Power grid engineering cost document processing method, device, equipment, medium and product

Abstract

This application relates to a method, apparatus, equipment, medium, and product for processing power grid engineering cost documents, and pertains to the field of intelligent cost analysis technology for power grid engineering, which can improve the comprehensiveness and efficiency of cost review. The method includes: determining cost-related information based on the cost documents of the target power grid engineering; determining target engineering feature information that characterizes the decisive features of the target power grid engineering based on the cost-related information; for each review dimension, determining the review benchmark information corresponding to the review dimension based on the target engineering feature information and extracting target cost information from the target engineering feature information; comparing the review benchmark information corresponding to the review dimension with the target cost information to obtain the cost review result of the target power grid engineering in the review dimension; and displaying the cost review results and cost-related information for each review dimension in a visual interface.

Description

Technical Field

[0001] This application relates to the field of intelligent analysis technology for power grid engineering costs, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for processing power grid engineering cost documents. Background Technology

[0002] For power grid engineering cost documents, traditional techniques mainly rely on manual review by experts. This involves manually checking the submitted cost documents, separating core cost data such as the quantity of each sub-item, quota fees, and unit prices of equipment and materials, and then subjectively judging the reasonableness of this data based on personal experience to arrive at a review conclusion.

[0003] However, this traditional technology is inefficient, susceptible to subjective influence from reviewers, and has poor consistency in review conclusions. It is also prone to failing to detect deep-seated problems such as logical contradictions in cost calculation and abnormal special unit prices due to the limitations of personal experience. Moreover, traditional technology lacks the ability to perform multi-dimensional quantitative comparisons and intelligent anomaly identification of power grid engineering cost documents, making it difficult to adapt to the cost control needs of high-quality development of power grid construction. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for processing power grid engineering cost documents to address the aforementioned technical problems.

[0005] Firstly, this application provides a method for processing power grid engineering cost documents, including:

[0006] Based on the cost documents of the target power grid project, determine the cost-related information associated with the project cost;

[0007] Based on the cost-related information, determine the target project characteristic information that characterizes the decisive features of the target power grid project;

[0008] For each review dimension, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0009] The cost review results and related cost information for each of the aforementioned review dimensions are displayed in the visual interface.

[0010] In one embodiment, the review dimensions include a similar project analogy review dimension;

[0011] The step of determining the review benchmark information corresponding to the review dimension based on the target project feature information and extracting target cost information from the target project feature information, comparing the review benchmark information corresponding to the review dimension with the target cost information, and obtaining the cost review result of the target power grid project in the review dimension includes:

[0012] In each completed power grid project, multiple similar project samples are identified; the comprehensive similarity between the target project feature information and the similar project sample feature information is greater than a similarity threshold.

[0013] The validity of each of the aforementioned similar engineering samples is verified; the validity verification includes verification of the number of samples and verification of the time span.

[0014] If any of the aforementioned similar engineering samples fail the validity check, an alert is triggered indicating insufficient historical data.

[0015] If each of the similar project samples passes the validity verification, target cost information is extracted from the target project feature information, and the cost information samples of each of the similar project samples are compared with the target cost information to obtain the cost review result of the target power grid project; the cost information samples of the similar project samples represent the actual cost of the similar project samples.

[0016] In one embodiment, the comprehensive similarity between the target engineering feature information and the engineering feature information samples of the similar engineering samples is determined through the following steps:

[0017] For each similar engineering sample,

[0018] Obtain the cosine similarity between the target project feature information and the project feature information samples of the similar project samples, and obtain the weighted Canberra distance between the target project feature information and the project feature information samples of the similar project samples;

[0019] Based on the weight information, the cosine similarity and the weighted Canberra distance are summed to obtain the comprehensive similarity between the target project feature information and the project feature information samples of the similar project samples; the weight information is learned by a random forest model based on the cost-related information of multiple completed power grid projects; the cost-related information of the completed power grid projects represents the decisive features and actual cost of the completed power grid projects.

[0020] In one embodiment, the review dimensions include the cost control line review dimension;

[0021] The step of determining the review benchmark information corresponding to the review dimension based on the target project feature information and extracting target cost information from the target project feature information, comparing the review benchmark information corresponding to the review dimension with the target cost information, and obtaining the cost review result of the target power grid project in the review dimension includes:

[0022] Based on cost control information related to cost control, the first cost control line information is determined; the cost control information is obtained from cost standard documents.

[0023] Multiple matching project samples are identified in each completed power grid project, and cost information samples of each matching project sample are obtained; the matching degree between the engineering feature information samples of the matching project samples and the feature information of the target project reaches the target condition.

[0024] Based on the cost information samples of each of the matched project samples, determine the second cost control line information;

[0025] Extract the target cost information from the target project feature information;

[0026] The target cost information is compared with the first cost control line information and the second cost control line information to obtain the cost review result of the target power grid project.

[0027] In one embodiment, the review dimensions include a rate compliance review dimension;

[0028] The step of determining the review benchmark information corresponding to the review dimension based on the target project feature information and extracting target cost information from the target project feature information, comparing the review benchmark information corresponding to the review dimension with the target cost information, and obtaining the cost review result of the target power grid project in the review dimension includes:

[0029] Based on the rate standard document, standard rates for multiple dimensions are determined, and target cost information representing multiple rate rates for the target power grid project is extracted from the target project feature information.

[0030] The target cost information is compared with each of the standard rates to obtain the cost review results that indicate whether each of the fee rates complies with regulations.

[0031] In one embodiment, determining the target project characteristic information that characterizes the decisive features of the target power grid project based on the cost-related information includes:

[0032] Feature extraction is performed on the cost-related information to obtain engineering feature information;

[0033] If the engineering feature information is missing or the cost document of the target power grid project is not fully parsed, the engineering feature information shall be supplemented based on the cost document.

[0034] Secondly, this application also provides a power grid engineering cost document processing device, comprising:

[0035] The first determining module is used to determine cost-related information related to the project cost based on the cost documents of the target power grid project.

[0036] The second determining module is used to determine target project feature information that characterizes the decisive features of the target power grid project based on the cost-related information.

[0037] The review module is used to determine the review benchmark information corresponding to the review dimension and extract the target cost information from the target project feature information for each review dimension, and compare the review benchmark information corresponding to the review dimension with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0038] The display module is used to display the cost review results and cost-related information for each of the review dimensions in a visual interface.

[0039] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0040] Based on the cost documents of the target power grid project, determine the cost-related information associated with the project cost;

[0041] Based on the cost-related information, determine the target project characteristic information that characterizes the decisive features of the target power grid project;

[0042] For each review dimension, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0043] The cost review results and related cost information for each of the aforementioned review dimensions are displayed in the visual interface.

[0044] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0045] Based on the cost documents of the target power grid project, determine the cost-related information associated with the project cost;

[0046] Based on the cost-related information, determine the target project characteristic information that characterizes the decisive features of the target power grid project;

[0047] For each review dimension, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0048] The cost review results and related cost information for each of the aforementioned review dimensions are displayed in the visual interface.

[0049] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0050] Based on the cost documents of the target power grid project, determine the cost-related information associated with the project cost;

[0051] Based on the cost-related information, determine the target project characteristic information that characterizes the decisive features of the target power grid project;

[0052] For each review dimension, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0053] The cost review results and related cost information for each of the aforementioned review dimensions are displayed in the visual interface.

[0054] The aforementioned method, apparatus, computer equipment, computer-readable storage medium, and computer program product for processing power grid engineering cost documents determine cost-related information based on the cost documents of the target power grid engineering, determine target engineering feature information that characterizes the decisive features of the target power grid engineering based on the cost-related information, determine the review benchmark information corresponding to the review dimension based on the target engineering feature information and extract target cost information from the target engineering feature information for each review dimension, compare the review benchmark information corresponding to the review dimension with the target cost information to obtain the cost review result of the target power grid engineering in the review dimension, and display the cost review result and cost-related information of each review dimension in a visualization interface. Compared to traditional technologies, this application automates the extraction of cost information and conducts intelligent comparative review, eliminating the need for manual item-by-item verification and significantly shortening the review cycle. It establishes multi-dimensional review benchmarks based on the target project's characteristic information. By comparing these benchmarks with the target cost information, it not only achieves quantitative comparison of cost information and intelligent anomaly detection, improving the comprehensiveness of cost review, but also avoids conclusion biases caused by subjective factors, ensuring consistent review standards for similar projects. Furthermore, this application provides a visual interface to intuitively display the multi-dimensional review results and cost information, facilitating managers' understanding of the project cost composition and anomalies, and providing data support for cost control. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is an application environment diagram of a power grid engineering cost document processing method in one embodiment;

[0057] Figure 2 This is a flowchart illustrating a method for processing power grid engineering cost documents in one embodiment;

[0058] Figure 3 This is a schematic diagram of the architecture of an intelligent assessment and analysis system for the rationality of power grid engineering costs in one embodiment;

[0059] Figure 4 Here is a block diagram of the power grid engineering cost document processing system in another embodiment;

[0060] Figure 5 This is a structural block diagram of a power grid engineering cost document processing device in one embodiment;

[0061] Figure 6This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0063] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various objects, but these objects are not limited by these terms. These terms are only used to distinguish the first object from the second object. The term "comprising" and any variations thereof, as used in this application, are intended to cover non-exclusive inclusion.

[0064] The power grid engineering cost document processing method provided in this application embodiment can be applied to, for example... Figure 1 The application environment is illustrated. The terminal communicates with the server via a network. The data storage system stores the data the server needs to process. The data storage system can be integrated onto the server or located on the cloud or other network servers. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, projection devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted displays, etc. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0065] In one exemplary embodiment, such as Figure 2 As shown, a method for processing power grid engineering cost documents is provided, which can be applied to... Figure 1 The terminal in the middle integrates such as Figure 3 The system architecture shown is a smart assessment and analysis system for the rationality of power grid project costs. This system architecture is driven by a cost document intelligent parsing and cleaning engine, an engineering feature vector generator, and a multi-dimensional review engine. The method may include the following steps:

[0066] Step S201: Based on the cost documents of the target power grid project, determine the cost-related information associated with the project cost.

[0067] The cost documents for the target power grid project can reflect cost-related information such as the project's workload, various expenses, and other costs.

[0068] For example, by parsing the cost estimate document of the target power grid project, target data such as the project quantity, various costs, and other expenses are extracted from the cost estimate document. The target data undergoes preprocessing such as text-to-numerical conversion, unit conversion (e.g., unifying "tons" to "t"), outlier detection (e.g., unit prices of 0 or negative values), and missing value imputation to obtain cleaned data. The cleaned data is stored in a database, constructing a hierarchical structure of "project-cost-quantity-equipment unit price" to ensure data traceability.

[0069] In one specific implementation of this embodiment, the cost document for the target power grid project is a power grid project budget document, and this budget document is submitted to the terminal in Bowei electronic format (.zwzj). After receiving the power grid project budget document, the terminal uses BCL (Budget Estimate Calculation Language) file parsing technology and the cost document data matching rules built into the cost document intelligent parsing and cleaning engine module to parse the document tabs such as building engineering, installation engineering, and labor, materials, and machinery, and uses data extraction templates to map cost-related information related to the project cost to the internal data model to obtain the target data.

[0070] In another specific implementation of this embodiment, the cost document for the target power grid project is a power grid project budget document, and this budget document is submitted to the terminal in Excel format. After receiving the power grid project budget document, the terminal adopts a dual-mode parsing strategy based on named region recognition and cell position pattern learning. Through the "header pattern library" of conventional cost documents built into the cost document intelligent parsing and cleaning engine module (such as recognizing the arrangement pattern of keywords such as "serial number", "project name", "unit", "total", "price excluding tax"), it automatically locates key sheets (worksheets) such as the bill of quantities, fees, and fee calculation tables for sub-items of the project, and extracts row and column data. For complex formats such as merged cells and multi-level headers, a recursive cell splitting and parent-child relationship reconstruction algorithm is used for structured transformation to obtain the target data.

[0071] Among them, such as Figure 3The architecture of the intelligent assessment and analysis system for the rationality of power grid engineering costs, as shown, can be divided into three core layers from bottom to top: the data layer, the analysis layer, and the results layer. The analysis layer is the intelligent core of the system, primarily responsible for extracting engineering features and performing multi-dimensional comparisons. This layer comprises two main branches: engineering indicators and reference indicators. Engineering indicators, based on the correlation between design parameters, engineering quantities, and quotas, automatically extract cost and quantity data from cost documents to calculate key cost indicators (such as unit cost and material consumption). Reference indicators, on the other hand, establish reference benchmarks for unit indicators (such as standard rates and experience-based cost indicators) based on design parameters, expert experience, and historical data. Both are calculated and compared using logical rules to determine differences and trigger bid evaluation reviews. Furthermore, the analysis layer supports auxiliary functions such as configuring statistical analysis tables and benchmarking rules, providing upper layers with flexible rule definition capabilities. The results layer is responsible for transforming the assessment results from the analysis layer into an intuitive user experience. It presents complex comparative data to technical and economic experts in the form of charts, lists, and other formats through visualized data viewing, intelligent review result display (such as automatic extraction and comparison of cost indicators, and benchmarking rule execution results), and in-depth technical control audit library. Users can review the quantity and cost of sub-items at this layer, and verify other costs calculated based on the ontology, achieving rapid human-machine collaborative review.

[0072] Step S202: Based on cost-related information, determine the target project characteristic information that characterizes the decisive features of the target power grid project.

[0073] The target project characteristic information may include multiple key parameters that can characterize the decisive features of the target power grid project.

[0074] For example, the project type of the target power grid project is first determined, and then key parameters are extracted from cost-related information based on the project type. The extracted key parameters are then used to construct the project feature vector (i.e., the target project feature information).

[0075] In one specific implementation of this embodiment, the main project types of power grid engineering include substation engineering, overhead line engineering, cable line engineering, and communication engineering. The key parameter extraction method does not rely on external drawings and technical documents; it only constructs data based on the cost estimates. The key parameter extraction methods for each project type are as follows:

[0076] Substation engineering: Automatically extract key parameters such as main transformer capacity, number of main transformers, high-voltage outgoing circuits, medium-voltage outgoing circuits, low-voltage outgoing circuits, voltage level, configuration method, type of power distribution equipment, and capacitor capacity from the building engineering and installation engineering sections.

[0077] Overhead line engineering: Automatically extract key parameters such as the number of poles and towers, conductor specifications, number of circuits, line length, and terrain proportion from the sub-projects of foundation engineering, pole and tower engineering, grounding engineering, and stringing engineering.

[0078] Cable line engineering: Automatically extract key parameters such as cable cross-section, cable trench length, jacking pipe length, and path length from the building engineering and installation engineering sections.

[0079] Telecommunications Engineering: Automatically extract key parameters such as optical cable length and pipeline type from the main body division.

[0080] Distribution network engineering: Automatically extract key parameters such as the length of medium-voltage distribution network cable lines, the length of medium-voltage distribution network overhead lines, the capacity of low-voltage distribution network, and the number of automated switch boxes from the building engineering and installation engineering sections.

[0081] Step S203: For each review dimension, determine the review benchmark information corresponding to the review dimension based on the target project feature information and extract the target cost information from the target project feature information. Compare the review benchmark information corresponding to the review dimension with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0082] Among these, reviewing the benchmark information can characterize the reasonable cost range of the target power grid project. The cost review results can characterize whether the target cost information in the cost documents of the target power grid project is reasonable.

[0083] For example, for each review dimension, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. Then, the review benchmark information corresponding to the review dimension is compared and matched with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0084] Step S204: Display the cost review results and cost-related information for each review dimension in the visualization interface.

[0085] For example, a visual, interactive review application for review experts is implemented using a JAVA architecture. This application primarily includes review database selection, review data viewing, and review data referencing. Specifically, based on the characteristics of the submitted cost documents and the target project's feature information, a database compatible with the target power grid project is matched from multiple backend databases. Alternatively, in response to a review expert's database switching operation, the review database is switched to the database corresponding to that switching operation. The databases include an engineering sample database, a standard rate database, and a document control line database. The engineering sample database stores cost information samples, project feature information samples, and detailed cost samples for completed power grid projects. The standard rate database stores standard rates across multiple dimensions. The document control line database stores static standard control line benchmarks. A visual interactive mode is used, displaying the cost review results and related cost information for each review dimension on a visual interface, facilitating quick data error retrieval and professional feedback from review experts. Furthermore, the terminal's system architecture also supports the introduction of system-assessed defects to form review minutes data, enabling rapid document review.

[0086] In the above-mentioned method for processing power grid engineering cost documents, cost-related information related to the project cost is determined based on the cost documents of the target power grid project. Based on the cost-related information, target project feature information that characterizes the decisive features of the target power grid project is determined. For each review dimension, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension. The cost review result and cost-related information of each review dimension are displayed in a visualization interface. Compared to traditional technologies, this embodiment automates the extraction of cost information and performs intelligent comparative review, eliminating the need for manual item-by-item verification and significantly shortening the review cycle. It establishes multi-dimensional review benchmarks based on the target project's characteristic information. By comparing these benchmarks with the target cost information, it not only achieves quantitative comparison of cost information and intelligent anomaly detection, improving the comprehensiveness of cost reviews, but also avoids conclusion biases caused by subjective factors, ensuring consistent review standards for similar projects. Furthermore, this embodiment visually displays the multi-dimensional review results and cost information through a graphical interface, allowing managers to intuitively grasp the composition of project costs and anomalies, providing data support for cost control.

[0087] In an exemplary embodiment, the review dimensions include a similar project analogy review dimension; in step S203, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension, which may include:

[0088] In each completed power grid project, multiple similar project samples are identified; the comprehensive similarity between the target project's feature information and the similar project samples' feature information is greater than a similarity threshold; each similar project sample undergoes validity verification; validity verification includes sample quantity verification and time span verification; if any similar project sample fails the validity verification, an early warning is triggered indicating insufficient historical data; if each similar project sample passes the validity verification, target cost information is extracted from the target project's feature information, and the cost information samples of each similar project sample are compared with the target cost information to obtain the cost review result of the target power grid project; the cost information samples of the similar project samples represent the actual cost of the similar project samples.

[0089] For example, the comprehensive similarity between historical project feature information and target project feature information of each completed power grid project is obtained; similar project samples with a comprehensive similarity greater than a similarity threshold (e.g., 0.75) are selected from each completed power grid project. The number of similar project samples can be limited to 5 to 10. Each similar project sample undergoes validity verification; validity verification includes sample quantity verification and time span verification; sample quantity verification includes whether the number of each similar project sample is greater than or equal to the target number (e.g., ≥3); time span verification includes whether the time span between each similar project sample covers a preset time period (e.g., the past 5 years). If any similar project sample fails the validity verification, an early warning is triggered indicating insufficient historical data; if any similar project sample passes the validity verification, target cost information is extracted from the target project feature information, and the cost information samples of each similar project sample are compared with the target cost information to obtain the cost review result of the target power grid project; the cost information samples of the similar project samples represent the actual cost of the similar project samples.

[0090] In an exemplary embodiment, the comprehensive similarity between the target engineering feature information and the engineering feature information samples of similar engineering samples can be determined through the following steps:

[0091] For each similar project sample, the cosine similarity between the target project feature information and the similar project feature information samples is obtained, as well as the weighted Canberra distance between the target project feature information and the similar project feature information samples. Based on the weight information, the cosine similarity and the weighted Canberra distance are summed to obtain the comprehensive similarity between the target project feature information and the similar project feature information samples. The weight information is learned by a random forest model based on the cost-related information of multiple completed power grid projects. The cost-related information of completed power grid projects represents the decisive characteristics and actual cost of the completed power grid projects.

[0092] For example, referring to Equation (1), the target engineering feature information is calculated using a fusion algorithm of cosine similarity and weighted Canberra distance. Engineering feature information samples with similar engineering samples Overall similarity .

[0093] (1)

[0094] In the formula, The weight information in the weight matrix. ∈[0.7,0.85], The cosine similarity between the target project's feature information and similar project feature information samples is given. The weighted Canberra distance between the target project feature information and similar project feature information samples.

[0095] In an exemplary embodiment, the review dimensions include the cost control line review dimension; in step S203, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension, which may include:

[0096] Based on cost control information related to cost control, the first cost control line is determined; the cost control information is obtained from cost standard documents; multiple matching project samples are identified in each completed power grid project, and cost information samples of each matching project sample are obtained; the matching degree between the engineering feature information samples of the matching project samples and the feature information of the target project meets the target condition; based on the cost information samples of each matching project sample, the second cost control line is determined; target cost information is extracted from the feature information of the target project; the target cost information is compared with the first cost control line information and the second cost control line information to obtain the cost review result of the target power grid project.

[0097] For example, semantic text understanding technology is used to extract cost control information related to cost control from official documents representing cost standards (i.e., cost standard documents). This cost control information is then categorized by subject name, province, etc., and stored in a standard control line library to form the first cost control line information (which can serve as a rigid constraint at the policy level). Multiple matching project samples are identified in each completed power grid project, and cost information samples for each matching project sample are obtained. The matching degree between the engineering feature information samples of the matching project samples and the feature information of the target project reaches the target value. Based on the cost information samples of each matching project sample, combined with a time factor, a second cost control line information is calculated using a normal distribution (e.g., by introducing a time factor to assign lower weights to the cost information of older matching project samples, or by making time-sensitive corrections to the cost information of more recent matching project samples, so that the cost control line better reflects the current cost level). The target cost information is extracted from the characteristic information of the target project; the target cost information is compared with the first cost control line information and the second cost control line information according to business characteristics such as voltage level, construction nature, and number of circuits to obtain the cost review result of the target power grid project.

[0098] In an exemplary embodiment, the review dimensions include a rate compliance review dimension; in step S203, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension, which may include:

[0099] Based on the rate standard document, standard rates for multiple dimensions are determined, and target cost information representing multiple fee rates for the target power grid project is extracted from the target project characteristic information. The target cost information is compared with each standard rate to obtain the cost review results representing whether each fee rate complies with the regulations.

[0100] For example, a standard rate library is constructed. The standard rate library is constructed according to project type, region, and voltage level based on official documents, budget quotas, and policy documents (i.e., rate standard documents). The standard rate library includes standard rates in multiple dimensions (such as winter and rainy season construction rates, provident fund rates, tax rates, etc.). Various fee rates are extracted from the characteristic information of the target project. Each fee rate is matched and compared with the standard rates in each dimension of the standard rate library to obtain the matching comparison results that indicate whether the fee rates comply with the regulations (i.e., the cost review results of the target power grid project).

[0101] In an exemplary embodiment, step S202, determining the target project characteristic information that characterizes the decisive features of the target power grid project based on cost-related information, may include:

[0102] Feature extraction is performed on cost-related information to obtain project feature information; if the project feature information is missing or the cost document of the target power grid project is not fully parsed, the project feature information is supplemented based on the cost document.

[0103] For example, feature extraction is performed on cost-related information to obtain project feature information. If there are missing features or incomplete parsing of the cost documents for the target power grid project, a completion algorithm based on cost ratio back-calculation is used to complete the target features in the project feature information.

[0104] In one specific implementation of this embodiment, if the main transformer capacity is not directly filled in for engineering feature information, the capacity is inferred from the ratio of "main transformer equipment cost" to "equipment purchase cost" and the historical statistical model.

[0105] Power grid construction plays a vital supporting role in economic development and social stability. With the accelerated pace of regional power grid construction and the continuous increase in project costs, project technical and economic reviews, as a key control link in the construction process, directly affect the overall quality level and cost control capabilities of power grid projects in terms of accuracy, timeliness, and efficiency.

[0106] Currently, the review of power grid project costs mainly relies on manual operations by review experts. Under the current model, experts need to check each submitted electronic cost document one by one, manually extracting key information, including the quantities of each item of work, quota numbers, fee standards, pricing basis, equipment unit prices, and material prices, and then judge the reasonableness of the project cost based on their personal experience. This type of review method has the following three main limitations:

[0107] First, there is a significant bottleneck in manual efficiency. Experts need to sift through cost estimation Excel spreadsheets or various electronic files, copying and pasting key data into their own spreadsheets. Processing a single 110kV substation project budget typically takes 3 to 5 working days. Because the data is scattered across multiple sheet tabs or different cost estimation documents, omissions or data entry errors are highly likely during the information integration process.

[0108] Second, subjective judgments lead to significant differences in conclusions. Influenced by factors such as personal experience and cognitive habits, different experts often reach different conclusions on the same project, with a deviation rate of 15% to 20%. At the same time, limited by accumulated experience, some deep-seated problems, such as contradictions in fee calculation logic and abnormal unit prices of special equipment, are difficult to detect in a timely manner.

[0109] Third, the level of intelligence is low. The existing review model lacks the ability to quantitatively compare data from similar historical projects, and it also lacks a dynamically adjusted cost benchmark. At the same time, it fails to effectively transform review rules into an executable logic engine, making it difficult to automatically identify deep-seated cost anomalies, and limiting the overall depth and comprehensiveness of the analysis.

[0110] In this regard, in one exemplary embodiment, such as Figure 4 As shown, a power grid engineering cost document processing system is also provided. This system may include: a cost document intelligent parsing and cleaning engine module 401, an engineering feature vector generator module 402, a multi-dimensional review engine module 403, and a review visualization module 404. Wherein:

[0111] The intelligent cost document parsing and cleaning engine module 401 is used to determine cost-related information related to the project cost based on the cost documents of the target power grid project.

[0112] For example, the intelligent cost document parsing and cleaning engine module 401 first receives the cost document of the target power grid project, supporting two mainstream formats: Excel and Bowei electronic files (.zwzj / .BDY3). For Excel files, the module adopts a dual-mode parsing strategy based on named region recognition and cell position pattern learning. It automatically locates key worksheets through the built-in "header pattern library" and uses recursive cell splitting and parent-child relationship reconstruction algorithms to process merged cells and multi-level headers, converting unstructured data into structured row data. For Bowei files, it directly maps to the internal data model based on BCL file parsing technology and built-in data matching rules. Subsequently, the module cleans and standardizes the extracted raw data, including text-to-numerical conversion, unit conversion, outlier detection, and missing value imputation, ultimately constructing a hierarchical cost-related information structure of "project-cost-quantity-equipment unit price" and storing it in the database.

[0113] The engineering feature vector generator module 402 is used to extract features from cost-related information to obtain engineering feature information; when there are missing engineering feature information or the parsing of the cost document of the target power grid project is incomplete, the engineering feature information is supplemented according to the cost document.

[0114] For example, the engineering feature vector generator module 402 extracts engineering feature information from the cleaned cost-related information. Based on the power grid project type (substation, overhead line, cable line, communication, distribution network), the module automatically identifies and extracts key design parameters, such as the main transformer capacity and number of circuits for substation projects, and the number of towers and conductor specifications for overhead lines. For incomplete or missing features, the module uses a cost-percentage-based completion algorithm, for example, using the ratio of main transformer equipment cost to equipment purchase cost to infer the main transformer capacity, thereby ensuring the integrity of the engineering feature vector. The final generated target engineering feature vector contains all the key information for subsequent review and is passed to the multi-dimensional review engine module 403.

[0115] The multi-dimensional review engine module 403 is used to determine the review benchmark information corresponding to the review dimension based on the target project feature information and extract the target cost information from the target project feature information for each review dimension. It compares the review benchmark information corresponding to the review dimension with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0116] For example, the multi-dimensional review engine module 403, based on the received target project feature vector, conducts reviews from three dimensions: rate compliance, analogy with similar projects, and cost control line. Each dimension involves dynamic comparison with external benchmark data. Specifically:

[0117] Regarding the rate compliance review dimension, the multi-dimensional review engine module 403 first retrieves multiple standard rates (such as winter and rainy season construction fees, housing provident fund, taxes, etc.) that match the target project's voltage level, construction nature, and year from the built-in dynamic standard rate list. Then, it extracts the corresponding actual fee rates from the target project's characteristic information, compares the two item by item, automatically marks rate items that exceed the reasonable fluctuation range, and generates the rate compliance review results.

[0118] For the comparative review of similar projects, the multi-dimensional review engine module 403 first filters similar project samples from the historical project database. A hybrid similarity algorithm, combining cosine similarity and weighted Canberra distance, is used to calculate the overall similarity. The weight α (0.7~0.85) is derived from a random forest model trained on historical data to ensure robustness against outliers. A similarity threshold S_min=0.75 is set, and Top-N (N=5~10) projects are selected as candidate samples for validity verification: the number of samples must be ≥3 and the time span must cover the most recent 5 years; otherwise, a "historical data insufficiency warning" is triggered. After verification, the module compares the cost information of the target project (such as main body cost and comprehensive unit price) with the corresponding cost information of each sample, analyzes the difference, and forms a comparative review conclusion.

[0119] Regarding the cost control line review dimensions, the multi-dimensional review engine module 403 comprehensively utilizes two types of control lines for dual verification. The first type of control line comes from official documents, automatically extracting cost control standards (such as the upper limit of the main body cost and the dynamic investment control line) issued by the provincial network company through semantic text understanding technology, and storing them in categories by region and voltage level. The second type of control line is the dynamic cost control line, calculated based on multiple historical project samples that highly match the characteristics of the target project, using a normal distribution combined with a year factor. The module compares the actual cost information of the target project with these two types of control lines respectively, identifies items that exceed the control lines, and outputs the control line review results.

[0120] The review visualization module 404 is used to display the cost review results and cost-related information for each review dimension in the visualization interface.

[0121] For example, the review visualization module 404 integrates all review results with the original cost information and implements an easily interactive graphical interface using a JAVA architecture. Experts can manually switch review libraries (the system automatically matches the optimal library based on project characteristics by default) to view the parsed cost data and the comparison results of each dimension side by side, with differences highlighted. The system also supports directly importing selected assessment deficiencies into the review minutes, enabling rapid archiving of review conclusions.

[0122] This embodiment proposes a power grid engineering cost document processing system. By introducing digital parsing and intelligent analysis methods, it can automatically complete in-depth analysis and multi-dimensional comparison of cost documents, significantly reducing the workload of manual screening and review, effectively shortening the review cycle. Furthermore, by integrating a multi-dimensional review engine model, it performs cross-validation from multiple dimensions such as project quantity, fee standards, and material prices, reducing biases and omissions caused by subjective judgment and ensuring more objective and reliable review conclusions, thereby significantly improving the efficiency and accuracy of technical and economic reviews. Based on this, the system uses a review visualization module to display the cost review results and related cost information for each review dimension in real time on a visual interface, generating detailed cost review result reports and intuitively presenting abnormal data and discrepancies, providing clear decision support for cost management personnel. In addition, the system constructs a technical and economic key point experience base (such as a two-dimensional relational database), transforming the implicit review experience of experts into quantifiable and reusable rules and models (such as a multi-dimensional review engine model), and supports continuous iterative updates, achieving effective accumulation and inheritance of experience. In summary, the application of this system not only helps to strengthen cost control and investment efficiency of individual projects, but also promotes the standardization and refinement of the power grid project cost review process, providing solid technical support for improving the overall cost management level of the industry.

[0123] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0124] Based on the same inventive concept, this application also provides a power grid engineering cost document processing device for implementing the above-mentioned power grid engineering cost document processing method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more power grid engineering cost document processing device embodiments provided below can be found in the limitations of the power grid engineering cost document processing method described above, and will not be repeated here.

[0125] In one exemplary embodiment, such as Figure 5 As shown, a power grid engineering cost document processing device is provided, comprising the following modules:

[0126] The first determining module 501 is used to determine cost-related information related to the project cost based on the cost documents of the target power grid project.

[0127] The second determining module 502 is used to determine the target project characteristic information that characterizes the decisive features of the target power grid project based on cost-related information.

[0128] The review module 503 is used to determine the review benchmark information corresponding to the review dimension based on the target project feature information and extract the target cost information from the target project feature information for each review dimension. It compares the review benchmark information corresponding to the review dimension with the target cost information to obtain the cost review result of the target power grid project in the review dimension.

[0129] The display module 504 is used to display the cost review results and cost-related information for each review dimension in a visual interface.

[0130] In an exemplary embodiment, the review dimension includes a similar project analogy review dimension. The review module 503 is further configured to identify multiple similar project samples among the various completed power grid projects; the comprehensive similarity between the target project feature information and the project feature information samples of the similar project samples is greater than a similarity threshold; perform validity verification on each similar project sample; the validity verification includes sample quantity verification and time span verification; if each similar project sample fails the validity verification, trigger an early warning indicating insufficient historical data; if each similar project sample passes the validity verification, extract target cost information from the target project feature information, compare the cost information samples of each similar project sample with the target cost information to obtain the cost review result of the target power grid project; the cost information samples of the similar project samples represent the actual cost of the similar project samples.

[0131] In an exemplary embodiment, the review module 503 is further configured to, for each similar project sample, obtain the cosine similarity between the target project feature information and the engineering feature information samples of the similar project samples, and obtain the weighted Canberra distance between the target project feature information and the engineering feature information samples of the similar project samples; based on the weight information, perform a weighted summation of the cosine similarity and the weighted Canberra distance to obtain the comprehensive similarity between the target project feature information and the engineering feature information samples of the similar project samples; the weight information is learned by a random forest model based on the cost-related information of multiple completed power grid projects; the cost-related information of completed power grid projects characterizes the decisive features and actual cost of the completed power grid projects.

[0132] In an exemplary embodiment, the review dimensions include a cost control line review dimension. The review module 503 is further configured to: determine first cost control line information based on cost control information related to cost control; obtain the cost control information according to cost standard documents; identify multiple matching project samples in each completed power grid project and obtain cost information samples for each matching project sample; ensure that the matching degree between the engineering feature information samples of the matching project samples and the target project feature information meets the target condition; determine second cost control line information based on the cost information samples of each matching project sample; extract target cost information from the target project feature information; and compare the target cost information with the first and second cost control line information to obtain the cost review result of the target power grid project.

[0133] In an exemplary embodiment, the review dimensions include a rate compliance review dimension. The review module 503 is further configured to determine standard rates for multiple dimensions based on the rate standard document, and to extract target cost information representing multiple fee rates for the target power grid project from the target project feature information; compare the target cost information with each standard rate to obtain a cost review result representing whether each fee rate complies with regulations.

[0134] In an exemplary embodiment, the second determining module 502 is further configured to extract features from cost-related information to obtain engineering feature information; and to complete the engineering feature information based on the cost document if the engineering feature information is missing or the cost document of the target power grid project is not fully parsed.

[0135] Each module in the aforementioned power grid engineering cost document processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0136] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a method for processing power grid engineering cost documents. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0137] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0138] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0139] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0140] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0141] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0142] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0143] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above 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 application.

[0144] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for processing power grid engineering cost documents, characterized in that, The method includes: Based on the cost documents of the target power grid project, determine the cost-related information associated with the project cost; Based on the cost-related information, determine the target project characteristic information that characterizes the decisive features of the target power grid project; For each review dimension, the review benchmark information corresponding to the review dimension is determined based on the target project feature information, and the target cost information is extracted from the target project feature information. The review benchmark information corresponding to the review dimension is compared with the target cost information to obtain the cost review result of the target power grid project in the review dimension. The cost review results and related cost information for each of the aforementioned review dimensions are displayed in the visual interface.

2. The method according to claim 1, characterized in that, The review dimensions include the review dimension based on similar engineering comparisons; The step of determining the review benchmark information corresponding to the review dimension based on the target project feature information and extracting target cost information from the target project feature information, comparing the review benchmark information corresponding to the review dimension with the target cost information, and obtaining the cost review result of the target power grid project in the review dimension includes: In each completed power grid project, multiple similar project samples are identified; the comprehensive similarity between the target project feature information and the similar project sample feature information is greater than a similarity threshold. The validity of each of the aforementioned similar engineering samples is verified; the validity verification includes verification of the number of samples and verification of the time span. If any of the aforementioned similar engineering samples fail the validity check, an alert is triggered indicating insufficient historical data. If each of the similar project samples passes the validity verification, target cost information is extracted from the target project feature information, and the cost information samples of each of the similar project samples are compared with the target cost information to obtain the cost review result of the target power grid project; the cost information samples of the similar project samples represent the actual cost of the similar project samples.

3. The method according to claim 2, characterized in that, The comprehensive similarity between the target project feature information and the project feature information samples of similar projects is determined through the following steps: For each similar engineering sample, Obtain the cosine similarity between the target project feature information and the project feature information samples of the similar project samples, and obtain the weighted Canberra distance between the target project feature information and the project feature information samples of the similar project samples; Based on the weight information, the cosine similarity and the weighted Canberra distance are summed to obtain the comprehensive similarity between the target project feature information and the project feature information samples of the similar project samples; the weight information is learned by a random forest model based on the cost-related information of multiple completed power grid projects; the cost-related information of the completed power grid projects represents the decisive features and actual cost of the completed power grid projects.

4. The method according to claim 1, characterized in that, The review dimensions include the cost control line review dimension; The step of determining the review benchmark information corresponding to the review dimension based on the target project feature information and extracting target cost information from the target project feature information, comparing the review benchmark information corresponding to the review dimension with the target cost information, and obtaining the cost review result of the target power grid project in the review dimension includes: Based on cost control information related to cost control, the first cost control line information is determined; the cost control information is obtained from cost standard documents. Multiple matching project samples are identified in each completed power grid project, and cost information samples of each matching project sample are obtained; the matching degree between the engineering feature information samples of the matching project samples and the feature information of the target project reaches the target condition. Based on the cost information samples of each of the matched project samples, determine the second cost control line information; Extract the target cost information from the target project feature information; The target cost information is compared with the first cost control line information and the second cost control line information to obtain the cost review result of the target power grid project.

5. The method according to claim 1, characterized in that, The review dimensions include the fee compliance review dimension; The step of determining the review benchmark information corresponding to the review dimension based on the target project feature information and extracting target cost information from the target project feature information, comparing the review benchmark information corresponding to the review dimension with the target cost information, and obtaining the cost review result of the target power grid project in the review dimension includes: Based on the rate standard document, standard rates for multiple dimensions are determined, and target cost information representing multiple rate rates for the target power grid project is extracted from the target project feature information. The target cost information is compared with each of the standard rates to obtain the cost review results that indicate whether each of the fee rates complies with regulations.

6. The method according to claim 1, characterized in that, The step of determining the target project characteristic information that characterizes the decisive features of the target power grid project based on the cost-related information includes: Feature extraction is performed on the cost-related information to obtain engineering feature information; If the engineering feature information is missing or the cost document of the target power grid project is not fully parsed, the engineering feature information shall be supplemented based on the cost document.

7. A power grid engineering cost document processing device, characterized in that, The device includes: The first determining module is used to determine cost-related information related to the project cost based on the cost documents of the target power grid project. The second determining module is used to determine target project feature information that characterizes the decisive features of the target power grid project based on the cost-related information. The review module is used to determine the review benchmark information corresponding to the review dimension and extract the target cost information from the target project feature information for each review dimension, and compare the review benchmark information corresponding to the review dimension with the target cost information to obtain the cost review result of the target power grid project in the review dimension. The display module is used to display the cost review results and cost-related information for each of the review dimensions in a visual interface.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images