Power transmission and transformation project cost risk whole-process dynamic early warning method and system

Abstract

The invention relates to a power transmission and transformation project cost risk whole-process dynamic early warning method and system. The method comprises the following steps: obtaining cost data and reference data of each stage of a power transmission and transformation project, and constructing a standardized cost database; based on a preset risk rule, constructing a risk index system comprising a static risk index, a dynamic risk index and a stage association index; performing real-time comparison and trend prediction on the cost data and reference data by using a preset deep learning model, and accurately identifying cost risk points; determining a risk level according to the deviation degree of the risk point, generating graded early warning information, and pushing the graded early warning information to a corresponding terminal; and the processing result is recorded and the database is updated, so that the whole-process dynamic monitoring and risk early warning are realized, the real-time performance, the accuracy and the decision-making efficiency of the cost risk management are effectively improved, and the problems of early warning lag and incomplete risk coverage in the traditional method are solved.

Description

Technical Field

[0001] The present invention belongs to the field of engineering cost risk management, and in particular relates to a dynamic early warning method and system for the entire process of power transmission and transformation engineering cost risk. Background Art

[0002] With technological advancements in power transmission and transformation projects, particularly driven by the large-scale construction and intelligent transformation of power infrastructure, traditional cost risk management techniques based on static indicators and historical data have emerged. This technology utilizes manual collection of cost data from various phases, combined with benchmarks from similar historical projects, to identify risks through simple threshold comparisons. Its core advantages lie in its intuitive operation and ease of implementation, providing a foundational framework for project cost control. Traditional cost risk early warning methods primarily rely on data aggregation by phase. For data processing, estimated cost data from the feasibility study phase, budgeted cost data from the preliminary design phase, and budgeted cost data from the construction drawing phase are collected and standardized to construct a preliminary database. For risk identification, actual data is manually compared against historical benchmarks based on preset static risk indicators (such as fixed deviation thresholds) to identify cost deviations within a single phase. Early warning mechanisms employ post-hoc analysis, generating warning information that is then pushed to decision-makers via a rules engine. However, these methods lack dynamic prediction and cross-phase correlation analysis.

[0003] However, current traditional methods suffer from the following issues: Early warning lag: Risk identification relies on static data comparisons and is unable to respond to dynamic changes (such as equipment and material price fluctuations or project schedule adjustments) in real time. For example, dynamic risk indicators (such as the cumulative value of sub-item costs based on project progress nodes) are not effectively integrated, resulting in delayed early warnings and affecting the timeliness of risk management. Incomplete risk coverage: Traditional methods focus on static risk indicators (such as construction project cost deviations) and ignore dynamic trends and stage-related risks. This leads to fragmented risk point identification, an inability to capture cross-stage transmission effects, and an increased risk of overall project overruns. Decision-making efficiency is low, the early warning information push path is rigid, lacks intelligent grading, and is not linked to historical management references, resulting in insufficient decision support and wasted resources. Summary of the Invention

[0004] Based on this, it is necessary to provide a dynamic early warning method and system for the entire process of power transmission and transformation project cost risks that can solve the above problems.

[0005] In the first aspect, the present application provides a dynamic early warning method for the entire process of power transmission and transformation project cost risk, including: Obtain cost data and benchmark data for each stage of power transmission and transformation projects and build a standardized cost database; Based on the preset risk rules, a risk indicator system is constructed, which includes static risk indicators, dynamic risk indicators and stage-related indicators; Based on the risk indicator system, a preset deep learning model is used to conduct real-time comparison and trend prediction of cost data in the cost database with benchmark data to identify cost risk points; According to the deviation degree of the cost risk point, the risk level is determined and graded warning information is generated and pushed to the corresponding terminal; Record the processing results of graded warning information and update the standardized cost database based on the processing results.

[0006] In one embodiment, obtaining cost data and benchmark data for each phase of a power transmission and transformation project and constructing a standardized cost database includes: Obtain estimated cost data for the feasibility study phase, budgeted cost data for the preliminary design phase, and budgeted cost data for the construction drawing phase. The cost data covers construction costs, equipment purchase costs, installation costs, and other expenses. Obtain cost data for similar historical projects, reference prices for equipment and materials in the project area, and industry standard rates as benchmark data; Normalize the cost data and benchmark data of each stage and build a standardized cost database.

[0007] In one embodiment, building a risk indicator system includes: Based on the benchmark data, corresponding static risk indicators are set for construction costs, equipment purchase costs, installation costs and other costs; Set dynamic risk indicators for equipment and material prices based on historical price fluctuation patterns, and set stage-by-stage dynamic risk indicators based on project progress nodes and the dynamic deviation rate calculated based on the accumulated value of the sub-item cost. For the paths from the feasibility study stage to the preliminary design stage and from the preliminary design stage to the construction drawing stage, a stage correlation threshold is set as the stage correlation indicator.

[0008] In one embodiment, a preset deep learning model is used to compare cost data in a cost database with benchmark data in real time and predict trends to identify cost risk points, including: Static risk indicators, dynamic risk indicators, and stage-related indicators are input into the deep learning model. The deep learning model then performs feature comparison between cost data and benchmark data and cross-stage trend deduction to generate a risk decision vector that includes static deviations, dynamic trends, and transition mutations. When it is detected that the static deviation exceeds the static risk index, or the dynamic trend exceeds the dynamic risk index, or the connection mutation coefficient breaks through the stage correlation index, the corresponding type of cost risk point is identified and the chapter identification position of the cost risk point in the cost list, the cost impact range and the current cost deviation are determined and output.

[0009] In one embodiment, based on the degree of deviation of the cost risk point, the risk level is determined and graded warning information is generated and pushed to the corresponding terminal, including: Based on the static deviation amount, dynamic trend probability and connection mutation coefficient in the risk decision vector, the compound deviation rate is calculated and the risk level is determined according to the preset mapping rules; the risk level can be minor, general or severe; Generate early warning information based on risk levels; the early warning information includes the chapter identification position of the risk point corresponding to the risk level in the cost list; a quantitative matrix reflecting the current cost deviation value and the predicted deviation value of the risk point; and the cost impact range of the risk point; The rule engine matches the warning information with the push path according to the risk level. When the risk level is serious, it is pushed to the first type of terminal and triggers the emergency response process preset on the first type of terminal. When the risk level is minor or general, it is pushed to the second type of terminal and associated with the preset historical risk disposal reference information.

[0010] In one embodiment, the composite deviation rate is calculated based on the static deviation amount, dynamic trend probability, and transition mutation coefficient in the risk decision vector, and is implemented by the following formula:

[0011] in is the composite deviation rate, is the static deviation, Indicates the actual value of the item cost at the current stage. Indicates the benchmark cost data, Indicates the deviation threshold corresponding to the static risk indicators set for construction project costs, equipment purchase costs, installation project costs and other costs; is the dynamic trend probability, which indicates the probability value of the future N periods of construction costs corresponding to the dynamic trend output by the deep learning model exceeding the dynamic risk index, and its value is ; is the transition coefficient, Indicates the actual total cost of the current project phase, Indicates the total actual cost of the previous project stage; , and is a dynamic weight factor, satisfying .

[0012] In one embodiment, the method further includes inputting the graded warning information into a visualization engine and performing the following steps: Based on the chapter mark position in the warning information, a heat map is generated, and the color depth of the heat map is mapped to the value of the composite deviation rate (CDR); Based on the current cost deviation value and the predicted deviation value in the early warning information, a deviation fluctuation map is generated, in which the actual deviation value is associated with the static deviation value. , the probability of dynamic trend associated with the prediction interval ; Based on the cost impact range and connection mutation coefficient in early warning information , generate a cross-stage risk transmission tree diagram, and the influence weights between nodes in the tree diagram are given by Value quantification.

[0013] Secondly, this application also provides a dynamic early warning system for the entire process of power transmission and transformation project cost risks, including: Data acquisition and processing module, used to obtain cost data and benchmark data for each stage of power transmission and transformation projects and build a standardized cost database; The risk indicator management module is used to build a risk indicator system based on preset risk rules. The risk indicator system includes static risk indicators, dynamic risk indicators and stage-related indicators; The intelligent risk analysis module is used to compare the cost data in the cost database with the benchmark data in real time and predict trends based on the risk indicator system and the preset deep learning model to identify cost risk points; The early warning generation and response module is used to determine the risk level based on the deviation degree of the cost risk point and generate graded early warning information and push it to the corresponding terminal; The feedback optimization module is used to record the processing results of graded warning information and update the standardized cost database based on the processing results.

[0014] In a third aspect, the present application also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the steps of the above-mentioned method for dynamic early warning of the entire process of cost risk of power transmission and transformation projects.

[0015] In a fourth aspect, the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned dynamic early warning method for the entire process of cost risk of power transmission and transformation projects are implemented.

[0016] The above-mentioned dynamic early warning method, system, computer equipment and storage medium for the entire process of cost risk of power transmission and transformation projects construct a standardized cost database by acquiring cost data and benchmark data of each stage of power transmission and transformation projects, and construct a risk indicator system covering static risk indicators, dynamic risk indicators and stage-related indicators based on preset risk rules. It uses a preset deep learning model to compare cost data and benchmark data in real time and predict trends to accurately identify cost risk points; determines the risk level according to the degree of deviation of the risk point, generates graded early warning information and pushes it to the corresponding terminal, and records the processing results and updates the database at the same time. The problem of delayed early warning has been solved: the real-time comparison and trend prediction mechanism of the deep learning model can dynamically respond to price fluctuations of equipment and materials or adjustments to project progress, avoiding delays caused by reliance on static data; the introduction of dynamic risk indicators and stage-related indicators in the risk indicator system solves the problem of incomplete risk coverage, and avoids the risk of project overruns caused by fragmented risk identification by capturing the patterns of price fluctuations and cross-stage correlation effects; the risk level determination and graded push mechanism based on the degree of deviation optimizes decision-making efficiency and realizes intelligent early warning information distribution, ensuring that emergency response processes are triggered in the event of serious risks, and historical disposal reference information is linked in the event of minor or general risks, thereby improving the real-time, accuracy and overall decision-making efficiency of risk management. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a flow chart of the dynamic early warning method for the entire process of power transmission and transformation project cost risk of the present invention; Figure 2 This is a structural diagram of the dynamic early warning system for the entire process of power transmission and transformation project cost risk according to the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0020] In one embodiment, Figure 1As shown, a dynamic early warning method for cost risks throughout the entire process of power transmission and transformation project construction is provided. This embodiment illustrates this method using a project site cost data collection terminal. This terminal has built-in sensors and a data interface, which captures real-time cost data (including construction costs and equipment purchase costs) from the feasibility study, preliminary design, and construction drawing stages of the power transmission and transformation project. It is understood that this method can also be deployed on a cloud-based analysis server, using a deep learning model to simultaneously process multiple project data streams. Alternatively, it can be applied to a terminal-server collaborative system: the terminal collects real-time cost data and receives early warning instructions, while the server runs a standardized cost database and risk analysis algorithm, interacting via a 4G / 5G private network. When a power transmission and transformation project faces the risk of dynamic price fluctuations (such as sudden price changes for equipment and materials during the construction phase) or when there are abnormalities in cost integration across different stages, the terminal uploads the current itemized cost data to the server. The server then utilizes a risk indicator system (including static deviation thresholds, dynamic trend probabilities, etc.) to generate a risk decision vector using a deep learning model. If the composite deviation rate exceeds the limit, the server immediately triggers a tiered push notification: A warning message containing a heat map (mapping the deviation location) is sent to the project manager's mobile terminal (first-class terminal) through the rule engine, activating the pre-installed emergency response process on the terminal. If it is a general risk, it is pushed to the cost engineer's terminal (second-class terminal) and linked to the historical disposal solution library. After the terminal feedbacks the disposal results, the server updates the database closed-loop risk management process. In this embodiment, the method includes the following steps: S01, obtain cost data and benchmark data for each stage of power transmission and transformation projects, and build a standardized cost database.

[0021] Among them, the cost data of each stage (estimated cost data of the feasibility study stage, budgeted cost data of the preliminary design stage and budgeted cost data of the construction drawing stage, which cover key sub-items such as construction project costs, equipment purchase costs, installation project costs and other costs); benchmark data (cost data of historical similar projects, reference prices of equipment and materials in the area where the project is located, and industry standard rates, used to provide a comparison benchmark); when implemented, the above data can be obtained in real time or asynchronously through terminals with built-in sensors or cloud interfaces, and standardized processing is carried out, including data cleaning, format unification and numerical standardization, to eliminate inconsistencies; the processed data is integrated through a database management system to build a standardized cost database with structured storage to support subsequent real-time access and updates.

[0022] S02. Based on the preset risk rules, a risk indicator system is constructed. The risk indicator system includes static risk indicators, dynamic risk indicators and stage-related indicators.

[0023] Among these, static risk indicators (based on benchmark data, such as historical data from similar projects and deviation thresholds set by industry standards, for sub-items such as construction costs, equipment purchase costs, installation costs, and other expenses) are used to quantify the static deviation between actual values ​​and benchmark values. Dynamic risk indicators (combining historical price fluctuation patterns with dynamic deviation rate indicators at project progress nodes, including risk probability models for equipment and material prices and indicators that calculate dynamic trends based on the cumulative values ​​of sub-item costs to capture the dynamic risks of price and progress changes) are used. Stage-related indicators (based on transition paths between project stages, such as transitions from the feasibility study stage to the preliminary design stage and from the preliminary design stage to the construction drawing stage, with transition mutation thresholds set to identify abnormal and sudden changes in the total cost between stages) can be used to construct indicators. Pre-set risk rules can be applied to implement these indicators: A rule engine is used to parse benchmark data and set static risk indicators for each sub-item, such as formulating them based on the maximum deviation threshold. A model is trained using historical data to generate dynamic risk indicators for equipment and material prices, and dynamic deviation rates are calculated based on progress nodes. Stage-related thresholds are defined for stage paths to quantify transition mutation coefficients.

[0024] S03, based on the risk indicator system, uses the preset deep learning model to conduct real-time comparison and trend prediction of cost data in the cost database with benchmark data to identify cost risk points.

[0025] Among them, a preset deep learning model (a neural network architecture formed by training the time series data of cost data and benchmark data in a large number of different engineering cases, which is used to simultaneously perform feature comparison and trend deduction); cost risk points (abnormal states generated when actual data deviate from the benchmark, specifically represented by risk decision vector quantization); during implementation, the risk indicator system is input into the deep learning model, and the model extracts the multi-scale features of cost data and benchmark data through the convolutional layer, and uses the recurrent neural network layer to perform cross-stage time series deduction to generate a risk decision vector containing static deviation, dynamic trend probability and connection mutation coefficient; when it is detected that the static deviation exceeds the static risk indicator threshold, or the dynamic trend probability exceeds the dynamic risk indicator range, or the connection mutation coefficient exceeds the stage association threshold, the model locates the risk point at the chapter identification position in the cost list, and simultaneously outputs its cost impact range and current deviation value.

[0026] S04: Determine the risk level based on the deviation degree of the cost risk point and generate graded warning information and push it to the corresponding terminal.

[0027] Among them, the risk level (based on the composite deviation rate (CDR) mapping of three levels: minor, general or severe, used to quantify the urgency of the risk point); graded warning information (including the chapter identification position of the risk point in the cost list, used to locate the risk source, a quantitative matrix reflecting the current cost deviation value and the predicted deviation value, showing the deviation trend and the cost impact range of the risk point through a numerical model, and describing the extent of the risk impact on the project sub-item costs); pushing to the corresponding terminal (using the rule engine to intelligently distribute the warning information to the designated terminal, specifically including pushing to the first type of terminal (such as the project manager's mobile terminal) when the risk level is severe and triggering the preset emergency response process, and pushing to the second type of terminal (such as the cost engineer terminal) when the risk level is minor or general and providing historical risk disposal reference information). When implemented, the composite deviation rate CDR value is calculated based on the static deviation amount, dynamic trend probability and connection mutation coefficient of the risk decision vector in the cost risk point; according to the preset mapping rules (such as the CDR threshold range: minor corresponds to CDR < 0.3, general corresponds to 0.3 ≤ CDR < 0.6, severe corresponds to CDR ≥ 0.6) Determine the risk level; generate graded warning information, integrating location identifiers, quantification matrices, and impact ranges; and distribute information through a rules engine that dynamically matches push paths.

[0028] S05, recording the processing results of the graded warning information, and updating the standardized cost database based on the processing results.

[0029] Among them, the processing results of graded warning information (the disposal measures executed by the terminal for the received warning information and the effect data generated, such as the cost data or risk mitigation records corrected after manual intervention); updating the standardized cost database (based on the processing results, dynamically optimize the benchmark data in the database, such as historical similar project data, equipment and material reference prices and industry standard rates, to improve the accuracy of subsequent risk analysis). When implemented, it can receive the warning processing results fed back from the terminal in real time, analyze the effectiveness data of the disposal measures, and dynamically correct the corresponding benchmark fields in the standardized cost database (such as when the processing results show that the equipment and material prices have undergone structural changes, update the reference price library; or when the stage connection mutation is manually corrected, adjust the calculation model of the stage correlation threshold); by forming a closed-loop feedback mechanism, the disposal experience is converted into continuous optimization of the benchmark data, ensuring the dynamic adaptability of the risk indicator system and realizing the self-iteration upgrade of the whole process warning system.

[0030] The above-mentioned dynamic early warning method for the entire process of cost risk of power transmission and transformation projects obtains cost data and benchmark data of each stage of power transmission and transformation projects and builds a standardized cost database; constructs a risk indicator system based on preset risk rules, covering static risk indicators, dynamic risk indicators (such as equipment and material price fluctuation trends and dynamic deviation rates based on project progress nodes) and stage-related indicators (such as the transition mutation thresholds from feasibility study to preliminary design, and from preliminary design to construction drawing stages), to ensure the comprehensiveness of the risk identification framework; uses a preset deep learning model (trained through historical time series data) to compare and predict trends between cost data and benchmark data in real time, simultaneously performs feature comparison and cross-stage deduction, generates a risk decision vector containing static deviations, dynamic trends and transition mutations, and identifies cost risk points; determines the risk level (minor, general or serious) according to the degree of deviation of the risk point and generates graded early warning information, which is pushed to the corresponding terminal through the rule engine; records the processing results and updates the database to form a closed-loop feedback mechanism. To address the problem of delayed early warning, the deep learning model integrates dynamic risk indicators (such as price fluctuation trends) to respond to changes in equipment and material prices or project progress adjustments in real time, avoid delays caused by relying on static data, and improve the timeliness of early warning; to address the problem of incomplete risk coverage, stage-related indicators capture the connection mutations between paths such as feasibility study and preliminary design, and combine the synergy of static and dynamic indicators to avoid fragmented risk identification and prevent the risk of project overruns caused by cross-stage transmission effects; to address the problem of low decision-making efficiency, risk level determination (based on the compound deviation rate) and hierarchical push mechanism optimize the early warning path, ensuring that emergency processes are triggered in the case of serious risks and historical references are linked in the case of minor risks, thereby improving the real-time, accuracy and decision-making efficiency of risk management.

[0031] In one embodiment, obtaining cost data and benchmark data for each phase of a power transmission and transformation project and constructing a standardized cost database includes: S11, respectively obtain the estimated cost data of the feasibility study stage, the estimated cost data of the preliminary design stage, and the budget cost data of the construction drawing stage. The cost data covers construction costs, equipment purchase costs, installation costs, and other costs; S12, obtaining cost data of historical similar projects, reference prices of equipment and materials in the project area, and industry standard rates as benchmark data; S13, normalize the cost data and benchmark data of each stage and build a standardized cost database.

[0032] Specifically, the feasibility study phase (the estimated cost data during the project feasibility study phase represents the construction costs, equipment purchase, and installation costs estimated based on the preliminary design, specifically covering construction costs (such as civil and structural engineering costs), equipment purchase costs (such as transformer and switchgear procurement costs), installation costs (such as equipment installation and commissioning costs), and other costs (such as land use fees, project management, and service fees)); the preliminary design phase (the preliminary design phase, where the estimated cost data is the estimated value compiled after the detailed design); and the construction drawing phase (the construction drawing design phase, where the budgeted cost data is the detailed construction drawing budget). This data can be collected in real time or asynchronously through terminals or cloud data interfaces to ensure coverage throughout the project lifecycle. Benchmark data (including cost data of similar historical projects, such as historical cost records of similar transmission and transformation projects, reference prices for equipment and materials in the project area, such as regional market prices for cables or insulation materials, and industry standard rates, such as standard rates for installation labor or management fees) can be obtained from project databases or external market information platforms to provide a comparison benchmark for risk analysis. During implementation, the estimated data of the feasibility study stage, the budget data of the preliminary design stage, and the budget data of the construction drawing stage are extracted respectively; the query engine can be used to retrieve benchmark data from the historical library and regional price library; the above data are normalized, including data cleaning (removing outliers or duplicates), format unification (such as conversion to standard JSON or CSV format) and numerical standardization (such as normalizing the numerical range through the z-score method), to build a standardized cost database with structured storage to support subsequent real-time access.

[0033] In one embodiment, building a risk indicator system includes: S21, based on the benchmark data, set corresponding static risk indicators for construction costs, equipment purchase costs, installation costs and other costs; S23: Set dynamic risk indicators for equipment and material prices based on historical price fluctuation patterns, and set stage-by-stage dynamic risk indicators based on the project progress, calculating the dynamic deviation rate based on the accumulated value of the sub-item cost, based on the project progress nodes. S24: For the path from the feasibility study stage to the preliminary design stage and from the preliminary design stage to the construction drawing stage, a stage correlation threshold is set as a stage correlation indicator.

[0034] For example, static risk indicators are deviation thresholds set for key sub-items of power transmission and transformation project costs, such as construction costs, equipment purchase costs, installation costs, and other expenses. Construction costs cover civil engineering costs, equipment purchase costs include transformers, and installation costs involve labor costs for equipment commissioning. Other costs include land acquisition, project management, pre-construction consultation, and production preparation. These indicators can be generated through a rule engine analysis based on benchmark data, such as historical similar project costs and industry standard rates. During implementation, static thresholds can be set for each sub-item using pre-set risk rules (such as a maximum deviation threshold formula). For example, the allowable deviation range can be calculated based on a database query and then a benchmark value, ensuring a static basis for risk identification. Dynamic risk indicators include a probability model for equipment and material price fluctuations and a dynamic deviation rate based on project progress milestones. Risk indicators for equipment and material prices (such as cables or insulation materials) are generated using a probability model trained with historical price data to capture cyclical fluctuations. The dynamic deviation rate is calculated in real time based on the accumulated value of the sub-item costs, combining project progress milestones, such as foundation completion or equipment installation, and includes the ratio of accumulated costs to the benchmark. During implementation, machine learning algorithms can be used for analysis. The historical price rules are integrated with progress sensor data to dynamically update the deviation rate model to cope with market changes and project delay risks. The stage-related indicators are the transition paths of the project stages, such as the connection mutation thresholds from the feasibility study stage to the preliminary design stage, and from the preliminary design stage to the construction drawing stage. The stage path represents the evolution of the project from feasibility study to preliminary design and then to the construction drawing. The connection mutation coefficient is quantified by the absolute value ratio of the total cost change between the previous stage and the current stage. When implementing, the stage-related thresholds such as the maximum mutation rate can be set, and the path boundaries can be defined using the rule engine. When the mutation coefficient exceeds the limit, an early warning is triggered. The indicators are dynamically optimized through a closed-loop feedback mechanism to achieve dynamic monitoring of the entire process.

[0035] In one embodiment, a preset deep learning model is used to compare cost data in a cost database with benchmark data in real time and predict trends to identify cost risk points, including: S31, inputting static risk indicators, dynamic risk indicators, and stage-related indicators into a deep learning model, and synchronously performing feature comparison between cost data and benchmark data and cross-stage trend deduction through the deep learning model to generate a risk decision vector containing static deviations, dynamic trends, and transition mutations; S32, when it is detected that the static deviation exceeds the static risk index, or the dynamic trend exceeds the dynamic risk index, or the connection mutation coefficient breaks through the stage correlation index, the corresponding type of cost risk point is identified and the chapter identification position of the cost risk point in the cost list, the cost impact range and the current cost deviation are determined and output.

[0036] Specifically, the preset deep learning model is a neural network architecture trained with time-series data from historical power transmission and transformation projects, which is used to simultaneously perform feature comparison (extracting the difference features between current cost data and benchmark data, such as the cost of similar historical projects) and cross-stage trend deduction (deducing future risk trends based on the evolution of project stages, such as feasibility study to preliminary design); the risk decision vector is the data structure output by the model, which includes static deviation (the deviation between actual sub-item costs, such as construction project costs and benchmark values), dynamic trends (future probability of equipment and material price fluctuations) and transition mutations (the coefficient of change in total cost between stages). During implementation, the static risk indicators (such as the construction project cost deviation threshold), dynamic risk indicators (such as the equipment and material price fluctuation probability) and stage-related indicators (such as the stage transition mutation threshold) in the risk indicator system are input into the model. The model extracts multi-scale features (such as the numerical distribution of sub-item costs) through the convolution layer and uses the recurrent neural network layer to perform time series analysis to generate a risk decision vector; the vector is detected in real time. When the static deviation exceeds the static risk indicator (such as the construction project cost deviation is greater than the threshold), or the dynamic trend exceeds the dynamic risk indicator (such as the price fluctuation probability exceeds the limit), or the transition mutation coefficient breaks through the stage-related indicator (such as the mutation rate from the feasibility study to the preliminary design stage exceeds the threshold), the model accurately identifies the corresponding risk point type (static, dynamic or transition type), and determines its chapter identification position in the cost list (such as the specific cost chapter code), the cost impact range (such as the equipment purchase cost range affected) and the current cost deviation value (such as the difference between the actual and the benchmark), and outputs an early warning. Through artificial intelligence technology to analyze the risk core, the real-time and accuracy of the early warning is improved, and the problems of early warning lag and incomplete risk coverage are solved.

[0037] In one embodiment, based on the degree of deviation of the cost risk point, the risk level is determined and graded warning information is generated and pushed to the corresponding terminal, including: S41, based on the static deviation amount, dynamic trend probability and connection mutation coefficient in the risk decision vector, calculate the composite deviation rate and determine the risk level according to the preset mapping rules; the risk level includes minor, general or severe; S42: Generate warning information based on the risk level; the warning information includes the chapter identification position of the risk point corresponding to the risk level in the cost list; a quantitative matrix reflecting the current cost deviation value and the predicted cost deviation value of the risk point; and the cost impact range of the risk point; S43, through the rule engine, the warning information is matched with the push path according to the risk level. When the risk level is serious, it is pushed to the first type of terminal and triggers the emergency response process preset on the first type of terminal. When the risk level is minor or general, it is pushed to the second type of terminal and associated with the preset historical risk disposal reference information.

[0038] For example, the composite deviation rate (CDR) is an indicator that comprehensively quantifies the degree of deviation of risk points, which is calculated through static deviation, dynamic trend probability and connection mutation coefficient; the risk level includes three levels: minor, general or severe, which can be determined based on preset mapping rules (such as CDR threshold interval: minor corresponds to CDR < 0.3, general corresponds to 0.3 ≤ CDR < 0.6, and severe corresponds to CDR ≥ 0.6); the early warning information includes the chapter identification position of the risk point in the cost list (such as the specific cost chapter code), a quantitative matrix reflecting the current cost deviation value and the predicted deviation value (displaying the deviation trend through a numerical model), and the cost impact range of the risk point (describing the degree of impact on the project sub-item costs); the first type of terminal refers to the terminal that receives serious risk warnings (such as the project manager's mobile terminal), which triggers the preset emergency response process, and the second type of terminal refers to the terminal that receives minor or general risk warnings (such as the cost engineer terminal), and provides associated preset historical risk disposal reference information. During implementation, the CDR value can be calculated based on the static deviation, dynamic trend probability and connection mutation coefficient in the risk decision vector (from the previous step), and the risk level can be determined by applying the preset mapping rules through the rule engine; early warning information is generated according to the risk level, integrating the chapter identification position, quantification matrix (actual deviation value association, prediction interval association) and impact range; the push path is matched through the rule engine, and in case of serious risks, it is pushed to the first type of terminal and triggers its emergency response process (such as automatic alarms and intervention instructions); in case of minor or general risks, it is pushed to the second type of terminal and associated with the historical disposal reference information library (such as risk resolution cases of similar projects), realizing a closed loop from risk identification to early warning distribution.

[0039] In one embodiment, S51 calculates the composite deviation rate based on the static deviation amount, dynamic trend probability, and transition mutation coefficient in the risk decision vector, which is implemented by the following formula:

[0040] in is the composite deviation rate, is the static deviation, Indicates the actual value of the item cost at the current stage. Indicates the benchmark cost data, Indicates the deviation threshold corresponding to the static risk indicators set for construction project costs, equipment purchase costs, installation project costs and other costs; is the dynamic trend probability, which indicates the probability value of the future N periods of construction costs corresponding to the dynamic trend output by the deep learning model exceeding the dynamic risk index, and its value is ; is the transition coefficient, Indicates the actual total cost of the current project phase, Indicates the total actual cost of the previous project stage; , and is a dynamic weight factor, satisfying .

[0041] Specifically, the compound deviation rate (CDR) can be calculated by the formula calculate, Static deviation, such as the actual value of construction project cost Benchmark cost data such as historical similar project data The absolute deviation ratio, Deviation thresholds (such as maximum allowable deviation values) corresponding to static risk indicators set for construction costs, equipment purchase costs, installation costs, and other costs, It is a probability value ranging from 0 to 1, indicating the probability that the cost in the next N periods (e.g., the next three months) corresponding to the dynamic trend output by the deep learning model will exceed the dynamic risk indicator (e.g., the equipment and material price fluctuation threshold). The transition coefficient is the actual total cost of the current project stage (e.g., the construction drawing stage). Compared with the actual total cost of the previous engineering stage (such as the initial design stage) The change ratio of quantifies the degree of cost mutation between stages. , and is a dynamic weight factor, satisfying Constraints are used to dynamically adjust the weights of each deviation component according to the engineering context (e.g. Focus on static deviation, Focus on dynamic trend stability). When implemented, the rule engine can be used to extract the risk decision vector (such as the data structure generated in the previous step) 、 and The CDR is calculated using the value of and Calculated based on the real-time itemized cost data (such as the actual value of equipment purchase costs) and stage total value (such as the transition total value from the feasibility study stage to the preliminary design stage) in the standardized cost database. The output is predicted by the deep learning model, and the weight factor is dynamically optimized through training of historical engineering data to ensure that the composite deviation rate accurately reflects the multi-dimensional deviation characteristics of the risk point, solving the problems of delayed warning and incomplete risk coverage.

[0042] In one embodiment, the method further includes inputting the graded warning information into a visualization engine and performing the following steps: S61, generating a heat map based on the chapter marker positions in the warning information, and mapping the color depth of the heat map to the value of the composite deviation rate (CDR); S62, based on the current cost deviation value and the predicted deviation value in the warning information, generate a deviation fluctuation map, in which the actual deviation value is associated with the static deviation value , the probability of dynamic trend associated with the prediction interval ; S63, based on the cost impact range and connection mutation coefficient in the early warning information , generate a cross-stage risk transmission tree diagram, and the influence weights between nodes in the tree diagram are given by Value quantification.

[0043] Specifically, the visualization engine is a software module dedicated to converting structured warning information into visual charts, assisting risk decision-making through a graphical interface. A heat map is a color-coded chart generated based on the location of chapter identifiers in the warning information (such as the specific chapter codes for construction project costs or equipment purchase costs in the cost list). The color depth of the chart is mapped to the value of the composite deviation rate (CDR) through an algorithm (for example, color scales from light green to dark red represent a low to high CDR, visually displaying the spatial distribution and severity of risk points). The deviation fluctuation map is a chart that displays the time-series changes of the current cost deviation value and the predicted deviation value. The actual deviation value is associated with the static deviation amount, representing the instantaneous deviation between the actual cost and the benchmark, and the prediction interval is associated with the dynamic trend probability (output by the deep learning model, representing the probability range of the cost exceeding the dynamic risk indicator within the next N cycles, such as 3 months). The cross-stage risk transmission tree diagram describes the propagation path of risk points between project stages (such as the feasibility study stage to the construction drawing stage). Nodes represent sub-items or project stages, edges represent transmission relationships, and the impact weight is quantified by the connection mutation coefficient, which is used to indicate the intensity of the impact of the mutation degree on adjacent nodes. During implementation, the chapter marker position in the warning information can be parsed and the rendering algorithm of the visualization engine can be called to generate a heat map. For example, the CDR value can be mapped to the HSV color scale model using the OpenCV library, and a color gradient image can be output to highlight high-risk chapters. Based on the deviation data in the warning information (including the current actual deviation value and the predicted deviation value), a deviation fluctuation map can be generated using a time series analysis library (such as Matplotlib). The actual deviation line is based on Values ​​are plotted, with prediction intervals based on Calculate confidence intervals; construct a tree diagram using graph theory algorithms (such as NetworkX), with the cost impact range in the warning information (such as the impact of equipment purchase cost fluctuations on installation project costs) as nodes. The value is the edge weight (such as >0.5, the edge thickness increases). Visualize the cross-stage transmission path of risk. Heat maps provide risk location, deviation fluctuation maps enhance the interpretability of trend forecasts, and tree diagrams reveal systemic transmission risks. Through intuitive display and feedback to the closed-loop system, the accuracy of full-process monitoring is improved.

[0044] The above-mentioned dynamic early warning method for the whole process of cost risk of power transmission and transformation project builds a standardized cost database by acquiring cost data and benchmark data of each stage of power transmission and transformation project, and builds a risk indicator system based on preset risk rules, covering static risk indicators (such as deviation thresholds of construction project costs, equipment purchase costs, installation project costs and other costs), dynamic risk indicators (such as equipment and material price fluctuation trends and dynamic deviation rates based on project progress nodes) and stage-related indicators (such as the connection mutation thresholds from feasibility study stage to preliminary design stage, and from preliminary design stage to construction drawing stage). The preset deep learning model is used to compare the cost data with the benchmark data in real time and predict trends, and the feature comparison is simultaneously executed. Generate risk decision vectors (including static deviation, dynamic trend probability and connection mutation coefficient) for cross-stage deductions, identify cost risk points and their chapter identification positions in the cost list, cost impact range and current deviation values; calculate the compound deviation rate according to the deviation degree of the risk point, determine the slight, general or serious risk level, and generate graded warning information (including chapter identification position, quantitative matrix and impact range) through the rule engine and push it to the corresponding terminal (in case of serious risk, it is pushed to the first type of terminal to trigger the emergency response process, and in case of slight or general risk, it is pushed to the second type of terminal to associate historical disposal reference information), and record the processing results and update the database to form a closed-loop feedback mechanism. To address the problem of delayed early warning: the deep learning model integrates dynamic risk indicators (such as price fluctuation trends) to respond to changes in equipment and material prices or project progress adjustments in real time, avoiding delays caused by reliance on static data; to address the problem of incomplete risk coverage: stage-related indicators capture cross-stage transmission effects, and combine the collaborative analysis of static and dynamic indicators to avoid the risk of project overruns caused by fragmented risk identification; to address the problem of low decision-making efficiency: based on the risk level determination of the composite deviation rate and the intelligent graded push mechanism, the early warning path is optimized to ensure that emergency processes are quickly triggered in the event of serious risks, and historical references are associated in the event of minor risks to improve decision support efficiency, significantly improving the real-time, accuracy and overall decision-making efficiency of cost risk management.

[0045] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0046] Based on the same inventive concept, embodiments of the present application also provide a system for implementing the aforementioned method for dynamically warning the cost risk of power transmission and transformation projects. The solution provided by this system is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the system for dynamically warning the cost risk of power transmission and transformation projects provided below can be found in the aforementioned definition of the method for dynamically warning the cost risk of power transmission and transformation projects, and will not be further elaborated here.

[0047] In an exemplary embodiment, Figure 2 As shown, a dynamic early warning system for the entire process of power transmission and transformation project cost risk is provided, including: The data acquisition and processing module 101 is used to obtain cost data and benchmark data for each stage of the power transmission and transformation project and to build a standardized cost database; The risk indicator management module 102 is used to build a risk indicator system based on preset risk rules. The risk indicator system includes static risk indicators, dynamic risk indicators, and stage-related indicators. Intelligent risk analysis module 103 is used to compare cost data in the cost database with benchmark data in real time and conduct trend prediction based on the risk indicator system and the preset deep learning model to identify cost risk points; The warning generation and response module 104 is used to determine the risk level according to the deviation degree of the cost risk point and generate graded warning information and push it to the corresponding terminal; The feedback optimization module 105 is used to record the processing results of the graded warning information and update the standardized cost database according to the processing results.

[0048] In one embodiment, the data acquisition and processing module 101 is further configured to: Obtain estimated cost data for the feasibility study phase, budgeted cost data for the preliminary design phase, and budgeted cost data for the construction drawing phase. The cost data covers construction costs, equipment purchase costs, installation costs, and other expenses. Obtain cost data for similar historical projects, reference prices for equipment and materials in the project area, and industry standard rates as benchmark data; Normalize the cost data and benchmark data of each stage and build a standardized cost database.

[0049] In one embodiment, the risk indicator management module 102 is further configured to: Based on the benchmark data, corresponding static risk indicators are set for construction costs, equipment purchase costs, installation costs and other costs; Set dynamic risk indicators for equipment and material prices based on historical price fluctuation patterns, and set stage-by-stage dynamic risk indicators based on project progress nodes and the dynamic deviation rate calculated based on the accumulated value of the sub-item cost. For the paths from the feasibility study stage to the preliminary design stage and from the preliminary design stage to the construction drawing stage, a stage correlation threshold is set as the stage correlation indicator.

[0050] In one embodiment, the intelligent risk analysis module 103 is further configured to: Static risk indicators, dynamic risk indicators, and stage-related indicators are input into the deep learning model. The deep learning model then performs feature comparison between cost data and benchmark data and cross-stage trend deduction to generate a risk decision vector that includes static deviations, dynamic trends, and transition mutations. When it is detected that the static deviation exceeds the static risk index, or the dynamic trend exceeds the dynamic risk index, or the connection mutation coefficient breaks through the stage correlation index, the corresponding type of cost risk point is identified and the chapter identification position of the cost risk point in the cost list, the cost impact range and the current cost deviation are determined and output.

[0051] In one embodiment, the warning generation and response module 104 is further configured to: Based on the static deviation amount, dynamic trend probability and connection mutation coefficient in the risk decision vector, the compound deviation rate is calculated and the risk level is determined according to the preset mapping rules; the risk level can be minor, general or severe; Generate early warning information based on risk levels; the early warning information includes the chapter identification position of the risk point corresponding to the risk level in the cost list; a quantitative matrix reflecting the current cost deviation value and the predicted deviation value of the risk point; and the cost impact range of the risk point; The rule engine matches the warning information with the push path according to the risk level. When the risk level is serious, it is pushed to the first type of terminal and triggers the emergency response process preset on the first type of terminal. When the risk level is minor or general, it is pushed to the second type of terminal and associated with the preset historical risk disposal reference information.

[0052] In one embodiment, the warning generation and response module 104 is further configured to calculate the composite deviation rate based on the static deviation amount, dynamic trend probability, and transition mutation coefficient in the risk decision vector using the following formula:

[0053] in is the composite deviation rate, is the static deviation, Indicates the actual value of the item cost at the current stage. Indicates the benchmark cost data, Indicates the deviation threshold corresponding to the static risk indicators set for construction project costs, equipment purchase costs, installation project costs and other costs; is the dynamic trend probability, which indicates the probability value of the future N periods of construction costs corresponding to the dynamic trend output by the deep learning model exceeding the dynamic risk index, and its value is ; is the transition coefficient, Indicates the actual total cost of the current project phase, Indicates the total actual cost of the previous project stage; , and is a dynamic weight factor, satisfying .

[0054] In one embodiment, the system further includes a visualization module for: Based on the chapter mark position in the warning information, a heat map is generated, and the color depth of the heat map is mapped to the value of the composite deviation rate (CDR); Based on the current cost deviation value and the predicted deviation value in the early warning information, a deviation fluctuation map is generated, in which the actual deviation value is associated with the static deviation value. , the probability of dynamic trend associated with the prediction interval ; Based on the cost impact range and connection mutation coefficient in early warning information , generate a cross-stage risk transmission tree diagram, and the influence weights between nodes in the tree diagram are given by Value quantification.

[0055] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the dynamic early warning method for the entire process of cost risk of power transmission and transformation projects as described above are implemented.

[0056] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0057] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0058] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A dynamic early warning method for the entire process of cost risk of power transmission and transformation projects, characterized by: The method comprises: Obtain cost data and benchmark data for each stage of power transmission and transformation projects and build a standardized cost database; Based on the preset risk rules, a risk indicator system is constructed, which includes static risk indicators, dynamic risk indicators and stage-related indicators; Based on the risk indicator system, a preset deep learning model is used to compare the cost data in the cost database with the benchmark data in real time and predict trends to identify cost risk points; According to the deviation degree of the cost risk point, the risk level is determined and graded warning information is generated and pushed to the corresponding terminal; Record the processing results of the graded warning information, and update the standardized cost database based on the processing results.

2. The method according to claim 1, characterized in that The acquisition of cost data and benchmark data for each stage of the power transmission and transformation project and the construction of a standardized cost database include: Obtain estimated cost data for the feasibility study phase, budgeted cost data for the preliminary design phase, and budgeted cost data for the construction drawing phase. The cost data covers construction costs, equipment purchase costs, installation costs, and other expenses. Obtaining cost data of historical similar projects, reference prices of equipment and materials in the region where the project is located, and industry standard rates as the benchmark data; The cost data of each stage and the benchmark data are normalized to construct the standardized cost database.

3. The method according to claim 2, characterized in that The risk indicator system is constructed, including: Based on the benchmark data, corresponding static risk indicators are set for the construction project costs, equipment purchase costs, installation project costs and other costs; Set dynamic risk indicators for equipment and material prices based on historical price fluctuation patterns, and set stage-by-stage dynamic risk indicators based on project progress nodes and the dynamic deviation rate calculated based on the accumulated value of the sub-item cost. For the path from the feasibility study stage to the preliminary design stage and from the preliminary design stage to the construction drawing stage, a stage correlation threshold is set as the stage correlation indicator.

4. The method according to claim 1, wherein The use of a preset deep learning model to perform real-time comparison and trend prediction of the cost data in the cost database with the benchmark data to identify cost risk points includes: Inputting the static risk indicators, dynamic risk indicators, and stage-related indicators into the deep learning model, and synchronously performing feature comparison and cross-stage trend deduction of cost data and benchmark data through the deep learning model to generate a risk decision vector containing static deviations, dynamic trends, and transition mutations; When it is detected that the static deviation exceeds the static risk index, or the dynamic trend exceeds the dynamic risk index, or the transition mutation coefficient breaks through the stage-related index, the corresponding type of cost risk point is identified and the chapter identification position, cost impact range and current cost deviation of the cost risk point in the cost list are determined and output.

5. The method according to claim 4, characterized in that The method of determining the risk level according to the deviation degree of the cost risk point and generating graded warning information and pushing it to the corresponding terminal includes: Based on the static deviation amount, dynamic trend probability and connection mutation coefficient in the risk decision vector, the composite deviation rate is calculated, and the risk level is determined according to the preset mapping rules; the risk level includes minor, general or severe; Based on the risk level, early warning information is generated; the early warning information includes the chapter identification position of the risk point corresponding to the risk level in the cost list; a quantitative matrix reflecting the current cost deviation value and the predicted deviation value of the risk point; and the cost impact range of the risk point; The rule engine matches the warning information with the push path according to the risk level. When the risk level is serious, it is pushed to the first type of terminal and triggers the emergency response process preset on the first type of terminal. When the risk level is minor or general, it is pushed to the second type of terminal and associated with the preset historical risk disposal reference information.

6. The method according to claim 5, characterized in that The calculation of the composite deviation rate based on the static deviation amount, dynamic trend probability and connection mutation coefficient in the risk decision vector is achieved by the following formula: ， in is the composite deviation rate, is the static deviation, Indicates the actual value of the item cost at the current stage. Indicates the benchmark cost data, Indicates the deviation threshold corresponding to the static risk indicators set for construction project costs, equipment purchase costs, installation project costs and other costs; is the dynamic trend probability, which indicates the probability value of the future N periods of construction costs corresponding to the dynamic trend output by the deep learning model exceeding the dynamic risk index, and its value is ; is the transition coefficient, Indicates the actual total cost of the current project phase, Indicates the total actual cost of the previous project stage; , and is a dynamic weight factor, satisfying .

7. The method according to claim 6, characterized in that The method further includes inputting the graded warning information into a visualization engine and performing the following steps: Based on the chapter mark position in the warning information, a heat map is generated, wherein the color scale depth of the heat map maps the value of the composite deviation rate CDR; Based on the current cost deviation value and the predicted deviation value in the warning information, a deviation fluctuation map is generated, wherein the actual deviation value is associated with the static deviation value. , the prediction interval is associated with the dynamic trend probability ; Based on the cost impact range in the warning information and the connection mutation coefficient , generating a cross-stage risk transmission tree diagram, in which the influence weights between nodes are given by Value quantification.

8. A dynamic early warning system for the entire process of cost risk of power transmission and transformation projects, characterized by: The system comprises: Data acquisition and processing module, used to obtain cost data and benchmark data for each stage of power transmission and transformation projects and build a standardized cost database; A risk indicator management module is used to construct a risk indicator system based on preset risk rules. The risk indicator system includes static risk indicators, dynamic risk indicators, and stage-related indicators; An intelligent risk analysis module is used to compare the cost data in the cost database with the benchmark data in real time and conduct trend prediction based on the risk indicator system and to identify cost risk points using a preset deep learning model; The early warning generation and response module is used to determine the risk level according to the deviation degree of the cost risk point and generate graded early warning information and push it to the corresponding terminal; The feedback optimization module is used to record the processing results of the graded warning information and update the standardized cost database according to the processing results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images