Dam safety evaluation method and system based on knowledge graph

Abstract

The invention provides a dam safety evaluation method and system based on a knowledge graph, and relates to the technical field of dam safety evaluation. The method comprises the following steps: collecting monitoring data of a typical dam section, and establishing a statistical model of related measuring points according to the collected monitoring data, so as to analyze the working state of the dam; establishing a finite element model of the typical dam section, and performing simulation calculation on the overall strength and stability of the typical dam section and the state of the new and old concrete joint surface to obtain a finite element simulation result; constructing a dam safety evaluation index system according to the collected monitoring data of the typical dam section, the finite element simulation result and the evaluation related data of the typical dam section; sorting a hierarchical system of the dam safety comprehensive evaluation index system, and performing knowledge modeling, knowledge extraction, knowledge fusion and knowledge storage processing according to a sorting result to construct a knowledge graph; and carrying out data calling and safety evaluation by utilizing the constructed knowledge graph. According to the scheme, dam safety evaluation work can be carried out more clearly, visually and efficiently.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of dam safety evaluation, in particular to a dam safety evaluation method and system based on a knowledge graph. BACKGROUND

[0002] Dam safety evaluation refers to a comprehensive evaluation of the structural integrity, stability and operation status of a dam through a series of technical means and methods, to determine whether there are safety hazards, and to take appropriate measures for repair or reinforcement.

[0003] At present, dam safety evaluation mainly relies on traditional monitoring methods and experience-based judgment. Specifically, it first collects monitoring data of the dam, such as displacement, stress, seepage, etc., and then uses statistical methods to process and analyze the data to evaluate the working behavior of the dam. In addition, it also combines the data of dam geological exploration, preliminary design, safety appraisal and operation inspection to comprehensively evaluate the overall safety of the dam. However, the traditional monitoring method often only focuses on the data changes of a single measuring point, lacking a comprehensive understanding of the overall working behavior of the dam. At the same time, the collected data is often not fully utilized, resulting in limited accuracy of the evaluation results. Moreover, the existing evaluation models are mostly based on experience-based judgment, lacking scientificity and systematicness. This leads to the evaluation results being influenced by the subjective factors of the evaluators, making it difficult to guarantee the objectivity and accuracy of the evaluation.

[0004] That is to say, the existing dam safety evaluation method has obvious deficiencies in data utilization, evaluation model, knowledge management and reaction speed, etc. Therefore, there is an urgent need for a more scientific, systematic and efficient dam safety evaluation method to fully utilize existing technology and data resources, improve the accuracy and reliability of the evaluation results, and meet the urgent needs of dam safety management. SUMMARY

[0005] The purpose of the present application is to provide a dam safety evaluation method and system based on a knowledge graph, which can be used to carry out dam safety evaluation work more clearly, intuitively and efficiently.

[0006] The present application is implemented as follows: In a first aspect, the application provides a dam safety evaluation method based on a knowledge graph, comprising the following steps: collecting and establishing a statistical model of relevant measuring points according to the collected monitoring data of a typical dam section to analyze the working behavior of the dam; establishing a finite element model of the typical dam section, and performing simulation calculation on the overall strength, stability and new and old concrete joint surface state of the typical dam section to obtain finite element simulation results; constructing a dam safety evaluation index system according to the collected monitoring data of the typical dam section, the finite element simulation results and the evaluation related data of the typical dam section, wherein the evaluation related data includes one or more of geological exploration, preliminary design, safety appraisal and operation inspection; combing the hierarchical system of the dam safety comprehensive evaluation index system, and performing knowledge modeling, knowledge extraction, knowledge fusion and knowledge storage processing according to the combing results to construct a knowledge graph; and using the constructed knowledge graph to perform data retrieval and safety evaluation.

[0007] In some implementations, the collecting and establishing a statistical model of relevant measuring points according to the collected monitoring data of a typical dam section to analyze the working behavior of the dam comprises: based on the collected measured data including displacement settlement, stress and temperature, analyzing the data by drawing process line graphs and / or characteristic value tables to establish a statistical model of relevant measuring points to analyze and evaluate the working behavior and change law of the dam.

[0008] In some implementations, the dam safety evaluation index system comprises: dam strength safety evaluation indexes including concrete stress and concrete autogenous volume deformation; dam overall stability safety evaluation indexes, the concerned physical quantities including dam body deformation, seepage and dam foundation anti-sliding stability, and the corresponding concerned measuring points including horizontal displacement, vertical displacement, dam foundation uplift pressure and anti-sliding stability safety factor; and new and old concrete joint surface stability safety evaluation indexes, the concerned physical quantities and information including joint surface opening degree, joint surface reinforcement stress and new and old concrete temperature.

[0009] In some implementations, the method further comprises perfecting the constructed dam safety evaluation index system by adding associated measuring point information, reasoning mechanism and operation safety evaluation rules.

[0010] In some implementations, the dam safety evaluation index system is constructed according to the collected monitoring data of typical dam sections, finite element simulation results, and evaluation related information of typical dam sections, including: establishing a dam safety state set, which is divided into normal, slight abnormal, general abnormal, and serious abnormal; establishing a dam safety evaluation index system according to the monitoring data of typical dam sections and the finite element simulation results, to screen the associated measuring point information of the safety evaluation index, the associated measuring point information including the associated measuring point information of the dam strength safety evaluation index, the associated measuring point information of the dam overall stability safety evaluation index, and the associated measuring point information of the new and old concrete joint stability safety evaluation index; establishing an associated measuring point information reasoning mechanism of the safety evaluation index; and establishing a dam safety comprehensive evaluation rule, including: if each associated measuring point information and the finite element simulation result are less than the allowable value specified in the design or specification, the dam operation state is recorded as normal; otherwise, each associated measuring point information reasoning mechanism of the safety evaluation index is further analyzed and processed to determine the dam safety comprehensive evaluation result according to the evaluation conclusion of the weakest link of each safety evaluation index.

[0011] In some implementations, the knowledge modeling includes presenting the required entities in the knowledge graph in the form of ontology model construction by using Protégé software, completing the ontology creation work by determining the ontology scope, listing the key items in the ontology, determining the class and structure of the class, the attributes of the class, and the characteristics of the attributes.

[0012] In some implementations, the knowledge extraction includes: obtaining related knowledge content including design reports, safety evaluation reports, operation reports, monitoring annual reports, and finite element simulation results of typical dam sections as data sources. Reading and organizing each related knowledge content, and combing the hierarchy and required content of the knowledge graph to obtain corresponding text data; wherein the text data includes basic conditions and related information of the safety evaluation object, key monitoring dam sections, dam strength, stability, and new and old concrete joint conditions, and related monitoring projects, monitoring physical quantities, and key monitoring measuring points. The obtained text data is annotated, and the annotated text data is input into the extraction model to extract corresponding entities, relationships, and attributes.

[0013] In some implementations, the knowledge fusion includes data integration, disambiguation, processing, and updating of synonymous knowledge with different description methods from different knowledge sources under the same framework specification, and fusion and unification of data, information, and methods.

[0014] In some implementations, the information in the knowledge graph is also retrieved and queried by using the built-in Cypher query language of Neo4j, the keywords in the query statement are extracted by entity recognition, and a structured query statement is generated, the parsed entity is searched corresponding to the entity in the knowledge base, and the corresponding result information is presented according to the entity and the corresponding relationship reasoning.

[0015] In a second aspect, the application provides a dam safety evaluation system based on a knowledge graph, which comprises: a model establishment module configured to collect and establish a statistical model of related measuring points according to the collected monitoring data of a typical dam section to analyze the working behavior of the dam. A simulation calculation module is configured to establish a finite element model of a typical dam section and perform simulation calculation on the overall strength, stability and new and old concrete joint surface state of the typical dam section to obtain finite element simulation results. An index system construction module is configured to construct a dam safety evaluation index system according to the collected monitoring data of a typical dam section, finite element simulation results, and evaluation related data of a typical dam section; the evaluation related data includes one or more of geological exploration, preliminary design, safety appraisal and operation inspection. A knowledge graph construction module is configured to sort out the hierarchical system of the dam safety comprehensive evaluation index system, and construct a knowledge graph according to the sorting results of knowledge modeling, knowledge extraction, knowledge fusion and knowledge storage processing. A safety evaluation module is configured to utilize the constructed knowledge graph for data retrieval and safety evaluation.

[0016] Compared with the prior art, the application has at least the following advantages or beneficial effects: The application provides a dam safety evaluation method based on a knowledge graph, which introduces knowledge graph technology to optimize the dam safety evaluation process, thereby efficiently integrating, managing and utilizing complex and diverse data and information, and improving the accuracy and reliability of dam safety evaluation. At the same time, this technical solution also solves the problems of insufficient data utilization, lack of scientificity and systematicness of evaluation models in the prior art, and provides strong support for dam safety management. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0018] Figure 1 A flowchart of an embodiment of the dam safety evaluation method based on a knowledge graph of the application; Figure 2Fig. 1 is a schematic diagram of a typical dam section model according to an embodiment of the present application; Figure 3 Fig. 2 is a schematic diagram of a dam model according to an embodiment of the present application; Figure 4 Fig. 3 is a schematic diagram of a new and old concrete contact surface model according to an embodiment of the present application; Figure 5 Fig. 4 is a schematic diagram of a part of the ontology of a dam safety evaluation system according to an embodiment of the present application; Figure 6 Fig. 5 is a schematic diagram of a knowledge reasoning query result according to an embodiment of the present application; Figure 7 Fig. 6 is a schematic diagram of a knowledge graph reasoning evaluation result according to an embodiment of the present application; Figure 8 Fig. 7 is a structural block diagram of an embodiment of a dam safety evaluation system based on a knowledge graph. DETAILED DESCRIPTION

[0019] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0020] Some embodiments of the present application will be described in detail below with reference to the drawings. In the case of no conflict, each of the following embodiments and each feature in the embodiments can be combined with each other.

[0021] Embodiment 1 Please refer to Figure 1 The dam safety evaluation method based on the knowledge graph includes the following steps: Step S101: Collecting monitoring data of a typical dam section and establishing a statistical model of related measuring points according to the collected monitoring data to analyze the working behavior of the dam; It should be noted that in step S101, professional equipment and methods are used to collect monitoring data of a typical dam section, such as displacement, settlement, stress, temperature and other physical quantities. These data are the basis for subsequent analysis. Then, statistical methods are used to process and analyze these data to establish a statistical model of related measuring points. These models can reflect the working behavior of the dam under different working conditions, such as deformation trend, stress distribution, etc. That is, by collecting and analyzing monitoring data, abnormal changes of the dam can be found in time, providing a scientific basis for subsequent safety evaluation. At the same time, the establishment of the statistical model makes the evaluation process more objective and quantitative, reducing the influence of subjective judgment.

[0022] Step S102: Establish a finite element model of the typical dam section and simulate the overall strength, stability, and new and old concrete joint surface state of the typical dam section to obtain the finite element simulation results. It should be noted that the above step S102 is based on step S101 to further establish a finite element model of the typical dam section. This model is a fine simulation of the dam structure, which can consider factors such as material properties, geometric shapes, and boundary conditions. By simulating the model, key information such as the overall strength, stability, and new and old concrete joint surface state of the dam can be obtained. The finite element simulation can reveal the stress distribution and deformation of the dam under complex conditions, providing more in-depth and accurate information for evaluating the safety of the dam. At the same time, this step also verifies the accuracy and reliability of the statistical model.

[0023] Step S103: According to the collected monitoring data of the typical dam section, the finite element simulation results, and the evaluation related materials of the typical dam section, construct a dam safety evaluation index system; the evaluation related materials include one or more of geological exploration, preliminary design, safety appraisal and operation inspection; It should be noted that the above step S103 combines the results of steps S101 and S102, as well as evaluation related materials such as geological exploration, preliminary design, safety appraisal and operation inspection, to construct a comprehensive and systematic dam safety evaluation index system. This system covers multiple evaluation dimensions and indicators, which can fully reflect the safety of the dam. The construction of the evaluation index system makes the evaluation process more systematic and standardized, improving the accuracy and operability of the evaluation. At the same time, this step also provides a data basis for the subsequent construction of the knowledge graph.

[0024] Step S104: Sort the hierarchical system of the dam safety comprehensive evaluation index system, and according to the sorting results, perform knowledge modeling, knowledge extraction, knowledge fusion and knowledge storage processing to construct a knowledge graph; It should be noted that the above step S104 is based on step S103 to sort the hierarchical system of the dam safety evaluation index system. This includes determining the logical relationship between the evaluation indicators, weight distribution, etc. Then, using knowledge modeling, knowledge extraction, knowledge fusion and knowledge storage techniques, a knowledge graph containing dam safety evaluation related knowledge and information is constructed. The construction of the knowledge graph makes the knowledge and information in the dam safety evaluation process more concentrated and easy to manage. At the same time, the hierarchical structure and logical relationship of the knowledge graph make the evaluation process clearer and more organized. This step provides strong support for subsequent data retrieval and safety evaluation.

[0025] Step S105: Use the constructed knowledge graph to retrieve data and perform safety evaluation.

[0026] It is to be noted that the above step S105 is based on the knowledge graph constructed in step S104, and relevant data and information are called according to the needs for safety evaluation. This includes comprehensive analysis using statistical models, finite element simulation results and evaluation indexes and other data to obtain the safety state evaluation result of the dam. The use of knowledge graph for data retrieval and safety evaluation can greatly improve the efficiency and accuracy of evaluation. At the same time, this step also makes the evaluation result more objective and reliable, and provides strong support for the safety management of the dam.

[0027] In summary, the above embodiments optimize the dam safety evaluation process by introducing knowledge graph technology, which can efficiently integrate, manage and utilize complex and diverse data and information, and improve the accuracy and reliability of dam safety evaluation. At the same time, this technical solution also solves the problems of insufficient data utilization, lack of scientificity and systematicness of evaluation model in the prior art, and provides strong support for dam safety management.

[0028] Based on the foregoing scheme, in some implementations of the present application, the statistical model of the relevant measuring point is established based on the collected monitoring data of the typical dam section to analyze the working behavior of the dam, including: based on the collected measured data including displacement settlement, stress and temperature, the data is analyzed by drawing process line graph and / or characteristic value table to establish the statistical model of the relevant measuring point to analyze and evaluate the working behavior and change law of the dam.

[0029] It should be noted that the working principle of the above implementation mode includes: 1) first, the monitoring data of the typical dam section is collected by various monitoring devices (such as displacement meters, stress meters, thermometers, etc.) arranged on the dam. These monitoring data should cover the key parts and possible weak areas of the dam to ensure the comprehensiveness and representativeness of the data. 2) Then, the collected monitoring data needs to be preliminarily processed and analyzed. This includes data cleaning, verification and format conversion, etc. to ensure the accuracy and consistency of the data. Next, the monitoring data is analyzed by drawing process line charts and / or feature value tables. Process line charts can intuitively show the trend of monitoring data over time, while feature value tables can extract key data features such as maximum value, minimum value, average value and standard deviation, etc. 3) Then, based on the results of data analysis, statistical models of related measuring points are established. These models can be linear, nonlinear, or complex models based on machine learning or deep learning. The selection of the model should be determined according to the characteristics of the data and the requirements of the analysis. Among them, the statistical model is used to describe and predict the variation law of the monitoring data, so as to analyze and evaluate the working state of the dam. For example, the displacement settlement trend of the dam in the future period of time can be predicted by the model, or the stress distribution state of the dam under certain working conditions can be evaluated. 4) Finally, the established statistical model is used to analyze and evaluate the working state of the dam. This includes evaluating the overall stability, deformation trend and abnormal change of the dam. By comparing historical data and current data, the variation law and potential safety hazards of the dam can be found, providing scientific basis for subsequent maintenance and reinforcement.

[0030] Based on the foregoing scheme, in some implementations of the present application, the dam safety evaluation index system includes: dam strength safety evaluation indexes, including concrete stress and concrete autogenous volume deformation; dam overall stability safety evaluation indexes, the physical quantities of concern include dam body deformation, seepage and dam foundation anti-slide stability, and the corresponding measuring points of concern include horizontal displacement, vertical displacement, dam foundation uplift pressure and anti-slide stability safety factor; and new and old concrete joint surface stability safety evaluation indexes, the physical quantities and information of concern include joint surface opening degree, joint surface steel stress and new and old concrete temperature.

[0031] It should be noted that according to the engineering structure characteristics, the dam safety comprehensive evaluation index system mainly includes three types of key indicators: dam strength safety evaluation index, dam overall stability safety evaluation index and new and old concrete joint surface stability safety evaluation index. The dam strength safety evaluation index mainly includes concrete stress and concrete autogenous volume deformation; the dam overall stability safety evaluation index mainly focuses on the physical quantities of dam body deformation, seepage and dam foundation anti-sliding stability, and the corresponding main concern measurement points are horizontal displacement, vertical displacement, dam foundation uplift pressure and anti-sliding stability safety factor and other information; the new and old concrete joint surface stability safety evaluation index mainly focuses on the physical quantities and information of joint surface opening degree, joint surface reinforcement stress, new and old concrete temperature, etc.

[0032] The index system covers multiple aspects of dam safety evaluation, including strength, overall stability and new and old concrete joint surface stability, etc., and can comprehensively reflect the safety of the dam. At the same time, the selection of evaluation indexes is based on physical principles and mechanical analysis, and has scientific basis and reliability. Moreover, the data source of the evaluation index is clear, which can be measured and recorded by monitoring equipment, and is convenient for actual operation and application. Thus, through regular monitoring and data analysis, abnormal changes of the dam can be found in time, which provides the possibility for timely response measures.

[0033] Based on the foregoing scheme, in some implementations of the present application, the method further comprises perfecting the constructed dam safety evaluation index system by adding associated measurement point information, reasoning mechanism and running safety evaluation rules.

[0034] Among them, by adding the measurement point information associated with the main evaluation index, the dam safety evaluation index system is further enriched and perfected. These associated measurement point information can provide more comprehensive and detailed data support, which helps to more accurately evaluate the safety of the dam. Illustratively, the positions and types of the measurement points associated with the main evaluation indexes can be determined according to the structural characteristics, operating conditions and historical data of the dam; the corresponding monitoring equipment such as displacement meter, stress meter, thermometer, seepage pressure meter, etc. is arranged at these measurement point positions to collect the data of the associated measurement points; the collected associated measurement point data is integrated into the dam safety evaluation index system for unified analysis and processing.

[0035] By introducing the reasoning mechanism, the known data and rules are used for logical reasoning to find potential safety hazards and abnormal changes. The reasoning mechanism can enhance the intelligence and automation level of the dam safety evaluation system, improve the accuracy and efficiency of the evaluation. Illustratively, a reasoning mechanism based on expert system, fuzzy theory or artificial neural network technology can be established; the data and rules in the dam safety evaluation index system are input into the reasoning mechanism for logical reasoning and judgment; according to the reasoning result, the potential safety hazards and abnormal change information are output to provide scientific basis for decision makers.

[0036] By formulating the operation safety evaluation rule, the standards and requirements of dam safety evaluation are clarified, and the objectivity and fairness of the evaluation are ensured. The operation safety evaluation rule can provide guidance and protection for the safe operation of the dam and reduce safety risks. Exemplarily, the operation safety evaluation rule can be formulated by referring to relevant standards and specifications at home and abroad and combining the actual situation of the dam; the operation safety evaluation rule should include the threshold value of the evaluation index, the selection of the evaluation method, the determination standard of the evaluation result, etc.; the operation safety evaluation rule is applied to the dam safety evaluation index system, and the safety of the dam is quantitatively and qualitatively evaluated.

[0037] Based on the foregoing scheme, in some implementations of the present application, the dam safety evaluation index system is constructed according to the collected monitoring data of the typical dam section, the finite element simulation results, and the evaluation related materials of the typical dam section, including: establishing a dam safety state set, which is divided into normal, slight abnormal, general abnormal and serious abnormal; according to the monitoring data and the finite element simulation results of the typical dam section, a dam safety evaluation index system is established to screen the associated measuring point information of the safety evaluation index, the associated measuring point information including the associated measuring point information of the dam strength safety evaluation index, the associated measuring point information of the dam overall stability safety evaluation index, and the associated measuring point information of the new and old concrete joint stability safety evaluation index; a reasoning mechanism of the associated measuring point information of the safety evaluation index is established; a dam safety comprehensive evaluation rule is established, including: if each associated measuring point information and the finite element simulation result is less than the allowable value specified in the design or specification, the dam operation state is recorded as normal; otherwise, each safety evaluation index is further analyzed and processed through the associated measuring point information reasoning mechanism to determine the dam safety comprehensive evaluation result according to the evaluation conclusion of the weakest link of each safety evaluation index.

[0038] In the above implementation, the process of constructing the dam safety evaluation index system is further described, and the screening of the associated measuring point information, the establishment of the reasoning mechanism and the formulation of the dam safety comprehensive evaluation rule are integrated.

[0039] Among them, for the construction of dam safety evaluation system, including the establishment of dam safety state set: the safety state of dam is divided into four levels: normal, slight abnormal, general abnormal and serious abnormal. These four levels can clearly reflect the safety situation of the dam, and provide the basis for subsequent safety evaluation. And including screening the related measuring point information of safety evaluation index: based on the monitoring data and finite element simulation results of typical dam section, the measuring point information related to the safety evaluation index of dam is screened out. These information includes the related measuring point information of dam strength safety evaluation index (such as concrete stress, concrete autogenous volume deformation, etc.), the related measuring point information of dam overall stability safety evaluation index (such as dam deformation, seepage, dam foundation anti sliding stability, etc.) and the related measuring point information of new and old concrete joint stability safety evaluation index (such as joint opening degree, joint steel stress, new and old concrete temperature, etc.).

[0040] For the establishment of related measuring point information reasoning mechanism, including establishing reasoning mechanism for the screened related measuring point information. This mechanism can use known data and rules to logically reason about related measuring point information to discover potential safety hazards and abnormal changes. The reasoning mechanism may be based on expert system, fuzzy theory, artificial neural network and other technologies, and the specific choice depends on the actual situation of dam and evaluation demand.

[0041] For the establishment of dam safety comprehensive evaluation rules, including: 1) basic rule: if each related measuring point information and finite element simulation result is less than the allowable value specified in the design or specification, the dam operation state is normal. This rule ensures that when each index of the dam is within the safe range, the dam is considered to be in normal operation. 2) Further analysis and processing: if there is related measuring point information or finite element simulation result exceeding the allowable value, further analysis and processing is carried out through the related measuring point information reasoning mechanism of each safety evaluation index. This step aims to dig deeper into the reasons behind abnormal data and assess its impact on dam safety. 3) Determine the comprehensive evaluation result of dam safety: according to the evaluation conclusion of the weakest link of each safety evaluation index, determine the comprehensive evaluation result of dam safety. This result will consider the evaluation of all safety evaluation indexes to obtain the overall safety situation of the dam.

[0042] In summary, by comprehensively considering the monitoring data, finite element simulation results and evaluation related information, the dam safety evaluation index system constructed can more accurately reflect the safety status of the dam. The screening of associated measuring point information and the establishment of the reasoning mechanism make the evaluation process more comprehensive and detailed, and can find potential safety hazards. The introduction of the reasoning mechanism makes the evaluation process more intelligent and automated, improving the evaluation efficiency. The results of the comprehensive evaluation of the dam safety can provide a scientific basis for decision-makers, helping them to take necessary measures in a timely manner to ensure the safe operation of the dam. That is, in the above implementation manners, by constructing the dam safety evaluation index system, establishing the reasoning mechanism of the associated measuring point information, and formulating the comprehensive evaluation rules of the dam safety, a comprehensive, accurate and intelligent solution for the safety evaluation of the dam is provided.

[0043] Based on the foregoing scheme, in some implementation manners of the present application, the knowledge modeling includes presenting the entities required in the knowledge graph in the form of ontology model constructed by using Protégé software, and completing the ontology creation work by determining the ontology scope, listing the key items in the ontology, determining the class and structure of the class, the attribute of the class, and the characteristics of the attribute.

[0044] Among them, Protégé is an open source ontology editor and knowledge base framework, which is widely used to construct ontology models. In the knowledge graph, entities are the basic units of information. Through the ontology model, these entities are endowed with richer semantics and relationships, thereby forming a structured knowledge system.

[0045] Among them, for determining the ontology scope: before constructing the ontology, the application field and scope of the ontology need to be determined first. This helps to ensure that the constructed ontology can accurately cover the knowledge of the target field, and avoid unnecessary complexity.

[0046] For listing the key items in the ontology: key items are the core elements in the ontology, including classes (or concepts), attributes, relationships, etc. By listing these key items, the content contained in the ontology can be clearly defined.

[0047] For determining the class and structure of the class: the class is a basic concept or entity type in the ontology. Determining the structure of the class involves defining the hierarchical relationship (such as parent-child relationship) of the class, the intersection relationship of the class, the complementary relationship of the class, etc. This helps to construct an ontology model with clear logic and reasonable structure.

[0048] For the attribute of the class: the attribute is an element that describes the characteristics of the class. Each class can have multiple attributes, which can be data attributes (such as age, height) or object attributes (such as parent-child relationship, friend relationship). Determining the attribute of the class helps to more comprehensively express the characteristics of the class For the determination of the characteristics of the attribute: in addition to defining the attribute itself, the characteristics of the attribute need to be determined, such as the value range of the attribute, the transitivity of the attribute, the symmetry, the reflexive property, etc. These characteristics help to further refine the semantics of the attribute and improve the expressiveness of the ontology.

[0049] Based on the foregoing scheme, in some implementations of the present application, the knowledge extraction includes: obtaining relevant knowledge contents including design reports of typical dam sections, safety appraisal reports, operation reports, monitoring annual reports and finite element simulation results as data sources. Each relevant knowledge content is read and sorted out to obtain the corresponding text data, and the hierarchy and required content of the knowledge graph are combed; wherein the text data includes the basic situation and related information of the safety evaluation object, the key monitoring dam section, the dam strength, the stability and the new and old concrete joint surface situation, and the related monitoring projects, monitoring physical quantities and key monitoring points. The obtained text data is labeled, and the labeled text data is input into the extraction model to extract the corresponding entities, relationships and attributes.

[0050] It should be noted that the knowledge extraction aims to extract key information from various data sources to construct and enrich the knowledge graph. Exemplarily, the main steps include: 1) Knowledge processing: the input data is further combed to obtain the hierarchy and required content of the knowledge graph, including the basic situation and related information of the safety evaluation object, the key monitoring dam section, the dam strength, the stability and the new and old concrete joint surface situation, and the related monitoring projects, monitoring physical quantities and key monitoring points. After completion, these contents are saved as txt text files for subsequent model training.

[0051] 2) Corpus annotation: the combed txt file is imported into the model, and after manually annotating a small amount of data, the system will filter out key information from all samples in the data set and perform annotation.

[0052] 3) Entity and relationship extraction: the annotated training data set is input into the extraction model according to the ratio of 8:2 of training set and validation set, so that the model learns features from the training set, extracts corresponding entities and relationships, and compares with the validation set to ensure the accuracy of the extraction.

[0053] 4) Model evaluation: after the model training is completed, F1-score, precision P (Presion) and recall R (Recall) are selected as the evaluation indicators of the evaluation results to comprehensively evaluate the extraction effect of the knowledge extraction model.

[0054] Based on the foregoing scheme, in some implementations of the present application, the knowledge fusion includes data integration, disambiguation, processing and updating of synonymous knowledge with different description methods from different knowledge sources under the same framework specification, and fusion and unification of data, information and methods.

[0055] It should be noted that through knowledge fusion, information from multiple knowledge sources such as design reports, monitoring data, and finite element simulation results can be integrated to build a comprehensive dam safety evaluation index system. Among them, data integration includes collecting, organizing and summarizing data from different knowledge sources to form a unified data set. In the integration process, problems such as inconsistent data formats, data redundancy, and data missing need to be solved to ensure the accuracy and integrity of the data. Disambiguation includes identifying and eliminating synonyms, near synonyms, and confusion between different description methods to ensure the accuracy and consistency of knowledge expression. Processing includes further processing of integrated data to extract valuable information and knowledge. For example, data cleaning, data conversion, data mining and other technical means are used to extract key information, build knowledge graphs or generate new knowledge representation forms. Updating includes adding new knowledge, deleting outdated knowledge, correcting incorrect knowledge, and other operations to ensure the timeliness and accuracy of the knowledge system.

[0056] Based on the foregoing scheme, in some implementations of the present application, it also includes using the built-in Cypher query language of Neo4j to search and query information in the knowledge graph, extracting keywords in the query statement through entity recognition and generating a structured query statement, searching the parsed entities in the knowledge base, and presenting the corresponding result information according to the entity and relationship correspondence reasoning.

[0057] It should be noted that Neo4j is a high-performance graph database that is specifically designed to store and query complex relational data. Cypher is a built-in query language for Neo4j that provides an intuitive and powerful way to retrieve, modify and traverse graph data. By querying and analyzing dam monitoring data, design parameters and other information, real-time monitoring and early warning analysis of dam safety status can be achieved. This helps to timely identify potential safety hazards and improve the safety and reliability of the dam.

[0058] In order to make the skilled in the art more intuitive understanding of the present application, hereinafter will be described with a specific example. In this example, the dam safety evaluation based on knowledge graph includes steps S1-S4: Step S1: Collect the monitoring data of the dam and analyze it by drawing process line graph, characteristic value table, etc., and establish statistical models of key measuring points in displacement, opening degree, seepage and other physical quantities based on measured data and analyze the main factors affecting each physical quantity.

[0059] Step S2: Select a typical dam section (take the overflow dam section as an example) to establish a finite element model of the dam section and the new and old concrete joint surface (as shown in FIG. 2) that meets the accuracy. Select the highest and lowest upstream water levels in a year to perform numerical simulation analysis and comparison, and focus on the stress and strain, displacement, settlement, anti-slide stability of the dam as a whole, and the stability and related changes of the new and old concrete joint surface. Figures 2-4

[0060] By comparing the finite element calculation results, monitoring data and statistical model analysis results of the dam section, the working behavior and change law of the dam section are preliminarily analyzed and evaluated, the change of the new and old concrete joint surface is focused on, and the working behavior and change law of the entire dam are analyzed, which provides data support and verification for the subsequent establishment of dam safety evaluation index system and knowledge graph to carry out evaluation work.

[0061] Step S3: According to the structure characteristics of the dam, combined with the analysis of preliminary design, geological exploration and other reports, the dam safety evaluation index system established this time mainly includes three types of key indicators: dam strength safety evaluation index, dam overall stability safety evaluation index and new and old concrete joint surface stability safety evaluation index. The dam strength safety evaluation index mainly includes concrete stress and concrete autogenous volume deformation; the main physical quantities concerned in the dam overall stability safety evaluation index are dam body displacement settlement, seepage and dam foundation stability, and the corresponding main concerned measuring points are horizontal displacement, vertical displacement, dam foundation uplift pressure, seepage water level and anti-slide stability safety factor information; the new and old concrete joint surface stability safety evaluation index mainly concerns the opening degree of the joint surface, the stress of the joint surface reinforcement, the temperature of the new and old concrete and other physical quantities and information.

[0062] Step S4: (1) If the data type is a monitoring result characteristic value statistical table that has been created, an automatic program can be used to complete the knowledge extraction and storage of the table in one key batch. Then place the program in the same directory as the table, and name the file according to the monitoring method and structure part.

[0063] (2) If it is other data types, the following steps need to be taken: ​The entities required in the knowledge graph are presented in the form of ontology model using Protégé software. First, determine the scope of the ontology. The ontology researched and created in this application is applied to the dam safety evaluation field in water conservancy engineering. The weak position of the dam is determined by analyzing the monitoring data, finite element simulation results and other contents. Entities such as key monitoring parts, projects, objects and measuring points are established. The dam safety evaluation index system is constructed to evaluate and judge the operation safety of the dam and ensure the safe operation of the dam. Second, list the key items in the ontology. Add the key information such as properties and descriptions of the ontology to the ontology to make the ontology have clearer concepts. Finally, determine the class and structure of the class, the properties of the class, and the characteristics of the properties. Repeat the above steps in Protégé to improve the class and property information in the ontology. After the above operations, the ontology creation work is completed, and part of the ontology is established as shown in Figure 5 .

[0064] The related materials sorted out in the above steps are annotated, and the annotated training data set is input into the extraction model according to the ratio of 8:2 of the training set and the validation set. The corresponding entities and relationships are extracted and compared with the validation set to ensure the accuracy of the extraction.

[0065] The extracted entities, relationships and properties are stored in the Neo4j database in the form of nodes and edges to realize the visualization of the knowledge structure and the storage of the knowledge network.

[0066] Using the built-in Cypher query language of Neo4j, according to the content and structure of the knowledge graph and the form and purpose of the sentence, a query template is established. When the knowledge graph receives a query sentence, the key words in the sentence are extracted by entity recognition, the appropriate query template is selected to generate a structured query sentence, and the result is returned after querying and matching in the knowledge base. For example, in the dam safety evaluation system knowledge graph, it is necessary to query "what is the monitoring project of the new and old concrete joint surface state?", the entity word "new and old concrete joint surface state" and the relationship word "monitoring project" are extracted, and the template sentence is called according to the two key information. The execution result is shown in Figure 6 .

[0067] Step S5: Develop dam safety evaluation criteria, the main steps are as follows: (1) Establish a dam safety state set, which is divided into normal, slight abnormal, general abnormal and serious abnormal.

[0068] (2) Based on monitoring and numerical simulation information, establish a dam operation safety evaluation index system, and screen the associated measuring point information of the safety evaluation index.

[0069] (3) Establish the correlation point information reasoning mechanism of the overall strength of the dam, the overall stability of the dam and the state of the new and old concrete joint surface.

[0070] (4) Establish the dam operation safety evaluation rule. If the monitoring data of the correlation points of each evaluation index and the numerical simulation results are less than the allowable values specified in the design or specification, the abnormal rate is zero, then the evaluation index is normal, the evaluation object is normal, and the dam operation safety state is normal. If the monitoring data of the correlation points of each evaluation index exceeds the allowable values specified in the design or specification, the correlation point information reasoning mechanism of the dam strength safety, the overall stability of the dam and the stability of the new and old concrete joint surface is further analyzed and processed, and the dam operation safety evaluation result is taken as the evaluation conclusion of the weakest link of each index.

[0071] The knowledge graph of the established dam safety comprehensive evaluation index system is fused in the dam safety monitoring evaluation management platform, based on the B / S mode and the micro-service architecture, the centralized management of engineering information and monitoring data is realized; the monitoring and early warning model is integrated, the online monitoring and early warning of engineering safety state is realized, and the platform support is provided for the safe operation of the engineering. On this basis, the knowledge graph is taken as a database, the dam safety comprehensive evaluation index system is taken as a calculation model, and is deployed in a private cloud platform, the powerful storage and reasoning capability of the knowledge graph is used to realize the data retrieval and safety evaluation. For example, according to the monitoring data analysis result, the measured values of some vertical displacement points of the dam exceed the historical extreme value on a certain day, which needs to be paid attention to. The measured data of the day stored in the knowledge graph is called, and according to the reasoning rule set in advance, the knowledge graph can store and display the normal and abnormal points (as shown in Figure 7 ), so that the abnormal conditions and safety hazards existing in the dam can be found in time, which helps to judge the dam operation condition faster and make response measures.

[0072] Similarly, the calculated values or monitoring values of other points are called and compared with the specification values or historical extreme values, and if the preset rule is met, it is safe, otherwise it is abnormal. After judging the conditions of all points, the results are summarized in the dam safety monitoring evaluation management platform for comprehensive evaluation.

[0073] Embodiment 2 Please refer to Figure 8 , the embodiment of the application provides a dam safety evaluation system based on a knowledge graph, which comprises: The model establishing module is configured to collect and establish a statistical model of a relevant measuring point according to the collected monitoring data of the typical dam section, so as to analyze the working behavior of the dam. The simulation calculation module is configured to establish a finite element model of the typical dam section, and perform simulation calculation on the overall strength, stability and new and old concrete joint surface state of the typical dam section, so as to obtain a finite element simulation result. The index system construction module is configured to construct a dam safety evaluation index system according to the collected monitoring data of the typical dam section, the finite element simulation result, and evaluation related data of the typical dam section; the evaluation related data includes one or more of geological exploration, preliminary design, safety appraisal and operation inspection. The knowledge graph construction module is configured to sort out a hierarchical system of the dam safety comprehensive evaluation index system, and construct a knowledge graph according to knowledge modeling, knowledge extraction, knowledge fusion and knowledge storage processing of the sorted out result. The safety evaluation module is configured to perform data retrieval and safety evaluation by using the constructed knowledge graph.

[0074] The specific implementation process of the system is please refer to the dam safety evaluation method based on knowledge graph provided in embodiment 1, which will not be repeated here.

[0075] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be regarded as exemplary and non-limiting, and the scope of the present application is defined by the appended claims rather than the above description, and all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. A dam safety evaluation method based on a knowledge graph, characterized in that, The method comprises the following steps: collecting monitoring data of typical dam sections and establishing a statistical model of relevant measuring points according to the collected monitoring data to analyze the working state of the dam; establishing a finite element model of the typical dam sections and simulating the overall strength, stability and new-old concrete joint surface state of the typical dam sections to obtain finite element simulation results; constructing a dam safety evaluation index system according to the collected monitoring data of the typical dam sections, the finite element simulation results and evaluation-related information of the typical dam sections; the evaluation-related information includes one or more of geological exploration, preliminary design, safety appraisal and operation inspection; combing the hierarchical system of the dam safety comprehensive evaluation index system and constructing a knowledge graph by knowledge modeling, knowledge extraction, knowledge fusion and knowledge storage processing according to the combing result; using the constructed knowledge graph to retrieve data and perform safety evaluation.

2. The method of claim 1, wherein, The collecting monitoring data of typical dam sections and establishing a statistical model of relevant measuring points according to the collected monitoring data to analyze the working state of the dam comprises: based on the collected measured data including displacement settlement, stress and temperature, analyzing the data by drawing process line graphs and / or characteristic value tables to establish a statistical model of relevant measuring points to analyze and evaluate the working state and change law of the dam.

3. The method of claim 1, wherein, The dam safety evaluation index system comprises: dam strength safety evaluation indexes including concrete stress and concrete autogenous volume deformation; dam overall stability safety evaluation indexes, the concerned physical quantities including dam body deformation, seepage and dam foundation anti-sliding stability, and the corresponding concerned measuring points including horizontal displacement, vertical displacement, dam foundation uplift pressure and anti-sliding stability safety factor; and new-old concrete joint surface stability safety evaluation indexes, the concerned physical quantities and information including joint surface opening degree, joint surface reinforcement stress and new-old concrete temperature.

4. The method of claim 1, wherein, The method further comprises perfecting the constructed dam safety evaluation index system by adding related measuring point information, reasoning mechanisms and operation safety evaluation rules.

5. The method of claim 1, wherein, The constructing a dam safety evaluation index system according to the collected monitoring data of the typical dam sections, the finite element simulation results and evaluation-related information of the typical dam sections comprises: establishing a dam safety state set, which is divided into normal, slight abnormal, general abnormal and serious abnormal; establishing a dam safety evaluation index system according to the monitoring data and the finite element simulation results of the typical dam sections to screen related measuring point information of the safety evaluation indexes; the related measuring point information includes related measuring point information of dam strength safety evaluation indexes, related measuring point information of dam overall stability safety evaluation indexes and related measuring point information of new-old concrete joint surface stability safety evaluation indexes; establishing a related measuring point information reasoning mechanism of the safety evaluation indexes; establishing dam safety comprehensive evaluation rules, which include: if each related measuring point information and the finite element simulation result is less than the allowable value specified in the design or specification, the dam is recorded as being in a normal operation state; otherwise, each related measuring point information is further analyzed and processed by the related measuring point information reasoning mechanism of each safety evaluation index to determine the dam safety comprehensive evaluation result according to the evaluation conclusion of the weakest link of each safety evaluation index.

6. The method of claim 1, wherein, The knowledge modeling includes presenting the required entities in the knowledge graph in the form of an ontology model by using Protégé software, completing the ontology creation by determining the ontology scope, listing the key items in the ontology, determining the class and structure of the class, the attributes of the class, and the characteristics of the attributes.

7. The method of claim 1, wherein, The knowledge extraction includes: acquiring related knowledge contents including design reports, safety assessment reports, operation reports, monitoring annual reports, and finite element simulation results of typical dam sections as data sources; reading and sorting the related knowledge contents to sort out the levels and required contents of the knowledge graph to obtain corresponding text data; the text data includes basic conditions and related information of the safety evaluation object, key monitoring dam sections, dam strength, stability, and new and old concrete joint surface conditions, and related monitoring projects, monitoring physical quantities, and key monitoring points; performing corpus annotation on the obtained text data, and feeding the annotated text data into an extraction model to extract corresponding entities, relationships, and attributes.

8. The method of claim 1, wherein, The knowledge fusion includes data integration, disambiguation, processing, and updating of synonymous knowledge with different description methods from different knowledge sources under the same framework specification, and fusion and unification of data, information, and methods.

9. The method of claim 1, wherein, It also includes searching and querying information in the knowledge graph by using the built-in Cypher query language of Neo4j, extracting keywords in the query statement by entity recognition and generating a structured query statement, searching the parsed entities in the knowledge base, and presenting the corresponding result information according to the entity and relationship correspondence reasoning. 10.A dam safety evaluation system based on a knowledge graph, characterized in that, It includes: a model establishment module configured to collect and establish statistical models of related monitoring points according to the collected monitoring data of the typical dam section to analyze the working behavior of the dam; a simulation calculation module configured to establish a finite element model of the typical dam section and perform simulation calculation on the overall strength, stability, and new and old concrete joint surface state of the typical dam section to obtain finite element simulation results; an index system construction module configured to construct a dam safety evaluation index system according to the collected monitoring data of the typical dam section, the finite element simulation results, and the evaluation related materials of the typical dam section; the evaluation related materials include one or more of geological exploration, preliminary design, safety assessment, and operation inspection; a knowledge graph construction module configured to sort out the hierarchical system of the dam safety comprehensive evaluation index system, and construct a knowledge graph according to the sorting results by knowledge modeling, knowledge extraction, knowledge fusion, and knowledge storage processing; a safety evaluation module configured to use the constructed knowledge graph to perform data retrieval and safety evaluation.

Images