Engineering cost multi-stage intelligent analysis and examination method, system, equipment and medium
By constructing a power grid engineering cost data warehouse, an OLAP multidimensional analysis model, and an intelligent decision engine, the problems of multi-source data integration and single analysis dimension in power grid engineering cost management have been solved. This has enabled efficient management of cost data throughout the entire process and intelligent decision support, thereby improving the scientific nature and efficiency of power grid engineering cost management.
Patent Information
- Application Number
- CN202511829844.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-05
- Publication Date
- 2026-05-01
AI Technical Summary
In power grid engineering cost management, there are problems such as difficulty in integrating multi-source data, single analysis dimensions, weak anomaly detection capabilities, and insufficient decision support, resulting in low data processing efficiency and difficulty in achieving unified management and analysis of cost data throughout the entire process.
We constructed a power grid engineering cost data warehouse, integrated multi-source data through ETL technology, established an OLAP multidimensional analysis model, designed an intelligent decision engine based on data mining, and developed a BI visualization dashboard to achieve multidimensional analysis, intelligent decision support, and visualization.
It has enabled unified management and full-process analysis of power grid engineering cost data, improved data processing efficiency and analysis depth, enhanced anomaly detection capabilities and the scientificity and accuracy of decision support, and lowered the threshold for data use.
Smart Images
Figure CN121960938A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of power grid engineering cost management and business intelligence technology, and in particular to methods, systems, equipment and media for multi-stage intelligent analysis and review of engineering costs. Background Technology
[0002] In recent years, with the continuous expansion of my country's power grid construction scale and the in-depth advancement of the smart grid strategy, power grid engineering projects have exhibited characteristics such as large investment scale, high technical complexity, diverse equipment and materials, and long construction periods. Against this backdrop, power grid engineering cost management faces unprecedented challenges: First, power grid project cost data exhibits typical characteristics of being multi-source and heterogeneous. Cost data is scattered across multiple independent information platforms such as ERP systems, material management systems, and project management systems, resulting in inconsistent data standards and significant format differences, creating severe data silos. Traditional data processing methods primarily rely on manual collection and organization, which is not only inefficient but also prone to errors, making it difficult to support unified management and analysis of cost data throughout the entire process.
[0003] Secondly, the cost analysis of power grid projects is complex and diverse. Power grid projects involve different voltage levels (such as ultra-high voltage, extra-high voltage, and high voltage), different project types (such as substation projects, line projects, and distribution network projects), and different equipment and materials (such as transformers, circuit breakers, and cables), requiring multi-dimensional and multi-level analysis and comparison. Most existing cost estimation software focuses on the cost preparation function of a single stage and lacks the ability to conduct correlation analysis and trend prediction of cost data at different stages such as feasibility study estimates, preliminary design estimates, and final settlement.
[0004] Third, detecting anomalies in power grid project costs is challenging. Due to the diverse types of equipment and materials used in power grid projects and the rapid pace of technological updates, discrepancies can easily arise between the purchase price and quantity of equipment and materials supplied by the client and the actual settlement. Traditional methods primarily rely on the experience of reviewers for manual judgment, lacking anomaly detection mechanisms based on historical data and intelligent algorithms, making it difficult to effectively identify hidden cost anomalies.
[0005] Fourth, the decision support capabilities for power grid engineering cost estimation are insufficient. Most existing cost management systems remain at the level of data storage and report generation, lacking deep data mining and analysis capabilities based on business intelligence (BI) technology. They are unable to discover potential patterns in cost data through association rule mining, nor can they intelligently predict cost review results using predictive models, resulting in a lack of sufficient data support in the decision-making process.
[0006] Furthermore, the visualization of power grid project cost data is low. Key cost indicators and technical and economic parameters are mostly presented in static tables, lacking intuitive visualization and interactive analysis functions. Reviewers find it difficult to quickly grasp the overall cost situation and cannot delve into the root causes of specific problems through drill-down analysis.
[0007] Currently, although some research attempts to apply information technology to power grid engineering cost management, most solutions are still limited to optimizing individual business processes and have failed to form an intelligent analysis system covering the entire cost process. In particular, systematic solutions have yet to emerge in areas such as multi-source data integration, multi-dimensional analysis, intelligent decision-making, and visualization.
[0008] Therefore, there is an urgent need for a comprehensive solution that can integrate data from the entire process of power grid engineering cost, and possess multi-dimensional analysis, intelligent decision support, and visualization capabilities, in order to improve the scientific nature, accuracy, and efficiency of power grid engineering cost management. Summary of the Invention
[0009] In view of the aforementioned existing problems, the present invention provides a method, system, equipment and medium for multi-stage intelligent analysis and review of engineering costs.
[0010] Therefore, the technical problem solved by the present invention is the difficulty in integrating multi-source data, the single dimension of analysis, the weak anomaly detection capability, and the lack of decision support in the prior art.
[0011] To address the aforementioned technical problems, this invention provides the following technical solution: a multi-stage intelligent analysis and review method for engineering cost, comprising: constructing a power grid engineering cost data warehouse; extracting, transforming, and loading preliminary estimate data, settlement data, material data, and list data using ETL technology to establish a unified data model; establishing an OLAP multidimensional analysis model to achieve general cost drill-down analysis and key indicator trend analysis in the preliminary estimate stage, and comparative analysis of feasibility study estimates, preliminary design estimates, and final settlement estimates, anomaly detection of materials supplied by the client, and consistency analysis of the list in the settlement stage; designing an intelligent decision engine based on data mining, using association rule mining, classification prediction, and anomaly detection algorithms to deeply mine the analysis results and automatically generate review opinions; and developing a BI visualization dashboard to achieve visualized display and interactive analysis of power grid cost data through KPI indicator dashboards, multi-dimensional analysis charts, and a real-time monitoring interface.
[0012] As a preferred embodiment of the multi-stage intelligent analysis and review method for engineering cost described in this invention, the construction of the power grid engineering cost data warehouse includes extracting multi-source cost data from the power grid business system through a data extraction layer, and controlling the time range of the extraction process. The data cleaning layer performs quality checks and standardization on the extracted data, and quantitatively evaluates the data quality based on three dimensions: data integrity, accuracy, and consistency. A star schema data model is constructed through the data modeling layer, which associates the fact table with multiple dimension tables, including voltage level, project type, equipment type, and regional distribution, to form a unified structured data storage.
[0013] As a preferred embodiment of the multi-stage intelligent analysis and review method for engineering cost described in this invention, the establishment of the OLAP multidimensional analysis model includes constructing a multidimensional data cube with time, project type, and cost item as dimensions, and defining a set of metrics including cost amount, indicator value, and deviation rate. Data can be filtered by specified dimension values through slicing operations, and drill-down operations can be used to perform drill-down analysis from summary data to detailed data. Based on a multidimensional data cube, we can achieve general cost comparison and key indicator trend analysis in the preliminary budget stage, as well as three-dimensional cost comparison, abnormal detection of materials supplied by the client, and consistency analysis of the bill of quantities in the settlement stage.
[0014] As a preferred embodiment of the multi-stage intelligent analysis and review method for engineering cost described in this invention, the design of the intelligent decision engine based on data mining includes discovering frequent itemsets and association rules from historical cost transaction data through association rule mining technology, with support and confidence as the basis for rule selection. Using a classification prediction model, based on the cost deviation rate and indicator completion feature vectors, it is predicted whether a project needs to be reviewed. Anomaly detection algorithms are used to calculate the distance between the data points to be tested and the cluster centers based on the clustering results, thereby identifying abnormal cost patterns. By using a decision fusion mechanism, the confidence level of association rules, the probability of classification prediction, and the anomaly score are weighted and fused together to output a comprehensive decision score.
[0015] As a preferred embodiment of the multi-stage intelligent analysis and review method for engineering cost described in this invention, the establishment of the OLAP multidimensional analysis model further includes establishing a multi-layer caching optimization mechanism to improve query efficiency through hot data caching. in, This represents the recent query set. Indicates the cache capacity.
[0016] By employing a fine-grained access control mechanism, data visibility control is achieved through the intersection operation of user permission sets and OLAP query requests. in, This indicates an OLAP query request. This represents a set of user permissions.
[0017] As a preferred embodiment of the multi-stage intelligent analysis and review method for engineering cost described in this invention, the development of the BI visualization dashboard includes calculating key performance indicators in real time based on basic cost data and parameter sets through a dynamic KPI calculation engine. in, This represents the basic cost data extracted from the data warehouse. This represents the set of parameters used to calculate the index. This represents the output value of the i-th key performance indicator. For calculation functions; The year-on-year and month-on-month analysis module calculates the annual growth rate and short-term fluctuation trend of cost indicators: in, Indicates the current period value. This represents the figure for the same period last year. Output the year-on-year growth rate to assess the annual changes in cost indicators; Through drill-down and roll-up operations, users can dynamically switch between summary data and detailed data analysis levels: in, Represents the current view dataset. Indicates the target detail level. This is the function for the drilling operation. For extended functions; in, This is the function for the roll-up operation. This is a folding function; An alarm is automatically generated when a monitored metric exceeds a preset threshold, triggered by a threshold-triggered mechanism. in, This is an alarm signal. Indicates the value of the monitoring indicator. Indicates the preset threshold. This is an indicator function that takes a value of 1 and triggers an alarm generation when a certain condition is met. This is the alarm generation function.
[0018] As a preferred embodiment of the multi-stage intelligent analysis and review method for engineering costs described in this invention, the development of the BI visualization dashboard further includes dynamically adjusting the layout of the visualization interface based on user roles, terminal devices, and preferences using adaptive layout technology. in, The output layout scheme, Indicates user role information, Indicates the type of terminal device. Indicates user personalized preferences, As an adaptive layout engine, the system dynamically adjusts the layout of visual components based on parameters; The event handling mechanism responds to user interactions and updates the data view. To address the needs of interactive analysis, the system implements an event handling mechanism: in, For the interactive result, Indicates user interaction events, Represents the current dataset. This is an event handling function. The system performs corresponding data processing and view update operations based on the event type.
[0019] This invention provides a multi-stage intelligent analysis and review system for engineering costs.
[0020] As a preferred embodiment of the multi-stage intelligent analysis and review system for engineering cost described in this invention, it includes a data integration and processing module, which is responsible for automatically collecting and organizing cost-related data from multiple business systems, cleaning and standardizing it, and storing it in a unified data warehouse. The multidimensional dynamic analysis module is responsible for flexibly querying and comparing cost data from multiple angles and levels, including cost comparison at different stages, key indicator analysis, and anomaly detection. The intelligent decision support module is responsible for using data mining algorithms to automatically discover data patterns, predict review risks, and identify abnormal patterns, providing quantitative basis for generating review opinions. The visualization and interaction module is responsible for displaying data and analysis results intuitively in the form of charts and dashboards, and supports users to interact and explore in real time through clicks and filtering operations.
[0021] The present invention 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 implement the steps of a multi-stage intelligent analysis and review method for engineering cost.
[0022] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of a multi-stage intelligent analysis and review method for engineering cost.
[0023] The beneficial effects of this invention are as follows: Through the innovative application of BI data technology, a complete technical system from data collection and intelligent analysis to decision support has been constructed, realizing a fundamental transformation of power grid engineering cost management from traditional experience-based to modern data analysis-based, and providing effective technical support for refined cost management in the context of smart grid construction. Attached Figure Description
[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0025] Figure 1 This is a schematic diagram of a multi-stage intelligent analysis and review method for engineering costs provided in one embodiment of the present invention. Detailed Implementation
[0026] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.
[0027] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a multi-stage intelligent analysis and review method for engineering costs, including: S1: Construct a power grid engineering cost data warehouse, and use ETL technology to extract, transform and load budget data, settlement data, material data and list data to establish a unified data model.
[0028] S2: Establish an OLAP multidimensional analysis model to achieve general cost drill-down analysis and key indicator trend analysis in the preliminary budget stage, and to achieve comparative analysis of feasibility study estimation, preliminary design budget, and final settlement in the settlement stage, as well as anomaly detection and consistency analysis of materials supplied by the client.
[0029] S3: Design an intelligent decision engine based on data mining, which uses association rule mining, classification prediction and anomaly detection algorithms to deeply mine the analysis results and automatically generate review opinions.
[0030] S4: Develop a BI visualization dashboard that enables the visualization and interactive analysis of power grid cost data through KPI indicator dashboards, multi-dimensional analysis charts, and real-time monitoring interfaces.
[0031] It should be noted that, compared to existing technologies, this invention achieves multiple breakthroughs in the field of power grid engineering cost management through the systematic integration and application of BI data technology: At the data layer, the power grid engineering cost data warehouse constructed by this invention breaks through the limitations of traditional data silos. By establishing a unified ETL data processing pipeline, it realizes automatic data collection and standardization from multiple systems such as power grid ERP, material management, and project management, solving the core pain point of scattered data sources and inconsistent formats in power grid cost data. The layered architecture design of the data warehouse ensures the integrity, accuracy, and consistency of the data, laying a solid foundation for subsequent in-depth analysis.
[0032] At the analysis level, the OLAP multidimensional analysis model of this invention provides unprecedented depth and flexibility in analysis. A specialized multidimensional data analysis architecture has been designed to address the unique characteristics of power grid engineering, including voltage level classification, project type differences, and the diversity of equipment and materials, supporting comprehensive analysis from grasping macro trends to pinpointing micro-level problems. Particularly noteworthy is that this invention, for the first time, elevates the three-dimensional cost comparison analysis from simple numerical comparison to trend correlation analysis, enabling precise identification of cost deviation patterns at each stage of power grid engineering projects.
[0033] At the decision-making level, the intelligent decision engine of this invention shifts power grid cost review from experience-based to data-driven. Through association rule mining, it can discover potential patterns such as the correlation between price fluctuations of certain equipment and materials and overall cost in line projects at specific voltage levels; through machine learning classification and prediction, it can intelligently assess the review risk level of the current project based on historical data; and through anomaly detection algorithms, it can promptly identify abnormal procurement and usage patterns in equipment and materials supplied by the client. This multi-layered, intelligent decision support system significantly improves the scientific rigor and accuracy of the review process.
[0034] At the application layer, the BI visualization dashboard of this invention effectively transforms technological achievements into practical value. Through KPI dashboards, managers can grasp the overall cost status of power grid projects in real time; through multi-dimensional drill-down analysis, reviewers can quickly pinpoint the root causes of problems; and through adaptive layout, the system provides the best user experience in different application scenarios. This intuitive and interactive visualization interface greatly lowers the barrier to data use and improves the work efficiency of the entire cost management team.
[0035] Example 2 is an embodiment of the present invention. Based on the above embodiment, a multi-stage intelligent analysis and review method for engineering cost is provided.
[0036] Furthermore, in this embodiment, step S1 involves constructing a power grid engineering cost data warehouse. This is achieved by using ETL technology to extract, transform, and load preliminary cost data, settlement data, material data, and list data to establish a unified data model. Specific steps include: The power grid engineering cost data warehouse adopts a layered architecture design, including: The data extraction layer extracts data from power grid business systems, including data from ERP systems, materials management systems, and project management systems, and defines data extraction functions: in, This represents the i-th power grid business data source, including power grid ERP, materials management, project management, and other systems. and This refers to the start and end times of data extraction. The data cleaning layer performs quality checks and standardization on the extracted data, and defines a data quality assessment function. in, This indicates the completeness score of the power grid cost data. This indicates the accuracy score of the power grid cost data. Indicates the data consistency score. , , Let be the weight coefficients for each dimension, and satisfy . .
[0037] In the data modeling layer, a star schema is established, with the fact table denoted as F and the dimension table set as follows: The dimensions include voltage level, project type, equipment type, and geographical distribution, defining the model relationships: in, This represents the fact table record. This represents the record in the i-th dimension table.
[0038] In an optional embodiment, the data modeling layer can further normalize some dimension tables based on the star schema, splitting them into multi-level relational tables. The core fact table is only connected to the finest-grained dimension tables, forming a snowflake-like branching structure.
[0039] In another alternative embodiment, the data modeling layer can also directly integrate all necessary dimensional attributes and core metrics into a large, comprehensive "cost wide table" during the data preprocessing stage (ETL process). Each record in this table contains all the fields required for analysis, such as project identifier, cost amount, voltage level, project type, equipment type, and region.
[0040] Furthermore, in the embodiments of this application, step S2 establishes an OLAP multidimensional analysis model to achieve general cost drill-down analysis and key indicator trend analysis in the preliminary estimate stage, and to achieve comparative analysis of feasibility study estimation, preliminary design estimate, and final settlement estimates, as well as anomaly detection and list consistency analysis of materials supplied by the client in the settlement stage. Specific steps include: OLAP multidimensional analysis models achieve multidimensional analysis of cost data by constructing multidimensional data cubes. Its core structure is defined as follows: in, Representing a multidimensional data cube, This represents a set of dimensions (including analysis dimensions such as time, project type, and cost items). It represents a set of metrics (including cost amount, indicator value, deviation rate, and other analytical indicators).
[0041] Based on this cube structure, the system defines a complete set of OLAP analysis operation functions. Data filtering by dimension value is achieved through slicing functions. in, As a dimension, For dimension values, Represents the measurement value Drill-down analysis of detailed data using drill-down functions: in, This indicates a more granular level of analysis. This indicates the metric value at this level.
[0042] Based on this multidimensional analysis framework, the system constructs specialized analysis functions tailored to the characteristics of power grid engineering costs. A general cost comparison function quantifies the degree of cost deviation. in, This represents the actual cost of the i-th project. This indicates the corresponding standard cost. Indicates the number of items.
[0043] Key indicator trend function analysis: Indicator change trend in, This represents the key indicator value for period t. Indicates the length of the time series. It is a symbolic function.
[0044] The three-dimensional comparison function evaluates the consistency of the estimated, preliminary, and settlement data: in, This represents an estimated value. This represents the estimated value. Indicates the settlement value. This represents the average of the three calculations.
[0045] To ensure analytical performance, the system establishes a multi-layered caching optimization mechanism, improving query efficiency through hot data caching: in, This represents the recent query set. Indicates the cache capacity.
[0046] It also integrates a fine-grained access control mechanism: in, This indicates an OLAP query request. This represents the set of user permissions. Through the combination of the above techniques, this multidimensional analysis model achieves comprehensive cost analysis capabilities, from grasping macro trends to pinpointing micro-level problems.
[0047] In an optional embodiment, hot data caching can also employ a FIFO caching strategy, where the system manages cache eviction according to the order in which data entries enter the cache. When the cache space is full and new data needs to be loaded, the system unconditionally evicts the earliest data entry that entered the cache, regardless of whether it has been frequently accessed recently.
[0048] In another alternative embodiment, managing the hot data cache can also employ an LFU (Live Validation) caching strategy, where the system manages the cache based on the frequency with which data entries are accessed within a historical period. When data needs to be evicted, the system prioritizes evicting data entries that have been accessed the least historically, thereby retaining the more "popular" data.
[0049] Furthermore, in this embodiment, step S3 designs an intelligent decision-making engine based on data mining. Through association rule mining, classification prediction, and anomaly detection algorithms, it performs in-depth analysis of the results and automatically generates review opinions. Specific steps include: The data mining-based intelligent decision engine achieves in-depth mining and intelligent decision-making of cost data through multi-level analysis techniques. Its core architecture includes the following technical implementations: The intelligent decision engine first establishes an association rule mining module, which discovers hidden patterns in cost data through support and confidence calculations. in, Represents a specific itemset, This represents a dataset of historical cost transactions. Represents the total number of transactions and support. Used to measure itemsets Frequency of occurrence in the dataset.
[0050] in, Represents association rules and confidence levels. Used to measure the itemset Itemsets when they appear The conditional probability of simultaneous occurrence.
[0051] Based on the association rule mining results, the engine further constructs a classification prediction module, which uses a logistic regression algorithm to achieve intelligent prediction of review results. in, This represents the input feature vector (containing features such as cost deviation rate and indicator completion rate). Represents the model weight vector. Indicates the bias term. This indicates the probability that the predicted result is "requires review".
[0052] Classification models optimize parameters by minimizing the loss function: in, Indicates the number of training samples. This represents the true label of the i-th sample. This represents the model's predicted output. This represents the cross-entropy loss function.
[0053] Meanwhile, the engine integrates an anomaly detection module, which identifies abnormal cost patterns based on clustering algorithms: in, This represents the cost data points to be tested. This represents the set of categories obtained through K-means clustering. This represents the cluster center of the c-th category. This indicates an abnormal score; the score exceeds a certain threshold. It is marked as abnormal at that time.
[0054] The intelligent decision engine establishes a decision fusion mechanism by integrating the above-mentioned technical modules: in, Indicates the confidence level of the association rule. Represents the classification prediction probability. Indicates an abnormal score. These are the weighting coefficients. The Sigmoid activation function is used to ultimately output a comprehensive decision score, providing a quantitative basis for generating review opinions.
[0055] In an optional embodiment, the decision fusion mechanism can also set independent judgment thresholds for the association rule module, classification prediction module, and anomaly detection module. Each module outputs a binary "yes / no" or "high risk / low risk" preliminary decision based on whether its output exceeds its threshold. The final fusion decision is generated using a "majority voting" principle, that is, adopting the consistent judgment results of two or three modules.
[0056] In another optional embodiment, the decision fusion mechanism can also use association rule confidence, classification prediction probability, and anomaly score as three input features to train a shallow decision tree model based on a batch of historically labeled training samples. The decision tree model will directly output the final classification decision or risk level according to the specific values of the three input features and through a preset "if-then" rule.
[0057] Furthermore, in this embodiment of the application, step S4 develops a BI visualization dashboard, which, through KPI indicator dashboards, multi-dimensional analysis charts, and a real-time monitoring interface, realizes the visualization and interactive analysis of power grid cost data. Specific steps include: The visualization dashboard built on BI technology achieves multi-dimensional display and intelligent monitoring of cost data through a multi-layered architecture. Its technical implementation includes: The visualization dashboard first establishes a dynamic KPI calculation engine, which uses indicator calculation functions to achieve real-time analysis of key cost indicators: in, This represents the basic cost data extracted from the data warehouse. This represents the set of parameters used to calculate the index. This represents the output value of the i-th key performance indicator.
[0058] Based on the indicator calculation results, the system constructs a year-on-year and month-on-month analysis module, which reveals the trend of cost changes through time series comparison functions: in, Indicates the current period value. This represents the figure for the same period last year. Output the year-on-year growth rate to assess the annual changes in cost indicators.
[0059] in, This indicates the value from the previous period. Output the month-on-month growth rate to analyze the short-term fluctuation trend of cost indicators.
[0060] The dashboard further integrates multi-dimensional drill-down analysis functions, enabling dynamic adjustment of data granularity through drill-down operation functions: in, Represents the current view dataset. This indicates the target detail level; this function enables drill-down analysis from summary data to detailed data.
[0061] This function enables roll-up aggregation from detailed data to summary data, allowing users to flexibly switch between different analysis levels.
[0062] To achieve real-time monitoring and early warning, the system establishes a threshold triggering mechanism, which automatically identifies abnormal situations through an alarm generation function: in, Indicates the value of the monitoring indicator. Indicates the preset threshold. This is an indicator function that takes a value of 1 and triggers an alarm generation when the condition is met.
[0063] The dashboard also employs adaptive layout technology, using layout optimization functions to intelligently adapt the display interface. in, Indicates user role information, Indicates the type of terminal device. These parameters represent users' personalized preferences, and the system dynamically adjusts the layout of the visualization components based on these parameters.
[0064] In an optional embodiment, intelligent adaptation of the display interface can also be achieved by pre-designing and storing several fixed layout templates for different combinations of user roles and terminal device types. When a user accesses the dashboard, the system directly calls and applies the corresponding preset template based on the identified user role and device information, and then fine-tunes some components according to user preferences.
[0065] In another optional embodiment, intelligent adaptation of the display interface can also be achieved by establishing a layout engine containing conditional judgment rules. Each rule defines the layout strategy to be adopted under specific conditions. During system runtime, parameters such as the current user role and device information are input into the rule engine, which then matches rules according to their priority and executes the layout strategy corresponding to the first matching rule.
[0066] To address the needs of interactive analysis, the system implements an event handling mechanism: in, This indicates user interaction events (such as clicks, filters, sorting, etc.). This represents the current dataset. The system performs corresponding data processing and view update operations based on the event type.
[0067] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that: This embodiment also provides a multi-stage intelligent analysis and review system for engineering cost, including: a data integration and processing module, which is responsible for automatically collecting and organizing cost-related data from multiple business systems, cleaning and standardizing it, and storing it in a unified data warehouse; The multidimensional dynamic analysis module is responsible for flexibly querying and comparing cost data from multiple angles and levels, including cost comparison at different stages, key indicator analysis, and anomaly detection. The intelligent decision support module is responsible for using data mining algorithms to automatically discover data patterns, predict review risks, and identify abnormal patterns, providing quantitative basis for generating review opinions. The visualization and interaction module is responsible for displaying data and analysis results intuitively in the form of charts and dashboards, and supports users to interact and explore in real time through clicks and filtering operations.
[0068] This embodiment also provides an electronic device, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the multi-stage intelligent analysis and review method for engineering cost as proposed in the above embodiment.
[0069] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the multi-stage intelligent analysis and review method for engineering cost as proposed in the above embodiments.
[0070] The storage medium proposed in this embodiment and the method for multi-stage intelligent analysis and review of engineering costs proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.
[0071] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.
[0072] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A multi-stage intelligent analysis and review method for engineering cost, characterized by: include, Construct a power grid engineering cost data warehouse, and use ETL technology to extract, transform and load budget data, settlement data, material data and list data to establish a unified data model; Establish an OLAP multidimensional analysis model to achieve general cost drilling analysis and key indicator trend analysis in the preliminary budget stage, and to achieve comparative analysis of feasibility study estimation, preliminary design budget, and final settlement in the settlement stage, as well as anomaly detection and consistency analysis of materials supplied by the client. Design an intelligent decision engine based on data mining, and use association rule mining, classification prediction and anomaly detection algorithms to deeply mine the analysis results and automatically generate review opinions; Develop a BI visualization dashboard that enables the visualization and interactive analysis of power grid cost data through KPI indicator dashboards, multi-dimensional analysis charts, and a real-time monitoring interface.
2. The multi-stage intelligent analysis and review method for engineering cost as described in claim 1, characterized in that: The construction of the power grid engineering cost data warehouse includes extracting multi-source cost data from the power grid business system through a data extraction layer, and controlling the time range of the extraction process. The data cleaning layer performs quality checks and standardization on the extracted data, and quantitatively evaluates the data quality based on three dimensions: data integrity, accuracy, and consistency. A star schema data model is constructed through the data modeling layer, which associates the fact table with multiple dimension tables, including voltage level, project type, equipment type, and regional distribution, to form a unified structured data storage.
3. The multi-stage intelligent analysis and review method for engineering cost as described in claim 2, characterized in that: The establishment of the OLAP multidimensional analysis model includes constructing a multidimensional data cube with time, project type, and cost item as dimensions, and defining a set of measures including cost amount, indicator value, and deviation rate. Data can be filtered by specified dimension values through slicing operations, and drill-down operations can be used to perform drill-down analysis from summary data to detailed data. Based on a multidimensional data cube, we can achieve general cost comparison and key indicator trend analysis in the preliminary budget stage, as well as three-dimensional cost comparison, abnormal detection of materials supplied by the client, and consistency analysis of the bill of quantities in the settlement stage.
4. The multi-stage intelligent analysis and review method for engineering cost as described in claim 3, characterized in that: The design of the intelligent decision engine based on data mining includes discovering frequent itemsets and association rules from historical cost transaction data through association rule mining technology, with support and confidence as the basis for rule selection. Using a classification prediction model, based on the cost deviation rate and indicator completion feature vectors, it is predicted whether a project needs to be reviewed. Anomaly detection algorithms are used to calculate the distance between the data points to be tested and the cluster centers based on the clustering results, thereby identifying abnormal cost patterns. By using a decision fusion mechanism, the confidence level of association rules, the probability of classification prediction, and the anomaly score are weighted and fused together to output a comprehensive decision score.
5. The multi-stage intelligent analysis and review method for engineering cost as described in claim 4, characterized in that: The establishment of the OLAP multidimensional analysis model also includes establishing a multi-layer caching optimization mechanism to improve query efficiency through hot data caching. in, This represents the recent query set. Indicates the cache capacity. By employing a fine-grained access control mechanism, data visibility control is achieved through the intersection operation of user permission sets and OLAP query requests. in, This indicates an OLAP query request. This represents the set of user permissions.
6. The multi-stage intelligent analysis and review method for engineering cost as described in claim 5, characterized in that: The development of the BI visualization dashboard includes the use of a dynamic KPI calculation engine to calculate key performance indicators in real time based on cost data and parameter sets. in, This represents the basic cost data extracted from the data warehouse. This represents the set of parameters used to calculate the index. This represents the output value of the i-th key performance indicator. For calculation functions; The year-on-year and month-on-month analysis module calculates the annual growth rate and short-term fluctuation trend of cost indicators: in, Indicates the current period value. This represents the figure for the same period last year. Output the year-on-year growth rate to assess the annual changes in cost indicators; Through drill-down and roll-up operations, users can dynamically switch between summary data and detailed data analysis levels: in, Represents the current view dataset. Indicates the target detail level. This is the function for the drilling operation. For extended functions; in, This is the function for the roll-up operation. This is a folding function; An alarm is automatically generated when a monitored metric exceeds a preset threshold, triggered by a threshold-triggered mechanism. in, This is an alarm signal. Indicates the value of the monitoring indicator. Indicates the preset threshold. This is an indicator function that takes a value of 1 and triggers an alarm generation when a certain condition is met. This is the alarm generation function.
7. The multi-stage intelligent analysis and review method for engineering cost as described in claim 6, characterized in that: The development of the BI visualization dashboard also includes dynamically adjusting the layout of the visualization interface based on user roles, terminal devices, and preferences using adaptive layout technology. in, The output layout scheme, Indicates user role information, Indicates the type of terminal device. Indicates user personalized preferences, As an adaptive layout engine, the system dynamically adjusts the layout of visual components based on parameters; The event handling mechanism responds to user interactions and updates the data view. To address the needs of interactive analysis, the system implements an event handling mechanism: in, For the interactive result, Indicates user interaction events, Represents the current dataset. This is an event handling function. The system performs corresponding data processing and view update operations based on the event type.
8. A multi-stage intelligent analysis and review system for engineering costs, employing the method for multi-stage intelligent analysis and review of engineering costs as described in any one of claims 1 to 7, characterized in that, include: The data integration and processing module is responsible for automatically collecting and organizing cost-related data from multiple business systems, cleaning and standardizing it, and storing it in a unified data warehouse. The multidimensional dynamic analysis module is responsible for flexibly querying and comparing cost data from multiple angles and levels, including cost comparison at different stages, key indicator analysis, and anomaly detection. The intelligent decision support module is responsible for using data mining algorithms to automatically discover data patterns, predict review risks, and identify abnormal patterns, providing quantitative basis for generating review opinions. The visualization and interaction module is responsible for displaying data and analysis results intuitively in the form of charts and dashboards, and supports users to interact and explore in real time through clicks and filtering operations.
9. 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 multi-stage intelligent analysis and review method for engineering cost as described in any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-stage intelligent analysis and review method for engineering cost as described in any one of claims 1 to 7.