0.81, 0.89] (L2-norm = 1.42);

[0182] Dynamic coupling weight: = 0.85;

[0183] Topological validation:

[0184] Betti number: = 2, = 2;

[0185] Persistent homology score: 0.78;

[0186] Final confidence: 0.81 → quantum decision VALID.

[0187] Case 3. Biological factor-induced dam breach path validation

[0188] Input path: termite nest → piping → structural hollowing → collapse → dam breach;

[0189] Topological validation process:

[0190] ;

[0191] Run output:

[0192] Quantum encoding time consumption: 11.2 ms;

[0193] spatiotemporal feature decomposition:

[0194] temporal component: [0.63, 0.59, 0.72] (L2-norm = 1.13);

[0195] spatial component: [0.81, 0.77, 0.83] (L2-norm = 1.39);

[0196] dynamic coupling weight: = 0.62;

[0197] topological verification:

[0198] Betti number: = 1, = 1;

[0199] persistent homology score: 0.64;

[0200] final reliability: 0.68 → classic correction PARTIAL.

[0201] All path results (summary table, such as Table 3).

[0202] Table 3

[0203]

[0204] Conclusion:

[0205] The model has a verification accuracy of more than 90% for meteorological / geological paths, proving that quantum encoding can effectively capture physical regularity; biological factor paths often require classic correction (reliability 0.4-0.7), reflecting their nonlinear characteristics.

[0206] 38% of human (engineering) factor paths are judged as INVALID, mainly due to:

[0207] low quantum coherence (average <0.3);

[0208] abnormal spatiotemporal coupling coefficient ( >1.2 or <0.4);

[0209] Low-reliability paths are manually reviewed.

[0210] In this embodiment, the dam collapse event entity is taken as the target node, and all paths from each disaster-causing factor to the dam collapse node are determined in the knowledge graph through the breadth-first search algorithm. All paths from the disaster-causing factor to the dam collapse event node and the relationship between them are displayed in the form of an intuitive graph using visualization technology, completing the entire analysis process.

[0211] For example, Figure 3As shown, the knowledge graph path diagram of dam collapse caused by meteorological factors includes:

[0212] 1. Rainfall characteristics trigger dam collapse path

[0213] Extreme rainfall threshold high → reservoir water level over limit → overtopping condition → water flow continues to erode the dam top, weakening the dam structure → the dam cannot withstand water pressure → dam collapse;

[0214] Large rainfall intensity → rapid increase in surface runoff → large amount of water flow rapidly impacting the dam, increasing the pressure on the dam → local structure of the dam is damaged, stability is reduced → dam collapse;

[0215] Long rainfall duration → continuous infiltration → increased permeability coefficient → piping formation → piping continuously expands, hollowing out the inside of the dam → dam structure instability → dam collapse.

[0216] 2. Temperature change triggers dam collapse path

[0217] Freeze-thaw cycle multiple → dam material cracking → increased permeability coefficient → piping formation → piping destroys the dam foundation, leading to structural imbalance → dam collapse;

[0218] Extreme high temperature high → concrete thermal expansion cracking → impermeable layer failure → concentrated leakage → long-term water erosion of the inside of the dam, reducing the strength → dam collapse.

[0219] 3. Extreme weather events trigger dam collapse path

[0220] Typhoon wind force strong → storm surge water level rise → abnormal rise of reservoir water level → overtopping condition → water flow erodes the dam top and foundation, causing erosion → dam structure is damaged → dam collapse;

[0221] High intensity thunderstorm → mountain torrents form → local erosion of the dam → local structure of the dam is damaged, affecting overall stability → dam collapse;

[0222] Extreme snowfall → snowmelt flood → sudden rise of reservoir water level → overtopping condition → water flow on the dam produces excessive pressure, exceeding the bearing limit → dam collapse;

[0223] Large hail diameter → concrete protective layer peeling off → accelerated corrosion of steel reinforcement → decay of dam structure strength → dam cannot withstand water pressure and external force → dam collapse;

[0224] High wind speed of sandstorm → surface material abrasion → increased porosity → accelerated seepage → abnormal seepage in the inside of the dam, structure damaged → dam collapse;

[0225] Freezing rain increases the self-weight of the structure → increases the stress of the dam foundation → uneven settlement → deformation of the dam structure, cracks appear → dam collapse;

[0226] High wind speed of straight-line wind → control system failure → unable to release flood → continuous rise of reservoir water level → overtopping condition → water flow erodes and damages the dam structure → dam collapse;

[0227] High temperature of hot dome lasts long → Evaporation of reservoir water increases sharply → Anti-seepage layer is exposed and cracks → Seepage channel is formed → Seepage in dam body is serious, and structural strength is reduced → Dam is breached.

[0228] 4. Dam breach path caused by flood factor

[0229] Large flood peak flow → Exceeds design flood discharge capacity → Overtopping condition → Water continuously erodes dam body, leading to structural damage → Dam is breached;

[0230] Long flood duration → Long-term high water level immersion → Anti-seepage layer aging → Seepage increases → Piping is formed → Piping destroys dam body structure, causing dam breach.

[0231] 5. Dam breach path caused by other meteorological variables

[0232] High air humidity → Steel gate corrosion → Flood discharge capacity decreases → Reservoir water level rises → Seepage increases → Piping is formed → Piping destroys dam body structure, leading to dam breach;

[0233] High wind speed → Wind wave erodes dam slope → Revetment structure is damaged → Local seepage increases → Dam body internal structure is damaged, and stability is reduced → Dam is breached.

[0234] As shown in FIG. 1, it is a knowledge graph path diagram of dam breach caused by geological factors, including: Figure 4

[0235] 1. Magmatic activity

[0236] Large thickness of weak interlayer → Insufficient bearing capacity of dam foundation → Foundation dam is breached;

[0237] Lava flow impact → Dam body structure is damaged → Structural instability → Dam is breached;

[0238] Large width of fault fracture zone → Seepage channel is formed → Piping dam is breached;

[0239] Volcanic ash accumulation → Seepage channel changes → Crack extension dam is breached;

[0240] Karst development → Concentrated seepage → Seepage dam is breached.

[0241] 2. Volcanic earthquake

[0242] Folding structure → Uneven settlement of dam body → Dam body tilting → Shear failure dam is breached;

[0243] Acid gas corrosion → Foundation deformation → Crack extension dam is breached;

[0244] Large joint density → Dam body crack development → Shear failure → Dam is breached;

[0245] Abnormal ground stress → Stress concentration damage → Dam is breached.

[0246] ​3. Dam-break path related to geotechnical properties

[0247] Low soil shear strength → soil sliding → sliding dam-break;

[0248] Low rock compressive strength → soil sliding → sliding dam-break;

[0249] Large permeability coefficient → increased seepage → seepage dam-break.

[0250] 4. Dam-break path related to seismic activity

[0251] High seismic intensity → seismic liquefaction → liquefaction dam-break;

[0252] High seismic intensity → landslide → landslide impact → impact dam-break;

[0253] High seismic intensity → abnormal ground stress → stress concentration and damage dam-break;

[0254] Large peak ground acceleration → wide fault fracture zone → seepage channel → piping dam-break;

[0255] Large peak ground acceleration → large joint density → crack development → crack propagation dam-break;

[0256] Landslide → landslide impact → impact dam-break;

[0257] Debris flow → accumulation obstruction → decreased flood discharge capacity → overtopping dam-break;

[0258] Large collapse depth → foundation instability → foundation dam-break;

[0259] Large reservoir capacity of barrier lake → rising water level of barrier lake → seepage channel → piping dam-break.

[0260] As shown in FIG. 1, it is a knowledge graph path diagram of dam-break caused by biological factors, including: Figure 5

[0261] 1. Dam-break path related to plant action

[0262] Large root penetration depth → dam body crack → concentrated seepage → local deformation → structural instability → dam-break;

[0263] Low vegetation coverage → weak soil shear resistance → increased risk of soil sliding → dam body landslide → dam-break;

[0264] Humus accumulation thickness → reduced permeability coefficient → changed seepage field → local deformation → structural instability → dam-break.

[0265] 2. Dam-break path related to animal activity

[0266] Large termite nest density → piping channel → piping → hollowing of dam body internal structure → dam body collapse → dam-break;

[0267] ​Large diameter of rodent burrow → local cavity → structural instability → dam body tilting or cracking → dam failure;

[0268] Large number of bird nests → increased structural load → increased additional stress → local deformation of dam body → structural instability → dam failure.

[0269] 3. Dam failure path related to microbial action

[0270] Large thickness of microbial membrane → increased seepage resistance → seepage anomaly → local water pressure imbalance of dam body → rupture of weak part of dam body → dam failure;

[0271] Large rate of biological decomposition → attenuation of material strength → structural instability → insufficient carrying capacity of dam body → collapse of dam body → dam failure;

[0272] Gas generation (high methane content) → internal gas pressure rise → explosion risk → instantaneous destruction of dam body → dam failure.

[0273] 4. Dam failure path related to algal influence

[0274] Algal outbreak (excessive biomass) → eutrophication of water body → deterioration of water quality → enhanced corrosion of water body → corrosion of dam body materials → structural instability of dam body → dam failure;

[0275] Large thickness of biological siltation → decreased flood discharge capacity → risk of overtopping → dam body overtopping → dam body scouring and destruction → dam failure.

[0276] 5. Dam failure path related to insect erosion

[0277] Large insect density → surface erosion → destruction of protective layer → exposure of main dam body materials → accelerated material damage → structural damage of dam body → dam failure;

[0278] Secretion corrosion (abnormal pH value) → accelerated carbonation of concrete → steel bar corrosion → reduced strength of steel bar → decreased carrying capacity of dam body structure → dam failure.

[0279] As shown in Figure 6 , the knowledge graph path diagram of dam failure caused by human (engineering) factors includes:

[0280] 1. Dam failure path related to design defects

[0281] Low flood control standard (low flood control level set during dam design, without fully considering regional flood historical data and future development trend) → insufficient flood control capacity (dam cannot resist flood impact when encountering large flood) → flood overtopping (flood level exceeds dam crest elevation, water flows over the dam crest) → dam failure (overtopping water flow continuously erodes the dam body, leading to structural damage and dam failure) ;

[0282] Improper material selection (not selecting appropriate construction materials according to dam design requirements and usage environment) → Material failure (premature performance degradation, damage, etc. of materials under the action of water, pressure, etc.) → Structure damage (material failure leads to the loss of support and protection of the dam structure, gradually damaging) → Dam collapse (structure damage is severe, unable to maintain the overall stability of the dam, resulting in dam collapse).

[0283] Design errors (defects in dam structure design) → Uneven stress distribution (during operation, the stress distribution of each part of the dam does not meet the design expectations, with stress concentration in some areas) → Crack generation (cracks appear in stress concentration areas due to excessive pressure) → Problem retention (cracks are not discovered and treated in time, and continue to expand) → Decision errors (based on incorrect design concepts or failure to recognize crack problems, making incorrect maintenance and management decisions) → Dam collapse (crack development eventually leads to the collapse of the dam structure, resulting in dam collapse).

[0284] 2. Dam collapse path related to construction problems

[0285] Poor construction quality (use of inferior materials, rough construction technology, and failure to strictly follow specifications during construction) → Increased permeability (inadequate material compaction and insecure joints allow water to more easily penetrate the dam body) → Dam seepage (large amounts of water seepage leads to surface seepage of the dam body) → Risk out of control (seepage continuously erodes the dam structure, and related risk control measures fail) → Dam collapse (dam structure is severely damaged and cannot withstand water pressure, resulting in dam collapse).

[0286] Construction process violations (violation of design scheme for layered filling, concrete pouring without vibration compaction, etc.) → Hidden dangers not eliminated (hidden dangers such as cavities and cracks caused by illegal construction are not detected and repaired) → Dam seepage (hidden danger sites become water seepage channels, with water seeping out of the dam body) → Risk out of control (seepage triggers a chain reaction, with continuous decline in dam stability and inability to control) → Dam collapse (the dam structure eventually collapses, resulting in a dam collapse accident).

[0287] Cutting corners (during dam construction, construction parties reduce material usage and simplify construction procedures for personal gain) → Weak structure (cutting corners leads to insufficient strength and stability of the dam structure, with weak links) → Reduced carrying capacity (the actual carrying capacity of the dam body cannot meet the design requirements, making it difficult to withstand water pressure and other external forces) → Dam collapse (weak structures gradually deteriorate under stress, eventually leading to dam collapse).

[0288] 3. Dam collapse path related to management omissions

[0289] Improper reservoir capacity management (unreasonable control of reservoir water level, over-reservoir capacity storage or too fast storage speed) → excessive storage (reservoir water level exceeds the safe water level range, causing excessive pressure on the dam) → risk out of control (dam cannot withstand excessive pressure, and relevant risk control measures cannot effectively alleviate the pressure) → dam collapse (dam cannot withstand excessive pressure and collapses);

[0290] Unreasonable dispatch (when flood comes, the reservoir flood discharge dispatch scheme is not scientific, and the flood discharge is not timely or the flood discharge volume is not properly controlled) → water level exceeds the limit (flood and unreasonable dispatch superimpose, leading to rapid rise of reservoir water level exceeding the safety value) → water flow out of control (excessive water level causes water flow impact and erosion force on the dam to be out of control) → dam collapse (dam structure is destroyed under the impact of out-of-control water flow);

[0291] Insufficient patrol → hidden danger not found → problem continues to worsen → dam damage → dam collapse.

[0292] 4. Dam collapse path related to insufficient maintenance

[0293] Old and unrepaired (long-term operation of the dam, dam material aging, structure component damage, and no timely maintenance and repair) → reduced stability (aging leads to reduced strength of dam materials, loose structure connection, and poor overall stability) → flood overtopping (under the action of flood, the dam with poor stability is more likely to be overtopped by flood) → dam collapse (flood overtopping further exacerbates dam damage, eventually leading to dam collapse);

[0294] Equipment failure (dam flood discharge equipment, monitoring equipment, etc. cannot normally operate due to aging, failure, etc.) → blocked flood discharge (flood discharge equipment failure causes flood to be unable to be discharged in time, and reservoir water level continues to rise) → water flow out of control (water level rise causes water flow pressure and erosion force on the dam to be out of control) → dam collapse (dam structure is damaged under the action of out-of-control water flow and eventually collapses);

[0295] Damaged anti-seepage layer → increased seepage → damaged dam foundation → dam collapse.

[0296] 5. Dam collapse path related to operation error

[0297] Misoperation (workers in dam operation and management process cause incorrect valve opening, incorrect water level adjustment, etc. due to operation error) → water level exceeds the limit (misoperation causes abnormal rise of reservoir water level, exceeding the safe water level range) → water flow out of control (excessive water level causes water flow impact and erosion force on the dam to be out of control) → dam collapse (dam structure is destroyed under the impact of out-of-control water flow).

[0298] 6. Dam collapse path related to third party damage

[0299] Tree cutting (large number of trees around the dam area are cut down, destroying the original vegetation protection system) → slope instability (without the root system of trees to fix the soil, the stability of the dam slope soil decreases) → dam collapse (after the slope instability, the dam body cannot be effectively protected, leading to the damage of the dam structure and the collapse of the dam);

[0300] Human destruction (intentional destruction of the dam structure, such as punching holes in the dam body, damaging flood discharge facilities, etc.) → structural damage (human destruction causes the integrity of the dam structure to be destroyed) → dam collapse (the structure is severely damaged and cannot withstand water pressure and other external forces, resulting in dam collapse);

[0301] Illegal sand mining (illegal sand mining in the downstream river of the dam or near the dam body, damaging the riverbed and dam foundation structure) → damage to the anti-seepage layer (sand mining activities cause the dam anti-seepage layer to be damaged, making it easier for water to penetrate into the dam body) → uncontrolled water flow (damage to the anti-seepage layer causes water flow to be turbulent, causing abnormal erosion and pressure on the dam body) → dam collapse (the dam structure is damaged by uncontrolled water flow and collapses).

[0302] 7. Monitoring missing related dam collapse path

[0303] Missing monitoring equipment (no installation of water level monitor, seepage monitor and other key monitoring equipment) → monitoring blind area (unable to obtain important data such as dam body water level and seepage in real time, leaving monitoring gaps) → crack generation (unable to timely detect stress changes in the dam body, leading to cracks being generated without detection) → problem retention (cracks are not discovered and treated in time, and continue to develop) → decision error (based on incomplete or incorrect information, making decisions without taking effective measures) → dam collapse (cracks develop seriously, eventually leading to dam collapse);

[0304] Data error (dam monitoring data has errors, incorrect entry or transmission failure, etc.) → decision error (based on incorrect data, making incorrect decisions on dam operation management and maintenance) → risk out of control (incorrect decisions fail to effectively address the actual risks of the dam, leading to escalating and out-of-control risks) → dam collapse (risk out of control eventually leads to dam collapse).

[0305] 8. Emergency weakness related dam collapse path

[0306] Emergency response delay (after discovering abnormal conditions in the dam, failing to timely initiate emergency plans and organize rescue work) → risk out of control (abnormal conditions are not handled in time, and risks continue to expand and worsen during the delay, exceeding the controllable range) → dam collapse (eventually, due to risk out of control, the dam structure is damaged and collapses);

[0307] Insufficient rescue capacity → maintenance lag → exacerbation of disaster impact after dam collapse.

[0308] 9. Economic factors related dam collapse path

[0309] Lack of funds (insufficient funding for dam maintenance, unable to carry out normal repair and maintenance work) → maintenance lag (due to lack of funds, dam aging, damage and other problems cannot be repaired in time, maintenance work is delayed) → quality decline (long-term lack of maintenance of the dam, structural performance gradually reduces, quality deteriorates) → hidden danger deterioration (existing hidden dangers continue to deteriorate without timely treatment, developing into serious problems) → dam break (hidden dangers deteriorate to a certain extent, leading to the dam unable to operate normally, resulting in dam break);

[0310] Cost over budget → quality decline → dam construction quality affected → increased risk of dam break.

[0311] 10. Policy-related dam break path

[0312] Policy defects (imperfect policies related to dam construction and management, such as lack of strict quality supervision policies, maintenance fund guarantee policies not in place) → reduced stability (due to policy gaps, dam quality cannot be guaranteed during construction and operation, stability gradually decreases) → flood overtopping (stability of the dam is insufficient, and it is difficult to withstand the pressure of the flood when the flood comes) → dam break (flood overtopping continues to damage the dam, eventually leading to dam break);

[0313] Lack of supervision (relevant departments lack effective supervision and inspection of dam construction and operation) → water level exceeds limit (lack of supervision leads to reservoir water level exceeding safety limit, no timely measures such as flood discharge taken) → water flow out of control (high water level causes water flow turbulence, abnormal erosion and pressure on the dam) → dam break (dam structure damaged under the action of out-of-control water flow, eventually leading to dam break).

Claims

1. A multi-source data driven hydropower dam breach path analysis and verification method, characterized in that, The method comprises the following steps: acquiring multi-source data related to dam break of a hydropower hub; dividing the multi-source data into a geological data set, a meteorological data set, a biological data set, and an engineering data set; performing entity extraction and relationship extraction based on the geological data set, the meteorological data set, the biological data set, and the engineering data set to obtain a final entity set and relationship information between entities, and converting the final entity set and the relationship information between entities into graph data to construct a dam break knowledge graph of the hydropower hub; respectively using a path ranking algorithm and a tensor decomposition algorithm to supplement and correct the incorrect relationships and the missing relationships in the dam break knowledge graph of the hydropower hub to obtain a corrected dam break knowledge graph of the hydropower hub; verifying the entity relationships in the corrected dam break knowledge graph of the hydropower hub to obtain a verified dam break knowledge graph of the hydropower hub; determining all paths from each disaster-causing factor to a dam break node by using a breadth-first search algorithm based on the verified dam break knowledge graph of the hydropower hub, visualizing each path, and completing node path analysis of a dam break event of the hydropower hub.

2. The multi-source data driven verification method for dam breach path analysis of a hydropower station according to claim 1, characterized in that, The entity extraction and relationship extraction are specifically as follows: determining entity types related to dam break, including disaster-causing factor entities, dam break event entities, time and space entities, attribute entities, and engineering part entities; designing a plurality of regular expressions with different complexities based on each entity type, and performing entity extraction and relationship extraction on the geological data set, the meteorological data set, the biological data set, and the engineering data set based on each regular expression to obtain entity sets extracted from different data and relationship information between entities in each entity set; integrating information of the same entity based on each entity set and the relationship information between entities in each entity set to obtain a final entity set and relationship information between entities.

3. The multi-source data driven verification method for dam breach path analysis of a hydropower station according to claim 1, wherein, The verified dam break knowledge graph of the hydropower hub is specifically obtained as follows: constructing a dam break path quantum topology verification model for verifying entity relationships; verifying the relationships of the corrected dam break knowledge graph of the hydropower hub by using the dam break path quantum topology verification model to obtain the verified dam break knowledge graph of the hydropower hub.

4. The multi-source data driven verification method for dam breach path analysis of a hydropower station according to claim 3, characterized in that, The dam break path quantum topology verification model comprises a path extraction module, a quantum encoder, a space-time decomposer, a dynamic gating fusion module, a topology verification network, a credibility evaluation module, and a hybrid decision module. The path extraction module is used to extract dam break paths from the corrected dam break knowledge graph of the hydropower hub. The quantum encoder is used to convert the dam break paths into quantum states, encode the paths by quantum entanglement, and output the quantum-encoded paths. The space-time decomposer is used to decompose the quantum-encoded paths into a time convolution branch and a space relationship branch to obtain time features and space features, respectively. The dynamic gating fusion module is used to fuse the time features and the space features to output space-time fused features. The topology verification network is used to construct a topology graph based on the space-time fused features, calculate a topology vulnerability index of the topology graph, and process and verify the topology graph by using a graph neural network and an attention mechanism to output a credibility evaluation result of the dam break path. The credibility evaluation module is configured to quantitatively evaluate the credibility evaluation results of the topological vulnerability index and the dam-break path, and obtain a quantitative result. The hybrid decision module is configured to make a decision according to the quantitative result, and output a verification result.

5. The multi-source data driven hydropower dam breach path analysis and verification method according to claim 4, wherein, An expression of the quantum-encoded path is: in, The path after quantum encoding; for The path states of the layered quantum circuit are connected to each layer through tensor products; It is an exponential function with the natural constant as its base; The imaginary unit; For the first Hermitian matrix dynamically generated by layered quantum circuits; To correct the linear unit; For the first Trainable qubit parameters of layered quantum circuits; For the first Normalized dam-break path of layered quantum circuits; It is the tensor product; It is the hyperbolic tangent function; For the first Trainable qubit parameters of +1 layer quantum circuit; For the first Normalized dam-break path of +1 layer quantum circuit; For layer index; It is a diagonal phase matrix; diagonal elements; For transpose; For the first Learnable entanglement gate parameters for layered quantum circuits; For the first Layered quantum circuits are projected onto classical features activated by ReLU; For the first Layered quantum circuits project classical features through tanh activation; For the first Layered quantum circuit path position encoding unitary matrix; For the first The dam-break path of layered quantum circuits; It is an L2 norm; It is a multilayer perceptron; This is a constructor for diagonal matrices.

6. The multi-source data driven verification method for dam breach path analysis of a hydropower station according to claim 4, wherein, An expression of the quantum-encoded path decomposed into a time convolution branch and a space relationship branch is: wherein, is a spatio-temporal feature; is a long short-term memory network, processing temporal dependency; is a real part of quantum path encoding, reflecting causal strength; is a path after quantum encoding; is an outer product operation; is a graph attention network, processing spatial correlation between nodes; is an imaginary part of quantum path encoding, reflecting phase relationship; is a Sigmoid function, compressing weights to interval (0, 1); is a trainable vector; is taking mean value of temporal feature, capturing global temporal pattern; is a temporal feature, input real part of quantum state; is taking maximum value of spatial feature, highlighting key region; is a spatial feature, input imaginary part of quantum state; is a dynamic coupling weight; is a normalized cross-correlation function; is a transpose; is an L2 norm.

7. The multi-source data driven verification method for dam breach path analysis of a hydropower station according to claim 4, characterized in that, An expression of the space-time fusion feature is: in, It features spatiotemporal fusion; The Sigmoid function compresses the weights to the (0,1) interval. It is a time-related feature; Spatial features; These are trainable weights; for and The splicing result; To pass Calculated weights; To smooth the ReLU function and ensure the output is positive; These are trainable weights; This is an outer product operation; For Hadema; To pass Calculate the nonlinear coupling coefficients.

8. The multi-source data driven verification method for dam breach path analysis of a hydropower station according to claim 4, wherein, An expression of the credibility evaluation results of the topological vulnerability index and the dam-break path is: wherein, is a topological vulnerability index; is a topological graph; is a connected component of diameter; is a branch; is the total number of vertices; is the 1st Betti number; is a vertex set; is an edge set; is the feature of the th topological graph vertex; is the spatiotemporal fusion feature corresponding to the th topological graph vertex; is the topological graph edge between the th topological graph vertex and the th topological graph vertex; is an indicator function, which is 1 if the condition is satisfied, otherwise, it is 0; is the spatiotemporal fusion feature corresponding to the th topological graph vertex; is the L2 norm; is the distance threshold; is the median of the upper triangular part of the distance matrix, excluding the diagonal part; is the credibility evaluation result of the dam-break path; is a normalization factor; is the total number of local subgraphs; is the Betti number corresponding to the th local subgraph; is a multi-head attention mechanism; is the feature representation of the th local subgraph; is the feature representation of the global graph; is a natural constant; is a Betti number calculation function of the homology group matrix; is the homology group matrix corresponding to the th local subgraph.

9. The multi-source data driven verification method for dam breach path analysis of a hydropower station according to claim 4, wherein, An expression of the quantitative result is: wherein, is a quantification result; is a function type; is a topological vulnerability index; is a topological graph; is a hyperbolic tangent function; is a credibility evaluation result of a dam-break path.

10. The multi-source data driven hydropower dam breach path analysis and verification method of claim 4, wherein, An expression of the verification result is: wherein, is a verification result; is a path valid; is a lower limit value of a quantization result for judging a path valid; is a path partially valid; is an upper limit value of a quantization result for judging a path invalid; is a path invalid; is a quantization result.

Images