A collaborative monitoring and obstacle risk quantification decision-making method for power grid planning projects

Abstract

This invention discloses a collaborative monitoring and obstacle risk quantification decision-making method for power grid planning projects, comprising: S1: collecting and standardizing five types of data related to power grid planning projects, and completing the quantification of regional characteristic parameters and multi-dimensional correlation mapping; S2: constructing three basic models: provincial and municipal level collaborative monitoring, project implementation evaluation, and obstacle-pre-feasibility study linkage, and completing parameter fitting and verification; S3: conducting provincial and municipal level collaborative dynamic monitoring simulation, four-dimensional quantitative transmission of obstacle risk, and pre-feasibility study constraint sensitivity analysis; S4: performing implementation-pre-feasibility study two-way linkage simulation, provincial and municipal level penetration analysis, and accurate calculation of regional characteristic impact; S5: constructing a multi-objective optimization model, solving and verifying the optimal solution based on the output results of S4, and outputting standardized decision results. This invention solves the problems of lack of regional adaptation, fragmented hierarchical data, and inability to accurately quantify risk transmission in existing power grid planning monitoring and decision-making technologies.

Description

Technical Field

[0001] This invention belongs to the field of power system planning technology, and specifically relates to a collaborative monitoring and obstacle risk quantification decision-making method for power grid planning projects. Background Technology

[0002] Currently, a mature technological system has been formed in the field of power grid planning, implementation monitoring, and project management. The mainstream existing technologies include three core product categories: online monitoring platforms for power grid planning projects, transmission and transformation engineering obstruction risk assessment systems, and intelligent auxiliary decision-making tools for power grid project pre-feasibility studies. For example, Chinese patent CN116566039B discloses a transmission line monitoring system and method based on cloud-edge-device collaborative sensing, which realizes multi-terminal data acquisition and status monitoring of power grid lines. Chinese patent CN115545477B discloses a transmission line congestion risk probability assessment method and product based on incremental interpolation, which utilizes a modified neural network to assess the congestion risk probability of target lines. Chinese patent CN119671307A discloses an intelligent review method and system for power grid infrastructure projects, providing intelligent review support for the feasibility study stage of power grid projects.

[0003] However, the existing technology has the following technical defects: 1. Existing power grid planning, monitoring and decision-making technologies all adopt generalized model design, without making specific quantitative adaptations for complex regional characteristics. The regional constraints are disconnected from the monitoring model, and cannot be adapted to differentiated regional conditions such as meteorology, geology, social coordination, and ecological management. 2. The lack of a two-level data collaboration mechanism results in significant information silos, preventing real-time communication and coordinated management between provincial-level coordinated data and municipal-level implementation data; 3. The obstruction risk can only be qualitatively described or simply scored, and the chain transmission of obstruction factors cannot be accurately quantified, resulting in inaccurate risk assessment results and delayed handling. 4. The planning and implementation monitoring stage and the pre-feasibility study constraint stage are independent of each other, without a two-way linkage closed loop. The pre-feasibility study conclusions cannot guide the implementation and control, and the implementation status cannot reverse the pre-feasibility study parameters. This results in insufficient accuracy of project control and low implementation efficiency, which cannot meet the actual needs of refined, regionalized, and full-process collaborative control of 220kV and above high voltage power grid planning projects. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention proposes a collaborative monitoring and obstacle risk quantification decision-making method for power grid planning projects.

[0005] The technical solution of the present invention is as follows: On the one hand, this invention provides a method for collaborative monitoring and quantification of obstruction risk in power grid planning projects, comprising the following steps: S1: Collect and standardize five types of data, including full life-cycle data of power grid planning projects, collaborative business data between provincial and local levels, regional constraint data, project obstruction risk data, and pre-feasibility study spatial constraint data. Perform regional feature parameter quantification and multi-dimensional correlation mapping on the five types of data, and output a standardized fusion dataset and a multi-dimensional correlation mapping matrix. S2: Based on the standardized fusion dataset, construct three basic models for provincial and local collaborative monitoring, project implementation evaluation, and obstacle-pre-feasibility study linkage, and complete the parameter fitting and verification of the three basic models; S3: Based on the three types of basic models, conduct provincial and local collaborative dynamic monitoring simulation, four-dimensional quantitative transmission of obstructed risks, and pre-feasibility study constraint sensitivity analysis; S4: Based on the output results of the four-dimensional quantitative transmission of obstructed risks and the sensitivity analysis of pre-feasibility study constraints, the pre-feasibility study bidirectional linkage simulation, provincial and municipal two-level penetration analysis, and precise calculation of regional characteristic impacts are performed to obtain the bidirectional linkage simulation results, provincial and municipal two-level penetration analysis results, and regional characteristic impact quantitative results. S5: Construct a multi-objective optimization model, solve and verify the optimal solution based on the two-way linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and output standardized decision results.

[0006] Preferably, in step S1, the quantization of regional feature parameters and multidimensional association mapping of the five types of data are performed, and the standardized fused dataset and multidimensional association mapping matrix are output as follows: The regional constraint data among the five types of data are quantified, including extreme weather characteristics, geological disaster characteristics, social coordination characteristics, and life constraint characteristics. Based on five types of data that have undergone standardized processing and quantification of regional characteristics, and based on the weighting rules determined by the expert scoring method, a multi-dimensional correlation mapping matrix is ​​established, which includes "project implementation - provincial and local level collaboration - regional characteristics - obstacles and risks - pre-feasibility study constraints".

[0007] Preferably, step S2 specifically comprises: The provincial and municipal level collaborative monitoring basic model consists of a perception layer, a data layer, a support layer, and an application layer. The perception layer is used to input standardized fusion datasets. The data layer, based on a multi-dimensional correlation mapping matrix, performs model operation format standardization on the standardized fusion datasets input from the perception layer. The support layer has built-in parameter resource libraries, algorithm resource libraries, and mapping resource libraries. By calling these three types of resource libraries, it provides parameter, algorithm, and correlation support for application layer simulation and indicator calculation. The application layer calls the data standardized by the data layer and the configuration resources of the support layer to perform parameter initialization and validity verification, and outputs the simulation input dataset. The project implementation evaluation model is composed of weighted indicators from four dimensions: progress, quality, benefits, and collaboration. The progress indicators include the completion rate of preliminary approvals, the construction progress deviation rate, and the on-time commissioning and acceptance rate. The quality indicators include the project acceptance pass rate, the equipment delivery pass rate, and the safety hazard rectification rate. The benefits indicators include clean energy consumption, power supply reliability improvement, and return on investment. The collaboration indicators include the timeliness of provincial and local information reporting, the cross-departmental approval collaboration rate, and the instruction response completion rate. The model calculates the scores for each dimension indicator by calling a standardized fusion dataset, and outputs the basic results of the project implementation evaluation. The hindered-pre-feasibility study linkage basic model adopts a dual-link closed-loop architecture of forward transmission link and backward correction link. The forward transmission link follows the flow of monitoring simulation → implementation evaluation → hindered quantification → pre-feasibility study constraint correction. The backward correction link follows the flow of pre-feasibility study constraint assessment → implementation risk calculation → monitoring parameter optimization → evaluation score correction. The standardized fusion dataset is substituted into the hindered-pre-feasibility study linkage basic model for iterative calculation and benchmarking against the actual implementation effect of the project to complete the model parameter fitting and output the correlation parameter table.

[0008] Preferably, S3 specifically includes: S31: Substitute the simulation input dataset into the provincial and municipal collaborative monitoring basic model, use the gradient descent algorithm to carry out iterative simulation, and output the provincial and municipal collaborative dynamic monitoring simulation results, including project progress, construction quality, and provincial and municipal collaborative status. S32: Construct a four-dimensional obstruction model with a three-level architecture of four-dimensional parallel input layer - sub-item feature conversion layer - comprehensive obstruction result output layer. The four-dimensional obstruction model decomposes and refines the standardized fusion dataset and multi-dimensional correlation mapping matrix into three categories of obstruction factors, including natural environment, policy constraints and external coordination, and internal management. The three categories of obstruction factors are combined with predetermined regional feature weights to perform individual risk scoring. The scores of the three categories of obstruction factors are aggregated and weighted to output the overall project obstruction quantitative score and the probability of obstruction. A risk transmission flow graph based on a four-dimensional hindered model is constructed using system dynamics, defining three types of model variables: level variables, rate variables, and auxiliary variables. The risk transmission flow graph adopts a three-layer hierarchical architecture: auxiliary variable input layer, rate-level variable transmission layer, and comprehensive risk output layer, outputting dynamic cumulative risk results. Based on the four-dimensional obstruction model, the hierarchical architecture of the analytic hierarchy process (AHP) criteria layer and the scheme layer is divided. The input data of the AHP consists of the scores of three categories of obstruction factors. The input data of the fuzzy comprehensive evaluation includes the index weights obtained by the AHP solution and the dynamic cumulative risk results. The four-dimensional quantitative result of obstruction risk is calculated by the AHP-fuzzy comprehensive evaluation combined algorithm. S33: Select four core elements of standardized fusion datasets: centralized site selection, corridor, ecological avoidance, and meteorological protection, to build a parameterized configuration framework. Perform two-way parameter adjustment on the data parameters corresponding to the four core elements: constraint tightening and constraint relaxation. Generate differentiated simulation scenario scheme parameter combinations through cross-combination of multiple elements and multiple control directions. The differentiated simulation scenario parameter combinations are substituted into the provincial and municipal collaborative monitoring basic model. Using the control variable method to keep other baseline parameters constant, the parameter impact magnitude of project feasibility under varying core element parameters is quantitatively calculated. Based on the impact magnitude of each core element parameter, the average impact magnitude of the four types of core elements under multiple operating conditions is calculated, and the core elements are ranked in descending order of sensitivity according to the magnitude of the average impact magnitude under multiple operating conditions. By fitting the correlation between the variation magnitude of each core element parameter and project feasibility, the parameter constraint critical values ​​for each core element to meet the pre-feasibility study feasibility requirements are obtained through reverse iterative solution. A pre-feasibility study constraint sensitivity analysis report containing the ranking results and parameter constraint critical values ​​is output.

[0009] Preferably, S4 specifically comprises: S41: Import the four-dimensional quantification result of the obstructed risk obtained from the four-dimensional quantification of the obstructed risk into the obstructed-pre-feasibility study linkage basic model. Based on the forward transmission link and backward correction link of the obstructed-pre-feasibility study linkage basic model, complete the bidirectional iterative substitution of the four-dimensional quantification result of the obstructed risk, execute the bidirectional linkage simulation of the obstructed-pre-feasibility study linkage basic model, call GraphSAGE graph neural network to mine and complete the potential correlation between elements, correct the linkage coefficient, and output the bidirectional linkage simulation results including linkage coefficient, risk transmission path, and rule correction record. S42: Perform quantitative processing of five types of data, including provincial and local level collaborative authority, regional constraint strength, and approval process duration, and combine the three types of influencing factors to perform weighted calculation based on preset influencing factor weights to obtain three levels of influencing values. The impact values ​​of each level are statistically analyzed at the provincial-level overall network coordination dimension and the prefecture-level single project implementation dimension. Then, using the provincial-level overall indicators as a benchmark, the single project indicators of each prefecture are compared with the provincial-level overall indicators item by item, the absolute difference and relative deviation of the indicators are calculated, the deviation of the indicators at the two levels are sorted out, the difference characteristics, difference range and distribution patterns are summarized, and the hierarchical difference analysis results at the provincial and prefecture levels are generated. In combination with the preset fluctuation threshold of the impact factors, business link tracing is carried out, and abnormal items exceeding the threshold are investigated layer by layer to complete the location of control breakpoints at each link. Finally, the penetrating analysis results at the provincial and prefecture levels are output, including the hierarchical difference analysis results and the location of control breakpoints. S43: Based on the standardized fusion dataset, the four-dimensional quantitative transmission of obstruction risk, and the output results of the pre-feasibility study constraint sensitivity analysis, the schedule delay rate, the probability of project obstruction, and the pre-feasibility study feasibility coefficient are weighted and calculated; the schedule delay rate, the probability of project obstruction, and the pre-feasibility study feasibility coefficient are normalized and combined with the optimized linkage coefficient for weighted fusion, and finally the quantitative results of regional feature influence are output.

[0010] Preferably, S5 constructs a multi-objective optimization model, solves and verifies the optimal solution based on the two-way linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and outputs standardized decision results as follows: The decision objectives and constraints are constructed as follows: the decision objectives are set as dual core objectives of maximizing project implementation efficiency and optimizing pre-feasibility study feasibility; a single-objective constraint function is synthesized based on the dual core decision objectives to optimize the feasible domain definition and the coarse screening of ineffective solutions; the constraints include the provincial and local level coordination cycle range, the regional cost increase range, the construction period delay ratio range, and the ecological constraint compliance rate. A hybrid algorithm combining scenario-optimized improved hierarchical analysis and fuzzy comprehensive evaluation was adopted to complete the construction of the index system and the basic weight calibration of the multi-objective optimization model. The improvements of the improved hierarchical analysis include three aspects: data-driven construction of the judgment matrix, addition of a dynamic weight correction mechanism, and linkage of fuzzy result optimization consistency verification. The equipment sample dataset was constructed using regional historical power grid planning project data and trained on the multi-objective model by dividing it into sets at a fixed ratio. The results of bidirectional linkage simulation, provincial and municipal penetration analysis, and regional characteristic influence quantification are integrated and input into the multi-objective optimization model solver for iterative solution, and the Pareto optimal solution set is generated as the output. Based on the Pareto optimal solution set, a combination of differentiated project optimization decision schemes is output. A comprehensive evaluation of the combination of decision-making schemes for differentiated projects is conducted from three dimensions: target achievement, regional adaptability, and implementation feasibility. The scheme with the highest comprehensive score is selected as the optimal scheme, and the effectiveness of the optimal scheme is verified. After the verification is successful, standardized decision-making results are output.

[0011] On the other hand, the present invention provides a collaborative monitoring and obstacle risk quantification decision-making system for power grid planning projects, including a data acquisition and preprocessing module, a basic model construction and verification module, a simulation and deduction module, a multi-dimensional linkage calculation module, and a standardized decision output module. The data acquisition and preprocessing module is used to collect and standardize five types of data, including full life-cycle data of power grid planning projects, collaborative business data between provincial and local levels, regional constraint data, project obstruction risk data, and pre-feasibility study spatial constraint data. It performs regional feature parameter quantification and multi-dimensional correlation mapping on the five types of data and outputs a standardized fusion dataset and a multi-dimensional correlation mapping matrix. The basic model construction and verification module is used to build three types of basic models based on standardized fusion datasets: provincial and local collaborative monitoring, project implementation evaluation, and obstacle-pre-feasibility study linkage, and to complete the parameter fitting and verification of the three types of basic models. The simulation and deduction module is used to carry out provincial and local collaborative dynamic monitoring simulation, four-dimensional quantitative transmission of obstructed risks, and pre-feasibility study constraint sensitivity analysis based on the three types of basic models. The multi-dimensional linkage calculation module is used to execute the implementation based on the output results of the four-dimensional quantitative transmission of obstructed risks and the constraint sensitivity analysis of the pre-feasibility study - two-way linkage simulation of the pre-feasibility study, two-level penetration analysis of the province and the region, and accurate calculation of the regional characteristic impact, to obtain the two-way linkage simulation results, the two-level penetration analysis results of the province and the region, and the quantitative results of the regional characteristic impact. The standardized decision output module is used to construct a multi-objective optimization model, solve and verify the optimal solution based on the two-way linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and output standardized decision results.

[0012] In another aspect, the present invention also provides an electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any embodiment of the present invention.

[0013] In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any embodiment of the present invention.

[0014] Compared with the prior art, the present invention has the following technical effects: The method described in this invention directly achieves accurate fusion, regional quantification, and structured correlation of all-dimensional data for 220kV and above power grid planning projects through standardized multi-source data acquisition, quantitative modeling of four types of regional characteristics, and construction of five-dimensional correlation mapping. Relying on a four-layer collaborative monitoring model, a four-dimensional obstruction risk quantification and transmission model, and a two-way linkage mechanism between implementation and pre-feasibility study, it automatically completes dynamic monitoring simulation of projects, obstruction risk classification quantification, and full-process linkage analysis. Through a regionalized multi-objective optimization algorithm, it calculates the optimal solution and completes compliance verification, directly outputting standardized project implementation optimization plans and pre-feasibility study auxiliary decision-making results.

[0015] Compared with existing power grid planning monitoring, risk assessment, and pre-feasibility study analysis technologies, this invention does not rely on human experience and general models. It can transform complex regional constraints into calculable parameters, achieve real-time data collaboration between provincial and local levels, and complete simulation of the cascading transmission of obstructed risks. It fundamentally solves the problems of poor regional adaptability, inaccurate risk quantification, and lack of linkage between planning implementation and pre-feasibility studies in traditional technologies. It significantly improves the real-time monitoring of power grid planning projects, the accuracy of risk assessment, and the scientific nature of decision output. At the same time, it simplifies the collaborative management process between provincial and local levels, reduces the probability of project implementation obstruction, and comprehensively improves the implementation efficiency and management quality of 220kV and above power grid planning projects. Attached Figure Description

[0016] Figure 1 This is an overall flowchart of the collaborative monitoring and obstacle risk quantification decision-making method for power grid planning projects described in this invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present application and with reference to the accompanying drawings.

[0018] Example 1 This embodiment provides a collaborative monitoring and obstacle risk quantification decision-making method for power grid planning projects. It solves data adaptation issues through comprehensive quantitative modeling of regional characteristics, achieves dynamic control based on a four-layer collaborative monitoring architecture, and completes accurate risk calculation using four-dimensional obstacle quantification and system dynamics. It establishes a two-way linkage closed loop to break down data barriers between implementation and pre-feasibility studies, and outputs compliant decisions through multi-objective optimization, forming a complete technical closed loop to solve core problems. (See reference...) Figure 1 As shown, it includes the following steps: S1: Collect and standardize five types of data, including full life cycle data of power grid planning projects, provincial and local collaborative business data, regional constraint data, project obstruction risk data, and pre-feasibility study spatial constraint data. Perform regional feature parameter quantification and multi-dimensional correlation mapping on the five types of data, and output a standardized fusion dataset and a multi-dimensional correlation mapping matrix.

[0019] S2: Based on standardized fusion datasets, construct three basic models: provincial and local collaborative monitoring, project implementation evaluation, and obstacle-pre-feasibility study linkage, and complete parameter fitting and verification of the three basic models.

[0020] S3: Based on the three types of basic models, conduct provincial and local collaborative dynamic monitoring simulation, four-dimensional quantitative transmission of obstructed risks, and pre-feasibility study constraint sensitivity analysis.

[0021] S4: Based on the output results of the four-dimensional quantitative transmission of obstructed risks and the sensitivity analysis of pre-feasibility study constraints, the pre-feasibility study bidirectional linkage simulation, provincial and municipal two-level penetration analysis, and precise calculation of regional characteristic impacts are executed to obtain the bidirectional linkage simulation results, provincial and municipal two-level penetration analysis results, and regional characteristic impact quantitative results.

[0022] S5: Construct a multi-objective optimization model, solve and verify the optimal solution based on the two-way linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and output standardized decision results.

[0023] As a preferred implementation of this embodiment, the data sources for the five types of data are the power grid planning and management system, the provincial power grid coordination platform, the municipal project execution terminal, the meteorological government data interface, the natural resources government data interface, the ecological environment government data interface, the power grid historical obstruction case database, and the 220kV and above power grid project pre-feasibility study database; the key parameters include the time stamps of the entire process nodes of the project planning stage, the preliminary stage, the construction stage, and the acceptance stage, the provincial and local data synchronization delay index, the extreme meteorological intensity parameter, the mountain geological disaster hazard level parameter, the ecological constraint threshold parameter, the project obstruction type coding parameter, the substation site selection coordinate range parameter, and the transmission line corridor path curvature parameter.

[0024] Data collection is performed on five types of data according to the power planning dedicated interface specification. The following numerical parameters are the preferred empirical values ​​for project implementation and can be flexibly modified according to different provincial regions, power grid scales, and operation and maintenance management requirements: External business data sources include the power grid planning and management system, the provincial power grid coordination platform, and municipal project execution terminals. Real-time operational data is accessed and retrieved from long-term, fixed, and rarely changed archived ledger data using a dedicated power grid TCP / IP communication protocol. The output includes five major data categories: real-time progress data for the entire lifecycle of projects under construction, real-time interactive data for provincial and municipal collaborative business, and real-time data on project site disruptions and anomalies. The data collection frequency is fixed at 10 minutes per collection, data encoding uses UA Binary encryption format, the interface timeout threshold is set to 60 seconds, and automatic retry is performed at 3-second intervals after a single failed collection, with a maximum of 3 retries. Specifically, the real-time operational data is responsible for collecting dynamically updated sub-fields from the five major categories: Project Lifecycle: real-time construction nodes and progress sequence data for projects under construction; Provincial and Municipal Collaborative Business: real-time reporting and instruction interaction dynamic data from provincial and municipal platforms; Project Disruption Risk: real-time dynamic information on sudden project disruptions at the project site.

[0025] External business data sources include the power grid planning and management system, the provincial power grid coordination platform, and municipal project execution terminals. Static ledger data is retrieved via HTTPS 2.0 through GET requests, using data that is long-term, fixed, and rarely changes. The output includes five categories: project initiation basic ledger data, provincial and municipal unit permission ledger data, project obstruction type coding ledger data, and pre-feasibility study spatial constraint substation and line fixed coordinate ledger data. The request header carries a unique project code token, and the response data format is enforced to UTF-8 encoded JSON, with missing fields automatically filled with null markers. Specifically, the static ledger data is responsible for collecting long-term, fixed, and rarely changes basic attribute data from the following five categories: Project lifecycle: basic information on project initiation and filing, project basic attribute ledger; Provincial and municipal collaborative business: fixed ledger of provincial and municipal unit permissions and approval structure; Project obstruction risk: static ledger of obstruction type coding and classification rules; Pre-feasibility study spatial constraints: fixed ledger of substation and line corridor basic coordinates.

[0026] External business data sources include a historical power grid obstruction case database and a pre-feasibility study database for 220kV and above power grid projects. Historical case and pre-feasibility study data connect to a MySQL 8.0 database via a JDBC 4.0 interface to read archived historical data within the system. The output consists of five categories: archived data for the entire lifecycle of completed projects, historical case data on project obstruction over the years, and archived space constraint data for previous pre-feasibility studies. The maximum number of data records exported per batch is 1000, sorted by project commissioning time in descending order. Specifically, the historical case and pre-feasibility study data collection includes the following five categories of completed and archived historical data: Project Lifecycle: Archived historical data for the entire process of completed projects; Project Obstruction Risk: Full historical case database of project obstruction over the years; Pre-feasibility Study Space Constraints: Archived pre-feasibility studies and historical space constraint data for previous projects.

[0027] External government data sources include meteorological, natural resources, and ecological environment government data interfaces. Regional constraint data is synchronized daily with data distributed through the government data sharing platform according to a preset cycle, outputting complete raw regional constraint data, including meteorological, geological, and ecological constraint parameters. The synchronized data includes a timestamp verification field, and duplicate data is automatically skipped.

[0028] All collected data is indexed using a four-dimensional combination index of "project unique code - province / region level code - regional type code - timestamp" and stored in a distributed time-series database, with each data record not exceeding 2KB in size.

[0029] Multi-dimensional data standardization preprocessing was performed on five types of data (the following numerical parameters are preferred empirical values ​​for project implementation and can be flexibly modified according to different provinces, power grid scales, and operation and maintenance management requirements), including: Using the project's unique code as the unique primary key, a hash deduplication algorithm is used to remove duplicate data. The rule for determining duplicates is that the primary keys must be completely identical and the time difference must be ≤1 second.

[0030] For numerical data, outliers are identified using the 3σ principle. The calculation formula is |x-μ|>3σ (where x is the original value, μ is the mean, and σ is the standard deviation). Outliers are identified by using this formula, and the identified outliers are marked separately and stored in isolation, and are not used in subsequent model calculations.

[0031] Missing data are filled using the K-nearest neighbor interpolation algorithm (e.g., K=5), with the distance calculated using the Euclidean distance formula. The final filled value is the arithmetic mean of the five nearest valid data points.

[0032] For unstructured text data such as project reports, approval documents, and government documents, the BERT fine-tuning model in the power industry is used for parsing. The optimal model parameters are set as follows: learning rate = 5e-5, number of iterations = 30 rounds, batch size = 16, which automatically converts unstructured text into standardized structured vector data.

[0033] Data labeling is completed using a four-dimensional tag system: “Project Type Code - Provincial / Regional Level Code - Regional Feature Code - Business Dimension Code”. The tag code length is fixed at 8 digits.

[0034] Numerical data are standardized and normalized using Z-Score to unify data dimensions.

[0035] Standardize the format of all data, enforce time data formatting as YYYY-MM-DD HH:MM:SS, convert text data to UTF-8 encoding, and use the WGS-84 coordinate system for spatial data.

[0036] Perform triple quality checks: completeness check (core parameter missing rate ≤5%), accuracy check (data matching degree ≥98%), and timeliness check (synchronization delay ≤15 minutes). Data that does not meet the standards is returned to S11 for re-collection.

[0037] In a preferred embodiment of this invention, step S1, which involves quantizing regional feature parameters and performing multidimensional association mapping on the five types of data to output a standardized fused dataset and a multidimensional association mapping matrix, specifically involves: The regional constraint data among the five types of data are quantified, including extreme weather characteristics, geological disaster characteristics, social coordination characteristics, and ecological constraint characteristics. Extreme weather characteristic quantification: The fluctuation coefficient is calculated using the sliding window method (window size = 1 hour), with the formula: CV = σ / μ, where CV is the fluctuation coefficient, μ is the overall mean within the window, and σ is the overall standard deviation within the window. CV is mapped to the [0,1] interval to obtain the extreme weather impact coefficient. Geological disaster characteristic quantification: A two-layer LSTM prediction model is constructed with parameters such as hidden_dim = 256, dropout = 0.3, and activation function = ReLU. The inputs are the collected topographic data and historical disaster data associated with the geological disaster hazard level in mountainous areas, as well as the detailed precipitation data corresponding to the extreme weather intensity parameters. The output is a geological disaster risk level of 1-5. Social coordination characteristic quantification: The probability distribution is fitted using the kernel density estimation (KDE) algorithm with a fixed algorithm bandwidth coefficient of 0.1. Based on the social coordination data attached to the project obstruction type coding parameters, the probability distribution of coordination difficulty is fitted to calculate the social coordination quantification coefficient in the 0-1 interval. Ecological constraint characteristic quantification: Based on the ecological control rules and the ecological constraint threshold parameters collected by S1, the parameters are mapped to ecological constraint levels of 1-5 according to the constraint intensity to generate a set of regional characteristic quantification parameters.

[0038] Based on five categories of structured data—project implementation, provincial and municipal level collaboration, regional characteristics, obstruction risks, and pre-feasibility study constraints—that have undergone standardized processing and quantification of regional characteristics, a multi-dimensional correlation mapping matrix of "Project Implementation - Provincial and Municipal Level Collaboration - Regional Characteristics - Obstruction Risks - Pre-feasibility Study Constraints" is established using weighting rules determined by expert scoring. The five mapping data sources include: Project Implementation Dimension Data: Standardized data from the entire lifecycle of 220kV and above power grid planning projects (time and progress-related parameters for each construction stage); Provincial and Municipal Level Collaboration Dimension Data: Business data from provincial and municipal level collaboration (collection parameters such as provincial and municipal data synchronization latency); Regional Characteristics Dimension Data: Quantified coefficients for four categories—extreme weather, geological disasters, social coordination, and ecological constraints—(converted from extreme weather intensity, mountainous geological hazards, ecological thresholds, and obstruction coding parameters); Obstruction Risk Dimension Data: Project obstruction risk data (original obstruction data corresponding to various obstruction types corresponding to project obstruction type coding parameters); Pre-feasibility Study Constraint Dimension Data: Spatial constraint data from pre-feasibility study (spatial parameters such as substation site selection and line corridor curvature).

[0039] Furthermore, the Fabric consortium blockchain is selected to store core sensitive data such as key project node data, regional constraint index data, and project obstruction record data from the aforementioned key parameters on the blockchain. Block generation intervals are set, and each piece of data uploaded to the blockchain is automatically generated with a unique SHA-256 hash identifier to ensure that the data is tamper-proof and traceable.

[0040] The final output is a standardized fusion dataset and a multidimensional association mapping matrix. The standardized fusion dataset includes on-chain evidence storage data and other non-on-chain business parameter data. The on-chain data is only a local subset of this dataset. The matrix dimension is N×5 (N is the data sample size). This N×5 matrix serves as the basic input data for the calculation of all steps S2, S3, S4 and S5.

[0041] In a preferred embodiment of this invention, step S2 specifically comprises: A provincial-level collaborative monitoring foundation model is constructed by integrating regional characteristic quantification parameters obtained from extreme meteorological intensity parameters, mountainous geological disaster hazard level parameters, ecological constraint threshold parameters, and project obstruction type coding parameters, as well as project data corresponding to the full-process node timestamps of the project planning stage, preliminary stage, construction stage, and acceptance stage, provincial and local data synchronization delay indicators, substation site selection coordinate range parameters, and transmission line corridor path curvature parameters. This provincial-level collaborative monitoring foundation model consists of a perception layer, a data layer, a support layer, and an application layer. The perception layer is used to input standardized fusion datasets. The data layer, based on a multi-dimensional correlation mapping matrix, performs model operation format standardization on the standardized fusion datasets input from the perception layer. The support layer has built-in parameter resource libraries, algorithm resource libraries, and mapping resource libraries. By calling these three types of resource libraries, it provides parameter, algorithm, and correlation support for application layer simulation and indicator calculation. The application layer calls the data standardized by the data layer and the configuration resources of the support layer to initialize parameters and verify validity, outputting the simulation input dataset.

[0042] Specifically, the perception layer is divided into three types of input units and functions, including: regional feature input unit: receiving regional feature quantitative parameters generated by extreme meteorological intensity parameters, mountain geological disaster hazard level parameters, ecological constraint threshold parameters, and project obstruction type coding parameters, responsible for collecting four types of regional constraint quantitative indicators; project time sequence input unit: receiving timestamps of the entire process nodes of the project planning stage, preliminary stage, construction stage, and acceptance stage, as well as provincial and local data synchronization delay indicators, collecting project construction progress and provincial and local collaborative timeliness data; and pre-feasibility study spatial input unit: receiving substation site selection coordinate range parameters and transmission line corridor path curvature parameters, collecting project site selection and line path constraint data.

[0043] The data layer receives data from the three types of units in the perception layer. Based on the association logic of the N×5 multidimensional association mapping matrix, it completes field classification indexing, dimension splitting, format adaptation and pruning within the model architecture, eliminates invalid and redundant fields in a single scenario, and uniformly adapts to the internal operation format of this four-layer model.

[0044] The support layer incorporates three types of resource libraries, configured and implemented based on prior research. These include: a parameter resource library (containing weights for four regional features: extreme weather, geological disasters, social coordination, and ecological constraints; the fitted model's linkage threshold and transmission coefficients; and initial values ​​and correction periods for the linkage coefficients); an algorithm resource library (configuring gradient descent iterative algorithms, feature weighting calculation formulas, and obstructed-pre-feasibility study linkage transmission rules); and a mapping resource library (importing an N×5 multidimensional association mapping matrix as a benchmark for cross-dimensional data association queries). By calling these three resource libraries, the support layer provides parameters, algorithms, and association relationships to support application-layer simulations and indicator calculations.

[0045] The application layer calls the support layer's configured resources and the data layer's structured data, which is then output in three categories: Provincial-regional collaborative dynamic monitoring results: Based on feature weights, the regional comprehensive constraint coefficient is calculated, and real-time project progress monitoring data is output; Project obstruction risk assessment results: Based on the obstruction type code and corresponding data, combined with the linkage coefficient, the probability of project obstruction under different regional conditions is quantified; Pre-feasibility study optimization constraint suggestions: Based on the risk assessment results, the substation site selection coordinate range parameters and transmission line corridor path curvature parameters are adjusted in reverse, and the pre-feasibility study optimization revision scheme is output.

[0046] Fixed parameters were initialized for the provincial and municipal collaborative monitoring basic model: the simulation calculation step size was set, and the calculation was carried out according to the power grid power balance equation ΣP_in=ΣP_out+P_loss, where P_in is the total power input of the power grid; P_out is the total power output of the power grid; P_loss is the total power loss of the power grid, and the spatial geographic resolution was set.

[0047] By substituting standardized datasets for trial calculations, benchmarking against historical measured data, and verifying layer by layer according to the error ≤2%, if the model fails to meet the requirements, the model parameters are backtracked and optimized. After passing the verification, the basic model for provincial and municipal collaborative monitoring is output in the form of an architecture configuration file. The simulation input dataset is output in the form of an N×5 five-dimensional structured matrix. The model has a four-layer architecture and fixed preset operation parameters. The simulation input dataset includes all processed key parameters in all dimensions. Finally, after completing the model validity verification, the trained basic model for provincial and municipal collaborative monitoring and the simulation input dataset required for model operation are output.

[0048] Furthermore, the provincial and municipal collaborative monitoring basic model is packaged and output in the form of an architecture configuration document and a standardized parameter configuration file. The content includes the solidified structure of four types of sub-modules and configuration parameters (such as simulation step size, geographic resolution, and power grid power balance calculation formula), and binding regional four types of regional feature weight configuration rules. The simulation input dataset is based on the generated N×5-dimensional correlation matrix and stored in the time series database in the form of a structured table. The content includes the project's full life cycle node data, provincial and municipal synchronization delay data, quantified regional feature coefficients, obstruction classification coding data, substation site selection and line corridor and other pre-feasibility study spatial parameters, which can be directly called and read in the subsequent S3 simulation steps.

[0049] The project implementation evaluation model consists of a weighted sum of indicators from four dimensions: progress, quality, benefits, and collaboration. The progress indicators include the completion rate of preliminary approvals, the construction progress deviation rate, and the on-time commissioning and acceptance rate. The quality indicators include the project acceptance pass rate, the equipment delivery pass rate, and the safety hazard rectification rate. The benefits indicators include clean energy consumption, power supply reliability improvement, and return on investment. The collaboration indicators include the timeliness of provincial and local information reporting, the cross-departmental approval collaboration rate, and the instruction response completion rate. The model calculates the scores for each dimension indicator by calling a standardized fusion dataset, outputting the basic results of the project implementation evaluation.

[0050] The obstacle-pre-feasibility study linkage basic model adopts a dual-link closed-loop architecture of forward transmission link and backward correction link. The forward transmission link follows the flow of monitoring simulation → implementation evaluation → obstacle quantification → pre-feasibility study constraint correction. The backward correction link follows the flow of pre-feasibility study constraint assessment → implementation risk calculation → monitoring parameter optimization → evaluation score correction.

[0051] The transmission rules are set up according to the two-level progressive logic of obstruction factors → degree of impact → pre-feasibility study constraints: the obstruction type is classified based on the coding parameters of the project obstruction type, the degree of impact is calculated in combination with extreme weather intensity parameters, mountain geological disaster hazard level parameters, and ecological constraint threshold parameters, and then the pre-feasibility study constraints corresponding to the substation site selection coordinate range parameters and transmission line corridor path curvature parameters are adjusted in stages according to the degree of impact. The project obstruction risk and pre-feasibility study constraint data are integrated to establish the transmission rules of obstruction factors → degree of impact → pre-feasibility study constraints. The rules for the obstruction factors → degree of impact are as follows: based on the project obstruction type coding parameters, various obstruction causes are distinguished, and combined with the quantitative results of extreme weather intensity parameters, mountain geological disaster hazard level parameters, and ecological constraint threshold parameters, the degree of impact corresponding to the obstruction is calculated; the rules for the degree of impact → pre-feasibility study constraints are as follows: using the critical threshold of the change of the preset linkage coefficient (e.g., 10%) as the judgment benchmark, the pre-feasibility study spatial constraints corresponding to the substation site selection coordinate range parameters and the transmission line corridor path curvature parameters are adjusted in a targeted manner according to the degree of impact; the higher the obstruction impact, the more stringent the constraints on the pre-feasibility study site selection and the line corridor.

[0052] Historical archived data from the standardized fusion dataset is substituted into the basic model for the linkage between obstruction and pre-feasibility study, and iterative calculations are performed. The model parameters are fitted by comparing the actual implementation effects with those of benchmark projects, and a table of correlation parameters is output. The transmission coefficient is used to characterize the strength of the impact of a single obstruction factor on the pre-feasibility study constraints, with a value range of 0-1. The linkage coefficient is defined as representing the overall correlation strength between changes in project obstruction fluctuations and adjustments to pre-feasibility study constraint parameters, with a value range of 0-1 and an initial configuration value of 0.5. The closer the coefficient is to 1, the stronger the adjustment effect of obstruction changes on pre-feasibility study constraints such as substation site selection coordinate range parameters and transmission line corridor path curvature parameters. The linkage coefficient is a weighted synthesis of the transmission coefficients of each sub-item. A critical threshold for the change in the linkage coefficient, i.e., the model linkage threshold, is set at 10%. If the linkage coefficient fluctuation exceeds this threshold, model parameter correction is triggered. Based on historical project sample data of the regional power grid, the model linkage threshold and transmission coefficient are fitted and determined.

[0053] In a preferred embodiment of this invention, S3 specifically comprises: S31: Substitute the simulation input dataset into the provincial and municipal collaborative monitoring basic model, dynamically allocate regional feature weights for the weighted calculation of the regional constraint comprehensive value of the collaborative monitoring simulation, and reuse them in the quantitative calculation of progress, obstacles, and feasibility indicators in subsequent steps. Iterative simulation is conducted using the gradient descent algorithm, with the convergence threshold being the critical value of the difference between the comprehensive constraint coefficients of two adjacent iterations (e.g., a limit of 0.01). Three scenarios are simulated: normal, constraint-enhanced, and extreme weather. A simulation error judgment standard is established: |simulated value - actual value| / actual value ≤ a set percentage. If the error exceeds the range, the simulation returns to the parameter calibration stage for iterative optimization. Finally, the simulation results of the provincial and municipal collaborative dynamic monitoring are output, including project progress, construction quality, and the collaborative status of the provincial and municipal levels.

[0054] S32: Construct a four-dimensional obstruction model with a three-level architecture: a four-dimensional parallel input layer, a sub-item feature conversion layer, and a comprehensive obstruction result output layer. The four-dimensional obstruction model decomposes and refines the standardized fusion dataset and multi-dimensional correlation mapping matrix into three categories of obstruction factors: natural environment, policy constraints and external coordination, and internal management. The three categories of obstruction factors are combined with predetermined regional feature weights to perform individual risk scoring. The scores of the three categories of obstruction factors are aggregated and weighted to output the overall project obstruction quantitative score and the probability of obstruction.

[0055] Specifically, the three-tiered architecture comprises a four-dimensional parallel input layer: divided into four independent input channels—policy constraints, external coordination, natural environment, and internal management. All four types of input data are derived from processed key parameter data. The natural environment channel receives environmental characteristic data quantified by extreme weather intensity parameters, mountainous geological disaster hazard level parameters, and ecological constraint threshold parameters. The policy constraints and external coordination channels receive detailed coordination data broken down by project obstruction type coding parameters. The internal management channel receives project control data corresponding to the time stamps of all nodes in the project planning, preliminary, construction, and acceptance phases, as well as provincial and regional data synchronization latency indicators. The sub-feature conversion layer, based on the input data and the predetermined weights of the four regional features, performs individual risk scoring on 18 sub-factors of obstruction, converting the individual obstruction coefficient. The comprehensive obstruction result output layer aggregates the weighted scores of the 18 sub-factors, outputting the overall project obstruction quantification score and the probability of obstruction occurrence.

[0056] The 18 subdivided obstruction factors were all derived from a standardized fusion dataset and an N×5 multidimensional correlation mapping matrix: Natural environment subdivisions: meteorological disasters, geological hazards, and ecological control restrictions were extracted from the original monitoring data corresponding to extreme meteorological intensity parameters, mountain geological disaster hazard level parameters, and ecological constraint threshold parameters; Policy constraints and external coordination subdivisions: land acquisition coordination, approval policies, and coordination with surrounding communities were extracted based on the historical obstruction ledgers corresponding to each category of the project obstruction type coding parameters; Internal management subdivisions: internal control-related subdivisions were extracted based on the project planning stage, preliminary stage, construction stage, and acceptance stage, as well as anomalies such as construction delays and data connection timeouts between provinces, cities, and counties. The above three categories of subdivisions were summarized to form a total of 18 subdivided obstruction factors.

[0057] A risk transmission flow graph based on a four-dimensional hindered model is constructed using system dynamics, defining three types of model variables: level variables, rate variables, and auxiliary variables. The risk transmission flow graph employs a three-layer hierarchical structure: an auxiliary variable input layer, a rate-level variable transmission layer, and a comprehensive risk output layer, outputting dynamic cumulative risk results. All auxiliary variables consist of 18 subdivided hindered factors obtained from the four-dimensional hindered model. The four-dimensional hindered model provides all input data sources for this risk transmission flow graph, used to complete the static quantitative analysis of single-type hindered factors. The risk transmission flow graph constructed in this step is used to realize the dynamic transmission and cumulative extrapolation of risks from multiple types of hindered factors.

[0058] Based on the aforementioned four-dimensional obstruction model, a hierarchical architecture for the analytic hierarchy process (AHP) criteria layer and solution layer is constructed. The input data for the AHP consists of scores from three categories of subdivided obstruction factors. The input data for the fuzzy comprehensive evaluation includes the indicator weights calculated by the AHP and the dynamic cumulative risk results. A combined AHP-fuzzy comprehensive evaluation algorithm is used to calculate the four-dimensional quantitative results of obstruction risk. Specifically, the four primary dimensions of the four-dimensional obstruction model—policy constraints, external coordination, natural environment, and internal management—serve as the AHP criteria layer, and the 18 subdivided obstruction factors serve as the AHP solution layer, directly constructing the hierarchical architecture of the AHP. The static individual risk scores of each subdivided factor serve as the basis for constructing the AHP judgment matrix, relying on AHP to solve for the indicator weights of each dimension and each subdivided obstruction factor. The system dynamics risk transmission flow diagram is used for dynamic extrapolation of the flow diagram through rate and level variables, outputting the dynamic cumulative risk values ​​after the superposition and transmission of each obstruction factor. This set of dynamic results serves as the original sample for the fuzzy comprehensive evaluation, used to determine the membership degree corresponding to each evaluation indicator and characterize the fluctuation characteristics after the risk linkage and superposition.

[0059] Specifically, the input data for the AHP (Analytic Hierarchy Process) includes hierarchical structure data: 4 major dimensions and 18 subdivided obstruction factors; the original basis for constructing the judgment matrix includes: extreme weather intensity parameters, mountain geological disaster hazard level parameters, ecological constraint threshold parameters, project obstruction type coding parameters, timestamps of the entire process nodes of the project planning stage, preliminary stage, construction stage, and acceptance stage, and single-factor historical risk statistics obtained by converting provincial and local data synchronization delay indicators, supplemented by predetermined weights of four types of regional characteristics to assist in verifying the rationality of the matrix.

[0060] The input data for fuzzy comprehensive evaluation includes: the weights of each indicator obtained from AHP solution; the dynamic cumulative risk values ​​of each sub-factor output by the risk transmission flow diagram; and auxiliary constraint references: the pre-feasibility study constraints corresponding to the substation site selection coordinate range parameters and the transmission line corridor path curvature parameters, which are used to correct the membership boundary under extreme constraint scenarios.

[0061] The AHP algorithm is used to construct pairwise comparison judgment matrices with a consistency ratio CR ≤ 0.1 to determine the hierarchical weights of each obstructing factor. The fuzzy comprehensive algorithm is used to construct a membership matrix with a comment set of {extremely low, low, medium, high, extremely high} to calculate the comprehensive risk value. Finally, the multidimensional quantitative results of obstruction risk in the range of 0 to 100 are output and divided into four levels: low risk, medium risk, high risk, and extremely high risk according to the score range.

[0062] S33: A parameterized configuration framework is built by selecting four core elements—site selection, corridor, ecological avoidance, and meteorological protection—from the standardized fusion dataset. The data parameters corresponding to these four core elements are adjusted bidirectionally by tightening and relaxing constraints. Differentiated simulation scenario parameter combinations are generated through cross-combinations of multiple elements and control directions. Specifically, the four core elements—site selection, corridor, ecological avoidance, and meteorological protection—are extracted from key parameters collected in S1 by classification and clustering within the standardized fusion dataset. The site selection element corresponds to the power station site coordinate range parameter; the corridor element corresponds to the transmission line corridor path curvature parameter; the ecological avoidance element corresponds to the ecological constraint threshold parameter; and the meteorological protection element corresponds to the extreme weather intensity parameter. By adjusting the step size of the preset fixed parameter gradient, the parameters corresponding to the substation site coordinate range, transmission line corridor path curvature, ecological constraint threshold, and extreme weather intensity for the four core elements of site selection, corridor, ecological avoidance, and meteorological protection are adjusted in two directions: constraint tightening and constraint relaxation. Through the differentiated cross-combination of the four core elements and the two-way control mode, the repetitive equivalent working conditions are eliminated, and finally, multiple combinations of differentiated simulation scenario schemes covering different constraint intensities and different risk conditions are generated iteratively.

[0063] The differentiated simulation scenario parameter combinations are substituted into the provincial and municipal collaborative monitoring basic model. Using the control variable method to keep other baseline parameters constant, the parameter impact magnitude of project feasibility under varying core element parameters is quantitatively calculated: Impact magnitude = (changed value - baseline value) / baseline value × 100%. Based on the impact magnitudes of four core element parameters (site selection, corridor, ecological avoidance, and meteorological protection), the average impact magnitude of each core element under multiple conditions is calculated, and the core elements are sorted in descending order of sensitivity according to the magnitude of the average impact magnitude under multiple conditions. By fitting the correlation between the parameter variation magnitude of each core element and project feasibility, the parameter constraint critical values ​​for each core element to meet the pre-feasibility study feasibility requirements are obtained through reverse iterative solution. A pre-feasibility study constraint sensitivity analysis report containing the ranking results and parameter constraint critical values ​​is output.

[0064] Furthermore, the specific principle of substituting the parameter combinations of differentiated simulation scenarios into the provincial and municipal collaborative monitoring basic model is as follows: the monitoring simulation module is the upstream data input and simulation start unit of the obstructed-pre-feasibility study two-way linkage basic model, and is the front-end basic port for full-link operation, supporting independent parameter assignment and single-point simulation operation. This step specifically conducts differentiated scenario simulations, and the specific substitution method is as follows: First, keep all the original baseline parameters unchanged, and then use the generated differentiated simulation scenario scheme combination (tightening / relaxing adjustment parameters of four elements: site selection, corridor, ecological avoidance, and meteorological protection) to update the monitoring and simulation module at the beginning of the linkage model in a targeted manner, so as to complete the replacement and assignment of simulation parameters. Second, by using the controlled variable method, only the element parameters corresponding to the current scenario are changed, while the remaining model parameters, association rules, and threshold coefficients remain in the baseline state, without the need to initiate the full-link closed-loop iteration of subsequent implementation evaluation, obstacle quantification, and pre-feasibility study constraints. Third, relying on the monitoring and simulation module to independently calculate and output the simulation results after parameter changes, and combined with the original baseline simulation results, the impact of changes in the parameters of each element on the feasibility of project implementation is quantified through preset formulas, so as to achieve accurate measurement of sensitivity of single element and single scenario.

[0065] The ranking calculation of multiple sets of impact magnitude data corresponding to the four core elements (site selection, corridor, ecological avoidance, and meteorological protection) obtained from multi-scenario simulation is as follows: First, for each core element, the average impact magnitude under multiple working conditions in multiple differentiated simulation scenarios is calculated to eliminate the random error of a single simulation scenario and objectively characterize the overall impact of the element on the feasibility of the project's pre-feasibility study. Then, the average impact magnitudes of the four elements are sorted in descending order according to their numerical values ​​to obtain the final priority ranking of the elements' impact, thus identifying the core elements that have the strongest constraints and the highest sensitivity to the power grid planning project.

[0066] The critical value is the maximum allowable threshold for changes in the parameters of the elements that ensure the feasibility and risk control of the project's pre-feasibility study plan. The specific calculation logic is as follows: based on the fitting correlation between the variation range of each element parameter and the feasibility of project implementation, a minimum qualified threshold for the project's pre-feasibility study is preset; the control variable method is used to perform marginal iterative simulation on each of the four types of core element parameters, continuously fine-tuning the variation range of the element parameters, and inversely solving for the parameter values ​​corresponding to when the project's feasibility just drops to the preset qualified threshold; this critical value is the risk constraint critical value of each element. Once the variation of the element parameter exceeds this critical value, it will lead to the project's pre-feasibility study failing to meet the standards and the risk of being hindered exceeding the limit.

[0067] In a preferred embodiment of this invention, S4 specifically comprises: S41: Import the four-dimensional quantification results of obstructed risk obtained from the four-dimensional quantification of obstructed risk into the obstructed-pre-feasibility study linkage basic model. Based on the forward transmission link and backward correction link of the obstructed-pre-feasibility study linkage basic model, complete the bidirectional iterative substitution of the four-dimensional quantification results of obstructed risk. Execute the bidirectional linkage simulation of the obstructed-pre-feasibility study linkage basic model, call the GraphSAG graph neural network to mine and complete the potential correlations between elements, correct the linkage coefficients, and output the bidirectional linkage simulation results including linkage coefficients, risk transmission paths, and rule correction records. Forward simulation: monitoring data → implementation evaluation → obstruction quantification → pre-feasibility study constraint correction; Backward simulation: pre-feasibility study constraints → implementation risk calculation → monitoring parameter optimization → evaluation score correction; A sliding window method (window size = 1 month) can be used to monitor changes in the correlation coefficients. When the change is greater than or equal to a set threshold, the linkage rule correction mechanism is automatically triggered.

[0068] Specifically, the method for substituting the multidimensional quantification results of the obstruction risk is as follows: the multidimensional obstruction risk quantification results calculated by the AHP-fuzzy comprehensive evaluation combination algorithm are used as the core input variables of the model and imported into the obstruction-pre-feasibility study bidirectional linkage basic model. Combined with the model's predetermined fixed parameters (0-1 interval transmission coefficient, critical threshold for correlation coefficient change, linkage coefficient, and automatic correction cycle of linkage rules), the precise substitution calculation of the two links is realized: First, the forward link substitution: using the multidimensional obstruction risk quantification value as the benchmark weight, the adjustment range of the pre-feasibility study constraint parameters corresponding to the four core elements of site selection, corridor, ecological avoidance, and meteorological protection is mapped and calculated to realize the positive constraint transmission of obstruction risk to the pre-feasibility study scheme; Second, the backward link substitution: the adaptation and adjustment results of the pre-feasibility study parameters are fed back into the model to iteratively correct the project obstruction risk quantification value, complete the reverse calibration of the risk assessment results, and finally realize the bidirectional closed-loop simulation calculation of the model.

[0069] Furthermore, the GraphSAGE graph neural network is invoked to construct a graph topology based on a standardized fusion dataset, a multi-dimensional association mapping matrix, and various business elements. Model training is conducted through node feature extraction and neighborhood feature aggregation. The number of model iterations is set, and through multiple iterations, new associations between elements are mined and supplemented. These supplemented associations are then mapped to the obstructed-pre-feasibility study bidirectional linkage basic model for computation, outputting the corrected linkage coefficients. The specific implementation process is as follows: 1. Constructing the Graph Topology and Samples to be Completed: Based on the standardized fusion dataset, an N×5 multidimensional association mapping matrix, four selected core elements, 18 subdivided obstruction factors, and various quantitative risk results, a GraphSAGE graph network topology is constructed. Four control elements—substation site selection, transmission line corridor, ecological avoidance, and meteorological protection—along with each subdivided obstruction factor, regional characteristic parameters, and pre-feasibility study constraint parameters, are defined as graph nodes. Existing relationships clearly defined in the bidirectional linkage model are set as known edges, while missing or unidentified potential relationships during model operation are marked as edges to be completed, forming the complete graph structure required for model training.

[0070] 2. Node Feature Extraction and Feature Aggregation: Extract the feature vectors corresponding to each graph node. The feature data are taken from the quantified values, risk scores, and feature weights of original parameters collected by S1, such as extreme meteorological intensity parameters, mountain geological disaster hazard level parameters, ecological constraint threshold parameters, project obstruction type coding parameters, substation site selection coordinate range parameters, and transmission line corridor path curvature parameters. Utilize GraphSAGE's neighborhood sampling and neighborhood feature aggregation mechanism to learn the inherent correlation patterns between nodes and their neighboring nodes.

[0071] 3. Iterative training to complete potential relationships: Set the number of model iteration rounds, continuously optimize node representations and the weights of relationships between nodes during each iteration, and deeply explore the hidden relationship logic between elements, obstacles, and constraint parameters; after completing full iteration training, the model converges and automatically identifies and completes all new relationships that need to be added.

[0072] 4. Mapping and correcting the linkage coefficients: The new linkages after completion are fed back into the basic model of bidirectional linkage between obstruction and pre-feasibility study. Combined with the original transmission coefficients, critical thresholds for changes in linkage coefficients, historical linkage coefficients and other parameters of the model, the linkage coefficients are recalculated and finally the corrected linkage coefficients are calculated and output.

[0073] S42: Perform quantification of the five types of data, including provincial and local level collaborative authority, regional constraint strength, and approval process duration. Combine the three types of influencing factors and perform weighted calculation based on preset influencing factor weights to obtain the three levels of influence values.

[0074] The analysis separately calculates the impact values ​​of various levels corresponding to the provincial-level overall network coordination dimension and the prefecture-level single project implementation dimension. Then, using the provincial-level overall indicators as a benchmark, it compares the single project indicators of each prefecture with the provincial-level overall indicators item by item, calculates the absolute difference and relative deviation of the indicators, and sorts out the deviation of the two levels in terms of collaborative operation efficiency, regional constraint enforcement intensity, and approval process time. It summarizes the difference characteristics, difference range and distribution patterns, and generates the provincial and prefecture-level level difference analysis results. Combined with the preset fluctuation threshold of the impact factors, it conducts business link tracing, and conducts layer-by-layer investigation of abnormal items exceeding the threshold to complete the location of control breakpoints in each link. Finally, it outputs the provincial and prefecture-level penetrating analysis results, including the level difference analysis results and the location of control breakpoints.

[0075] Specifically, at the provincial level, the analysis corresponds to the overall macro-level indicators of the entire network, relying on project data from across the province to conduct overall assessment of planned power grid projects. At the prefecture-level, the analysis corresponds to the micro-level implementation indicators of individual projects, focusing on the implementation constraints and risk indicators of individual planned projects, achieving data penetration and hierarchical differentiated analysis from top to bottom at both the provincial and prefecture levels. The provincial-level overall network indicators belong to the macro-level indicators of the entire network, taking batch projects of 220kV and above power grid planning across the province as the analysis object. Relying on quantitative parameters of regional characteristics across the province, time-series data of the entire network project process, and provincial and prefecture data synchronization delay indicators, the analysis statistically analyzes the overall distribution of project obstruction risks, regional constraint adaptability, and provincial and municipal collaborative dispatch and coordination capabilities, among other macro-level indicators. The prefecture-level individual project implementation indicators belong to the micro-level indicators of project implementation, taking individual power grid planning projects in each prefecture as the analysis object. Relying on parameters such as substation site selection coordinate range, transmission line corridor path curvature, ecological constraint threshold, extreme weather intensity, and quantitative obstruction risk results, the analysis analyzes the implementation feasibility, local regional constraint matching degree, and project implementation deviation risk, among other implementation control indicators.

[0076] Furthermore, among the three influencing factors, the provincial and municipal level collaborative authority is quantified by the provincial and municipal data synchronization latency index and the project obstruction type coding parameter. This index directly reflects the smoothness of the flow of business, data, and authority between the provincial and municipal levels. Combined with the collaborative obstruction classification data in the project obstruction type coding parameter, the implementation effect of the provincial and municipal level collaborative authority is comprehensively quantified. The regional constraint strength is comprehensively calculated by extreme meteorological intensity parameters, mountain geological disaster hazard level parameters, and ecological constraint threshold parameters. These three parameters correspond to the inherent constraints of meteorological, geological, and ecological regions, respectively. After integration and conversion, the overall regional constraint strength is obtained. The approval process duration is obtained based on the statistics of the time stamps of the entire process nodes in the project planning stage, preliminary stage, construction stage, and acceptance stage. By extracting the start and end time nodes of each stage of the project and statistically analyzing the stage interval duration, the overall time consumption of the approval process is quantified.

[0077] Specifically, the method for locating the control breakpoints is as follows: First, reasonable fluctuation thresholds for three types of influencing factors are set in advance, combining historical project samples and model operation rules. If the deviation value of a certain influencing factor exceeds the preset threshold, a full-business-link tracing is initiated, linking and verifying original data such as provincial and regional data synchronization latency indicators, timestamps of all process nodes from project planning to acceptance, extreme weather intensity parameters, mountainous geological disaster hazard level parameters, ecological constraint threshold parameters, and project obstruction type coding parameters. Breakpoints are then located specifically for different anomaly types: For anomalies in collaborative permission indicators, control breakpoints in the permission connection and data interaction links between provinces and regions are located; for anomalies in regional constraint intensity indicators, control breakpoints in the top-down transmission and on-site implementation of regional constraint standards are located; for anomalies in approval process duration indicators, control breakpoints in the process connection and hierarchical approval flow links at each stage of the project are located. Finally, the level, specific business node, and cause of the anomaly are determined.

[0078] S43: Based on the standardized fusion dataset, the four-dimensional quantitative transmission of obstruction risk, and the output results of the pre-feasibility study constraint sensitivity analysis, the schedule delay rate, the probability of project obstruction, and the pre-feasibility study feasibility coefficient are weighted and calculated; the schedule delay rate, the probability of project obstruction, and the pre-feasibility study feasibility coefficient are normalized and combined with the optimized linkage coefficient for weighted fusion, and finally the quantitative results of regional feature influence are output.

[0079] Specifically, the weighted calculation of the schedule delay rate, project obstruction probability, and pre-feasibility study feasibility coefficient based on the output results of the standardized fusion dataset, the four-dimensional quantitative transmission of obstruction risk, and the pre-feasibility study constraint sensitivity analysis is as follows: Based on four types of regional characteristics—extreme weather, geological disasters, social coordination, and ecological constraints—and their corresponding weight parameters, combined with the project time-series data, risk quantification data, pre-feasibility study constraint parameters, and the corrected linkage coefficient generated in the previous steps, the impact values ​​of the four types of regional characteristics—project schedule delay rate, project obstruction probability, pre-feasibility study feasibility coefficient, and the final corrected score of project implementation evaluation—are calculated to complete the full-dimensional quantitative evaluation of the impact of regional multi-dimensional constraints on the implementation of power grid planning projects.

[0080] The acquisition and calculation of the schedule delay rate parameter are as follows: the basic parameters are taken from the collected timestamps of the project planning, preliminary, construction and acceptance process nodes and the synchronization delay index of provincial and local data; based on the industry standard construction period, the deviation value between the actual time consumption of each stage of the project and the standard construction period is calculated, and combined with the weighted correction of the four types of regional characteristics preset in S312, namely extreme weather, geological disasters, social coordination and ecological constraints, the project schedule delay rate under the coupling effect of regional constraints is finally calculated.

[0081] The project obstruction probability parameters are obtained and calculated as follows: the basic parameters are taken from the 18 subdivided obstruction factors of the four-dimensional obstruction model, the dynamic cumulative risk data of the risk transmission flow diagram, and the basic risk quantification value obtained from AHP-fuzzy comprehensive evaluation; the static risk value and the dynamic transmission risk value are weighted and fitted based on the four types of regional feature weights, the superposition effect of multi-regional constraints is corrected, and the overall project obstruction probability is quantified.

[0082] The acquisition and calculation of the pre-feasibility study feasibility coefficient parameters are as follows: the basic parameters are taken from the collected substation site selection coordinate range parameters, transmission line corridor path curvature parameters, ecological constraint threshold parameters, extreme weather intensity parameters, and the transmission coefficient and linkage threshold parameters of the S23 obstruction-pre-feasibility study two-way linkage model; based on the constraint intensity of the four types of regional characteristics, the degree of adaptation of the pre-feasibility study site selection, line, avoidance, and protection core elements is matched, and the feasibility coefficient of the pre-feasibility study scheme is quantitatively solved by combining the linkage transmission rules.

[0083] The parameters for obtaining and calculating the final revised score of the project implementation evaluation are as follows: the basic parameters are taken from the aforementioned calculated progress delay rate, project obstruction probability, pre-feasibility study feasibility coefficient, and linkage coefficient after iterative correction by the GraphSAGE graph neural network; the three types of special quantitative indicators are normalized and combined with the optimized linkage coefficient for weighted fusion, and finally the final revised score of the project implementation evaluation adapted to the regional characteristics is output.

[0084] Specifically, based on the calculated schedule delay rate, project obstruction probability, and pre-feasibility study feasibility coefficient, and combined with pre-set regional characteristic weights for weighted normalization, a single regional comprehensive constraint impact value is obtained. The project implementation evaluation score is then corrected using the formula E_final = E - original × (1 - impact value), where E_original is the original project implementation evaluation score without the overlay of regional constraints and obstruction risk corrections, initially calculated based on project infrastructure conditions and planning adaptability parameters. The impact value uses the calculated schedule delay rate, project obstruction probability, and pre-feasibility study feasibility coefficient as core data sources, coupled with weighted fusion and normalization of four types of regional characteristics: extreme weather, geological disasters, social coordination, and ecological constraints. Its value ranges from 0 to 1, representing the overall negative impact of multi-dimensional regional constraints on project implementation. E_final is the corrected final project implementation evaluation score, i.e., the corrected evaluation score obtained after deducting the impact of regional comprehensive constraints, representing an accurate assessment result adapted to regional obstruction risks.

[0085] As a preferred implementation method in this embodiment, S5 constructs a multi-objective optimization model, solves and verifies the optimal solution based on the bidirectional linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and outputs standardized decision results as follows: The decision objectives and constraints are constructed as follows: the decision objectives are set as dual core objectives of maximizing project implementation efficiency and optimizing pre-feasibility study feasibility. Based on the dual core decision objectives, a single objective constraint function is synthesized by weighting the dual core decision objectives to optimize the feasible region definition and the coarse screening of invalid solutions. By fusing the dual objective dimensions with fixed weights, the pre-constraints and convergence bounds of the optimization solution space are completed. The constraints include the provincial and local level coordination cycle range, the regional cost increase range, the construction period delay ratio range, and the ecological constraint compliance rate. The constraint threshold is a hard limit, and if the threshold is exceeded, the solution is directly determined to be invalid.

[0086] A hybrid algorithm combining scenario-optimized improved hierarchical analysis and fuzzy comprehensive evaluation was adopted to complete the construction of the index system and the calibration of the basic weights for the multi-objective optimization model. The improved hierarchical analysis method mainly completes the optimization from three aspects: data-driven construction of judgment matrix, addition of dynamic weight correction mechanism, and linkage of fuzzy result optimization consistency verification. The project sample dataset was constructed using regional historical power grid planning project data, and the multi-objective model was trained by set distribution according to a fixed ratio.

[0087] Furthermore, the improvements of the analytic hierarchy process (AHP) combined with fuzzy comprehensive evaluation include: integrating the previously output quantified results of obstruction risk, provincial and regional penetration analysis data, and measured and extrapolated data such as the linkage coefficients corrected by the GraphSAGE graph neural network, as the core basis for constructing the judgment matrix, realizing matrix quantification and significantly reducing human bias; combining multiple differentiated simulation scenarios, adding weight iteration update logic, which can dynamically adjust the indicator weights according to changes in regional constraint strength and project obstruction risk, adapting to the differentiated evaluation needs of regional power grid projects; relying on the membership matrix output by fuzzy comprehensive evaluation to assist in consistency verification, and appropriately optimizing the verification threshold in combination with business scenarios, improving the algorithm's adaptability to complex indicator systems. The improved AHP combined with fuzzy comprehensive evaluation is used to complete the construction of regional multi-objective optimization models, indicator system configuration, dynamic weight calibration, and model basic parameter initialization; after the model is fully constructed and the parameters are calibrated, subsequent steps use a dedicated multi-objective optimization algorithm to perform optimization calculations based on the model, and then output the Pareto optimal solution set.

[0088] The system integrates two-way simulation results, provincial and municipal level penetration analysis results, and regional characteristic influence quantification results into a multi-objective optimization model solver for iterative solution, outputting a Pareto optimal solution set. Based on the Pareto optimal solution set, it outputs a combination of differentiated project optimization decision-making schemes. This combination of differentiated project optimization decision-making schemes specifically includes three core deliverables: a project implementation optimization plan, pre-feasibility study auxiliary suggestions, and detailed investment cost estimates. Specifically, the project implementation optimization plan, based on provincial and municipal level differences, control breakpoint location, and schedule delay characteristics, outputs full-process time-series optimization, provincial-municipal coordination mechanism optimization, and risk pre-emptive prevention strategies. The pre-feasibility study auxiliary suggestions, based on four types of regional core constraints and multi-scenario simulation results, output adaptive optimization suggestions for substation site selection, transmission corridor paths, ecological avoidance, and meteorological protection. The detailed investment cost estimates, combined with optimized construction schemes and risk prevention and control requirements, quantitatively output multi-dimensional detailed investment cost calculation results, completing the collaborative monitoring and risk quantification decision optimization output for the entire power grid planning project.

[0089] Furthermore, the project implementation optimization plan specifically relies on the aforementioned Pareto optimal solution set, project schedule delay rate, hierarchical differences and control breakpoint location results from the provincial and municipal level penetration analysis, and combines the corrected linkage coefficients to generate a refined implementation optimization scheme. Specifically, this includes: optimization of the entire project planning, pre-construction, construction, and acceptance process; optimization of the provincial and municipal level collaborative authority flow; a strategy to compress approval process time; a proactive risk prevention and control plan for obstructed nodes; and a dynamic adjustment scheme for project schedule under multiple scenarios. This addresses the problems of rigid project implementation timelines, disconnected provincial and municipal collaboration, and delayed risk handling in traditional projects, maximizing project implementation efficiency.

[0090] The pre-feasibility study auxiliary suggestions are specifically designed based on four types of regional characteristic constraints, combinations of differentiated simulation scenario results, and optimization results of the pre-feasibility study feasibility coefficient. These suggestions are tailored to the specific characteristics of the region and provide targeted optimization recommendations. Specifically, these include: optimization suggestions for substation site coordinate ranges, fine-tuning schemes for transmission line corridor path curvature, adaptation and avoidance strategies for regional ecological constraints, and optimization schemes for line meteorological protection under extreme weather scenarios. This achieves refined adaptation and optimization of the four core pre-feasibility study elements—site selection, corridor, ecology, and meteorology—ensuring optimal feasibility for project implementation.

[0091] The detailed investment cost estimate is specifically calculated by combining the optimized project implementation period, construction plan, pre-feasibility study adjustment parameters, and regional obstruction risk prevention costs to quantify the investment details across all dimensions. Specifically, it includes: basic construction costs, costs resulting from timeline optimization due to increased or decreased construction period, incremental costs from ecological avoidance and meteorological protection upgrades, costs reduced through provincial and local collaborative process optimization, and contingency costs for risk obstruction prevention, among other sub-items. This results in a precise and actionable detailed project investment cost estimate, providing data support for power grid planning project investment decisions.

[0092] A comprehensive evaluation of the combination of differentiated project optimization decision-making schemes is conducted by weighted summation from three dimensions: target achievement rate, regional adaptability, and implementation feasibility. Target achievement rate is calculated as actual value / target value; regional adaptability is calculated based on the constraint satisfaction rate; and implementation feasibility is calculated based on the project implementation probability. The scheme with the highest comprehensive score is selected as the optimal scheme, and its effectiveness is verified (e.g., by incorporating it into a provincial and municipal collaborative dynamic monitoring model for compliance verification, with verification indicators including simulation error ≤ preset threshold and ecological constraint compliance rate reaching 100%). After verification, standardized decision-making results are output.

[0093] After the effectiveness of the solution is verified and the verification meets the standards, three types of standardized decision-making outcome documents are solidified and output, including an independent project implementation optimization plan, independent pre-feasibility study auxiliary suggestions, and a standardized decision-making report integrating all data. Among them, the standardized decision-making report embeds core supporting content such as detailed investment cost estimates, quantitative results of regional obstruction risks, multi-scenario simulation data, solution optimization basis and verification conclusions. All outcome documents can be exported in common formats such as Excel and PDF.

[0094] Example 2 Accordingly, this embodiment provides a collaborative monitoring and obstacle risk quantification decision-making system for power grid planning projects, used to implement the method described in any embodiment of the present invention, including a data acquisition and preprocessing module, a basic model construction and verification module, a simulation and deduction module, a multi-dimensional linkage calculation module, and a standardized decision output module; The data acquisition and preprocessing module is used to collect and standardize five types of data, including full life-cycle data of power grid planning projects, collaborative business data between provincial and local levels, regional constraint data, project obstruction risk data, and pre-feasibility study spatial constraint data. It performs regional feature parameter quantification and multi-dimensional correlation mapping on the five types of data and outputs a standardized fusion dataset and a multi-dimensional correlation mapping matrix. The basic model construction and verification module is used to build three types of basic models based on standardized fusion datasets: provincial and local collaborative monitoring, project implementation evaluation, and obstacle-pre-feasibility study linkage, and to complete the parameter fitting and verification of the three types of basic models. The simulation and deduction module is used to carry out provincial and local collaborative dynamic monitoring simulation, four-dimensional quantitative transmission of obstructed risks, and pre-feasibility study constraint sensitivity analysis based on the three types of basic models. The multi-dimensional linkage calculation module is used to execute the implementation based on the output results of the four-dimensional quantitative transmission of obstructed risks and the constraint sensitivity analysis of the pre-feasibility study - two-way linkage simulation of the pre-feasibility study, two-level penetration analysis of the province and the region, and accurate calculation of the regional characteristic impact, to obtain the two-way linkage simulation results, the two-level penetration analysis results of the province and the region, and the quantitative results of the regional characteristic impact. The standardized decision output module is used to construct a multi-objective optimization model, solve and verify the optimal solution based on the two-way linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and output standardized decision results.

[0095] Example 3 This embodiment provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method described in any embodiment of the present invention.

[0096] Example 4 This embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any embodiment of the present invention.

[0097] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0098] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0099] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0100] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0101] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for collaborative monitoring and quantitative decision-making regarding obstruction risks in power grid planning projects, characterized in that, Includes the following steps: S1: Collect and standardize five types of data, including full life-cycle data of power grid planning projects, collaborative business data between provincial and local levels, regional constraint data, project obstruction risk data, and pre-feasibility study spatial constraint data. Perform regional feature parameter quantification and multi-dimensional correlation mapping on the five types of data, and output a standardized fusion dataset and a multi-dimensional correlation mapping matrix. S2: Based on the standardized fusion dataset, construct three basic models for provincial and local collaborative monitoring, project implementation evaluation, and obstacle-pre-feasibility study linkage, and complete the parameter fitting and verification of the three basic models; S3: Based on the three types of basic models, conduct provincial and local collaborative dynamic monitoring simulation, four-dimensional quantitative transmission of obstructed risks, and pre-feasibility study constraint sensitivity analysis; S4: Based on the output results of the four-dimensional quantitative transmission of obstructed risks and the sensitivity analysis of pre-feasibility study constraints, the pre-feasibility study bidirectional linkage simulation, provincial and municipal two-level penetration analysis, and precise calculation of regional characteristic impacts are performed to obtain the bidirectional linkage simulation results, provincial and municipal two-level penetration analysis results, and regional characteristic impact quantitative results. S5: Construct a multi-objective optimization model, solve and verify the optimal solution based on the two-way linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and output standardized decision results.

2. The method for collaborative monitoring and quantified decision-making regarding obstruction risks in power grid planning projects according to claim 1, characterized in that, In step S1, the quantization of regional feature parameters and multidimensional association mapping of the five types of data are performed, and the standardized fused dataset and multidimensional association mapping matrix are output as follows: The regional constraint data among the five types of data are quantified, including extreme weather characteristics, geological disaster characteristics, social coordination characteristics, and life constraint characteristics; Based on five types of data that have undergone standardized processing and quantification of regional characteristics, and based on the weighting rules determined by the expert scoring method, a multi-dimensional correlation mapping matrix is ​​established, which includes "project implementation - provincial and local level collaboration - regional characteristics - obstacles and risks - pre-feasibility study constraints".

3. The method for collaborative monitoring and quantified decision-making regarding obstruction risks in power grid planning projects according to claim 1, characterized in that, Step S2 specifically involves: The provincial and municipal level collaborative monitoring basic model consists of a perception layer, a data layer, a support layer, and an application layer: The perception layer is used to input the standardized fusion dataset; the data layer performs model operation format standardization on the standardized fusion dataset input by the perception layer based on the multidimensional association mapping matrix. The support layer has a built-in parameter resource library, algorithm resource library, and mapping resource library. By calling these three types of resource libraries, it provides parameters, algorithms, and correlation support for application layer simulation and indicator calculation. The application layer calls the data from the data layer (which has been normalized) and the configuration resources from the support layer to initialize parameters and verify their validity, and outputs the simulation input dataset. The project implementation evaluation model is composed of weighted indicators from four dimensions: progress, quality, benefits, and collaboration. The progress indicators include the completion rate of preliminary approvals, the construction progress deviation rate, and the on-time commissioning and acceptance rate. The quality indicators include the project acceptance pass rate, the equipment delivery pass rate, and the safety hazard rectification rate. The benefits indicators include clean energy consumption, power supply reliability improvement, and return on investment. The collaboration indicators include the timeliness of provincial and local information reporting, the cross-departmental approval collaboration rate, and the instruction response completion rate. The model calculates the scores for each dimension indicator by calling a standardized fusion dataset, and outputs the basic results of the project implementation evaluation. The hindered-pre-feasibility study linkage basic model adopts a dual-link closed-loop architecture of forward transmission link and backward correction link. The forward transmission link follows the flow of monitoring simulation → implementation evaluation → hindered quantification → pre-feasibility study constraint correction. The backward correction link follows the flow of pre-feasibility study constraint assessment → implementation risk calculation → monitoring parameter optimization → evaluation score correction. The standardized fusion dataset is substituted into the hindered-pre-feasibility study linkage basic model for iterative calculation and benchmarking against the actual implementation effect of the project to complete the model parameter fitting and output the correlation parameter table.

4. The method for collaborative monitoring and quantification of obstruction risk in power grid planning projects according to claim 3, characterized in that, Specifically, S3 is: S31: Substitute the simulation input dataset into the provincial and municipal collaborative monitoring basic model, use the gradient descent algorithm to carry out iterative simulation, and output the provincial and municipal collaborative dynamic monitoring simulation results, including project progress, construction quality, and provincial and municipal collaborative status. S32: Construct a four-dimensional obstruction model with a three-level architecture of four-dimensional parallel input layer - sub-item feature conversion layer - comprehensive obstruction result output layer. The four-dimensional obstruction model decomposes and refines the standardized fusion dataset and multi-dimensional correlation mapping matrix into three categories of obstruction factors, including natural environment, policy constraints and external coordination, and internal management. The three categories of obstruction factors are combined with predetermined regional feature weights to perform individual risk scoring. The scores of the three categories of obstruction factors are aggregated and weighted to output the overall project obstruction quantitative score and the probability of obstruction. A risk transmission flow graph based on a four-dimensional hindered model is constructed using system dynamics, defining three types of model variables: level variables, rate variables, and auxiliary variables. The risk transmission flow graph adopts a three-layer hierarchical architecture: auxiliary variable input layer, rate-level variable transmission layer, and comprehensive risk output layer, outputting dynamic cumulative risk results. Based on the four-dimensional obstruction model, the hierarchical architecture of the analytic hierarchy process (AHP) criteria layer and the scheme layer is divided. The input data of the AHP consists of the scores of three categories of obstruction factors. The input data of the fuzzy comprehensive evaluation includes the index weights obtained by the AHP solution and the dynamic cumulative risk results. The four-dimensional quantitative result of obstruction risk is calculated by the AHP-fuzzy comprehensive evaluation combined algorithm. S33: Select four core elements of standardized fusion datasets: centralized site selection, corridor, ecological avoidance, and meteorological protection, to build a parameterized configuration framework. Perform two-way parameter adjustment on the data parameters corresponding to the four core elements: constraint tightening and constraint relaxation. Generate differentiated simulation scenario scheme parameter combinations through cross-combination of multiple elements and multiple control directions. The differentiated simulation scenario parameter combinations are substituted into the provincial and municipal collaborative monitoring basic model. Using the control variable method to keep other baseline parameters constant, the parameter impact magnitude of project feasibility under varying core element parameters is quantitatively calculated. Based on the impact magnitude of each core element parameter, the average impact magnitude of the four types of core elements under multiple operating conditions is calculated, and the core elements are ranked in descending order of sensitivity according to the magnitude of the average impact magnitude under multiple operating conditions. By fitting the correlation between the variation magnitude of each core element parameter and project feasibility, the parameter constraint critical values ​​for each core element to meet the pre-feasibility study feasibility requirements are obtained through reverse iterative solution. A pre-feasibility study constraint sensitivity analysis report containing the ranking results and parameter constraint critical values ​​is output.

5. The method for collaborative monitoring and quantified decision-making regarding obstruction risks in power grid planning projects according to claim 1, characterized in that, Specifically, S4 is: S41: Import the four-dimensional quantification result of the obstructed risk obtained from the four-dimensional quantification of the obstructed risk into the obstructed-pre-feasibility study linkage basic model. Based on the forward transmission link and backward correction link of the obstructed-pre-feasibility study linkage basic model, complete the bidirectional iterative substitution of the four-dimensional quantification result of the obstructed risk, execute the bidirectional linkage simulation of the obstructed-pre-feasibility study linkage basic model, call the GraphSAGE neural network to mine and complete the potential correlation between elements, correct the linkage coefficient, and output the bidirectional linkage simulation results including the linkage coefficient, risk transmission path, and rule correction record. S42: Perform quantitative processing of five types of data, including provincial and local level collaborative authority, regional constraint strength, and approval process duration, and calculate the three types of impact factors by combining them with preset impact factor weights to obtain the three levels of impact values. The impact values ​​of each level are statistically analyzed at the provincial-level overall network coordination dimension and the prefecture-level single project implementation dimension. Then, using the provincial-level overall indicators as a benchmark, the single project indicators of each prefecture-level city are compared with the provincial-level overall indicators item by item. The absolute difference and relative deviation of the indicators are calculated, the deviation of the indicators at the two levels are sorted out, and the characteristics, range and distribution patterns of the differences are summarized to generate the hierarchical difference analysis results at the provincial and prefecture levels. In combination with the preset fluctuation threshold of the impact factors, the business link is traced, and abnormal items exceeding the threshold are investigated layer by layer to complete the location of control breakpoints at each link. Finally, the output includes the results of the provincial and municipal level penetration analysis, including the results of the hierarchical difference analysis and the location of control breakpoints. S43: Based on the standardized fusion dataset, the four-dimensional quantitative transmission of obstruction risk, and the output results of the pre-feasibility study constraint sensitivity analysis, the schedule delay rate, the probability of project obstruction, and the pre-feasibility study feasibility coefficient are weighted and calculated; the schedule delay rate, the probability of project obstruction, and the pre-feasibility study feasibility coefficient are normalized and combined with the optimized linkage coefficient for weighted fusion, and finally the quantitative results of regional feature influence are output.

6. The method for collaborative monitoring and quantification of obstruction risk in power grid planning projects according to claim 3, characterized in that, The S5 constructs a multi-objective optimization model, solves and verifies the optimal solution based on bidirectional linkage simulation results, provincial and regional two-level penetration analysis results, and regional characteristic influence quantification results, and outputs standardized decision-making results as follows: The decision objectives and constraints are constructed as follows: the decision objectives are set as dual core objectives of maximizing project implementation efficiency and optimizing pre-feasibility study feasibility; a single-objective constraint function is synthesized based on the dual core decision objectives to optimize the feasible domain definition and the coarse screening of ineffective solutions; the constraints include the provincial and local level coordination cycle range, the regional cost increase range, the construction period delay ratio range, and the ecological constraint compliance rate. A hybrid algorithm combining scenario-optimized improved hierarchical analysis and fuzzy comprehensive evaluation was adopted to complete the construction of the index system and the basic weight calibration of the multi-objective optimization model. The improvements of the improved hierarchical analysis include three aspects: data-driven construction of the judgment matrix, the addition of a dynamic weight correction mechanism, and linkage fuzzy result optimization consistency verification. A multi-objective model was trained by constructing an equipment sample dataset using historical power grid planning project data from the region and training it by subsetting the datasets at a fixed ratio. The results of bidirectional linkage simulation, provincial and municipal penetration analysis, and regional characteristic influence quantification are integrated and input into the multi-objective optimization model solver for iterative solution, and the Pareto optimal solution set is generated as the output. Based on the Pareto optimal solution set, a combination of differentiated project optimization decision schemes is output. A comprehensive evaluation of the combination of decision-making schemes for differentiated projects is conducted from three dimensions: target achievement, regional adaptability, and implementation feasibility. The scheme with the highest comprehensive score is selected as the optimal scheme, and the effectiveness of the optimal scheme is verified. After the verification is successful, standardized decision-making results are output.

7. A collaborative monitoring and obstacle risk quantification decision-making system for power grid planning projects, characterized in that, The system is used to implement the method as described in any one of claims 1 to 6, and includes a data acquisition and preprocessing module, a basic model construction and verification module, a simulation and deduction module, a multi-dimensional linkage calculation module, and a standardized decision output module; The data acquisition and preprocessing module is used to collect and standardize five types of data, including full life-cycle data of power grid planning projects, collaborative business data between provincial and local levels, regional constraint data, project obstruction risk data, and pre-feasibility study spatial constraint data. It performs regional feature parameter quantification and multi-dimensional correlation mapping on the five types of data and outputs a standardized fusion dataset and a multi-dimensional correlation mapping matrix. The basic model construction and verification module is used to build three types of basic models based on standardized fusion datasets: provincial and local collaborative monitoring, project implementation evaluation, and obstacle-pre-feasibility study linkage, and to complete the parameter fitting and verification of the three types of basic models. The simulation and deduction module is used to carry out provincial and local collaborative dynamic monitoring simulation, four-dimensional quantitative transmission of obstructed risks, and pre-feasibility study constraint sensitivity analysis based on the three types of basic models. The multi-dimensional linkage calculation module is used to execute the implementation based on the output results of the four-dimensional quantitative transmission of obstructed risks and the constraint sensitivity analysis of the pre-feasibility study - two-way linkage simulation of the pre-feasibility study, two-level penetration analysis of the province and the region, and accurate calculation of the regional characteristic impact, to obtain the two-way linkage simulation results, the two-level penetration analysis results of the province and the region, and the quantitative results of the regional characteristic impact. The standardized decision output module is used to construct a multi-objective optimization model, solve and verify the optimal solution based on the two-way linkage simulation results, the provincial and regional two-level penetration analysis results, and the quantitative results of the regional characteristic influence, and output standardized decision results.

8. An electronic device, the electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the method as described in any one of claims 1 to 6.

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

Images