Electric power engineering material technical review method and system based on rule graph driving

Abstract

The invention belongs to the field of engineering material review, and discloses an electric power engineering material technical review method and system based on rule graph driving, and the method comprises the steps: constructing an initial rule graph according to expert experience; utilizing a large language model to extract a relationship among materials, specifications and the materials in a historical technical examination result, converting the relationship among the materials into rule constraints, and perfecting edges and attributes of the edges in an initial rule graph; verifying wrong material items according to the similarity between the to-be-examined materials and the materials in the rule graph and generating an engineering material graph, verifying missing / redundant material items based on the engineering material graph and the rule graph, and verifying material specification wrong items according to rule constraints in the rule graph; and generating a final technical review result according to the verification results of the wrong material items, the missing / redundant material items and the material specification error items. According to the invention, engineering design review work can be better promoted, and the accuracy and efficiency of technical review of electric power engineering materials are improved.

Description

Technical Field

[0001] The present invention belongs to the field of engineering material review, and in particular relates to a method and system for technical review of power engineering materials driven by a rule graph. Background Art

[0002] In power engineering construction, especially substation projects, the accuracy of material allocation is crucial to project quality and safe operation. Traditional material review relies mainly on manual experience and static rule templates, which has the following problems: (1) Delayed rule updates: Rule templates are not updated in a timely manner and cannot reflect the latest engineering requirements and material configuration standards, resulting in low audit accuracy.

[0003] (2) Lack of intelligent analysis: Existing review methods are unable to conduct intelligent analysis of material configurations, making it difficult to identify potential configuration errors or omissions, resulting in low audit accuracy.

[0004] (3) Inability to utilize historical data: The material configuration data in the historical review results are not fully utilized, and it is impossible to extract potential rules and patterns from them to guide the material configuration of the current project.

[0005] However, with the increasingly fierce competition in the power design market, it is urgent to further improve labor productivity and product quality. Therefore, a better technical review method for engineering materials is urgently needed. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for technical review of power engineering materials based on rule graph drive, which can better promote engineering design review work and improve the accuracy and efficiency of technical review of power engineering materials.

[0007] To achieve the above object, the technical solution adopted by the present invention is: First, a rule-graph-driven technical review method for power engineering materials is provided, comprising: An initial rule graph is constructed based on expert experience, wherein the rule graph includes an entity set and an edge set, where entities in the entity set are materials with specifications; Obtain historical technical review results and use a large language model to extract materials, specifications, and relationships between materials from these results. Convert these relationships between materials into rule constraints and refine the edges and edge attributes in the initial rule graph. Obtain the list of power engineering materials to be reviewed, check for incorrect material items based on the similarity between the materials to be reviewed and the materials in the rule map, and generate a project material map. Verify missing / redundant material items based on the project material map and the rule map, and verify material specification errors based on the rule constraints in the rule map. The final technical review results are generated based on the verification results of incorrect material items, missing / excess material items, and material specification errors.

[0008] Several optional methods are also provided below, but they are not intended to be additional limitations on the above-mentioned overall solution. They are merely further supplements or optimizations. Under the premise that there are no technical or logical contradictions, each optional method can be combined separately for the above-mentioned overall solution, or multiple optional methods can be combined.

[0009] Preferably, converting the relationship between materials into rule constraints includes: Preconditions and result conditions are extracted from the relationships between materials, and a relationship rule constraint is generated as one of the rule constraints. The relationship rule constraint is: if the precondition is met, the result condition must be met.

[0010] Preferably, the rule constraints further include joint rule constraints, exclusive rule constraints and integrity rule constraints; The joint rule constraint is: if there is a material as a prerequisite, there must be multiple materials as the result conditions; The exclusion rule constraint is: if the material used as the prerequisite exists, the material used as the result condition must not exist; The integrity rule constraint is: if there is a material as a prerequisite, there must be a material as a result condition.

[0011] Preferably, the improving the edges and edge attributes in the initial regular graph includes: Calculate the similarity between entities in the historical technical review results and entities in the rule graph, and select the entities in the rule graph that are most similar to the entities in the historical technical review results; For the filtered entities, if there is no connecting edge between any two entities, add an edge between the two entities; For the filtered entities, if there are rule constraints related to the entities, the rule constraints are added as attributes of the edges connecting the corresponding entities.

[0012] Preferably, the similarity between the materials to be reviewed and the materials in the rule graph is calculated as follows: Calculate the similarity between the names of the materials to be reviewed and those in the rule graph; Calculate the similarity of specifications between the materials to be reviewed and the materials in the rule diagram; The similarity between the materials to be reviewed and the materials in the rule diagram is obtained by combining the similarity of the names and the similarity of the specifications.

[0013] Preferably, the step of reviewing the erroneous material items and generating the engineering material map based on the similarity between the materials to be reviewed and the materials in the rule map includes: If the similarity between the material to be reviewed and the material in the rule graph is less than the threshold, the material to be reviewed in the power engineering material list to be reviewed is marked as an incorrect material; otherwise, the material to be reviewed is added as a node to the engineering material graph; After completing the traversal of the list of power engineering materials to be reviewed, add edges to the corresponding nodes in the engineering materials graph based on the edges between the entities in the rule graph.

[0014] Preferably, the review of missing / redundant material items based on the engineering material drawing and the rule drawing includes: If there is an isolated node in the engineering material graph, it is determined that there are missing materials in the power engineering material list to be reviewed or the materials to be reviewed corresponding to the isolated node are redundant materials; Alternatively, if there is a non-connected area between the engineering material map and the rule map, it is determined that there are missing materials in the power engineering material list to be reviewed; Alternatively, if the material to be reviewed meets the prerequisites in the joint rule constraint, but other materials in the list of electric power engineering materials to be reviewed do not meet the result conditions in the joint rule constraint, then it is determined that there are missing materials in the list of electric power engineering materials to be reviewed; Alternatively, if the materials to be reviewed meet the preconditions in the exclusionary rule constraints, but other materials in the list of electric power engineering materials to be reviewed do not meet the result conditions in the exclusionary rule constraints, then it is determined that there are redundant materials in the list of electric power engineering materials to be reviewed; Alternatively, if the materials to be reviewed meet the prerequisites in the integrity rule constraints, but other materials in the list of electric power engineering materials to be reviewed do not meet the result conditions in the integrity rule constraints, it is determined that there are missing materials in the list of electric power engineering materials to be reviewed.

[0015] Preferably, checking material specification error items according to rule constraints in the rule graph includes: Get the specifications of the materials at both ends of an edge in the engineering material graph. If the relationship between the two specifications satisfies the relationship rule constraints, there is no material specification error; otherwise, there is a material specification error.

[0016] Secondly, a rule-graph-driven technical review system for power engineering materials is provided, including: A rule graph initial construction module is used to construct an initial rule graph based on expert experience. The rule graph includes an entity set and an edge set. The entities in the entity set are materials with specifications. The rule graph improvement module is used to obtain historical technical review results, use the large language model to extract the materials, specifications, and relationships between materials in the historical technical review results, convert the relationships between materials into rule constraints, and improve the edges and edge attributes in the initial rule graph; The material verification module is used to obtain the list of materials for the power project to be reviewed, verify incorrect material items based on the similarity between the materials to be reviewed and the materials in the rule map, and generate a project material map. It verifies missing / redundant material items based on the project material map and the rule map, and verifies incorrect material specification items based on the rule constraints in the rule map. The result output module is used to generate the final technical review results based on the verification results of incorrect material items, missing / excess material items and material specification errors.

[0017] The present invention provides a method and system for technical review of power engineering materials based on rule graph drive. The method makes full use of historical review data and expert experience, adopts a large language model to perform semantic analysis on review rules in natural language form, automatically extracts material entities, attributes and the dependencies between them, and thus constructs a unified rule graph model. During the review process, for the list of materials in the project to be reviewed, a project material graph is first constructed and matched with the rule graph as a subgraph to identify its structural integrity and topological consistency. Then, combined with the attribute mapping function defined on the edge of the rule, the attribute configuration relationship between each material is checked for consistency, thereby identifying missing material units and nodes with incorrect attribute configurations. The entire review process does not require human intervention, and the review is accurate and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of the technical review method for power engineering materials driven by rule graphs according to the present invention; Figure 2 A schematic diagram of an embodiment of a rule diagram of the present invention; Figure 3 This is a schematic diagram of an embodiment of the engineering material diagram of the present invention. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0021] Example 1: In order to solve the problem of low efficiency and low accuracy in the review of power engineering materials in the prior art, this embodiment proposes a technical review method for power engineering materials driven by rule graphs, such as Figure 1 As shown, the specific steps include: Step 1: Construct an initial rule graph based on expert experience.

[0022] The regular graph is an undirected topological structure, represented as ,in Represents a collection of entities, is the total number of entities, each entity A material with a set of specifications belonging to it , is the total number of specifications corresponding to the entity, . Represents a set of edges between entities.

[0023] It should be noted that, in the initial regular graph, edges may have been added between any two entities, or edges may have been added only between some entities, and the edges at this time do not contain attributes.

[0024] Step 2: Obtain historical technical review results, use the large language model to extract materials, specifications, and the relationships between materials in the historical technical review results, convert the relationships between materials into rule constraints, and improve the edges and edge attributes in the initial rule graph.

[0025] For edges in a regular graph , To connect at the edge Entities on both sides, this example examines key points from historical corpus Find the attribute corresponding to the edge, that is, the rule constraint function .

[0026] Step 2-1: Targeting a historical technical review result , the prompt words for building a large model are as follows: Get the output of the large model, including historical technical review results Entities in and the specifications carried by the entity.

[0027] Step 2-2, entity similarity detection: Considering that the names of material entities in the historical technical review results may be slightly different from the names of standard material entities, this embodiment uses an entity similarity detection algorithm to determine the entities in the historical technical review results.

[0028] Where, Represents the string edit distance, which is defined as the minimum number of single-character edit operations required to transform one string into another. These operations include insertion, deletion, and substitution. Indicates the length of the string. Indicates taking the maximum value, To calculate the similarity of two materials, Indicates supplies Name, Indicates supplies Name of the .

[0029] By traversing historical technical review results Entities in The similarity with the names of all entities in the rule graph, corresponding to the most similar entity in the rule graph , if the most similar entity obtained If there is no edge between two entities, add an edge between the corresponding entities in the rule graph. For example, the entities in the filtered rule graph are , then respectively in 、 、 Add edges between them.

[0030] Step 2-3: Construct edge attributes, i.e. rule constraint functions .

[0031] For the entities in the rule graph filtered in the previous step , take the two entities corresponding to an edge and the entity specifications, and use the big model to extract the relationship corresponding to these two entities and specifications in the historical technical review results.

[0032] For example, if the two entities corresponding to an edge are OPGW optical cable and optical cable tension string, and the entity specifications are wire diameter and range, the following prompt words are used to construct the formal logic of the large model: enter Get the formal logic of the large model output: , as a relationship between two entities.

[0033] Then, the relationship between materials is further transformed into rule constraints, specifically extracting the prerequisites from the relationship between materials ( ) and the result condition ( ), and generate a relational rule constraint as one of the rule constraints, where the relational rule constraint is: if the premise condition is met, then the result condition must be met. For example, if an entity of an edge meets of , then the other entity of the edge must satisfy of .

[0034] It should be noted that in actual design, the above rules need to be further converted into computer language, such as computer program logic: IF . = 120 THEN . >=12.9&& . <= 14.1.

[0035] In addition to the relationship rule constraints directly related to the edges, there are also rules for the relationship between materials. In this embodiment, the rule constraints also include joint rule constraints, exclusive rule constraints, and integrity rule constraints.

[0036] The joint rule constraint states that the simultaneous existence of multiple materials constitutes a legal configuration. Specifically, if the material that serves as the prerequisite exists, then multiple materials that serve as the result condition must exist. For example, if material A exists, then materials B and C must also exist. The formal logic is expressed as , the corresponding computer program logic is IF THEN && .

[0037] Exclusive rule constraints that two materials cannot exist at the same time. Specifically, if the material that is the prerequisite exists, the material that is the result condition must not exist. For example, material A and material B cannot exist at the same time. The formal logic is expressed as , the corresponding computer program logic is IF A = ​​true THEN B = false.

[0038] The integrity rule constraint states that a certain material cannot be missing. Specifically, if a material exists as a prerequisite, then a material as a result condition must also exist. For example, as long as material A exists, material B must also exist. The formal logic is expressed as , the corresponding computer program logic is IF A = ​​true THEN B = true.

[0039] For the filtered entities, if there are rule constraints related to the entities, that is, if the entities appear in the premise of the rule constraints, then add the rule constraints as the attributes of the edges connecting the corresponding entities. For example, for the entities filtered out from the rule graph in step 2-2, for an edge between two entities, first add the relationship rule constraints, and then judge the two entities in this edge. If one entity is material A in the joint rule constraint, exclusive rule constraint or integrity rule constraint, then add the corresponding joint rule constraint, exclusive rule constraint or integrity rule constraint to this edge. Finally, get all the rule constraints added to this edge and complete the construction of the edge attributes. The final rule graph structure is as follows: Figure 2 As shown in the figure, the circles in the figure represent entities, the lines between the circles represent edges, and the content on the lines represents the attributes of the edges.

[0040] Step 3: Obtain the list of power engineering materials to be reviewed, check for incorrect material items based on the similarity between the materials to be reviewed and the materials in the rule diagram, and generate an engineering material diagram. Check for missing / redundant material items based on the engineering material diagram and the rule diagram, and check for incorrect material specification items based on the rule constraints in the rule diagram.

[0041] Get the list of electrical engineering materials to be reviewed (without structural diagram), represented as , is the total number of materials in the list of power engineering materials to be reviewed, For the A material to be reviewed is a structured material entity with an attribute vector, including fields such as name, specification, and number. .

[0042] Step 3-1: Node fuzzy matching.

[0043] For material entities in the rule diagram , and materials awaiting review , the similarity calculation process is defined as follows: Where, Indicates materials awaiting review and the materials in the rule diagram The similarity between The similarity between the name of the material to be reviewed and the material in the rule diagram is calculated using the same formula as in step 2-2. same, The similarity between the specifications of the materials to be reviewed and those in the rule diagram is the comparison of the field names of the specifications. The specific calculation formula is the same as in step 2-2. Same, and and are the weights of the name and specification respectively, In this embodiment, , , which can be adjusted according to actual needs.

[0044] Step 3-2: Construction of engineering material diagram.

[0045] Based on the topology of the rule graph and the current material list, try to reconstruct a maximum possible matching subgraph, i.e., the engineering material graph, without topology. The specific process is: if the similarity between the material to be reviewed and the material in the rule diagram is Both are less than the threshold , then the materials to be reviewed in the list of materials for power engineering to be reviewed are marked as wrong materials; otherwise, the materials to be reviewed are added as nodes to the engineering materials graph, and the similarity in the rule graph is taken. The largest entity is the entity corresponding to the materials to be reviewed.

[0046] After completing the traversal of the list of power engineering materials to be reviewed, add edges to the corresponding nodes in the engineering material graph according to the edges between the entities in the rule graph to obtain an engineering material diagram such as Figure 3 As shown in the figure, the circles represent nodes, the characters in the nodes are the system numbers of the materials, and the lines between the circles represent the edges between the nodes, which do not have attributes. In this embodiment, incorrect materials refer to incorrect material names (such as OPGW optical cable) or incorrect specifications (such as wire diameter).

[0047] Step 3-3: Determine missing / excess materials.

[0048] If there is an isolated node in the engineering material graph, and the node is not connected to any other node in the graph, it is determined that there are missing materials in the power engineering material list to be reviewed or the materials to be reviewed corresponding to the isolated node are redundant materials.

[0049] Alternatively, if there is a non-connected area between the engineering material map and the rule map, it is determined that there are missing materials in the power engineering material list to be reviewed.

[0050] Alternatively, if the material to be reviewed meets the preconditions of the joint rule constraint, but other materials in the power engineering material list to be reviewed do not meet the result conditions of the joint rule constraint, then it is determined that there are missing materials in the power engineering material list to be reviewed. For example, if the material to be reviewed is material A, and material B and / or material C are not in the power engineering material list to be reviewed, then it is determined that there are missing materials, and that material B and / or material C are missing.

[0051] Alternatively, if the material to be reviewed meets the preconditions of the exclusionary rule, but other materials in the list of power engineering materials to be reviewed do not meet the result conditions of the exclusionary rule, then it is determined that there are redundant materials in the list of power engineering materials to be reviewed. For example, if the material to be reviewed is material A, and material B also exists in the list of power engineering materials to be reviewed, then it is determined that there are redundant materials, and the redundant material is material B.

[0052] Alternatively, if the material to be reviewed meets the prerequisites of the integrity rule constraint, but other materials in the power engineering material list to be reviewed do not meet the consequence conditions of the integrity rule constraint, then the power engineering material list to be reviewed is considered to contain missing materials. For example, if the material to be reviewed is material A, and material B does not exist in the power engineering material list to be reviewed, then it is determined that there are missing materials, and material B is missing.

[0053] Step 3-4: Incorrect judgment of material specifications.

[0054] If the rule constraint function of an edge in the engineering material graph If it fails, it is determined that there is an error in the specifications of the materials connecting the two end nodes. Specifically: obtain the specifications of the materials at both ends of an edge in the engineering material graph. If the relationship between the two specifications meets the relationship rule constraints, there is no material specification error; otherwise, there is a material specification error, specifically the value of the specification is wrong. For example, the nodes at both ends of an edge are OPGW optical cable and optical cable tension string, and the node specifications are wire diameter 120 and range 12.9 to 13.1. The relationship rule constraint is that if an entity of an edge meets of , then the other entity of the edge must satisfy of If the range of the optical cable tension string is not met, it is judged that there is a material specification error, and the value of the range of the optical cable tension string is wrong.

[0055] Step 4: Generate the final technical review results based on the verification results of incorrect material items, missing / excess material items, and material specification errors.

[0056] The technical review results include error information and corresponding correction information. The correction information can be generated manually or output by the big model. For example, if it is determined that material B is missing based on the integrity rule constraint, the big model can output the correction information as follows: as long as material A exists, material B must exist; for example, if it is determined that the range of the optical cable tension string does not meet the requirements based on the relationship rule constraint, the big model can output the correction information as follows: an entity on an edge meets the requirements. of , then the other entity of the edge must satisfy of If the missing / surplus materials are unclear, the correction information may not be generated, or the correction information may be generated by the large model after the missing / surplus materials are manually clarified.

[0057] Example 2: This embodiment provides a rule-graph-driven system for technical review of power engineering materials, including: A rule graph initial construction module is used to construct an initial rule graph based on expert experience. The rule graph includes an entity set and an edge set. The entities in the entity set are materials with specifications. The rule graph improvement module is used to obtain historical technical review results, use the large language model to extract the materials, specifications, and relationships between materials in the historical technical review results, convert the relationships between materials into rule constraints, and improve the edges and edge attributes in the initial rule graph; The material verification module is used to obtain the list of materials for the power project to be reviewed, verify incorrect material items based on the similarity between the materials to be reviewed and the materials in the rule map, and generate a project material map. It verifies missing / redundant material items based on the project material map and the rule map, and verifies incorrect material specification items based on the rule constraints in the rule map. The result output module is used to generate the final technical review results based on the verification results of incorrect material items, missing / excess material items and material specification errors.

[0058] For the specific limitations of the rule-graph-driven technical review system for electric power engineering materials, please refer to the limitations of the rule-graph-driven technical review method for electric power engineering materials in the above text, which will not be repeated here.

[0059] It should be noted that Figure 2 and Figure 3 They are mainly used to show the effects of rule diagrams and engineering material diagrams. Figure 2 and Figure 3 The graphic characters are only elements in the running interface of the software, and do not involve the focus of improvement of this application. In addition, the clarity of the running interface is related to pixels and scaling, so the presentation effect is relatively limited.

[0060] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The above-described embodiments merely illustrate several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for technical review of power engineering materials based on rule graph drive, characterized in that: The rule-graph-driven method for technical review of power engineering materials includes: An initial rule graph is constructed based on expert experience, wherein the rule graph includes an entity set and an edge set, where entities in the entity set are materials with specifications; Obtain historical technical review results and use a large language model to extract materials, specifications, and relationships between materials from these results. Convert these relationships between materials into rule constraints and refine the edges and edge attributes in the initial rule graph. Obtain the list of power engineering materials to be reviewed, check for incorrect material items based on the similarity between the materials to be reviewed and the materials in the rule map, and generate a project material map. Verify missing / redundant material items based on the project material map and the rule map, and verify material specification errors based on the rule constraints in the rule map. The final technical review results are generated based on the verification results of incorrect material items, missing / excess material items, and material specification errors.

2. The method for technical review of power engineering materials based on rule graph drive according to claim 1 is characterized in that: The conversion of the relationship between materials into rule constraints includes: Preconditions and result conditions are extracted from the relationships between materials, and a relationship rule constraint is generated as one of the rule constraints. The relationship rule constraint is: if the precondition is met, the result condition must be met.

3. The method for technical review of electric power engineering materials based on rule graph drive according to claim 1 or 2, characterized in that: The rule constraints also include joint rule constraints, exclusive rule constraints and integrity rule constraints; The joint rule constraint is: if there is a material as a prerequisite, there must be multiple materials as the result conditions; The exclusion rule constraint is: if the material used as the prerequisite exists, the material used as the result condition must not exist; The integrity rule constraint is: if there is a material as a prerequisite, there must be a material as a result condition.

4. The method for technical review of electric power engineering materials based on rule graph drive according to claim 1, characterized in that: The improvement of the edges and edge attributes in the initial regular graph includes: Calculate the similarity between entities in the historical technical review results and entities in the rule graph, and select the entities in the rule graph that are most similar to the entities in the historical technical review results; For the filtered entities, if there is no connecting edge between any two entities, add an edge between the two entities; For the filtered entities, if there are rule constraints related to the entities, the rule constraints are added as attributes of the edges connecting the corresponding entities.

5. The method for technical review of electric power engineering materials based on rule graph drive according to claim 1 is characterized in that: The similarity between the materials to be reviewed and the materials in the rule graph is calculated as follows: Calculate the similarity between the names of the materials to be reviewed and those in the rule graph; Calculate the similarity of specifications between the materials to be reviewed and the materials in the rule diagram; The similarity between the materials to be reviewed and the materials in the rule diagram is obtained by combining the similarity of the names and the similarity of the specifications.

6. The method for technical review of power engineering materials based on rule graph drive according to claim 1, characterized in that: The step of reviewing erroneous material items and generating an engineering material diagram based on the similarity between the materials to be reviewed and the materials in the rule diagram includes: If the similarity between the material to be reviewed and the material in the rule graph is less than the threshold, the material to be reviewed in the power engineering material list to be reviewed is marked as an incorrect material; otherwise, the material to be reviewed is added as a node to the engineering material graph; After completing the traversal of the list of power engineering materials to be reviewed, add edges to the corresponding nodes in the engineering materials graph based on the edges between the entities in the rule graph.

7. The method for technical review of power engineering materials based on rule graph drive according to claim 3 is characterized in that: The review of missing / redundant material items based on engineering material drawings and rule drawings includes: If there is an isolated node in the engineering material graph, it is determined that there are missing materials in the power engineering material list to be reviewed or the materials to be reviewed corresponding to the isolated node are redundant materials; Alternatively, if there is a non-connected area between the engineering material map and the rule map, it is determined that there are missing materials in the power engineering material list to be reviewed; Alternatively, if the material to be reviewed meets the prerequisites in the joint rule constraint, but other materials in the list of electric power engineering materials to be reviewed do not meet the result conditions in the joint rule constraint, then it is determined that there are missing materials in the list of electric power engineering materials to be reviewed; Alternatively, if the materials to be reviewed meet the preconditions in the exclusionary rule constraints, but other materials in the list of electric power engineering materials to be reviewed do not meet the result conditions in the exclusionary rule constraints, then it is determined that there are redundant materials in the list of electric power engineering materials to be reviewed; Alternatively, if the materials to be reviewed meet the prerequisites in the integrity rule constraints, but other materials in the list of electric power engineering materials to be reviewed do not meet the result conditions in the integrity rule constraints, it is determined that there are missing materials in the list of electric power engineering materials to be reviewed.

8. The method for technical review of electric power engineering materials based on rule graph drive according to claim 2, characterized in that: The checking of material specification errors according to the rule constraints in the rule graph includes: Get the specifications of the materials at both ends of an edge in the engineering material graph. If the relationship between the two specifications satisfies the relationship rule constraints, there is no material specification error; otherwise, there is a material specification error.

9. A rule-graph driven power engineering material technical review system, characterized in that: The rule-graph-driven power engineering material technical review system includes: A rule graph initial construction module is used to construct an initial rule graph based on expert experience. The rule graph includes an entity set and an edge set. The entities in the entity set are materials with specifications. The rule graph improvement module is used to obtain historical technical review results, use the large language model to extract the materials, specifications, and relationships between materials in the historical technical review results, convert the relationships between materials into rule constraints, and improve the edges and edge attributes in the initial rule graph; The material verification module is used to obtain the list of materials for the power project to be reviewed, verify incorrect material items based on the similarity between the materials to be reviewed and the materials in the rule map, and generate a project material map. It verifies missing / redundant material items based on the project material map and the rule map, and verifies incorrect material specification items based on the rule constraints in the rule map. The result output module is used to generate the final technical review results based on the verification results of incorrect material items, missing / excess material items and material specification errors.

Images