Engineering budget checking method and system based on knowledge graph and dynamic rule learning

Abstract

The invention provides a project budget checking method and system based on a knowledge graph and dynamic rule learning, and the method comprises the steps: building a parameter correlation graph based on an anomaly detection model of a graph neural network, so as to recognize abnormal data, and positioning the root data which generates cost deviation in each power grid project through a multi-dimensional traceability algorithm; decoding the specific coding system of each power grid project; utilizing the terrain coefficient calculation rule and the transportation distance judgment rule to decompose and construct a terrain coefficient dynamic calculation model and a transportation distance judgment model, and respectively setting a preset terrain threshold value and a distance threshold value in the terrain coefficient dynamic calculation model and the transportation distance judgment model; and according to the root data, the specific coding system, the terrain coefficient dynamic calculation model and the transportation distance determination model, constructing an engineering budget check model so as to realize budget check of the power grid engineering. According to the method, intelligent analysis support is provided for quota optimization and investment decision making based on accumulated engineering data, and the scientificity of management decision making is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for verifying engineering budget estimates based on knowledge graphs and dynamic rule learning. Background Art

[0002] Technical and economic analysis during the early planning stages of power grid projects presents unique challenges. Technical documentation from feasibility studies and preliminary designs often contains large amounts of unstructured data, including critical information such as equipment parameters, bills of quantities, and technical specifications. Traditionally, cost engineers spend a significant amount of time manually sifting through disparate design specifications, bills of materials, and technical briefing documents, comparing them item by item with budget documents. This manual process is not only inefficient but also prone to cost inaccuracies due to misunderstandings of specialized terminology and data omissions.

[0003] In recent years, breakthroughs in cognitive computing technology have brought innovative opportunities to this field. In particular, deep learning-driven text understanding technology has demonstrated significant advantages in semantic parsing and structural processing of complex engineering documents. Current methods for checking project budgets typically employ the following two approaches: 1. Manual review mode: Technical and economic personnel manually compare design drawings, quota books, and capital proposals, and verify each item item by item using tools such as Excel. 2. Traditional software tools: Pricing software only implements basic engineering quantity calculations, and document management systems provide simple file retrieval functions; However, the current methods have the following drawbacks: 1. Serious data fragmentation: Design parameters are scattered across heterogeneous files such as drawings and proposals, lacking an automatic association mechanism. 2. High professional barriers: Transmission tower terminology (e.g., "ZC straight tower" and "JC tension tower") requires manual identification; the basis for applying quotas is difficult to trace (e.g., the calculation rules for the number of piles and hole diameters in rock anchor foundations). 3. Poor dynamic adaptability: Unable to automatically adapt to newly issued quota rules (such as the 2023 version of the line stringing quota adjustment); parameters such as terrain coefficient and transportation distance must be calculated manually. Summary of the Invention

[0004] The embodiments of the present application provide a method and system for verifying engineering budget estimates based on knowledge graphs and dynamic rule learning, so as to at least address the deficiencies in the above-mentioned related technologies.

[0005] In a first aspect, an embodiment of the present application provides a method for verifying an engineering budget based on a knowledge graph and dynamic rule learning, comprising the following steps: Step 1: Real-time analysis of several professional rules for power grid projects, including terrain coefficient calculation rules, transportation distance determination rules, and quota standard matching rules; Step 2: Construct a parameter association map based on the graph neural network anomaly detection model to identify abnormal data in the engineering data of each power grid project, and locate the root cause of cost deviations in each power grid project through a multi-dimensional tracing algorithm; Step 3: Decode the unique coding system of each power grid project, and analyze the tower model by hierarchically decomposing the coded string to extract the voltage level, tower type representation, and call height value; Step 4: Decomposing and constructing a terrain coefficient dynamic calculation model and a transportation distance determination model using the terrain coefficient calculation rule and the transportation distance determination rule, and setting a preset terrain threshold and distance threshold in the terrain coefficient dynamic calculation model and the transportation distance determination model respectively; Step 5: Constructing a project budget verification model based on the source data, the unique coding system, the terrain coefficient dynamic calculation model, and the transportation distance determination model, and implementing budget verification of the power grid project using the project budget verification model.

[0006] Furthermore, the step 1 includes: Processing each of the power grid projects based on a rule management framework and conditional judgment logic to obtain engineering specification data in each of the power grid projects; A semantic mapping algorithm is used to convert the engineering specification data into executable logical rules to obtain professional rules for each power grid project.

[0007] Furthermore, the step 2 includes: Construct parameter association graphs, mapping engineering parameters, quota standards, and construction conditions into graph structure nodes; A graph convolutional network is used to deeply mine the node features of the graph structure nodes to identify abnormal correlations between parameters; By using a difference analysis algorithm and abnormal correlations between the parameters, a deviation report of the engineering data in each of the power grid projects is identified, and correction suggestions are provided. The report includes the deviation type, impact range and priority labeling.

[0008] Furthermore, the step three includes: For tower model analysis, the model code is split into voltage segment, tower type segment and call height segment according to preset rules; The first digit code is mapped to the voltage level, where the first digit is 2 and the corresponding voltage level is 220kV, and the first digit is 1 and the corresponding voltage level is 110kV; The tower type is determined by the last letter, where the last letter is Z for a straight tower and J for a tension tower. The last digit is extracted and converted into a height value.

[0009] Furthermore, the method further comprises: Real-time monitoring of industry standard data and extraction of key parameters of the industry standard data through natural language processing algorithms; Evaluate the impact of the key parameters on historical power grid engineering projects to generate corresponding sensitivity analysis reports; Based on the sensitivity analysis report, the impact of the adjustment range of the key parameters on the total cost is quantified, and a professional difference analysis report is generated to achieve adjustments in engineering management decisions.

[0010] In a second aspect, the present invention further proposes an engineering budget verification system based on knowledge graph and dynamic rule learning, comprising: A rule parsing module is used to parse out several professional rules of power grid projects in real time, wherein the professional rules include terrain coefficient calculation rules, transportation distance determination rules and quota standard matching rules; A data analysis module is used to construct a parameter association map based on an anomaly detection model of a graph neural network to identify abnormal data in the engineering data of each of the power grid projects, and locate the root cause of cost deviations in each of the power grid projects through a multi-dimensional tracing algorithm; A data decoding module is used to decode the unique coding system of each power grid project, and to parse the tower model by hierarchically decomposing the coded string to extract the voltage level, tower type representation, and call height value; a model construction module for decomposing and constructing a terrain coefficient dynamic calculation model and a transportation distance determination model by using the terrain coefficient calculation rule and the transportation distance determination rule, and setting a preset terrain threshold and distance threshold in the terrain coefficient dynamic calculation model and the transportation distance determination model respectively; The budget verification module is used to construct a project budget verification model based on the source data, the unique coding system, the terrain coefficient dynamic calculation model and the transportation distance determination model, and use the project budget verification model to realize the budget verification of the power grid project.

[0011] Furthermore, the rule parsing module is specifically used to: Processing each of the power grid projects based on a rule management framework and conditional judgment logic to obtain engineering specification data in each of the power grid projects; A semantic mapping algorithm is used to convert the engineering specification data into executable logical rules to obtain professional rules for each power grid project.

[0012] Furthermore, the data analysis module is specifically used to: Construct parameter association graphs, mapping engineering parameters, quota standards, and construction conditions into graph structure nodes; A graph convolutional network is used to deeply mine the node features of the graph structure nodes to identify abnormal correlations between parameters; By using a difference analysis algorithm and abnormal correlations between the parameters, a deviation report of the engineering data in each of the power grid projects is identified, and correction suggestions are provided. The report includes the deviation type, impact range and priority labeling.

[0013] Furthermore, the data decoding module is specifically used to: For tower model analysis, the model code is split into voltage segment, tower type segment and call height segment according to preset rules; The first digit code is mapped to the voltage level, where the first digit is 2 and the corresponding voltage level is 220kV, and the first digit is 1 and the corresponding voltage level is 110kV; The tower type is determined by the last letter, where the last letter is Z for a straight tower and J for a tension tower. The last digit is extracted and converted into a height value.

[0014] Furthermore, the system further comprises: A data monitoring module is used to monitor industry standard data in real time and extract key parameters of the industry standard data through natural language processing algorithms; A project evaluation module, used to evaluate the impact of the key parameters on historical power grid engineering projects to generate corresponding sensitivity analysis reports; The decision adjustment module is used to quantify the impact of the adjustment range of the key parameters on the total cost based on the sensitivity analysis report, and generate a professional difference analysis report to achieve the adjustment of engineering management decisions.

[0015] Compared with related technologies, the embodiment of the present application provides a method and system for engineering budget verification based on knowledge graph and dynamic rule learning, which greatly shortens the time required for traditional manual review through intelligent processing, thereby shortening the engineering budget review cycle and improving review efficiency; through systematic inspection, it can avoid hidden errors in manual review and improve the accuracy and reliability of review results; it transforms scattered engineering experience and industry standards into a structured, reusable professional knowledge base, providing an effective carrier for industry knowledge; based on accumulated engineering data, it provides intelligent analysis support for quota optimization and investment decisions, thereby improving the scientific nature of management decisions.

[0016] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 This is a flow chart of a method for checking engineering budget estimates based on knowledge graph and dynamic rule learning in the first embodiment of the present invention; Figure 2 This is a structural block diagram of the engineering budget verification system based on knowledge graph and dynamic rule learning in the second embodiment of the present invention.

[0018] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without making any creative efforts are within the scope of protection of this application.

[0020] Obviously, the drawings described below are merely examples or embodiments of the present application. Those skilled in the art can, without inventive effort, apply the present application to other similar scenarios based on these drawings. Furthermore, it is also understood that, although the effort involved in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, changes in design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as an insufficiency of the content disclosed in this application.

[0021] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0022] Unless otherwise defined, the technical or scientific terms involved in this application should have the usual meaning understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "the" and the like involved in this application do not indicate a quantitative limitation and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Example 1

[0023] See also Figure 1 , which shows a method for checking engineering budget estimates based on knowledge graph and dynamic rule learning in the first embodiment of the present invention. The method specifically includes steps S101 to S103: S101, analyzing in real time several professional rules of power grid projects, wherein the professional rules include terrain coefficient calculation rules, transportation distance determination rules, and quota standard matching rules; Furthermore, the step S101 specifically includes steps S1011 to S1012: S1011, processing each of the power grid projects based on a rule management framework and conditional judgment logic to obtain engineering specification data in each of the power grid projects; S1012: Using a semantic mapping algorithm, each of the engineering specification data is converted into executable logical rules to obtain professional rules for each of the power grid projects.

[0024] In implementation, a rule management framework is employed to support complex conditional logic (IF-THEN-ELSE rule chains) and to process specialized rules for power grid engineering in real time, including but not limited to terrain coefficient calculations and transport distance determination. The rule base utilizes a modular design and supports online updates and maintenance, ensuring that audit standards remain synchronized with the latest industry regulations.

[0025] S102, constructing a parameter association map based on a graph neural network anomaly detection model to identify abnormal data in the engineering data of each of the power grid projects, and locating the root cause data of cost deviations in each of the power grid projects through a multi-dimensional tracing algorithm; Furthermore, the step S102 specifically includes steps S1021 to S1023: S1021, constructing a parameter association graph, mapping engineering parameters, quota standards, and construction conditions into graph structure nodes; S1022, using a graph convolutional network to perform deep mining on node features of the graph structure nodes to identify abnormal correlations between parameters; S1023, identifying deviation reports of engineering data in each of the power grid projects through a difference analysis algorithm and abnormal correlations between the parameters, and providing correction suggestions, wherein the reports include deviation types, impact scopes, and priority labels.

[0026] In specific implementation, the anomaly detection model based on graph neural network (GNN) automatically identifies logical contradictions and unreasonable parameter combinations in engineering data by constructing parameter association maps.

[0027] Through the difference analysis algorithm and the use of multi-dimensional traceability technology, the root cause of cost deviation can be accurately located and targeted modification suggestions can be provided to auditors.

[0028] S103, decoding the unique coding system of each power grid project, parsing the tower model by hierarchically decomposing the coded character string to extract the voltage level, tower type representation, and call height value; Furthermore, the step S103 specifically includes steps S1031 to S1033: S1031, for tower model analysis, the model code is split into voltage segment, tower type segment and call height segment according to preset rules; S1032: Map the first digit code to a voltage level. A first digit of 2 corresponds to a voltage level of 220 kV, and a first digit of 1 corresponds to a voltage level of 110 kV. S1033, determine the tower type by the last letter, where the last letter is Z for a straight tower and the last letter is J for a tension tower, and extract the last digit and convert it into a height value.

[0029] In the specific implementation, a parameter parser is constructed to automatically decode the coding system unique to power grid projects. Taking tower model parsing as an example, the following processing flow is used: Taking the model parsing of transmission line towers as an example, the system decomposes the coded string hierarchically, intelligently extracting the engineering meaning of each segment. First, the model is segmented, identifying the first digit representing the voltage level. "2" corresponds to 220kV, and "1" corresponds to 110kV. The system then analyzes the characteristic letters in the model's tail. If a "Z" is included, it identifies a straight tower; if a "J" is included, it identifies a tension tower. Finally, the numerical portion is extracted and converted into a specific call height value.

[0030] S104, decomposing and constructing a terrain coefficient dynamic calculation model and a transportation distance determination model using the terrain coefficient calculation rule and the transportation distance determination rule, and setting a preset terrain threshold and distance threshold in the terrain coefficient dynamic calculation model and the transportation distance determination model, respectively; In the specific implementation, by building a professional calculation model, including: 1. Dynamic calculation model of terrain coefficient: mountain coefficient = 0.8 × mountain ratio + 0.5 × hill ratio; 2. Intelligent transportation distance determination model: Trucking distance = min(5km, route length / 2); The above model automatically adjusts the calculation parameters according to the actual project conditions to ensure the accuracy of the audit results.

[0031] S105 , constructing a project budget verification model based on the source data, the unique coding system, the terrain coefficient dynamic calculation model, and the transportation distance determination model, and implementing budget verification of the power grid project using the project budget verification model.

[0032] In this example, the system receives the foundation model code "2424B22" and automatically identifies the technical meaning of each code segment through data analysis. The first four digits "2424" are interpreted as a foundation length of 2.4 meters and a width of 2.4 meters. The letter "B" corresponds to a specific reinforcement specification, and the last two digits "22" indicate a standard pit depth of 2.2 meters. The system also automatically links the funding document to verify supporting parameters such as the foundation pit quantity.

[0033] Based on the structural parameters obtained through analysis, the calculation process is automatically triggered. Earthwork calculations utilize a 3D geometric algorithm, taking into account the actual shape of the foundation pit. Rebar calculations incorporate a dynamic adjustment factor, automatically integrating the project's topographical characteristics. All calculations maintain a complete formula traceability chain.

[0034] A multi-level early warning mechanism has been established. When the calculated concrete volume exceeds the set threshold for the capital contribution unit, a graded warning is automatically triggered. High-level warnings activate the root cause analysis module, automatically tracing possible causes, such as parameter entry errors and improper application of calculation rules, and providing corrective suggestions.

[0035] Using deep learning-based text understanding technology, the system automatically parses engineering description text. For specialized phrases like "crossing 220kV lines twice," the system accurately identifies key parameters such as crossing type and voltage level, automatically matching them to industry standard safety protection requirements. The parsed results are output in a structured data format to ensure consistency in subsequent processing.

[0036] Automatically verify the compliance of protective measures based on a pre-set knowledge base of wiring regulations. The verification process uses a rule-based reasoning engine to support combined judgment of complex conditions. For special working conditions, similar engineering cases can be automatically retrieved to provide reference solutions. Establish an engineering parameter association network to automatically check the geometric relationship between the crossing point and the adjacent tower. Use a three-dimensional spatial analysis algorithm to verify the compliance of the clearance requirements and consider the influence of dynamic factors such as conductor sag and wind deflection to ensure the engineering reliability of the analysis results. Specifically, by unifying the data processing pipeline, engineering data from different sources can be collaboratively analyzed: 1. Design drawing data: Extract geometric parameters such as basic structure dimensions through CAD parsing engine 2. Capital Proposal Document: Use NLP technology to automatically obtain technical parameters such as terrain scale 3. Construction Log: Verify construction parameters such as actual transport distances. This data cross-validation significantly improves the reliability of audit results.

[0037] Furthermore, the method further comprises: Real-time monitoring of industry standard data and extraction of key parameters of the industry standard data through natural language processing algorithms; Evaluate the impact of the key parameters on historical power grid engineering projects to generate corresponding sensitivity analysis reports; Based on the sensitivity analysis report, the impact of the adjustment range of the key parameters on the total cost is quantified, and a professional difference analysis report is generated to achieve adjustments in engineering management decisions.

[0038] During implementation, the system continuously monitors updates to industry standards and automatically identifies changes to key parameters using natural language processing. Adjustments to core parameters like the rock bolt aperture coefficient can accurately capture version differences and assess their impact.

[0039] A dual-track operation mechanism automatically generates a test environment for new rules while maintaining the operation of existing rules. Shadow testing verifies the applicability of new rules to ensure a smooth transition. Key parameter adjustments support multi-version comparative analysis to assist in engineering decision-making.

[0040] Automatically scan historical engineering projects to assess the potential impact of rule changes on cost indicators. Use sensitivity analysis to quantify the impact of key parameter adjustments on total cost and generate professional variance analysis reports to support project management decisions.

[0041] In summary, the engineering budget verification method based on knowledge graph and dynamic rule learning in the above embodiments of the present invention greatly shortens the time required for traditional manual review through intelligent processing, so that the engineering budget review cycle is shortened and the review efficiency is improved; through systematic inspection, it can avoid hidden errors in manual review and improve the accuracy and reliability of the review results; it transforms scattered engineering experience and industry standards into a structured and reusable professional knowledge base, providing an effective carrier for industry knowledge; based on accumulated engineering data, it provides intelligent analysis support for quota optimization and investment decisions, and improves the scientific nature of management decisions. Example 2

[0042] On the other hand, the present invention also proposes an engineering budget verification system based on knowledge graph and dynamic rule learning, please refer to Figure 2 , shown is a system for checking engineering budget estimates based on knowledge graph and dynamic rule learning in a second embodiment of the present invention, comprising: A rule parsing module 11 is used to parse out several professional rules of power grid projects in real time, wherein the professional rules include terrain coefficient calculation rules, transportation distance determination rules and quota standard matching rules; A data analysis module 12 is configured to construct a parameter association map based on a graph neural network anomaly detection model to identify anomalies in the engineering data of each power grid project and locate the root cause of cost deviations in each power grid project through a multi-dimensional tracing algorithm; The data decoding module 13 is used to decode the unique coding system of each power grid project, and extract the voltage level, tower type representation and call height value by hierarchically decomposing the coded string of the tower model; A model construction module 14 is configured to decompose and construct a terrain coefficient dynamic calculation model and a transportation distance determination model using the terrain coefficient calculation rule and the transportation distance determination rule, and to set a preset terrain threshold and distance threshold in the terrain coefficient dynamic calculation model and the transportation distance determination model, respectively; The budget verification module 15 is used to construct a project budget verification model based on the source data, the unique coding system, the terrain coefficient dynamic calculation model and the transportation distance determination model, and use the project budget verification model to implement budget verification of the power grid project.

[0043] Furthermore, the rule parsing module 11 is specifically used to: Processing each of the power grid projects based on a rule management framework and conditional judgment logic to obtain engineering specification data in each of the power grid projects; A semantic mapping algorithm is used to convert the engineering specification data into executable logical rules to obtain professional rules for each power grid project.

[0044] Furthermore, the data analysis module 12 is specifically configured to: Construct parameter association graphs, mapping engineering parameters, quota standards, and construction conditions into graph structure nodes; A graph convolutional network is used to deeply mine the node features of the graph structure nodes to identify abnormal correlations between parameters; By using a difference analysis algorithm and abnormal correlations between the parameters, a deviation report of the engineering data in each of the power grid projects is identified, and correction suggestions are provided. The report includes the deviation type, impact range and priority labeling.

[0045] Furthermore, the data decoding module 13 is specifically configured to: For tower model analysis, the model code is split into voltage segment, tower type segment and call height segment according to preset rules; The first digit code is mapped to the voltage level, where the first digit is 2 and the corresponding voltage level is 220kV, and the first digit is 1 and the corresponding voltage level is 110kV; The tower type is determined by the last letter, where the last letter is Z for a straight tower and J for a tension tower. The last digit is extracted and converted into a height value.

[0046] Furthermore, the system further comprises: A data monitoring module is used to monitor industry standard data in real time and extract key parameters of the industry standard data through natural language processing algorithms; A project evaluation module, used to evaluate the impact of the key parameters on historical power grid engineering projects to generate corresponding sensitivity analysis reports; The decision adjustment module is used to quantify the impact of the adjustment range of the key parameters on the total cost based on the sensitivity analysis report, and generate a professional difference analysis report to achieve the adjustment of engineering management decisions.

[0047] The functions or operation steps implemented when the above modules are executed are substantially the same as those in the above method embodiments and will not be repeated here.

[0048] An embodiment of the present invention provides an engineering budget verification system based on knowledge graph and dynamic rule learning. Its implementation principle and technical effects are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, please refer to the corresponding content in the aforementioned method embodiment.

[0049] 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.

[0050] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for checking engineering budget estimates based on knowledge graph and dynamic rule learning, characterized in that: The following steps are involved: Step 1: Real-time analysis of several professional rules for power grid projects, including terrain coefficient calculation rules, transportation distance determination rules, and quota standard matching rules; Step 2: Construct a parameter association map based on the graph neural network anomaly detection model to identify abnormal data in the engineering data of each power grid project, and locate the root cause of cost deviations in each power grid project through a multi-dimensional tracing algorithm; Step 3: Decode the unique coding system of each power grid project, and analyze the tower model by hierarchically decomposing the coded string to extract the voltage level, tower type representation, and call height value; Step 4: Decomposing and constructing a terrain coefficient dynamic calculation model and a transportation distance determination model using the terrain coefficient calculation rule and the transportation distance determination rule, and setting a preset terrain threshold and distance threshold in the terrain coefficient dynamic calculation model and the transportation distance determination model respectively; Step 5: Constructing a project budget verification model based on the source data, the unique coding system, the terrain coefficient dynamic calculation model, and the transportation distance determination model, and implementing budget verification of the power grid project using the project budget verification model.

2. The engineering budget verification method based on knowledge graph and dynamic rule learning according to claim 1 is characterized in that: The step one comprises: Processing each of the power grid projects based on a rule management framework and conditional judgment logic to obtain engineering specification data in each of the power grid projects; A semantic mapping algorithm is used to convert the engineering specification data into executable logical rules to obtain professional rules for each power grid project.

3. The engineering budget verification method based on knowledge graph and dynamic rule learning according to claim 1 is characterized in that: The second step includes: Construct parameter association graphs, mapping engineering parameters, quota standards, and construction conditions into graph structure nodes; A graph convolutional network is used to deeply mine the node features of the graph structure nodes to identify abnormal correlations between parameters; By using a difference analysis algorithm and abnormal correlations between the parameters, a deviation report of the engineering data in each of the power grid projects is identified, and correction suggestions are provided. The report includes the deviation type, impact range and priority labeling.

4. The engineering budget verification method based on knowledge graph and dynamic rule learning according to claim 1 is characterized in that: The step three includes: For tower model analysis, the model code is split into voltage segment, tower type segment and call height segment according to preset rules; The first digit code is mapped to the voltage level, where the voltage level corresponding to the first digit being 2 is 220kV, and the voltage level corresponding to the first digit being 1 is 110kV; The tower type is determined by the last letter, where the last letter is Z for a straight tower and J for a tension tower. The last digit is extracted and converted into a height value.

5. The engineering budget verification method based on knowledge graph and dynamic rule learning according to claim 1 is characterized in that: The method further comprises: Real-time monitoring of industry standard data and extraction of key parameters of the industry standard data through natural language processing algorithms; Evaluate the impact of the key parameters on historical power grid engineering projects to generate corresponding sensitivity analysis reports; Based on the sensitivity analysis report, the impact of the adjustment range of the key parameters on the total cost is quantified, and a professional difference analysis report is generated to achieve adjustments in engineering management decisions.

6. A system for checking engineering budget estimates based on knowledge graph and dynamic rule learning, characterized by: include: A rule parsing module is used to parse out several professional rules of power grid projects in real time, wherein the professional rules include terrain coefficient calculation rules, transportation distance determination rules and quota standard matching rules; A data analysis module is used to construct a parameter association map based on an anomaly detection model of a graph neural network to identify abnormal data in the engineering data of each of the power grid projects, and locate the root cause of cost deviations in each of the power grid projects through a multi-dimensional tracing algorithm; A data decoding module is used to decode the unique coding system of each power grid project, and to parse the tower model by hierarchically decomposing the coded string to extract the voltage level, tower type representation and call height value; a model construction module for decomposing and constructing a terrain coefficient dynamic calculation model and a transportation distance determination model by using the terrain coefficient calculation rule and the transportation distance determination rule, and setting a preset terrain threshold and distance threshold in the terrain coefficient dynamic calculation model and the transportation distance determination model respectively; The budget verification module is used to construct a project budget verification model based on the source data, the unique coding system, the terrain coefficient dynamic calculation model and the transportation distance determination model, and use the project budget verification model to realize the budget verification of the power grid project.

7. The engineering budget verification system based on knowledge graph and dynamic rule learning according to claim 6 is characterized in that: The rule parsing module is specifically used for: Processing each of the power grid projects based on a rule management framework and conditional judgment logic to obtain engineering specification data in each of the power grid projects; A semantic mapping algorithm is used to convert the engineering specification data into executable logical rules to obtain professional rules for each power grid project.

8. The engineering budget verification system based on knowledge graph and dynamic rule learning according to claim 6 is characterized in that: The data analysis module is specifically used for: Construct parameter association graphs, mapping engineering parameters, quota standards, and construction conditions into graph structure nodes; A graph convolutional network is used to deeply mine the node features of the graph structure nodes to identify abnormal correlations between parameters; By using a difference analysis algorithm and abnormal correlations between the parameters, a deviation report of the engineering data in each of the power grid projects is identified, and correction suggestions are provided. The report includes the deviation type, impact range and priority labeling.

9. The engineering budget verification system based on knowledge graph and dynamic rule learning according to claim 6 is characterized in that: The data decoding module is specifically used for: For tower model analysis, the model code is split into voltage segment, tower type segment and call height segment according to preset rules; The first digit code is mapped to the voltage level, where the voltage level corresponding to the first digit being 2 is 220kV, and the voltage level corresponding to the first digit being 1 is 110kV; The tower type is determined by the last letter, where the last letter is Z for a straight tower and J for a tension tower. The last digit is extracted and converted into a height value.

10. The engineering budget verification system based on knowledge graph and dynamic rule learning according to claim 6 is characterized in that: The system further comprises: A data monitoring module is used to monitor industry standard data in real time and extract key parameters of the industry standard data through natural language processing algorithms; A project evaluation module, used to evaluate the impact of the key parameters on historical power grid engineering projects to generate corresponding sensitivity analysis reports; The decision adjustment module is used to quantify the impact of the adjustment range of the key parameters on the total cost based on the sensitivity analysis report, and generate a professional difference analysis report to achieve the adjustment of engineering management decisions.

Images