Knowledge graph-based earthquake-damaged building structure reinforcement scheme recommendation method and device

Abstract

The application relates to a knowledge graph-based seismic-damaged building structure reinforcement scheme recommendation method and equipment, and belongs to the technical field of computers.The method comprises the following steps: based on a knowledge graph, rule reasoning and similar case matching are performed to determine a first reinforcement technology node candidate set corresponding to a damage mode node of a building structure subjected to seismic damage; constraint condition nodes triggered by the damage mode node are determined according to engineering input parameters, the first reinforcement technology node candidate set is filtered according to the constraint condition nodes, and a second reinforcement technology node candidate set is obtained; the applicability score of each reinforcement technology node in the second reinforcement technology node candidate set is calculated for the reinforcement of the building structure; the second reinforcement technology node candidate set is sorted according to the applicability score, at least one second reinforcement technology node ranking at the front is determined according to the sorting result, and a reinforcement scheme recommendation list is further determined.The application realizes intelligent reinforcement scheme recommendation based on a knowledge graph.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method and device for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs. Background Technology

[0002] Earthquake-damaged building structures include masonry structures, which encompass brick masonry, stone masonry, and mixed masonry, and constitute a significant portion of the existing urban and rural building stock in my country. Masonry structures have relatively low seismic resistance. Therefore, the ability to quickly assess the damage to masonry structures and select appropriate reinforcement techniques after an earthquake directly impacts post-disaster recovery efficiency and resident safety.

[0003] Masonry reinforcement methods have established written standards in China. For example, the "Technical Specification for Reinforcement of Masonry Structures" (GB50702-2011) specifies the applicable conditions for construction techniques such as reinforced concrete mortar surface layer reinforcement, fiber-reinforced polymer (FRP) reinforcement, replacement brickwork, and encasing reinforcement. Existing reinforcement technology databases or management systems mainly fall into the following two categories:

[0004] Static technical manual type database: Various standard design atlases (such as "15G611") store reinforcement technologies in the form of charts and text descriptions, but the information is static. Engineers need to manually select technologies by comparing them with the site conditions, which is inefficient and relies on experience.

[0005] Building Information Modeling (BIM)-driven reinforcement information management: Reinforcement-related data in BIM platforms mostly rely on manual input and annotation, and lack intelligent recommendation capabilities. Summary of the Invention

[0006] Based on the above analysis, the present invention aims to provide a method and device for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs, in order to solve the problem that existing technologies cannot achieve intelligent recommendation of reinforcement schemes based on earthquake damage conditions.

[0007] On one hand, this invention provides a method for recommending reinforcement schemes for earthquake-damaged building structures based on a knowledge graph. The knowledge graph includes building structure nodes, damage mode nodes, reinforcement technology nodes, constraint condition nodes, historical case nodes, and effect evaluation nodes. The method includes: constructing corresponding building structure nodes and damage mode nodes based on the earthquake-damaged building structure and damage mode of the reinforcement scheme to be determined; determining a first set of candidate reinforcement technology nodes corresponding to the damage mode nodes of the building structure based on rule reasoning and similar case matching using the knowledge graph; determining the constraint condition nodes triggered by the damage mode nodes of the building structure according to engineering input parameters; filtering the first set of candidate reinforcement technology nodes according to the constraint condition nodes to obtain a second set of candidate reinforcement technology nodes; calculating the applicability score of the reinforcement technology corresponding to each reinforcement technology node in the second set of candidate reinforcement technology nodes for the reinforcement of the building structure; sorting the second set of candidate reinforcement technology nodes according to the applicability score; determining at least one second reinforcement technology node with the highest ranking according to the sorting result; and determining a recommended list of reinforcement schemes based on the at least one second reinforcement technology node.

[0008] Based on further improvements to the above method, in the knowledge graph, the building structure node points to the damage pattern node, indicating that the building structure corresponding to the building structure node has the damage pattern corresponding to the damage pattern node; the damage pattern node points to the historical case node, indicating that the damage pattern corresponding to the damage pattern node is similar to the damage pattern in the historical case corresponding to the historical case node; the reinforcement technology node points to the constraint condition node, indicating that the reinforcement technology corresponding to the reinforcement technology node satisfies the constraint condition corresponding to the constraint condition node; the damage pattern node points to the constraint condition node, indicating that the damage pattern corresponding to the damage pattern node triggers the constraint condition corresponding to the constraint condition node; the damage pattern node points to the reinforcement technology node, indicating that the reinforcement technology corresponding to the reinforcement technology node is a recommended solution for the damage pattern of the damage pattern node; the historical case node points to the reinforcement technology node. The following statements indicate that the historical case node corresponds to a historical case that uses the reinforcement technology corresponding to the reinforcement technology node; the constraint condition node pointing to the historical case node indicates that the constraint condition corresponding to the constraint condition node is effective in the historical case corresponding to the historical case node; the constraint condition node pointing to the reinforcement technology node indicates that the constraint condition corresponding to the constraint condition node and the reinforcement technology corresponding to the reinforcement technology node have a constraint relationship; the historical case node pointing to the effect evaluation node indicates that the historical case corresponding to the historical case node has the reinforcement effect corresponding to the effect evaluation node; the reinforcement technology node pointing to the historical case node indicates that the reinforcement technology corresponding to the reinforcement technology node is applied in the historical case corresponding to the historical case node; the historical case node pointing to the damage mode node indicates that the damage mode of the historical case corresponding to the historical case node is consistent with the damage mode corresponding to the damage mode node.

[0009] A further improvement to the above method, the step of determining the first set of candidate reinforcement technology nodes corresponding to the damage pattern nodes of the building structure based on the knowledge graph through rule reasoning and similar case matching, includes: determining the historical case nodes pointed to by the damage pattern nodes of the building structure, and determining the similarity between the damage patterns corresponding to the damage pattern nodes and the damage patterns of the historical cases corresponding to the historical case nodes; performing rule reasoning using a rule engine based on at least one of the damage patterns corresponding to the damage pattern nodes in the knowledge graph, the constraint conditions corresponding to the constraint condition nodes triggered by the damage pattern nodes of the building structure, and the similarity, to obtain a third set of candidate reinforcement technology nodes; determining the historical case nodes pointed to by the damage pattern nodes of the building structure, determining the effect evaluation nodes pointed to by the historical case nodes, and obtaining a fourth set of candidate reinforcement technology nodes based on the historical case nodes corresponding to the effect evaluation nodes whose reinforcement effects meet preset requirements; and obtaining the first set of candidate reinforcement technology nodes based on the third and fourth set of candidate reinforcement technology nodes.

[0010] Based on a further improvement of the above method, the calculation of the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure includes: determining at least one historical case node pointed to by the reinforcement technology node; determining a historical case similarity score based on the similarity between the damage mode of the historical case corresponding to the at least one historical case node and the damage mode corresponding to the damage mode node of the building structure; determining a constraint compliance score based on the situation where the reinforcement technology node satisfies the constraint condition node triggered by the damage mode node of the building structure; determining an effect evaluation node pointed to by at least one historical case node respectively; determining a historical reinforcement effect evaluation score based on the reinforcement effect corresponding to each effect evaluation node; determining an expert experience score corresponding to the reinforcement technology node; determining a preset additional factor score corresponding to the reinforcement technology node; and performing a weighted summation based on the historical case similarity score, the constraint compliance score, the historical reinforcement effect evaluation score, the expert experience score, and the preset additional factor score to obtain the applicability score.

[0011] Based on a further improvement of the above method, determining the historical case similarity score based on the similarity between the damage mode of the historical case corresponding to at least one of the historical case nodes and the damage mode corresponding to the damage mode node of the building structure includes: determining the historical case similarity score by calculating the average of the similarities between the damage mode of the historical case corresponding to at least one of the historical case nodes and the damage mode corresponding to the damage mode node of the building structure; and / or, determining the historical reinforcement effect evaluation score based on the reinforcement effect corresponding to each of the effect evaluation nodes includes: determining the effect score of each effect evaluation node based on the reinforcement effect corresponding to each of the effect evaluation nodes, and obtaining the historical reinforcement effect evaluation score by calculating the average of the effect scores of each of the effect evaluation nodes.

[0012] Based on a further improvement of the above method, after determining the recommended list of reinforcement schemes according to the at least one second reinforcement technology node, the method further includes: obtaining the node relationship between the constraint condition node and the historical case node in the knowledge graph; determining the effect evaluation node pointed to by the historical case node; determining the success rate of the historical case node pointed to by the constraint condition node when a certain constraint condition occurs based on the reinforcement effect in the effect evaluation node; and adjusting the contribution of the constraint condition in calculating the constraint compliance score based on the success rate.

[0013] Based on a further improvement of the above method, after calculating the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure, and before sorting the second set of candidate reinforcement technology nodes according to the applicability score, the method further includes: inputting the damage mode features of the damage mode node of the building structure into a pre-trained reinforcement scheme recommendation model, and improving the applicability score of the corresponding reinforcement technology node according to the output reinforcement scheme.

[0014] Based on a further improvement of the above method, after determining the reinforcement scheme recommendation list according to the at least one second reinforcement technology node, the method further includes: pushing the reinforcement scheme recommendation list to the user interaction interface; wherein, the reinforcement scheme recommendation list includes reinforcement schemes corresponding to the at least one second reinforcement technology node respectively; and receiving the user's final confirmation of the reinforcement scheme based on the reinforcement scheme recommendation list.

[0015] Based on a further improvement of the above method, after receiving the final confirmed reinforcement scheme from the user based on the reinforcement scheme recommendation list, the method further includes: generating historical case nodes and effect evaluation nodes of the building structure according to the reinforcement effect of the final confirmed reinforcement scheme, and updating the knowledge graph based on the relevant nodes of the building structure.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs as described above.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs as described above.

[0018] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures as described above.

[0019] The present invention provides a method and device for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs. This method determines a first set of candidate reinforcement technology nodes corresponding to damage pattern nodes through rule-based reasoning and similar case matching based on knowledge graphs. It then filters the first set of candidate reinforcement technology nodes according to constraint condition nodes to obtain a second set of candidate reinforcement technology nodes. Finally, it sorts the second set of candidate reinforcement technology nodes based on applicability scores and determines a recommended list of reinforcement schemes based on the sorting results, thus achieving intelligent recommendation of reinforcement schemes based on knowledge graphs.

[0020] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings.

[0021] The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs provided by this invention has the following technical effects:

[0022] 1. Data structuring and semantic representation improve the standardization of information.

[0023] This invention achieves unified parsing by converting raw information into structured fields, such as numerical attributes like crack width, length, and material type, and storing them as knowledge graph nodes.

[0024] After structured transformation, the data can be directly imported into the reasoning rule base for analysis, reducing the need for manual data re-entry compared to traditional manual recording.

[0025] 2. Integration of assessment and reinforcement technology databases enables second-level solution generation.

[0026] This invention automatically performs rule matching and case retrieval after data parsing, enabling direct association of reinforcement technology nodes with damage mode nodes. Attached Figure Description

[0027] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0028] Figure 1 This is a flowchart illustrating the method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs provided by the present invention.

[0029] Figure 2 This is a schematic diagram of the knowledge graph in the method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs provided by the present invention;

[0030] Figure 3 This is a schematic diagram of the structure of the knowledge graph-based reinforcement scheme recommendation device for earthquake-damaged building structures provided by the present invention;

[0031] Figure 4 A schematic diagram of the physical structure of an electronic device is provided.

[0032] Figure label:

[0033] 410 - Processor; 420 - Communication interface; 430 - Memory; 440 - Communication bus. Detailed Implementation

[0034] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0035] Figure 1 This is a flowchart illustrating the method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs provided by this invention. The knowledge graph includes building structure nodes, damage mode nodes, reinforcement technology nodes, constraint condition nodes, historical case nodes, and effect evaluation nodes, such as... Figure 1 As shown, the method includes:

[0036] Step S1: Based on the earthquake-damaged building structure and damage mode of the reinforcement scheme to be determined, construct the corresponding building structure nodes and damage mode nodes.

[0037] The knowledge graph-based device for recommending reinforcement schemes for earthquake-damaged building structures collects on-site data including images, videos, quantitative measurements, and text records. The collected data undergoes data cleaning and structure transformation to construct and update the knowledge graph, and then performs rule-based reasoning and reinforcement technology recommendations. A reinforcement technology database stores detailed processes, material parameters, and applicable conditions for reinforcement schemes. Through a user-interactive terminal, it provides users with evaluation results, recommended schemes, and selection tools.

[0038] After a building structure suffers earthquake damage, a reinforcement scheme needs to be determined. Based on the damaged building structure and damage mode for which a reinforcement scheme needs to be determined, corresponding building structure nodes and damage mode nodes are constructed, and the pointing relationship between the building structure nodes and damage mode nodes is established.

[0039] When collecting data on earthquake damage, data acquisition terminals can be used to collect damage data of building structures after the earthquake, including: (1) crack location and width; (2) wall tilt; (3) material type, including brick, stone, and mixed masonry; and (4) environmental constraints, including site dimensions and construction accessibility. Implementation method: Mobile devices or fixed-point acquisition devices are used, and the data is transmitted to a local server via a wireless network. Among them, crack location and width can be identified using digital camera + ruler image recognition; wall tilt can be collected based on accelerometer or laser ranging.

[0040] When converting the collected unstructured information into structured data, the methods include: (1) extracting the crack length, width, and orientation through image segmentation and feature extraction algorithms; (2) converting the engineer's on-site records into standard damage descriptions through text parsing.

[0041] The nodes and relationships of a knowledge graph can be stored using RDF / OWL format or a graph database.

[0042] Table 1 shows an example of the fields for building structure nodes.

[0043] Table 1

[0044]

[0045] Table 2 shows examples of fields for the damage mode node.

[0046] Table 2

[0047]

[0048] Table 3 shows examples of fields for the reinforcement technology nodes.

[0049] Table 3

[0050]

[0051] Table 4 shows an example of the fields in the constraint node.

[0052] Table 4

[0053]

[0054] The constraint nodes in Table 4 belong to the "Project Implementation and Resource Constraints" category. They can also include other constraint types such as those listed below:

[0055] Performance Constraints:

[0056] Examples of fields: Target_Strength_Increase (target strength increase rate), Ductility_Requirement (ductility requirement), Life_Expectancy (design service life).

[0057] Key focus: The technical specifications that must be met after reinforcement.

[0058] Heritage / Aesthetic Constraints:

[0059] Examples of fields: Facade_Alteration_Allowed (whether facade alteration is allowed), Reversibility (whether reversible construction is required), Color_Match (color matching degree).

[0060] Key point: For historical buildings, there is a requirement to maintain the original appearance.

[0061] Environmental and Safety Constraints:

[0062] Examples of fields: Noise_Limit_dB (maximum decibel level for construction noise), Dust_Control_Level (dust control level), Toxic_Material_Ban (ban on toxic materials).

[0063] Key concerns: Environmental interference restrictions during construction in hospitals, schools, or residential areas.

[0064] Table 5 shows examples of fields for historical case nodes.

[0065] Table 5

[0066]

[0067] Table 6 shows an example of the fields for the effect evaluation node.

[0068] Table 6

[0069]

[0070] Step S2: Based on the knowledge graph, perform rule reasoning and similar case matching to determine the first set of reinforcement technology nodes corresponding to the damage mode nodes of the building structure.

[0071] Based on knowledge graphs, rule-based reasoning and similar case matching are used to determine the first set of candidate reinforcement technology nodes corresponding to the damage mode nodes of earthquake-damaged building structures. In determining the first set of candidate reinforcement technology nodes, not only rule matching but also the reference value of similar cases are considered.

[0072] Step S3: Determine the constraint condition node triggered by the damage mode node of the building structure according to the engineering input parameters, and filter the first reinforcement technology node candidate set according to the constraint condition node to obtain the second reinforcement technology node candidate set.

[0073] The project input parameters can be hard specifications that users enter into the system interface, specific to the current renovation project. These may include: budget limits, completion time requirements, equipment requirements, etc. For example, if the project budget limit is 500,000 yuan, it must be completed within 30 days, and due to the narrow site and lack of access roads for large machinery, large reinforcement equipment cannot be used.

[0074] The engineering input parameters define the constraints on strengthening the earthquake-damaged building structure for which a specific strengthening scheme is to be determined. Corresponding constraint condition nodes are determined based on these parameters. These constraint condition nodes, triggered by the damage mode nodes of the building structure, represent the constraints that must be satisfied when strengthening the damaged structure. The first set of candidate strengthening technology nodes is filtered based on these constraint condition nodes triggered by the damage mode nodes. Strengthening technologies that satisfy the constraints of all constraint condition nodes triggered by the damage mode nodes are retained, forming the second set of candidate strengthening technology nodes.

[0075] Step S4: For each reinforcement technology node in the second reinforcement technology node candidate set, calculate the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure.

[0076] After obtaining the second set of candidate reinforcement technology nodes, for each reinforcement technology node in the candidate set, the suitability score of the corresponding reinforcement technology for strengthening the building structure is calculated. The rules for the suitability score can be preset. The higher the suitability score, the more suitable the reinforcement technology corresponding to the reinforcement technology node is for strengthening the building structure that has suffered earthquake damage.

[0077] Step S5: Sort the candidate set of the second reinforcement technology nodes according to the applicability score, determine at least one second reinforcement technology node with the highest ranking according to the sorting result, and determine a recommended list of reinforcement schemes according to the at least one second reinforcement technology node.

[0078] The candidate set of second reinforcement technology nodes is ranked according to their applicability scores. Based on the ranking, at least one top-ranked second reinforcement technology node is determined. This top-ranked node can be a predetermined number of nodes or a predetermined proportion of nodes. A recommended list of reinforcement schemes is then formed based on these top-ranked nodes. Specifically, the reinforcement technologies for the top-ranked nodes are identified, and the recommended list of reinforcement schemes is created based on these technologies. In this recommended list, the higher the applicability score of the second reinforcement technology node for strengthening the building structure, the higher its ranking can be.

[0079] The present invention provides a knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures. This method determines a first set of candidate reinforcement technology nodes corresponding to damage pattern nodes by performing rule-based reasoning and similar case matching based on the knowledge graph. The first set of candidate reinforcement technology nodes is then filtered according to constraint condition nodes to obtain a second set of candidate reinforcement technology nodes. The second set of candidate reinforcement technology nodes is then ranked according to applicability scores. Finally, a recommended list of reinforcement schemes is determined based on the ranking results, thus realizing intelligent recommendation of reinforcement schemes based on knowledge graphs.

[0080] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on a knowledge graph is provided. In the knowledge graph, the building structure node points to the damage mode node, indicating that the building structure corresponding to the building structure node has the damage mode corresponding to the damage mode node; the damage mode node points to the historical case node, indicating that the damage mode corresponding to the damage mode node is similar to the damage mode in the historical case corresponding to the historical case node; the reinforcement technology node points to the constraint condition node, indicating that the reinforcement technology corresponding to the reinforcement technology node satisfies the constraint condition corresponding to the constraint condition node; the damage mode node points to the constraint condition node, indicating that the damage mode corresponding to the damage mode node triggers the constraint condition corresponding to the constraint condition node; the damage mode node points to the reinforcement technology node, indicating that the reinforcement technology corresponding to the reinforcement technology node is a recommended scheme for the damage mode of the damage mode node; the historical case node points to... The reinforcement technology node indicates that the historical case corresponding to the historical case node used the reinforcement technology corresponding to the reinforcement technology node; the constraint condition node pointing to the historical case node indicates that the constraint condition corresponding to the constraint condition node is effective in the historical case corresponding to the historical case node; the constraint condition node pointing to the reinforcement technology node indicates that the constraint condition corresponding to the constraint condition node has a constraint relationship with the reinforcement technology corresponding to the reinforcement technology node; the historical case node pointing to the effect evaluation node indicates that the historical case corresponding to the historical case node has the reinforcement effect corresponding to the effect evaluation node; the reinforcement technology node pointing to the historical case node indicates that the reinforcement technology corresponding to the reinforcement technology node is applied in the historical case corresponding to the historical case node; the historical case node pointing to the damage mode node indicates that the damage mode of the historical case corresponding to the historical case node is consistent with the damage mode corresponding to the damage mode node.

[0081] Figure 2 This is a schematic diagram of the knowledge graph in the method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs provided by this invention. For example... Figure 2 As shown, the knowledge graph includes building structure nodes, damage mode nodes, reinforcement technology nodes, constraint condition nodes, historical case nodes, and effect evaluation nodes, and the pointers between nodes have defined node relationships.

[0082] Table 7 is a table defining the node relationship types in the knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provided by this invention.

[0083] Table 7

[0084]

[0085] The knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provides convenience for recommending reinforcement schemes by defining the relationship types between nodes.

[0086] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs includes determining a first set of candidate reinforcement technology nodes corresponding to the damage pattern nodes of the building structure by performing rule reasoning and similar case matching based on the knowledge graph. This includes: determining historical case nodes pointed to by the damage pattern nodes of the building structure, and determining the similarity between the damage patterns corresponding to the damage pattern nodes and the damage patterns of the historical cases corresponding to the historical case nodes; performing rule reasoning using a rule engine based on at least one of the damage patterns corresponding to the damage pattern nodes in the knowledge graph, the constraint conditions corresponding to the constraint condition nodes triggered by the damage pattern nodes of the building structure, and the similarity, to obtain a third set of candidate reinforcement technology nodes; determining the historical case nodes pointed to by the damage pattern nodes of the building structure, determining the effect evaluation nodes pointed to by the historical case nodes, and obtaining a fourth set of candidate reinforcement technology nodes based on the historical case nodes corresponding to the effect evaluation nodes whose reinforcement effects meet preset requirements; and obtaining a first set of candidate reinforcement technology nodes based on the third and fourth set of candidate reinforcement technology nodes.

[0087] When determining the first set of candidate reinforcement technology nodes corresponding to the damage pattern nodes of a building structure based on rule reasoning and similar case matching using knowledge graphs, the historical case nodes pointed to by the damage pattern nodes of the building structure are determined. This can be achieved by calculating the distance between the damage pattern of the damage pattern node and the damage pattern of the historical case in the historical case node, matching similar historical case nodes, thereby determining the historical case nodes pointed to by the damage pattern nodes, and determining the similarity between the damage pattern corresponding to the damage pattern node and the damage pattern of the historical case corresponding to the historical case node.

[0088] A rule engine is used for rule reasoning to obtain a candidate set of third reinforcement technology nodes. The rule engine reasons based on at least one of the following: the damage pattern corresponding to the damage pattern node in the knowledge graph, the constraint condition corresponding to the constraint condition node triggered by the damage pattern node of the building structure, and similarity.

[0089] Table 8 is an example table of reasoning rules in the knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provided by this invention.

[0090] Table 8

[0091]

[0092] The process involves identifying historical case nodes corresponding to the damage mode nodes of the earthquake-damaged building structure, determining the effect evaluation nodes corresponding to these historical case nodes, and obtaining a fourth set of candidate reinforcement technology nodes based on the historical case nodes corresponding to the effect evaluation nodes that meet preset requirements. When obtaining the fourth set of candidate reinforcement technology nodes based on the historical case nodes corresponding to the effect evaluation nodes that meet preset requirements, the reinforcement technology nodes pointed to by these historical case nodes are then acquired, thus completing the fourth set of candidate reinforcement technology nodes. Effect evaluation nodes that meet preset requirements include nodes with repair effects rated as "good" or above.

[0093] The first set of candidate reinforcement technology nodes is obtained based on the third and fourth candidate sets of candidate reinforcement technology nodes. The first set of candidate reinforcement technology nodes can be considered as the union of the third and fourth candidate sets.

[0094] The knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provides this invention improves the accuracy of the first reinforcement technology node candidate set by determining the candidate set based on rule engines and similar cases.

[0095] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs includes calculating the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure. This includes: determining at least one historical case node pointed to by the reinforcement technology node; determining a historical case similarity score based on the similarity between the damage mode of the historical case corresponding to the at least one historical case node and the damage mode corresponding to the damage mode node of the building structure; determining a constraint compliance score based on the situation where the reinforcement technology node satisfies the constraint conditions of the constraint condition node triggered by the damage mode node of the building structure; determining an effect evaluation node pointed to by at least one historical case node; determining a historical reinforcement effect evaluation score based on the reinforcement effect corresponding to each effect evaluation node; determining an expert experience score corresponding to the reinforcement technology node; determining a preset additional factor score corresponding to the reinforcement technology node; and performing a weighted summation of the historical case similarity score, the constraint compliance score, the historical reinforcement effect evaluation score, the expert experience score, and the preset additional factor score to obtain the applicability score.

[0096] When calculating the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of building structure, a comprehensive evaluation is conducted from multiple aspects such as similarity of historical cases, constraint compliance, historical reinforcement effect, expert experience, and pre-set additional factors related to the scenario to obtain the applicability score.

[0097] At least one historical case node is identified as the node to which the reinforcement technology node points. The reinforcement technology node pointing to the historical case node indicates that the reinforcement technology corresponding to the reinforcement technology node is applied in the historical case corresponding to the historical case node. The historical case similarity score is determined based on the similarity between the damage mode of the historical case corresponding to at least one historical case node and the damage mode corresponding to the damage mode node of the building structure.

[0098] The constraint compliance score is determined based on whether the reinforcement technology node meets the constraint conditions triggered by the damage mode node of the earthquake-damaged building structure. A higher compliance score results in a higher compliance score. When comparing multiple constraints, the constraint satisfaction score for each constraint can be obtained separately. The constraint compliance score is then obtained by weighted summing of these constraint satisfaction scores. The weights for each constraint can be pre-defined.

[0099] The process involves identifying at least one historical case node linked to by the reinforcement technology node, and then identifying the effect evaluation node linked to by each of these historical case nodes. Based on the reinforcement effect corresponding to each effect evaluation node, a historical reinforcement effect evaluation score is determined. The effect evaluation node may include a pre-calculated comprehensive reinforcement effect score, which can be directly used as the reinforcement effect evaluation score for each effect evaluation node. Finally, the historical reinforcement effect evaluation score is obtained based on the reinforcement effect evaluation scores corresponding to the effect evaluation nodes linked to by each historical case node.

[0100] The expert experience scores corresponding to the reinforcement technology nodes are determined. These expert experience scores are preset and can be adjusted according to the actual situation.

[0101] Determine the preset additional factor scores corresponding to the reinforcement technology nodes. The preset additional factor scores are related to the specific scenario requirements. For example, some scenarios consider the magnitude of noise generated by the reinforcement technology and include it in the calculation of the preset additional factor scores. If there are no particularly important factors to consider, the preset additional factor scores can be set directly to 0.

[0102] After obtaining the historical case similarity score, constraint compliance score, historical reinforcement effect evaluation score, expert experience score, and preset additional factor score, a weighted sum is calculated based on these scores to obtain the applicability score. The applicability score is expressed as follows:

[0103] Applicability_Score=w 1 ×Similariry+w 2 ×Constraint_Fit+w 3 ×Historical_ Effect+w 4 ×ExpertWeight+w 5 ×OtherFactors ;

[0104] in, Applicability Score Indicates the applicability rating. Similariry This represents the similarity score between historical cases. w 1 indicates the weight of the historical case similarity score. Constraint_Fit This indicates the constraint compliance score. w 2 indicates the weight of the constraint compliance score. Historical_Effect This indicates the score for evaluating the effectiveness of historical reinforcement. w 3 indicates the weight of the historical reinforcement effect evaluation score. ExpertWeight Indicates expert experience rating. w 4 indicates the weight of the expert experience score. Other Factors This indicates the pre-set additional factor score. w 5 indicates the weight of the preset additional factor score.

[0105] The weights for historical case similarity score, constraint compliance score, historical reinforcement effect evaluation score, expert experience score, and preset additional factor score are pre-set and can be adjusted according to actual conditions.

[0106] Table 9 provides explanations of the sub-indicators for calculating the applicability score in the knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provided by this invention.

[0107] Table 9

[0108]

[0109] The knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provides an applicability score by weighting and summing historical case similarity scores, constraint compliance scores, historical reinforcement effect evaluation scores, expert experience scores, and preset additional factor scores, thereby improving the accuracy of the applicability score.

[0110] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs includes determining a historical case similarity score based on the similarity between the damage patterns of historical cases corresponding to at least one historical case node and the damage patterns corresponding to the damage pattern nodes of the building structure. This includes: determining the historical case similarity score by calculating the average of the similarities between the damage patterns of historical cases corresponding to at least one historical case node and the damage patterns corresponding to the damage pattern nodes of the building structure; and / or, determining a historical reinforcement effect evaluation score based on the reinforcement effect corresponding to each effect evaluation node includes: determining the effect score of each effect evaluation node based on the reinforcement effect corresponding to each effect evaluation node, and obtaining the historical reinforcement effect evaluation score by calculating the average of the effect scores of each effect evaluation node.

[0111] When determining the historical case similarity score based on the similarity between the damage mode of the historical case corresponding to at least one historical case node and the damage mode corresponding to the damage mode node of the building structure, the similarity between the damage mode of the historical case corresponding to at least one historical case node and the damage mode corresponding to the damage mode node of the building structure is calculated, and the historical case similarity score is obtained by calculating the average of each similarity.

[0112] When determining the historical reinforcement effect evaluation score based on the reinforcement effect corresponding to each effect evaluation node, the effect score of each effect evaluation node is determined based on the reinforcement effect corresponding to each effect evaluation node. For example, the effect score of each effect evaluation node can be taken as the comprehensive reinforcement effect score contained in the node; the historical reinforcement effect evaluation score is obtained by calculating the average of the effect scores of each effect evaluation node.

[0113] The knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provided by this invention improves the accuracy of historical case similarity scores and historical reinforcement effect evaluation scores.

[0114] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on a knowledge graph, after determining the recommended list of reinforcement schemes based on at least one second reinforcement technology node, the method further includes: obtaining the node relationships between the constraint condition nodes and the historical case nodes in the knowledge graph; determining the effect evaluation nodes pointed to by the historical case nodes; determining the success rate of the historical case nodes pointed to by the constraint condition nodes when a certain constraint condition occurs based on the reinforcement effect in the effect evaluation nodes; and adjusting the contribution of the constraint condition in calculating the constraint compliance score based on the success rate.

[0115] The process involves obtaining the node relationships between constraint nodes and historical case nodes in the knowledge graph, i.e., scanning cases with the `affect_by` relationship, identifying the effect evaluation nodes pointed to by historical case nodes, and determining the success rate of historical case nodes pointed to by constraint nodes when a certain constraint occurs based on the reinforcement effect in the effect evaluation nodes, according to different constraints. The contribution of constraints in calculating constraint compliance scores is adjusted based on the success rate, such as adjusting the weight of constraints in calculating constraint compliance scores. For example, if it is found that when the constraint "extremely short construction period" occurs, although a certain technology is theoretically the most applicable, the actual success rate of historical cases is generally low, then the weight of the "extremely short construction period" constraint in calculating constraint compliance can be adjusted.

[0116] The knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provides this invention. By learning the influence of constraints in reverse according to the pointing relationship between constraint nodes and historical case nodes, the contribution of constraints in calculating constraint compliance is adjusted, thereby improving the accuracy of constraint compliance.

[0117] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs, after calculating the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure, and before sorting the second set of candidate reinforcement technology nodes according to the applicability score, the method further includes: inputting the damage mode features of the damage mode nodes of the building structure into a pre-trained reinforcement scheme recommendation model, and improving the applicability score of the corresponding reinforcement technology node according to the output reinforcement scheme.

[0118] The final reinforcement scheme recommendation can be obtained by combining knowledge graph reasoning and machine learning models. A reinforcement scheme recommendation model is pre-trained. A training set is constructed: historical case data is used as samples, input features are loss pattern-related parameters such as crack width, earthquake damage level, and wall material, and output labels are the corresponding reinforcement technology types. The model can use decision trees, random forests, support vector machines (SVM), or deep learning multilayer perceptrons (MLP). After training, the model can directly provide the most likely reinforcement scheme category based on the damage pattern-related features of new inputs, representing a data-driven intelligent recommendation.

[0119] After calculating the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure, the damage mode features of the damage mode node of the building structure can be input into the pre-trained reinforcement scheme recommendation model to obtain the reinforcement scheme recommended by the machine learning model, thereby improving the applicability score of the reinforcement technology node corresponding to the machine learning recommended reinforcement scheme.

[0120] The knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provides this invention. It adjusts the applicability scores of corresponding reinforcement technology nodes based on the reinforcement schemes output by the machine learning model, thereby further improving the accuracy of the applicability scores.

[0121] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on a knowledge graph, after determining a recommended list of reinforcement schemes based on at least one second reinforcement technology node, the method further includes: pushing the recommended list of reinforcement schemes to a user interface; wherein, the recommended list of reinforcement schemes includes reinforcement schemes corresponding to the at least one second reinforcement technology node; and receiving the user's final confirmation of the reinforcement scheme based on the recommended list of reinforcement schemes.

[0122] After obtaining the recommended reinforcement scheme list, it is pushed to the user interface. The recommended reinforcement scheme list includes reinforcement schemes corresponding to at least one second reinforcement technology node. The user interface may include a visualized seismic damage model built based on WebGL or OpenGL, a knowledge graph browsing interface, and a reinforcement scheme comparison interface. The recommended reinforcement scheme list is displayed on the reinforcement scheme comparison interface, supporting operations such as sorting and conditional filtering.

[0123] Users determine their final reinforcement solution based on a recommended list of reinforcement options.

[0124] For example, in a brick masonry house, the crack width is 8mm, the wall tilt is 3°, the construction site is limited, the budget is no more than 250,000 yuan, and the construction period is required to be ≤20 days.

[0125] Knowledge graph reasoning conclusion: The reinforced concrete mortar surface layer reinforcement scheme has the highest priority, followed by FRP reinforcement.

[0126] Recommended output:

[0127] Material list, process schedule, and budget estimate.

[0128] Historical case link (xx% matching of similar damage patterns, "Post-earthquake reinforcement in a certain county in xxx year" case).

[0129] The present invention provides a method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs, which realizes the push of a list of recommended reinforcement schemes and the filtering of schemes.

[0130] According to the present invention, a method for recommending reinforcement schemes for earthquake-damaged building structures based on a knowledge graph, after receiving the final confirmed reinforcement scheme from the user based on the reinforcement scheme recommendation list, the method further includes: generating historical case nodes and effect evaluation nodes of the building structure based on the reinforcement effect of the final confirmed reinforcement scheme, and updating the knowledge graph based on the relevant nodes of the building structure.

[0131] After obtaining the final confirmed reinforcement plan from the recommended list, users use this plan to reinforce the earthquake-damaged building structure and then evaluate the reinforcement effect after completion. Correspondingly, historical case nodes and effect evaluation nodes can be constructed.

[0132] The knowledge graph is updated based on relevant nodes of the earthquake-damaged building structure, including:

[0133] Establish the pointing relationship between building structure nodes and damage mode nodes;

[0134] Establish the pointing relationship between damage mode nodes and the triggered constraint condition nodes;

[0135] Establish the pointing relationship between damage pattern nodes and similar historical case nodes;

[0136] Establish the pointing relationship between damage mode nodes and reinforcement technology nodes corresponding to reinforcement schemes in the recommended reinforcement scheme list;

[0137] Establish the relationship between historical case nodes and the nodes of the hardening technologies used;

[0138] Establish the pointing relationship between constraint condition nodes and historical case nodes;

[0139] Establish the relationship between historical case nodes and effect evaluation nodes;

[0140] Establish the pointing relationship between historical case nodes and damage mode nodes.

[0141] Each time a recommendation result is actually adopted, new engineering data can be automatically written into the historical case node, and the weights of relevant parameters in the reinforcement scheme recommendation can be adjusted according to the reinforcement effect of the new historical cases, forming a closed-loop learning mechanism.

[0142] For example, if it is found that there are more cases where the "similarity of historical cases" is high but the final reinforcement effect is poor, the weight of the historical case similarity score will be automatically reduced, or the weight of the historical reinforcement effect evaluation score will be increased, so that the recommendation will pay more attention to the actual reinforcement effect rather than the superficial similarity.

[0143] The present invention provides a method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs, which realizes the updating of knowledge graphs.

[0144] The following describes the knowledge graph-based reinforcement scheme recommendation device for earthquake-damaged building structures provided by the present invention. The knowledge graph-based reinforcement scheme recommendation device described below can be referred to in correspondence with the knowledge graph-based reinforcement scheme recommendation method for earthquake-damaged building structures described above.

[0145] Figure 3 This is a structural schematic diagram of the knowledge graph-based reinforcement scheme recommendation device for earthquake-damaged building structures provided by the present invention. The knowledge graph includes building structure nodes, damage mode nodes, reinforcement technology nodes, constraint condition nodes, historical case nodes, and effect evaluation nodes, such as… Figure 3As shown, the device includes a construction module 10, a rule-based reasoning and similar case matching module 20, a constraint filtering module 30, an applicability evaluation module 40, and a reinforcement scheme recommendation module 50. The construction module 10 is used to: construct corresponding building structure nodes and damage mode nodes based on the earthquake-damaged building structure and damage mode of the reinforcement scheme to be determined; the rule-based reasoning and similar case matching module 20 is used to: determine the first set of candidate reinforcement technology nodes corresponding to the damage mode nodes of the building structure based on the knowledge graph through rule-based reasoning and similar case matching; the constraint filtering module 30 is used to: determine the damage mode nodes of the building structure according to engineering input parameters. The constraint condition node triggered by the point is used to filter the first set of candidate reinforcement technology nodes to obtain a second set of candidate reinforcement technology nodes. The applicability evaluation module 40 is used to calculate the applicability score of the reinforcement technology corresponding to each reinforcement technology node in the second set of candidate reinforcement technology nodes for the reinforcement of the building structure. The reinforcement scheme recommendation module 50 is used to sort the second set of candidate reinforcement technology nodes according to the applicability score, determine at least one second reinforcement technology node with a high ranking according to the sorting result, and determine a reinforcement scheme recommendation list according to the at least one second reinforcement technology node.

[0146] The knowledge graph-based reinforcement scheme recommendation device for earthquake-damaged building structures provided by this invention determines a first set of candidate reinforcement technology nodes corresponding to damage pattern nodes through rule reasoning and similar case matching based on the knowledge graph. The first set of candidate reinforcement technology nodes is filtered according to constraint condition nodes to obtain a second set of candidate reinforcement technology nodes. The second set of candidate reinforcement technology nodes is sorted according to applicability scores, and a reinforcement scheme recommendation list is determined based on the sorting results, thus realizing intelligent recommendation of reinforcement schemes based on knowledge graph.

[0147] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include a processor 410, a communications interface 420, a memory 430, and a communication bus 440. The processor 410, communications interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions stored in the memory 430 to execute the knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provided in the above embodiments.

[0148] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0149] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provided in the above embodiments.

[0150] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the knowledge graph-based method for recommending reinforcement schemes for earthquake-damaged building structures provided in the above embodiments.

[0151] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0152] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0153] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0155] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs, characterized in that, The knowledge graph includes building structure nodes, damage mode nodes, reinforcement technology nodes, constraint condition nodes, historical case nodes, and effect evaluation nodes. The method includes: Based on the earthquake-damaged building structure and damage mode of the reinforcement scheme to be determined, the corresponding building structure nodes and damage mode nodes are constructed. Based on the knowledge graph, rule-based reasoning and similar case matching are used to determine the first set of candidate reinforcement technology nodes corresponding to the damage mode nodes of the building structure. Based on the engineering input parameters, the constraint condition nodes triggered by the damage mode nodes of the building structure are determined, and the first set of candidate reinforcement technology nodes is filtered based on the constraint condition nodes to obtain the second set of candidate reinforcement technology nodes. For each reinforcement technology node in the second set of candidate reinforcement technology nodes, calculate the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure. The candidate set of the second reinforcement technology nodes is sorted according to the applicability score, and at least one second reinforcement technology node with the highest ranking is determined according to the sorting result. A recommended list of reinforcement schemes is determined according to the at least one second reinforcement technology node. In the knowledge graph, the building structure node points to the damage mode node, indicating that the building structure corresponding to the building structure node has the damage mode corresponding to the damage mode node. The damage pattern node points to the historical case node, indicating that the damage pattern corresponding to the damage pattern node is similar to the damage pattern in the historical case corresponding to the historical case node. The reinforcement technology node points to the constraint condition node, indicating that the reinforcement technology corresponding to the reinforcement technology node satisfies the constraint condition corresponding to the constraint condition node. The damage mode node points to the constraint condition node, indicating that the damage mode corresponding to the damage mode node triggers the constraint condition corresponding to the constraint condition node; The damage mode node points to the reinforcement technology node, indicating that the reinforcement technology corresponding to the reinforcement technology node is the recommended solution for the damage mode of the damage mode node; The historical case node pointing to the reinforcement technology node indicates that the historical case corresponding to the historical case node used the reinforcement technology corresponding to the reinforcement technology node; The constraint node points to the historical case node, indicating that the constraint corresponding to the constraint node plays a role in the historical case corresponding to the historical case node. The fact that the constraint condition node points to the reinforcement technology node indicates that the constraint condition corresponding to the constraint condition node has a constraint relationship with the reinforcement technology corresponding to the reinforcement technology node. The historical case node pointing to the effect evaluation node indicates that the historical case corresponding to the historical case node has the reinforcement effect corresponding to the effect evaluation node; The reinforcement technology node points to the historical case node, indicating that the reinforcement technology corresponding to the reinforcement technology node is applied in the historical case corresponding to the historical case node; The fact that the historical case node points to the damage pattern node indicates that the damage pattern of the historical case corresponding to the historical case node is consistent with the damage pattern corresponding to the damage pattern node.

2. The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs according to claim 1, characterized in that, The process of determining the first set of candidate reinforcement technology nodes corresponding to the damage mode nodes of the building structure based on the knowledge graph through rule reasoning and similar case matching includes: Determine the historical case node pointed to by the damage mode node of the building structure, and determine the similarity between the damage mode corresponding to the damage mode node and the damage mode of the historical case corresponding to the historical case node. Based on at least one of the damage patterns corresponding to the damage pattern nodes in the knowledge graph, the constraint conditions corresponding to the constraint condition nodes triggered by the damage pattern nodes of the building structure, and the similarity, a rule engine is used to perform rule reasoning to obtain a candidate set of third reinforcement technology nodes. Determine the historical case node to which the damage mode node of the building structure points, determine the effect evaluation node to which the historical case node points, and obtain the fourth reinforcement technology node candidate set based on the historical case node corresponding to the effect evaluation node in which the reinforcement effect meets the preset requirements. The first reinforcement technology node candidate set is obtained based on the third reinforcement technology node candidate set and the fourth reinforcement technology node candidate set.

3. The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs according to claim 1, characterized in that, The calculation of the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure includes: Determine at least one of the historical case nodes pointed to by the reinforcement technology node, and determine the historical case similarity score based on the similarity between the damage mode of the historical case corresponding to the at least one historical case node and the damage mode corresponding to the damage mode node of the building structure. The constraint compliance score is determined based on the constraint conditions of the reinforcement technology node that meet the constraint conditions triggered by the damage mode node of the building structure. Determine the effect evaluation node that each of the at least one historical case node points to, and determine the historical reinforcement effect evaluation score based on the reinforcement effect corresponding to each of the effect evaluation nodes; Determine the expert experience score corresponding to the reinforcement technology node; Determine the preset additional factor scores corresponding to the reinforcement technology nodes; The applicability score is obtained by weighting and summing the historical case similarity score, the constraint compliance score, the historical reinforcement effect evaluation score, the expert experience score, and the preset additional factor score.

4. The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs according to claim 3, characterized in that, The step of determining the historical case similarity score based on the similarity between the damage pattern of the historical case corresponding to at least one of the historical case nodes and the damage pattern corresponding to the damage pattern node of the building structure includes: determining the historical case similarity score by calculating the average value of the similarity between the damage pattern of the historical case corresponding to at least one of the historical case nodes and the damage pattern corresponding to the damage pattern node of the building structure. And / or, The step of determining the historical reinforcement effect evaluation score based on the reinforcement effect corresponding to each of the effect evaluation nodes includes: determining the effect score of each effect evaluation node based on the reinforcement effect corresponding to each of the effect evaluation nodes, and obtaining the historical reinforcement effect evaluation score by calculating the average of the effect scores of each of the effect evaluation nodes.

5. The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs according to claim 3, characterized in that, After determining the recommended list of reinforcement schemes based on the at least one second reinforcement technology node, the method further includes: Obtain the node relationships in the knowledge graph from the constraint condition nodes to the historical case nodes; Determine the effect evaluation node that the historical case node points to; The success rate of the historical case node pointed to by the constraint condition node when a certain constraint condition occurs is determined based on the reinforcement effect in the effect evaluation node. The contribution of the constraint in calculating the constraint compliance score is adjusted based on the success rate.

6. The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs according to claim 2, characterized in that, After calculating the applicability score of the reinforcement technology corresponding to the reinforcement technology node for the reinforcement of the building structure, and before sorting the second set of candidate reinforcement technology nodes according to the applicability score, the method further includes: The damage mode features of the damage mode nodes of the building structure are input into the pre-trained reinforcement scheme recommendation model, and the applicability score of the corresponding reinforcement technology node is improved according to the output reinforcement scheme.

7. The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs according to claim 1, characterized in that, After determining the recommended list of reinforcement schemes based on the at least one second reinforcement technology node, the method further includes: The reinforcement scheme recommendation list is pushed to the user interface; wherein, the reinforcement scheme recommendation list includes reinforcement schemes corresponding to the at least one second reinforcement technology node; Receive the user's final confirmation of the reinforcement scheme based on the reinforcement scheme recommendation list.

8. The method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs according to claim 7, characterized in that, After receiving the user's final confirmation of the reinforcement solution based on the reinforcement solution recommendation list, the method further includes: Based on the reinforcement effect of the finally confirmed reinforcement scheme, historical case nodes and effect evaluation nodes of the building structure are generated, and the knowledge graph is updated based on the relevant nodes of the building structure.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the method for recommending reinforcement schemes for earthquake-damaged building structures based on knowledge graphs as described in any one of claims 1 to 8.

Images