Knowledge graph-based buttress dam break pre-evaluation method, equipment and medium

Abstract

The invention discloses a buttress dam break pre-evaluation method and device based on a knowledge graph, and a medium, and relates to the technical field of dam break pre-estimation. The method comprises the steps of obtaining multi-source data of buttress dam monitoring points, wherein the monitoring points are located at a dam body and a joint of the buttress dam; fusing the multi-source data, extracting entities and entity relationships of the multi-source data through natural language processing, and constructing a knowledge graph; performing mode layer updating and data layer updating on the knowledge graph based on preset time; selecting a basic event and an intermediate event according to an entity relationship and entities in the knowledge graph; performing sequence modeling on the basic events by using the improved recurrent neural network, and identifying dam break types; and performing fault tree evaluation on the dam break type to obtain a risk index of the dam break type. By means of the method, comprehensive consideration of the logic relation among complex environmental factors, faults and reasons is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of dam-break prediction, and in particular to a knowledge graph-based dam-break prediction method, device and medium. Background Art

[0002] As an important water conservancy project facility, the safety of reservoir pier dams is directly related to the safety of life and property of the people in the downstream area and the stability of the ecological environment. Due to the influence of natural factors (such as extreme weather and earthquakes) and human factors (such as design defects and improper maintenance), pier dams are at risk of dam failure, especially since pier dams are composed of water retaining panels and a series of piers, and these panels and piers, as well as the piers and foundations, need to be connected.

[0003] However, traditional dam break risk analysis methods mostly rely on single static data analysis and manual experience, lack comprehensive consideration of complex environmental factors, and are difficult to fully and real-time reflect the operating status and potential risks of buttress dams.

[0004] Through the above analysis, the problems and defects of the prior art are as follows:

[0005] The dam break risk in existing technologies relies on single static data analysis and manual experience, and lacks comprehensive consideration of the logical relationship between complex environmental factors, faults and causes. Summary of the invention

[0006] The embodiments of the present application provide a knowledge graph-based dam breach pre-assessment method, device, and medium, which can solve the problem in the prior art that the risk of dam breach depends on single static data analysis and manual experience, and lacks comprehensive consideration of complex environmental factors and the logical relationship between faults and causes.

[0007] In a first aspect, an embodiment of the present application provides a method for pre-assessment of dam break of a buttress dam based on a knowledge graph, the method comprising: obtaining multi-source data of monitoring points of the buttress dam, the multi-source data comprising coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature, the monitoring points being located at the dam body and joints of the buttress dam; fusing the multi-source data, and extracting entities and entity relationships of the multi-source data through natural language processing to construct a knowledge graph; updating the knowledge graph at the pattern layer and the data layer based on a preset time; selecting basic events and intermediate events according to the entity relationships and entities in the knowledge graph; using an improved recurrent neural network to perform sequence modeling on basic events and identify dam break types; performing a fault tree evaluation on the dam break type to obtain a risk index for the dam break type.

[0008] In one implementation of the present application, multi-source data of the buttress dam monitoring points are obtained, specifically including: collecting real-time coordinates, stresses, deformations, seepage rates, water levels and water temperatures through sensors installed on the buttress dam body and joints; obtaining geological and geomorphological information and meteorological forecast data of the buttress dam through remote sensing technology; and obtaining historical data of the buttress dam, which includes design information, construction records, operation logs, flood records and maintenance records.

[0009] In one implementation of the present application, entities and entity relationships of detection data are extracted through natural language processing to construct a knowledge graph, specifically including: using natural language processing and information extraction technology to identify entities, entities include monitoring point numbers, monitoring equipment and monitoring indicators; extracting entity relationships between entities in multi-source data, entity relationships include rainfall and water level, stress and deformation, monitoring equipment and monitoring indicators; using graph database technology to organize entities and entity relationships into a knowledge graph.

[0010] In one implementation of the present application, model layer updates include concept updates, relationship updates and attribute updates, and data layer updates include entity updates, relationship and attribute value updates; based on a preset time, model layer updates and data layer updates are performed on the knowledge graph, specifically including: when monitoring indicators change, concept updates are performed in the knowledge graph; when new monitoring indicators are added, relationship updates are performed in the knowledge graph; when new monitoring equipment is added, entity updates are performed in the knowledge graph.

[0011] In one implementation of the present application, the method also includes: using natural language processing and information extraction technology to automatically extract online data containing pier dam keywords in an online database, the online database including news reports; integrating the extracted online data with the knowledge graph, and updating the entities and entity relationships in the knowledge graph.

[0012] In one implementation of the present application, an improved recurrent neural network is used to perform sequence modeling on entity relationships and features in a knowledge graph to identify dam break types, specifically including: encoding entities and entity relationships into vectors to form a sequence input into the recurrent neural network, and combining it with the hidden state of the previous time step; introducing an attention mechanism into the recurrent neural network to calculate the attention score of the hidden state of each time step; extracting time dependency and dynamic change features in the sequence through the hidden layer; using a softmax classifier to classify the extracted features to obtain attention weights; using the attention weights to perform weighted summation of the hidden state of each time step in the sequence to obtain a weighted representation; using the weighted representation and the hidden state of the current time step for calculation in the output layer to identify the dam break type.

[0013] In one implementation of the present application, a fault tree evaluation is performed on the dam break type to obtain a risk index of the dam break type, which specifically includes: taking the dam break risk as the top event, connecting the basic events and the intermediate events through logic gates, and assigning a failure probability to each basic event based on historical data; traversing the fault tree using the minimum cut set method to obtain a minimum cut set combination; calculating the product of the failure probabilities of the basic events to obtain the probability of occurrence of the minimum cut set; calculating the weighted sum of the probability of occurrence of the minimum cut set to obtain the risk index of the dam break type.

[0014] In one implementation of the present application, after obtaining the risk index of the dam breach type, the method further includes: when the risk index reaches a warning threshold, triggering a warning mechanism, and notifying staff via SMS, email or a dedicated application.

[0015] In the second aspect, an embodiment of the present application also provides a pier dam break pre-assessment device based on a knowledge graph, the device comprising at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: obtain multi-source data of the monitoring points of the pier dam, the multi-source data including coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature, the monitoring points being located at the dam body and connections of the pier dam; fuse the multi-source data, and extract entities and entity relationships of the multi-source data through natural language processing to construct a knowledge graph; update the knowledge graph at the pattern layer and the data layer based on a preset time; select basic events and intermediate events based on the entity relationships and entities in the knowledge graph; use the improved recurrent neural network to perform sequence modeling on the basic events and identify the dam break type; perform fault tree evaluation on the dam break type to obtain the risk index of the dam break type.

[0016] In the third aspect, the embodiment of the present application also provides a non-volatile computer storage medium for pre-assessment of pier dam break based on a knowledge graph, storing computer executable instructions, and the computer executable instructions are set to: obtain multi-source data of the monitoring points of the pier dam, the multi-source data including coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature, and the monitoring points are located at the dam body and connections of the pier dam; fuse the multi-source data, and extract the entities and entity relationships of the multi-source data through natural language processing to construct a knowledge graph; based on a preset time, update the knowledge graph at the pattern layer and the data layer; select basic events and intermediate events according to the entity relationships and entities in the knowledge graph; use the improved recurrent neural network to perform sequence modeling on the basic events and identify the type of dam break; perform fault tree evaluation on the dam break type to obtain the risk index of the dam break type.

[0017] The embodiments of the present application provide a dam-break pre-assessment method, device and medium based on a knowledge graph. Fault tree analysis can help reveal the logical relationship between faults and causes, clarify which combination of factors is most likely to cause dam break, and achieve comprehensive consideration of the logical relationship between complex environmental factors, faults and causes. By introducing more comprehensive data collection and analysis methods, including multi-source information such as real-time data, historical data, and environmental factors, the risk of dam break can be assessed more accurately. This helps to reduce risk assessment errors caused by insufficient data or deviations from a single data source. Enhancing the consideration of complex environmental factors can more accurately predict and assess dam break risks. Create an intelligent platform that integrates multi-source data fusion, knowledge graph construction, risk analysis construction and early warning, build a comprehensive knowledge graph for safe operation of pier dams, achieve early warning and dynamic management of dam break risks, and improve the accuracy and efficiency of pier dam break risk prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 A flowchart of a pier dam breach pre-assessment method based on a knowledge graph provided in an embodiment of the present application;

[0020] Figure 2 A logical schematic diagram of a pier dam breach pre-assessment method based on a knowledge graph provided in an embodiment of the present application;

[0021] Figure 3 A fault tree diagram of a pier dam failure pre-assessment method based on a knowledge graph provided in an embodiment of the present application;

[0022] Figure 4 A schematic diagram of the internal structure of a knowledge graph-based pier dam break pre-assessment device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0024] The embodiments of the present application provide a knowledge graph-based dam breach pre-assessment method, device, and medium, which solves the problem in the prior art that the dam breach risk relies on single static data analysis and manual experience, and lacks a comprehensive consideration of complex environmental factors and the logical relationship between faults and causes.

[0025] The technical solution proposed in the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0026] Figure 1 A flow chart of a dam failure pre-assessment method based on a knowledge graph provided in an embodiment of the present application. Figure 1 As shown, a pier dam breach pre-assessment method based on a knowledge graph provided in an embodiment of the present application specifically includes the following steps:

[0027] Step 10: Obtain multi-source data of the monitoring points of the buttress dam, including coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature. The monitoring points are located at the dam body and joints of the buttress dam.

[0028] In this step, the monitoring points are located at the dam body and the connection of the buttress dam, because the dam body and the connection are the weak areas of the buttress dam; Figure 2 As shown, socioeconomic data can also be collected, such as population distribution, economic output, industrial structure, etc. in the downstream area of ​​the buttress dam to assess the socioeconomic losses that may be caused by dam failure.

[0029] As an optional embodiment, obtaining multi-source data of the buttress dam monitoring points may specifically include: Step 101: collecting real-time coordinates, stresses, deformations, seepage rates, water levels and water temperatures through sensors installed on the buttress dam body and joints; Step 102: obtaining geological and geomorphological information and meteorological forecast data of the buttress dam through remote sensing technology; Step 103: obtaining historical data of the buttress dam, which includes design data, construction records, operation logs, flood records and maintenance records.

[0030] Step 20: Fusion multi-source data, and extract entities and entity relationships from multi-source data through natural language processing to build a knowledge graph.

[0031] In this step, based on the collected data, the data is first preprocessed, and then entities such as pier dam parameters, operating indicators, environmental variables and their relationships are extracted through technical means such as text analysis and relational database query to construct a safety knowledge graph containing pier dam entities, entity attributes and relationships between entities.

[0032] As an optional embodiment, entities and entity relationships of detection data are extracted through natural language processing to construct a knowledge graph, which may specifically include: Step 201: using natural language processing and information extraction technology to identify entities, where the entities include monitoring point numbers, monitoring equipment, and monitoring indicators;

[0033] In this step, entities related to the safety of the buttress dam are identified, such as the buttress dam name, structure, monitoring equipment, indicators, geological conditions, climate conditions, etc.

[0034] Step 202: extracting entity relationships between entities in multi-source data, wherein the entity relationships include rainfall and water level, stress and deformation, monitoring equipment and monitoring indicators.

[0035] In this step, the relationships between entities are extracted from the data, such as the relationship between rainfall and reservoir water level, the correspondence between monitoring equipment and monitoring indicators, the correlation between monitoring indicators and the safety status of buttress dams, and the causal relationship between risk factors and dam breach consequences.

[0036] Step 203: Use graph database technology to organize entities and entity relationships into a knowledge graph.

[0037] In this step, graph database technology is used to organize entities and their relationships into a knowledge graph, which includes entity nodes, relationship edges, and attribute information. The constructed knowledge graph is verified to ensure the accuracy of entities and relationships and the integrity of the knowledge graph.

[0038] Step 30: Based on the preset time, update the model layer and data layer of the knowledge graph;

[0039] In this step, as the buttress dam operates over time, its structure and performance may change. At the same time, external environmental factors such as climate and geological conditions may also have an impact on the safety of the buttress dam. Therefore, the information in the knowledge graph needs to be updated regularly or in real time to reflect these changes and ensure the accuracy and timeliness of the risk analysis.

[0040] As an optional embodiment, the model layer update includes concept update, relationship update and attribute update, and the data layer update includes entity update, relationship and attribute value update; based on the preset time, the model layer update and data layer update of the knowledge graph may specifically include: Step 301: when the monitoring indicators change, perform concept update in the knowledge graph; Step 302: when new monitoring indicators are added, perform relationship update in the knowledge graph; Step 303: when new monitoring equipment is added, perform entity update in the knowledge graph.

[0041] In specific implementation, the model layer update includes concept update, relationship update, and attribute update; concept update: including the addition, modification or deletion of concepts related to the buttress dam. For example, with the development of buttress dam monitoring technology, it may be necessary to introduce new monitoring indicators or equipment, thereby adding corresponding concepts to the knowledge graph; relationship update: updating or adjusting the relationship between concepts, for example, it is necessary to establish or modify the corresponding relationship in the knowledge graph when a new influencing factor is found to be associated with the safety of the buttress dam; attribute update: updating the attribute information of the concept, for example, some physical parameters of the buttress dam, such as water level, seepage velocity, etc., will change over time, and these changes need to be reflected in the knowledge graph in a timely manner. Data layer update: including entity update, relationship and attribute value update; entity update: including the addition, modification or deletion of buttress dams and their related entities, for example, adding a new monitoring device or deleting an abandoned monitoring point; relationship and attribute value update: updating the relationship and attribute value between entities, for example, updating the attribute values ​​of the buttress dam such as water level and seepage according to real-time monitoring data, or adjusting the relationship strength between the buttress dam and certain risk factors according to new research results.

[0042] As an optional embodiment, the method may also include: Step 304: using natural language processing and information extraction technology, automatically extracting online data containing pier dam keywords from an online database, the online database including news reports; Step 305: integrating the extracted online data with the knowledge graph, and updating the entities and entity relationships in the knowledge graph.

[0043] In specific implementation, there are two update methods: automatic update and manual intervention; automatic update: including data extraction, knowledge fusion, and incremental update; data extraction: using natural language processing and information extraction technology to automatically extract data related to the safety of pier dams from diversified data sources such as real-time monitoring systems, online databases, news reports, etc.; knowledge fusion: integrating the extracted data with the existing knowledge in the knowledge graph to update or supplement the entities, relationships and attributes in the knowledge graph; incremental update: based on the update content published by the data source or the user's behavior on the system, such as feedback, query, etc., the knowledge graph is partially updated to maintain its timeliness and accuracy; manual intervention: including expert review, rule definition, and complete update; expert review: for errors or omissions that may exist in the automatic update process, experts are required to conduct manual review and correction; rule definition: in the process of updating the model layer, it is necessary to manually define the update rules to deal with possible conflicts and abnormal situations; complete update: in specific circumstances, such as system upgrades, data source changes, etc., it may be necessary to completely rebuild the knowledge graph to ensure its accuracy and completeness.

[0044] Step 40: Select basic events and intermediate events based on entity relationships and entities in the knowledge graph.

[0045] The basic events and intermediate events in this step are the basis of the fault tree.

[0046] Step 50: Use the improved recurrent neural network to perform sequence modeling on basic events and identify the type of dam failure.

[0047] In this step, the attention mechanism is used in conjunction with the recurrent neural network to enhance its ability to process sequence data.

[0048] As an optional embodiment, an improved recurrent neural network is used to perform sequence modeling on entity relationships and features in a knowledge graph to identify dam break types, which may specifically include: Step 501: Encode entities and entity relationships into vectors to form a sequence input to the recurrent neural network, and combine it with the hidden state of the previous time step; Step 502: Introduce an attention mechanism into the recurrent neural network to calculate the attention score of the hidden state of each time step; Step 503: Extract the time dependency and dynamic change features in the sequence through the hidden layer; Step 504: Use a softmax classifier to classify the extracted features to obtain attention weights; Step 505: Use the attention weights to perform weighted summation on the hidden state of each time step in the sequence to obtain a weighted representation; Step 506: Use the weighted representation and the hidden state of the current time step to calculate the output layer to identify the dam break type.

[0049] For example, the coordinates, stress, deformation, seepage, water level, and water temperature of the buttress dam are all within the normal range. Geographical location: downstream of a mountain river; rainfall: reaching a historical peak in the past week; soil type: loose and easily eroded clay; maintenance record: no major maintenance in the past five years, but there have been multiple small-scale repairs. First, encode these entities and their relationships into vectors: geographic location: [0.1, 0.5, 0.3, ...] (assuming it is a high-dimensional vector), rainfall: [0.9, 0.05, 0.0, ...] (indicating an extremely high level), soil type: [0.2, 0.8, 0.0, ...] (indicating loose and easily eroded), maintenance record: [0.4, 0.1, 0.5, ...] (indicating no major maintenance in the recent period). These vectors will be encoded in order. The sequence is formed into a sequence and input into the recurrent neural network, and combined with the hidden state of the previous time step. For example, if the rainfall at this time is heavy, the corresponding hidden state is a higher attention score; the hidden layer of the recurrent neural network will gradually process each element in the sequence while keeping the memory of the previous element information. In this way, the model can capture the time dependence and dynamic change characteristics such as soil saturation caused by the continuous increase in rainfall; the hidden state of each time step is weighted and summed to obtain a weighted representation, which combines the information of all elements in the sequence; the weighted representation and the hidden state of the current time step (or the last time step) are passed to the output layer as input, followed by a softmax activation function to identify the type of dam break, such as flood-type dam break, infiltration-type dam break, etc.

[0050] Step 60: Perform a fault tree assessment on the dam failure type to obtain a risk index for the dam failure type.

[0051] In this step, the cause-effect relationship and consequences of different dam failure paths are analyzed through event tree or fault tree methods to assess the dam failure risk level and potential losses.

[0052] As an optional embodiment, a fault tree evaluation is performed on the dam break type to obtain a risk index of the dam break type, which may specifically include: step 601: taking the dam break risk as the top event, connecting the basic events and the intermediate events through logic gates, and assigning a failure probability to each basic event based on historical data; step 602: traversing the fault tree using the minimum cut set method to obtain a minimum cut set combination; step 603: calculating the product of the failure probabilities of the basic events to obtain the probability of occurrence of the minimum cut set; step 604: calculating the weighted sum of the probability of occurrence of the minimum cut set to obtain the risk index of the dam break type.

[0053] In this step, if Figure 3As shown in the figure, T represents the top event (dam break event), A, B, C represent the intermediate events, and X represents the basic event. The dam break factor analysis and probability statistical analysis under different situations are analyzed to obtain the risk level, risk probability or event priority, which provides support for the subsequent risk warning and emergency measures.

[0054] As an optional embodiment, after obtaining the risk index of the dam breach type, the method may further include: when the risk index reaches a warning threshold, triggering a warning mechanism, and notifying staff via SMS, email or a dedicated application.

[0055] The above is an embodiment of the method proposed in this application. Based on the same inventive concept, this application embodiment also provides a pier dam break pre-assessment device based on a knowledge graph, and its structure is as follows: Figure 4 shown.

[0056] Figure 4 A schematic diagram of the internal structure of a pier dam breach pre-assessment device based on a knowledge graph provided in an embodiment of the present application. Figure 4 As shown, the device includes:

[0057] at least one processor 401;

[0058] and, a memory 402 communicatively coupled to the at least one processor;

[0059] Among them, the memory 402 stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor 401 so that the at least one processor 401 can: obtain multi-source data of the monitoring points of the pier dam, the multi-source data including coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature, and the monitoring points are located at the dam body and connection of the pier dam; fuse the multi-source data, and extract the entities and entity relationships of the multi-source data through natural language processing to construct a knowledge graph; based on a preset time, update the knowledge graph at the pattern layer and the data layer; select basic events and intermediate events according to the entity relationships and entities in the knowledge graph; use the improved recurrent neural network to perform sequence modeling on the basic events and identify the dam break type; perform fault tree evaluation on the dam break type to obtain the risk index of the dam break type.

[0060] Some embodiments of the present application provide corresponding Figure 1A non-volatile computer storage medium for pre-assessment of pier dam break based on knowledge graph stores computer executable instructions, wherein the computer executable instructions are set to: obtain multi-source data of monitoring points of the pier dam, the multi-source data including coordinates, stress, deformation, seepage, water level, rainfall and water temperature, the monitoring points are located at the dam body and connection of the pier dam; fuse the multi-source data, extract entities and entity relationships of the multi-source data through natural language processing, and construct a knowledge graph; update the knowledge graph at the pattern layer and the data layer based on the preset time; select basic events and intermediate events according to the entity relationships and entities in the knowledge graph; use the improved recurrent neural network to perform sequence modeling on the basic events and identify the dam break type; perform fault tree evaluation on the dam break type to obtain the risk index of the dam break type.

[0061] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the IoT device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0062] The system and medium provided in the embodiments of the present application correspond one-to-one to the method. Therefore, the system and medium also have similar beneficial technical effects to the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the system and medium will not be repeated here.

[0063] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0064] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0065] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0066] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0067] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0068] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0069] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0070] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0071] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A pier dam breach pre-assessment method based on knowledge graph, characterized in that: The method comprises: Acquire multi-source data of monitoring points of the buttress dam, the multi-source data including coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature, the monitoring points being located at the dam body and the connection of the buttress dam; The multi-source data are integrated, and entities and entity relationships of the multi-source data are extracted through natural language processing to construct a knowledge graph; Based on a preset time, updating the model layer and the data layer of the knowledge graph; Selecting basic events and intermediate events according to entity relationships and entities in the knowledge graph; Using the improved recurrent neural network, the basic events are sequence modeled to identify the dam failure type; A fault tree assessment is performed on the dam failure type to obtain a risk index of the dam failure type.

2. The method for pre-assessment of pier dam failure based on knowledge graph according to claim 1 is characterized in that: Obtain multi-source data of the monitoring points of the buttress dam, including: Sensors installed on the pier dam and its joints collect real-time coordinates, stress, deformation, seepage, water level and water temperature; Obtain geological and geomorphological information and meteorological forecast data of the buttress dam through remote sensing technology; The historical data of the buttress dam is obtained, wherein the historical data includes design information, construction records, operation logs, flood records, and maintenance records.

3. The method for pre-assessment of pier dam failure based on knowledge graph according to claim 1 is characterized in that: The entities and entity relationships of the detection data are extracted through natural language processing to construct a knowledge graph, which specifically includes: Using natural language processing and information extraction technology, the entity is identified, and the entity includes a monitoring point number, monitoring equipment, and monitoring indicators; Extracting entity relationships between the entities in the multi-source data, wherein the entity relationships include rainfall and water level, stress and deformation, monitoring equipment and monitoring indicators; The entities and entity relationships are organized into a knowledge graph using graph database technology.

4. The method for pre-assessment of pier dam failure based on knowledge graph according to claim 3 is characterized in that: The model layer update includes concept update, relationship update and attribute update, and the data layer update includes entity update, relationship and attribute value update; Based on the preset time, the knowledge graph is updated at the model layer and the data layer, specifically including: When monitoring indicators change, concepts are updated in the knowledge graph; When new monitoring indicators are added, the relationships are updated in the knowledge graph; When new monitoring equipment is added, entity updates are performed in the knowledge graph.

5. The method for pre-assessment of pier dam failure based on knowledge graph according to claim 4 is characterized in that: The method further comprises: Using natural language processing and information extraction technology, automatically extracting online data containing key words of buttress dam from an online database, wherein the online database includes news reports; The extracted online data is integrated with the knowledge graph to update the entities and entity relationships in the knowledge graph.

6. The method for pre-assessment of pier dam failure based on knowledge graph according to claim 1, characterized in that: The improved recurrent neural network is used to perform sequence modeling on entity relationships and features in the knowledge graph to identify dam breach types, including: Encode the entities and entity relationships into vectors to form a sequence input into a recurrent neural network and combine it with the hidden state of the previous time step; Introducing an attention mechanism into the recurrent neural network to calculate the attention score of the hidden state at each time step; Extract the time dependency and dynamic change features in the sequence through hidden layers; Use the softmax classifier to classify the extracted features and obtain the attention weights; Using the attention weights, weighted sum is performed on the hidden state of each time step in the sequence to obtain a weighted representation; The weighted representation and the hidden state of the current time step are used for calculation of the output layer to identify the type of dam breach.

7. The method for pre-assessment of pier dam failure based on knowledge graph according to claim 2 is characterized in that: A fault tree assessment is performed on the dam failure type to obtain a risk index of the dam failure type, specifically including: Taking the dam break risk as the top event, connecting the basic events and intermediate events through logic gates, and assigning a failure probability to each basic event based on historical data; Use the minimum cut set method to traverse the fault tree and obtain the minimum cut set combination; Calculate the product of the basic event failure probabilities to obtain the probability of occurrence of the minimum cut set; The weighted sum of the probability of occurrence of the minimum cut set is calculated to obtain the risk index of the dam break type.

8. The method for pre-assessment of pier dam failure based on knowledge graph according to claim 1, characterized in that: After obtaining the risk index of the dam failure type, the method further includes: When the risk index reaches the warning threshold, the warning mechanism is triggered to notify staff via SMS, email or dedicated application.

9. A knowledge graph-based pier dam breach pre-assessment device, characterized in that: The device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: Acquire multi-source data of monitoring points of the buttress dam, the multi-source data including coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature, the monitoring points being located at the dam body and the connection of the buttress dam; The multi-source data are integrated, and entities and entity relationships of the multi-source data are extracted through natural language processing to construct a knowledge graph; Based on a preset time, updating the model layer and the data layer of the knowledge graph; Selecting basic events and intermediate events according to entity relationships and entities in the knowledge graph; Using the improved recurrent neural network, the basic events are sequence modeled to identify the dam failure type; A fault tree assessment is performed on the dam failure type to obtain a risk index of the dam failure type.

10. A non-volatile computer storage medium for pre-assessment of pier dam failure based on knowledge graph, storing computer executable instructions, characterized in that: The computer executable instructions are configured to: Acquire multi-source data of monitoring points of the buttress dam, the multi-source data including coordinates, stress, deformation, seepage, as well as water level, rainfall and water temperature, the monitoring points being located at the dam body and the connection of the buttress dam; The multi-source data are integrated, and entities and entity relationships of the multi-source data are extracted through natural language processing to construct a knowledge graph; Based on a preset time, updating the model layer and the data layer of the knowledge graph; Selecting basic events and intermediate events according to entity relationships and entities in the knowledge graph; Using the improved recurrent neural network, the basic events are sequence modeled to identify the dam failure type; A fault tree assessment is performed on the dam failure type to obtain a risk index of the dam failure type.

Images