Construction risk management and control knowledge graph construction method and device

Abstract

The invention relates to the technical field of engineering safety, and discloses a construction risk management and control knowledge graph construction method and device, and the method comprises the steps: obtaining a construction risk management and control material; decomposing the construction risk management and control material to obtain a plurality of knowledge modules, and constructing a corresponding ontology for each knowledge module in the plurality of knowledge modules to obtain a plurality of ontologies; a semantic relationship between the different ontologies is constructed, a plurality of construction risk management and control sub-maps are constructed according to the plurality of ontologies and the semantic relationship between the different ontologies, and the plurality of construction risk management and control sub-maps comprise a construction risk management sub-map, a typical violation library sub-map and a specification sub-map; according to the construction risk management and control knowledge graph, the multiple construction risk management and control sub-graphs are fused to form the construction risk management and control knowledge graph, the integrity and consistency of the knowledge graph are further enhanced, association and fusion between bodies are enhanced, and the timeliness and accuracy of dynamic changes of construction risk factors are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering safety, and particularly relates to a method and device for constructing a knowledge graph for construction risk control and management. Background Art

[0002] With the acceleration of the urbanization process and the continuous expansion of infrastructure construction, construction risk control and management has become an important issue. Traditional construction risk control and management methods often rely on manual experience and some simple tools, and it is difficult to cope with complex and changeable construction environments and risk factors. With the development of engineering construction, it has become an urgent problem to predict, monitor, make decisions and conduct emergency management of construction safety risks on a unified platform. In order to improve the efficiency and accuracy of construction risk control and management, knowledge graph technology has been introduced into this field.

[0003] The existing knowledge graphs for construction safety risk control and management often have the following technical problems:

[0004] First, for integrated construction in specific application scenarios, some secondary units are often missing, resulting in incomplete and inconsistent information in the knowledge graph;

[0005] Second, the existing knowledge graph construction methods often only focus on the design and construction of ontologies, ignoring the associations and integrations between ontologies;

[0006] Third, the existing knowledge graph technology lacks effective updates and is difficult to adapt to the dynamic changes of construction risk factors. Summary of the Invention

[0007] This part of the present invention is used to briefly introduce the concepts, which will be described in detail in the following specific implementation part. This part of the present invention is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0008] Some embodiments of the present invention propose a method for constructing a knowledge graph for construction risk control and management to solve one or more of the technical problems mentioned in the above background art part.

[0009] Some embodiments of the present invention provide a method for constructing a knowledge graph for construction risk control and management, including:

[0010] Obtain construction risk control and management materials;

[0011] Decompose the construction risk control and management materials into multiple knowledge modules, and the multiple knowledge modules include a construction object knowledge module, a construction process knowledge module, a construction risk knowledge module, a construction violation event knowledge module, and a construction specification knowledge module;

[0012] Construct corresponding ontologies for each of the multiple knowledge modules to obtain multiple ontologies, where the multiple ontologies include an object ontology corresponding to the construction object knowledge module, a construction ontology corresponding to the construction process knowledge module, a risk ontology corresponding to the construction risk knowledge module, a violation ontology corresponding to the construction violation event knowledge module, and a specification ontology corresponding to the construction specification knowledge module;

[0013] Construct semantic relationships between different ontologies, and based on the multiple ontologies and the semantic relationships between different ontologies, construct multiple construction risk control sub-graphs, where the multiple construction risk control sub-graphs include a construction risk management sub-graph, a typical violation library sub-graph, and a specification sub-graph;

[0014] Fuse multiple construction risk control sub-graphs to form a construction risk control knowledge graph.

[0015] Optionally, constructing a corresponding ontology for each of the multiple knowledge modules includes:

[0016] Encode the construction object knowledge module based on a pre-set multi-level classification system to obtain the object ontology corresponding to the construction object knowledge module.

[0017] Optionally, constructing a corresponding ontology for each of the multiple knowledge modules includes:

[0018] Extract knowledge sources from the pre-collected information of multiple engineering construction projects to obtain prior knowledge information, where the prior knowledge information includes multiple constituent elements in the construction process, and each constituent element in the multiple constituent elements is implemented as a class;

[0019] Define multiple relationships between the multiple constituent elements, where the multiple relationships include inclusion relationships, sequential relationships, and participation relationships;

[0020] Generate the construction ontology corresponding to the construction process knowledge module based on the classes corresponding to the multiple constituent elements and the multiple relationships between the multiple constituent elements.

[0021] Optionally, constructing a corresponding ontology for each of the multiple knowledge modules includes:

[0022] Define multiple classes and multiple object properties included in the risk ontology, where the multiple classes include risk accidents, risk causes, and risk management, and the multiple object properties include the causes of risk accidents and management measures for controlling risk accidents;

[0023] Extract the first number of risk accident types, the second number of accident causes, and the third number of risk management measures as the knowledge sources for constructing the risk ontology;

[0024] Define the first number of risk accident types as instances of risk accidents;

[0025] Define the second quantity of accident causes as instances of risk causes;

[0026] Define the third quantity of risk management measures as instances of risk management.

[0027] Optionally, the construction risk management sub-graph includes an object ontology, a risk ontology, and a construction ontology.

[0028] Optionally, the relationships between risk management measures, risk accident types, and accident causes are created according to a risk rating table.

[0029] Optionally, the entities involved in risk management measures are created through the following steps:

[0030] Using the entities in the object ontology as a user dictionary, and applying the word segmentation algorithm in natural language processing technology to segment the risk management measures, obtaining the entities included in the risk management measures.

[0031] Optionally, fuse multiple construction risk control sub-graphs to form a construction risk control knowledge graph, including:

[0032] Fuse multiple construction risk control sub-graphs through the graph interfaces reserved in each sub-graph to obtain a construction risk control knowledge graph.

[0033] Some embodiments of the present invention propose a construction risk control knowledge graph construction device, including:

[0034] A material acquisition module for acquiring construction risk control materials;

[0035] A decomposition module for decomposing the construction risk control materials into multiple knowledge modules, where the multiple knowledge modules include a construction object knowledge module, a construction process knowledge module, a construction risk knowledge module, a construction violation event knowledge module, and a construction specification knowledge module;

[0036] An ontology construction module for constructing corresponding ontologies for each of the multiple knowledge modules to obtain multiple ontologies, where the multiple ontologies include an object ontology corresponding to the construction object knowledge module, a construction ontology corresponding to the construction process knowledge module, a risk ontology corresponding to the construction risk knowledge module, a violation ontology corresponding to the construction violation event knowledge module, and a specification ontology corresponding to the construction specification knowledge module;

[0037] A semantic relationship construction module for constructing semantic relationships between different ontologies, and constructing multiple construction risk control sub-graphs according to the multiple ontologies and the semantic relationships between different ontologies, where the multiple construction risk control sub-graphs include a construction risk management sub-graph, a typical violation library sub-graph, and a specification sub-graph;

[0038] A fusion module for fusing multiple sub-graphs of construction risk control to form a knowledge graph of construction risk control

[0039] The present invention has the following beneficial effects: By constructing the obtained multiple construction risk control knowledge modules into corresponding multiple ontologies, the fusion mechanism between ontologies further strengthens the integrity and consistency of the knowledge graph; By constructing the semantic relationships between different ontologies, through the semantic relationships between ontologies, sub-graphs of construction risk control are formed, and deep fusion is achieved through the semantic associations between sub-graphs, strengthening the association and fusion between ontologies; Through the graph interfaces reserved for each sub-graph, the knowledge graph has good scalability and real-time update capabilities. When the construction risk factors change dynamically, relevant sub-graphs and even the entire knowledge graph can be adjusted and supplemented through the graph interfaces, thereby enhancing the timeliness and accuracy of the dynamic changes of construction risk factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages and aspects of the embodiments of the present invention will become more obvious. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the elements and elements are not necessarily drawn to scale.

[0041] Figure 1 is a flowchart of the method for constructing the knowledge graph of construction risk control of the present invention;

[0042] Figure 2 is a schematic diagram of the module of the device for constructing the knowledge graph of construction risk control of the present invention;

[0043] Figure 3 is a coding diagram of the power grid project classification system in the method for constructing the knowledge graph of construction risk control of the present invention;

[0044] Figure 4 is a schematic diagram of constructing a risk ontology in the method for constructing the knowledge graph of construction risk control of the present invention;

[0045] Figure 5 is a specific example of the risk cause category in the method for constructing the knowledge graph of construction risk control of the present invention;

[0046] Figure 6 is a specific example of the risk accident category in the method for constructing the knowledge graph of construction risk control of the present invention;

[0047] Figure 7 A specific example of the material category in the method for constructing the knowledge graph of construction risk control of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The present invention will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0049] In addition, it should be noted that for the sake of convenience of description, only the parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0050] It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence relationship of the functions performed by these devices, modules or units.

[0051] It should be noted that the modifications of "one" and "a plurality of" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0052] The names of the messages or information exchanged between multiple devices of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0053] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0054] As Figure 1 shown, a flowchart of some embodiments of a method for constructing a construction risk control knowledge graph according to the present invention is shown, which specifically includes the following steps:

[0055] Step 101, obtaining construction risk control materials;

[0056] In some embodiments, the execution subject of the method for constructing a construction risk control knowledge graph of the present invention can be various electronic devices. On this basis, the execution subject can obtain construction risk control materials from a database or online. The construction risk control materials include, but are not limited to, laws and regulations and standards, national laws and regulations, industry standards, local regulations, etc. related to the implementation of engineering construction.

[0057] Step 102, decomposing the construction risk control materials into a plurality of knowledge modules, and the plurality of knowledge modules include a construction object knowledge module, a construction process knowledge module, a construction risk knowledge module, a construction violation event knowledge module, and a construction specification knowledge module;

[0058] In some embodiments, the obtained construction risk control materials are systematically sorted out and subdivided into several knowledge modules. The sorted knowledge modules can be divided into a construction object knowledge module, a construction process knowledge module, a construction risk knowledge module, a construction violation event knowledge module, and a construction specification knowledge module.

[0059] Step 103, construct corresponding ontologies for each of the multiple knowledge modules to obtain multiple ontologies, where the multiple ontologies include an object ontology corresponding to the construction object knowledge module, a construction ontology corresponding to the construction process knowledge module, a risk ontology corresponding to the construction risk knowledge module, a violation ontology corresponding to the construction violation event knowledge module, and a specification ontology corresponding to the construction specification knowledge module;

[0060] In some embodiments, based on the ontology building tool, the obtained multiple knowledge modules are used to establish corresponding multiple ontologies. The multiple ontologies include an object ontology corresponding to the construction object knowledge module, a construction ontology corresponding to the construction process knowledge module, a risk ontology corresponding to the construction risk knowledge module, a violation ontology corresponding to the construction violation event knowledge module, and a specification ontology corresponding to the construction specification knowledge module.

[0061] Optionally, constructing corresponding ontologies for each of the multiple knowledge modules includes:

[0062] Encoding the construction object knowledge module based on a preset multi-level classification system to obtain the object ontology corresponding to the construction object knowledge module.

[0063] In some embodiments, a reasonable number of classification levels (such as three levels, four levels, five levels, etc.) is set according to the characteristics of the engineering project. In this embodiment, a six-level classification system for power grid projects is taken as an example for illustration. Such as Figure 3As shown, there are a total of 6 groups of digital sequence codes for the encoding. Each group of digital sequence codes takes values from 00 to 99, where 00 represents default. Classification labels are set for each level of classification layer. As an example, the first-level classification layer is "substation project", and its encoding is "01.00.00.00.00.00". The second-level classification layer is further refined based on the first-level classification layer. As an example, under the first-level classification layer "construction project", there is "substation project civil engineering", and its encoding is "01.02.00.00.00.00". The subsequent classification layers follow this pattern until it is stratified to the most refined construction object category. As an example, under the second-level classification layer "substation project civil engineering", there is a third-level classification layer "architecture and structure", and its encoding is "01.02.01.00.00.00". Under the third-level classification layer "architecture and structure", there is a fourth-level classification layer "building", and its encoding is "01.02.01.01.00.00". Under the fourth-level classification layer "building", there is a fifth-level classification layer "accident oil pool", "shear wall", "sound-absorbing structure", "basement wall", "diaphragm wall", "ground ring beam", "foundation slab", "foundation pit structure", "foundation" and "wall". Among them, under the fifth-level classification layer "foundation wall", there are sixth-level classification layers "foundation wall" and "foundation beam". For the fifth-level classification layer "wall", its encoding is "01.02.01.01.01.00". Under the fifth-level classification layer "wall", there is a sixth-level classification layer "retaining wall", and its encoding is "01.02.01.01.01.01". For each knowledge module of the constructed object after encoding, a corresponding object ontology is built.

[0064] Optionally, for each of multiple knowledge modules, constructing a corresponding ontology includes the following steps:

[0065] Step 1, extracting knowledge sources from the pre-collected information of multiple engineering construction projects to obtain prior knowledge information. The prior knowledge information includes multiple constituent elements in the construction process, and each constituent element in the multiple constituent elements is implemented as a class.

[0066] In some embodiments, the multiple constituent elements include construction projects, construction processes, construction operations, participants, machinery, and materials. One class implemented by each constituent element includes a construction project class, a construction process class, a construction operation class, a participant class, a machinery class, and a materials class. As an example, the construction project class contains the name, location, scale, design drawings, construction period, etc. of the project. As an example, the construction process class represents the stages or steps arranged in a logical order in an engineering project. As an example, the construction operation class contains the specific operations during the construction process, such as earth excavation, steel reinforcement work, formwork work, concrete pouring structure work, etc. As an example, the participant class can be subdivided into organizations, personnel, and job types. As an example, the participant class is further subdivided into organizations, personnel, and job types, covering various enterprise and individual roles participating in the engineering project. As an example, the machinery class is divided into two categories: measuring instruments and construction tools. As an example, the materials class contains different types of materials during the construction process, such as Figure 7 As shown, the materials include crushed stones, fiberglass meshes, sands, blocks, concrete admixtures, mortars, channel steels, formworks, water, cement, concrete glass fibers, bricks, concrete, backfill soil, wooden beams, and concrete curing agents.

[0067] Step 2, define multiple relationships between the multiple constituent elements. The multiple relationships include containment relationships, sequential relationships, and participation relationships;

[0068] In some embodiments, as an example, "construction project" contains "construction process", "construction process" contains "construction operations", "construction tasks", and "construction operations" contain "materials"

[0069] . The sequential relationship represents the sequential execution order between operations. As an example, the next operation after "formwork installation" is "formwork removal". The participation relationship represents the participants in the construction operations and construction tasks. As an example, "construction organization" participates in "construction tasks".

[0070] Step 3, generate a construction ontology corresponding to the construction process knowledge module based on the classes corresponding to the multiple constituent elements and the multiple relationships between the multiple constituent elements.

[0071] In some embodiments, according to the classes corresponding to each element and the multiple relationships between each element, instantiate the above-mentioned classes according to the actual engineering project, and define the multiple relationships between the classes to generate a construction ontology corresponding to the construction process knowledge module.

[0072] Optionally, for each knowledge module in the multiple knowledge modules, constructing a corresponding ontology includes the following steps:

[0073] Step 1: Define multiple classes and multiple object properties included in the risk ontology. The multiple classes include risk accident class, risk cause class, and risk management class. The multiple object properties include the cause of risk accidents and management measures for controlling risk accidents.

[0074] In some embodiments, as Figure 4 shown, define the risk ontology framework corresponding to the construction risk knowledge module. As an example, the risk ontology includes three categories: risk accident class, risk cause class, and risk management class. As Figure 5 shown, the risk cause class includes climate change, environmental change, equipment anomaly, equipment maintenance, cross-operation, personnel anomaly, near-electrical operation, and geological anomaly. As Figure 6 shown, the risk accident class includes fire, power grid accident, collapse, explosion, equipment accident, toppling, drowning, traffic accident, mechanical injury, risk accident class, landslide, fall from height, other injuries, suspension of electrified railway operation, poisoning, plane crash, epidemic spread, interruption of highway traffic, asphyxiation, scald, lifting injury, and object striking. At the same time, define the object property. The cause of risk accidents is used to link the risk accident class with the corresponding risk cause class. The management measures for controlling risk accidents are used to associate the risk management class with risk accidents.

[0075] Step 2: Extract the first quantity of risk accident types, the second quantity of accident causes, and the third quantity of risk management measures and use them as the knowledge sources for constructing the risk ontology.

[0076] Step 3: Define the first quantity of risk accident types as instances of risk accidents.

[0077] Step 4: Define the second quantity of accident causes as instances of risk causes.

[0078] Step 5: Define the third quantity of risk management measures as instances of risk management.

[0079] In some embodiments, collect and organize the data of the construction risk knowledge module, including the first quantity of risk accident type instances corresponding to the risk accident class. The second quantity of risk accident cause instances corresponding to the risk cause class. The third quantity of specific risk management measure instances corresponding to the risk management class.

[0080] Step 104: Construct the semantic relationships between different ontologies. According to multiple ontologies and the semantic relationships between different ontologies, construct multiple construction risk control sub-graphs, and the multiple construction risk control sub-graphs include construction risk management sub-graph, typical violation library sub-graph, and specification sub-graph.

[0081] In some embodiments, the above-mentioned execution entity may receive information input by professionals to establish semantic relationships between ontologies, where the semantic relationships represent the logical relationships between different ontologies. As an example, a causal relationship is established between a certain process (such as scaffolding erection) in the construction ontology corresponding to the construction process knowledge module and the risk ontology (such as the risk of falling from height) corresponding to the construction risk knowledge module. At the same time, the process is also restricted by the specific regulations on the erection and use of scaffolding in the specification ontology corresponding to the construction specification knowledge module.

[0082] On this basis, based on the semantic relationships established between multiple ontologies, three main sub-graphs for construction risk control are constructed: the construction risk management sub-graph, the typical violation library sub-graph, and the specification sub-graph. Among them, the construction risk management sub-graph is used to record and display the types of risk accidents, the causes of risk accidents, construction processes, and corresponding risk management measures in each link of the entire construction process. The typical violation library sub-graph collects various violation behaviors that may occur during construction, the reasons for violations, the possible consequences, and corresponding rectification and prevention measures. The specification sub-graph is based on existing construction specifications and relevant industry specifications, and clarifies the application and association relationships of the specifications in specific construction activities.

[0083] Optionally, the construction risk management sub-graph includes an object ontology, a risk ontology, and a construction ontology.

[0084] In some embodiments, the construction risk management sub-graph includes three components, namely an object ontology, a risk ontology, and a construction ontology.

[0085] Optionally, the relationship between the risk management measures and the types of risk accidents and accident causes is created according to the risk rating table.

[0086] In some embodiments, referring to the risk rating table, clarify the types of risk accidents that each construction process may encounter and the accident causes that trigger the accidents, and match corresponding risk management measures for each construction process. Secondly, the parent process corresponding to the sub-process inherits the types of risk accidents, accident causes, and corresponding risk management measures of the sub-process. As an example, in the case where "steel support installation" is a sub-process of "steel support construction", for the types of risk accidents, accident causes, and corresponding risk management measures not recorded in the risk rating table of "steel support construction", the types of risk accidents, accident causes, and corresponding risk management measures of "steel support construction" can be inferred based on the sub-process "steel support installation" of "steel support construction".

[0087] Optionally, the entities involved in the risk management measures are created through the following steps:

[0088] Using the entities in the object ontology as the user dictionary, and applying the word segmentation algorithm in natural language processing technology to segment the risk management measures, the entities included in the risk management measures are obtained.

[0089] In some embodiments, by selecting the entities defined in the object ontology as the user dictionary, the risk management measures are segmented using the word segmentation algorithm in natural language processing technology to obtain the entities included in the risk management measures.

[0090] Step 105: Integrate multiple construction risk control sub-graphs to form a construction risk control knowledge graph.

[0091] In some embodiments, the multiple established construction risk control sub-graphs are integrated to form a construction risk control knowledge graph.

[0092] Optionally, integrating multiple construction risk control sub-graphs to form a construction risk control knowledge graph includes:

[0093] Integrate multiple construction risk control sub-graphs through the graph interfaces reserved in each sub-graph to obtain a construction risk control knowledge graph.

[0094] In some embodiments, corresponding sub-graphs are created for construction risk management, typical violation libraries, and specifications. The construction risk control sub-graphs contain entities of construction risk management, typical violation libraries, and specifications. By pre-designing a common graph interface in each sub-graph, the construction risk control sub-graphs can be conveniently integrated into one, and finally a construction risk control knowledge graph is formed.

[0095] Further referring to Figure 2 , as an implementation of the methods shown in the above figures, the present invention provides some embodiments of a construction risk control knowledge graph construction device. These device embodiments correspond to Figure 1 the method embodiments shown, and the device can be specifically applied to various electronic devices.

[0096] As Figure 2 shown, some construction risk control knowledge graph construction devices in this embodiment include: a material acquisition module 201, a decomposition module 202, an ontology construction module 203, a semantic relationship construction module 204, and a fusion module 205. Among them, the material acquisition module 201 is used to acquire construction risk control materials;

[0097] The decomposition module 202 is used to decompose the construction risk control materials into multiple knowledge modules, and the multiple knowledge modules include a construction object knowledge module, a construction process knowledge module, a construction risk knowledge module, a construction violation event knowledge module, and a construction specification knowledge module;

[0098] The ontology construction module 203 is used to construct a corresponding ontology for each of the multiple knowledge modules, obtaining multiple ontologies. The multiple ontologies include the object ontology corresponding to the construction object knowledge module, the construction ontology corresponding to the construction process knowledge module, the risk ontology corresponding to the construction risk knowledge module, the violation ontology corresponding to the construction violation event knowledge module, and the specification ontology corresponding to the construction specification knowledge module;

[0099] The semantic relationship construction module 204 is used to construct the semantic relationships between different ontologies. Based on the multiple ontologies and the semantic relationships between different ontologies, multiple construction risk control sub-graphs are constructed. The multiple construction risk control sub-graphs include the construction risk management sub-graph, the typical violation library sub-graph, and the specification sub-graph;

[0100] The fusion module 205 is used to fuse the multiple construction risk control sub-graphs to form a construction risk control knowledge graph.

[0101] It can be understood that the various units described in the construction risk control knowledge graph construction device correspond to the respective steps in the method described in the reference Figure 1 description. Therefore, the operations, features, and beneficial effects described above for the method also apply to the device and the units contained therein, and will not be elaborated here.

[0102] The above description is only some preferred embodiments of the present invention and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, but also covers other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features having similar functions disclosed in the present invention.

Claims

1. A method for constructing a construction risk management knowledge graph, characterized in that: include: Obtain construction risk management materials; Decomposing the construction risk control material to obtain a plurality of knowledge modules, wherein the plurality of knowledge modules include a construction object knowledge module, a construction process knowledge module, a construction risk knowledge module, a construction violation event knowledge module, and a construction specification knowledge module; Constructing a corresponding ontology for each of the multiple knowledge modules to obtain multiple ontologies, wherein the multiple ontologies include an object ontology corresponding to the construction object knowledge module, a construction ontology corresponding to the construction process knowledge module, a risk ontology corresponding to the construction risk knowledge module, a violation ontology corresponding to the construction violation event knowledge module, and a specification ontology corresponding to the construction specification knowledge module; Constructing semantic relationships between different ontologies, and constructing multiple construction risk control sub-graphs according to the multiple ontologies and the semantic relationships between the different ontologies, wherein the multiple construction risk control sub-graphs include a construction risk management sub-graph, a typical violation library sub-graph, and a specification sub-graph; The multiple construction risk management sub-graphs are integrated to form a construction risk management knowledge graph.

2. The method for constructing a construction risk management knowledge graph according to claim 1, characterized in that: The constructing a corresponding ontology for each of the plurality of knowledge modules comprises: The construction object knowledge module is encoded based on a pre-set multi-level classification system to obtain an object ontology corresponding to the construction object knowledge module.

3. The method for constructing a construction risk management knowledge graph according to claim 1, characterized in that: The constructing a corresponding ontology for each of the plurality of knowledge modules comprises: Extracting knowledge sources from pre-collected information on multiple engineering construction projects to obtain prior knowledge information, wherein the prior knowledge information includes multiple components in the construction process, and each of the multiple components is implemented as a class; defining a plurality of relationships between the plurality of components, the plurality of relationships comprising relationship, sequence relationship and participation relationship; Based on the classes corresponding to the multiple components and the multiple relationships between the multiple components, a construction ontology corresponding to the construction process knowledge module is generated.

4. The method for constructing a construction risk management knowledge graph according to claim 1, characterized in that: The constructing a corresponding ontology for each of the plurality of knowledge modules comprises: Defining a plurality of classes and a plurality of object attributes included in the risk ontology, wherein the plurality of classes include a risk accident class, a risk cause class, and a risk management class, and the plurality of object attributes include the cause of the risk accident and the management measures for controlling the risk accident; Extracting a first number of risk accident types, a second number of accident causes, and a third number of risk management measures as knowledge sources for constructing the risk ontology; The first number of risk accident types are defined as instances of the risk accident; the second number of accident causes are defined as instances of the risk cause; and the third number of risk management measures are defined as instances of the risk management.

5. The method for constructing a construction risk management knowledge graph according to claim 1, characterized in that: The construction risk management sub-graph includes the object ontology, the risk ontology and the construction ontology.

6. The method for constructing a construction risk management knowledge graph according to claim 4, characterized in that: The relationship between the risk management measures and the risk accident types and accident causes is created based on the risk level table.

7. The method for constructing a construction risk management knowledge graph according to claim 4, characterized in that: The entities involved in the risk management measures are created through the following steps: The entities in the object ontology are used as user dictionaries, and the word segmentation algorithm in the natural language processing technology is used to segment the risk management measures to obtain the entities included in the risk management measures.

8. The method for constructing a construction risk management knowledge graph according to claim 4, characterized in that: The fusion of the multiple construction risk management sub-graphs to form a construction risk management knowledge graph includes: The multiple construction risk management sub-graphs are integrated through the graph interface reserved in each sub-graph to obtain a construction risk management knowledge graph.

9. A device for constructing a construction risk management knowledge graph, characterized in that: include: Material acquisition module, used to obtain construction risk management materials; A decomposition module, used for decomposing the construction risk management materials into a plurality of knowledge modules, wherein the plurality of knowledge modules include a construction object knowledge module, a construction process knowledge module, a construction risk knowledge module, a construction violation event knowledge module and a construction specification knowledge module; An ontology construction module is used to construct a corresponding ontology for each of the multiple knowledge modules to obtain multiple ontologies, wherein the multiple ontologies include an object ontology corresponding to the construction object knowledge module, a construction ontology corresponding to the construction process knowledge module, a risk ontology corresponding to the construction risk knowledge module, a violation ontology corresponding to the construction violation event knowledge module, and a specification ontology corresponding to the construction specification knowledge module; A semantic relationship construction module, used to construct semantic relationships between different ontologies, and to construct multiple construction risk control sub-graphs according to the multiple ontologies and the semantic relationships between the different ontologies, wherein the multiple construction risk control sub-graphs include a construction risk management sub-graph, a typical violation library sub-graph, and a specification sub-graph; The fusion module is used to fuse the multiple construction risk management sub-graphs to form a construction risk management knowledge graph.

Images