An intelligent management and control method and system for field operation risks of a coal-fired power plant

Abstract

The application provides a coal-fired power plant field operation risk intelligent management and control method and system, belonging to the technical field of safety production risk management and control and gas monitoring. The method builds a digital space model, combines real-time wind speed data and wind direction data, uses finite element analysis to simulate airflow field distribution, accurately identifies low ventilation areas, and predicts harmful gas diffusion trajectory and accumulation area through a particle tracking algorithm. A risk heat map is generated for high concentration areas to determine priority intervention points, and ventilation configuration is iteratively optimized to significantly reduce gas accumulation risk, ultimately forming a safety access permission evaluation report. The application seamlessly connects spatial modeling, dynamic simulation and risk assessment, accurately predicts and intelligently controls gas risks in confined spaces, and significantly improves the safety of coal-fired power plant field operations.

Description

Technical Field

[0001] This application relates to the fields of safety production risk management and gas monitoring technology, specifically to an intelligent management method and system for on-site operation risks in coal-fired power plants. Background Technology

[0002] As a crucial pillar of energy production, the safety of coal-fired power plants directly impacts personnel lives and equipment operational stability, especially in confined space operations where risks and hazards are particularly pronounced. Effectively identifying and managing potential hazards in the work environment has become a critical issue that cannot be ignored in ensuring the safe operation of power plants. Research in this area is not only related to production efficiency but also crucial to the safety of workers, urgently requiring innovative technological means to enhance risk management capabilities.

[0003] Currently, although many power plants have implemented some safety monitoring and protection measures, these methods often have limitations and are difficult to fully adapt to complex operating environments. Especially when dealing with gas diffusion issues in confined spaces, existing methods rely more on post-event detection or experience-based judgment, lacking the ability to anticipate dynamic environmental changes. This passive approach often leads to delayed risk warnings, making it impossible to take effective measures before danger occurs, a problem particularly pronounced in scenarios with complex and variable ventilation conditions.

[0004] Against this backdrop, the problem of gas diffusion in confined spaces at coal-fired power plants has become a key and challenging technical issue. The core challenge lies in accurately determining the distribution patterns of harmful gases within a specific space, and these patterns are particularly significantly affected by ventilation conditions. Different ventilation conditions directly alter the flow paths and accumulation areas of gases. For example, inside coal bunkers or boiler furnaces, insufficient ventilation or uneven airflow can lead to high concentrations of harmful gases such as carbon monoxide accumulating in certain corners, threatening the safety of workers. This uncertainty in gas distribution caused by ventilation conditions has become a critical technical obstacle that urgently needs to be addressed. Summary of the Invention

[0005] To address the above issues, this application provides an intelligent risk management method and system for on-site operations in coal-fired power plants. This system is designed to accurately identify potential areas of hazardous gas accumulation under changing ventilation conditions before personnel enter confined spaces, thereby providing a reliable basis and more comprehensive protection for operational safety.

[0006] To achieve the above objectives, the technical solution adopted in this application is as follows:

[0007] As the first major aspect of this application, this application provides a method for intelligent management and control of on-site operational risks in coal-fired power plants, the method comprising:

[0008] S1) Obtain the geometric structure data of the confined space through a pre-established three-dimensional scanning method, and combine it with the initial gas concentration distribution data collected by the field sensors to construct an initial spatial model that includes boundary conditions and grid distribution;

[0009] S2) Import the real-time monitored wind speed and wind direction data according to the initial spatial model, and use finite element analysis to parameterize the ventilation conditions to obtain ventilation parameters;

[0010] S3) Extract the dominant airflow vector data from the ventilation parameters. If the dominant airflow vector data is lower than the preset airflow threshold, it is marked as a low ventilation area. The diffusion trajectory of harmful gas after being released from the source is simulated by the particle tracking algorithm to determine the location of potential harmful gas accumulation areas and generate a harmful gas concentration distribution map.

[0011] S4) Obtain the concentration distribution map of harmful gases in the simulation results of harmful gas diffusion, perform cluster analysis on high concentration nodes, and obtain the boundary contour and peak concentration value of the harmful gas accumulation area;

[0012] S5) Based on the boundary contour of the harmful gas accumulation area and the initial spatial model, if the peak concentration value exceeds the safe gas concentration threshold, a risk heat map is generated.

[0013] S6) Extract high-risk coordinates from the risk heat map, and use iterative optimization to adjust and simulate ventilation conditions to obtain an optimized harmful gas diffusion mode and a ventilation configuration scheme to reduce accumulation;

[0014] S7) By comparing the optimized ventilation configuration with the initial space model, the degree of reduction in the area where harmful gases accumulate is determined, and the final safe entry permit assessment report is obtained.

[0015] Further, step S1) involves acquiring confined space geometric data using a pre-established three-dimensional scanning method, and combining this data with initial gas concentration distribution data collected by on-site sensors to construct an initial spatial model that includes boundary conditions and grid distribution. Specifically, this may include the following steps:

[0016] A comprehensive data collection was conducted in the confined space using a pre-established 3D scanning method. The collected geometric data was then formatted to obtain a preliminary spatial geometric dataset.

[0017] Based on the preliminary spatial geometry dataset, the restricted space is structurally segmented using a grid partitioning method to generate a refined spatial framework that includes boundary conditions and grid distribution.

[0018] By using on-site sensor equipment, the initial distribution data of gas concentration in the confined space is obtained. The collected concentration data is standardized to determine a unified concentration distribution matrix. If there are missing data or outliers in the concentration distribution matrix, the missing areas are filled in by interpolation to obtain a complete concentration distribution dataset.

[0019] Based on the refined spatial framework and the complete concentration distribution dataset, data fusion processing is performed to spatially map the geometric structure with gas concentration information, generating an initial spatial model with concentration distribution.

[0020] Using the initial spatial model with concentration distribution, the finite element analysis method is used to conduct a preliminary simulation of the gas diffusion trend in the confined space and determine the distribution of potential high-concentration areas.

[0021] Based on the high-concentration region distribution obtained from the simulation, the initial spatial model with concentration distribution is subjected to local mesh refinement processing for subsequent in-depth analysis.

[0022] Further, in step S2), the real-time monitored wind speed and direction data are imported based on the initial spatial model, and the ventilation conditions are parameterized using finite element analysis to obtain ventilation parameters, which may specifically include:

[0023] By using a real-time monitoring system, wind speed and direction data are acquired from environmental sensors to construct an initial airflow input dataset. After data preprocessing, standardized input parameters are obtained.

[0024] Standardized input parameters are imported into a pre-established initial spatial model, and the airflow distribution under different ventilation scenarios is simulated using the finite element analysis method to calculate the corresponding fluid dynamic parameters.

[0025] Based on the fluid dynamics parameters, the gas flow path under different ventilation scenarios is analyzed, the key nodes of gas flow path changes are extracted, and the main influencing factors of gas flow path changes are determined.

[0026] To address the main influencing factors of gas flow path changes, the simulation boundary conditions of the ventilation scenario were adjusted, the dynamic changes in airflow distribution were recalculated, and updated fluid dynamics data were obtained.

[0027] By comparing the differences in gas flow paths under different ventilation scenarios using updated fluid dynamics data, the specific effects of adjusting ventilation parameters are determined. If the differences in gas flow paths exceed the preset threshold, a secondary parametric simulation of the ventilation scenario is performed to iteratively optimize the gas flow distribution model and obtain the final gas flow path impact analysis results.

[0028] Further, in step S3), the dominant airflow vector data is extracted from the ventilation parameters. If the dominant airflow vector data is lower than a preset airflow threshold, it is marked as a low-ventilation area. A particle tracking algorithm is used to simulate the diffusion trajectory of harmful gases after release from the source, determining the location of potential harmful gas accumulation areas and generating a harmful gas concentration distribution map. Specifically, this may include:

[0029] By extracting the dominant airflow vector data from the ventilation parameters, and comparing the velocity magnitude in the dominant airflow vector data with a pre-established airflow vector analysis model, if the dominant airflow vector data is lower than a preset airflow threshold, the area is marked as a low ventilation area.

[0030] Data on the source release location of harmful gases in low-ventilation areas are obtained, and particle tracking algorithms are used to simulate the diffusion path of harmful gases to determine the distribution range of the diffusion trajectory of harmful gases.

[0031] Based on the distribution range of the harmful gas diffusion trajectory, the residence time of the harmful gas in each grid cell through which the harmful gas diffusion trajectory passes is obtained, the cumulative amount of harmful gas in each grid cell is calculated, and the specific location of the potential harmful gas accumulation area is determined.

[0032] From the specific location of potential hazardous gas accumulation areas, extract grid cells with high accumulation of hazardous gases, analyze the changes in hazardous gas concentration between adjacent grid cells, and determine the hazardous gas concentration distribution map;

[0033] Based on the distribution map of harmful gas concentrations, the overlapping area between high-concentration areas and low-ventilation areas is obtained. The changes in airflow vectors in the overlapping area are analyzed to determine whether there is a risk of continuous accumulation.

[0034] Based on the assessment results of continuous risk accumulation, the locations of high-risk areas are extracted. Combined with real-time updated data of ventilation parameters, the changes in diffusion paths after the adjustment of airflow vectors are simulated to obtain the final location of the harmful gas accumulation area.

[0035] For the final location of the hazardous gas accumulation area, the interaction between the hazardous gas concentration gradient and the airflow vector is analyzed to determine the priority ranking of high-risk areas and complete the risk assessment output.

[0036] Further, step S4) involves obtaining the hazardous gas concentration distribution map from the hazardous gas diffusion simulation results, performing cluster analysis on high-concentration nodes to obtain the boundary contours and peak concentration values ​​of hazardous gas accumulation regions. This may specifically include the following steps:

[0037] The concentration distribution map of harmful gases is extracted from the simulation results of harmful gas diffusion and preliminarily cleaned to remove outliers, resulting in a standardized dataset of harmful gas concentration distribution.

[0038] For a standardized dataset of hazardous gas concentration distribution, if the hazardous gas concentration is higher than a preset hazardous gas concentration threshold, it is selected as a node with high hazardous gas concentration.

[0039] For the nodes with high concentrations of harmful gases, the K-means clustering method is used to group adjacent nodes with high concentrations of harmful gases, and the preliminary division results of multiple harmful gas accumulation regions are obtained.

[0040] Based on the preliminary division of harmful gas accumulation areas, the spatial distribution characteristics of the concentration values ​​within each harmful gas accumulation area are calculated, and the boundary contour of each harmful gas accumulation area is obtained.

[0041] By sorting the concentration values ​​of each harmful gas accumulation region, the highest concentration value in each harmful gas accumulation region is extracted to obtain the corresponding peak concentration value.

[0042] Based on the boundary contour data and peak concentration values, an overlay analysis is performed according to the initial spatial model to generate a comprehensive mapping map showing the concentration distribution and the accumulation area of ​​harmful gases.

[0043] If there is overlapping boundary contours in the comprehensive mapping map, the concentration values ​​of the overlapping areas are weighted and averaged, and the boundary contours are redefined to obtain the final distribution results of the harmful gas accumulation areas.

[0044] Further, in step S5), based on the superposition of the boundary contour of the hazardous gas accumulation region with the initial spatial model, if the peak concentration value exceeds the safe gas concentration threshold, a risk heat map is generated, which may specifically include:

[0045] By superimposing the boundary contour of the harmful gas accumulation region with the initial spatial model, the distribution characteristics of the overlapping region are obtained.

[0046] Based on the distribution characteristics of the overlapping area, the concentration values ​​in the overlapping area are collected in real time to obtain the peak concentration value. If the detected peak concentration value exceeds the preset safe gas concentration threshold, the risk assessment mechanism is triggered to generate the corresponding risk heat map.

[0047] By using risk heat maps, high-risk areas can be identified, and priority intervention points can be identified.

[0048] For the locked work site, acquire the surrounding environmental data of the work site in the initial spatial model to determine the boundary contour of the intervention range;

[0049] By using a pre-established decision tree model, combined with the risk heat map and boundary contour of the locked operation points, specific intervention sequences and resource allocation plans are generated.

[0050] By adjusting the intervention sequence and resource allocation plan, the risk assessment data in the initial spatial model is updated, and key areas for subsequent monitoring are identified.

[0051] Furthermore, step S6) extracts high-risk coordinates from the risk heat map and uses iterative optimization to simulate and adjust ventilation conditions, obtaining an optimized harmful gas diffusion pattern and a ventilation configuration scheme to reduce accumulation. Specifically, this may include:

[0052] The distribution of high-risk coordinates can be obtained through risk heatmaps;

[0053] Based on the location of high-risk coordinates, and using preset ventilation scenario parameters, the initial state of harmful gas diffusion is simulated to obtain preliminary diffusion pattern data.

[0054] If the accumulation risk is found to be higher than the preset accumulation risk threshold based on the preliminary diffusion pattern data, the ventilation scenario parameters are adjusted through an iterative method to obtain updated diffusion simulation results, and the distribution changes of harmful gas diffusion are extracted from them to determine whether the accumulation risk has been reduced.

[0055] If the risk of accumulation is reduced, the optimized ventilation scenario parameters are used, and the output of the simulation analysis is combined to generate the final configuration scheme to reduce the risk.

[0056] Obtain the simulation results data of the final configuration scheme, determine whether it meets the preset risk accumulation control standards, and output a complete solution.

[0057] Further, step S7) compares the optimized ventilation configuration with the initial space model to determine the degree of reduction in the area where harmful gases accumulate, and obtains the final safe entry permit assessment report, which may specifically include:

[0058] By obtaining the optimized ventilation configuration scheme and the initial space model, preliminary data format unification processing is performed to obtain the first standardized dataset.

[0059] Based on the standardized first dataset, a pre-set comparative analysis module is used to compare the distribution range of the harmful gas accumulation area layer by layer to determine the specific range of reduction of the harmful gas accumulation area.

[0060] If the range value of the reduction of the harmful gas accumulation area meets the preset safety permit standard threshold, it is determined that the safe entry conditions are met, and the first judgment result is generated.

[0061] Based on the first determination result, the relevant gas accumulation situation is obtained, and the accumulation situation is classified by the support vector machine algorithm to obtain the classified second dataset;

[0062] Using the second dataset after classification, a second verification is performed on the matching degree between the classification results and the security permission standards. If the matching degree reaches the preset matching degree standard, the final evaluation state is confirmed.

[0063] Based on the final status of the assessment and in conjunction with the preset assessment report template, the relevant data fields are automatically filled in to generate the final conclusion of secure access permission;

[0064] The final conclusions are stored in a preset database, and the real-time status data of the initial spatial model is updated synchronously to complete the entire evaluation process.

[0065] As a second major aspect of this application, this application provides an intelligent risk management and control system for on-site operations in coal-fired power plants, the system comprising:

[0066] The spatial modeling module is used to acquire confined space geometry data through a pre-established 3D scanning method, and combine it with initial gas concentration distribution data collected by on-site sensors to construct an initial spatial model that includes boundary conditions and grid distribution.

[0067] The airflow field simulation module is used to import real-time monitored wind speed and wind direction data based on the initial spatial model, and to perform parameterization processing of ventilation conditions using finite element analysis to obtain ventilation parameters.

[0068] The accumulation area prediction module is used to extract the dominant airflow vector data from the ventilation parameters. If the dominant airflow vector data is lower than the preset airflow threshold, it is marked as a low ventilation area. The particle tracking algorithm is used to simulate the diffusion trajectory of harmful gases after they are released from the source, determine the location of potential harmful gas accumulation areas, and generate a harmful gas concentration distribution map.

[0069] The accumulation region analysis module is used to obtain the concentration distribution map of harmful gases in the simulation results of harmful gas diffusion, and to perform cluster analysis on high concentration nodes to obtain the boundary contour and peak concentration value of the harmful gas accumulation region.

[0070] The risk location module is used to generate a risk heat map by superimposing the boundary contour of the harmful gas accumulation area with the initial spatial model. If the peak concentration value exceeds the safe gas concentration threshold, the heat map is generated.

[0071] The ventilation optimization module is used to extract high-risk coordinates from the risk heat map, and to perform iterative optimization to adjust and simulate ventilation conditions, thereby obtaining an optimized harmful gas diffusion mode and a ventilation configuration scheme to reduce accumulation.

[0072] The safety assessment module is used to compare the optimized ventilation configuration with the initial space model to determine the extent to which the area where harmful gases accumulate is reduced, and to obtain the final safety access permit assessment report.

[0073] In summary, this application provides an intelligent risk management method and system for on-site operations in coal-fired power plants. Addressing the safety risks caused by the accumulation of harmful gases in confined spaces, it innovatively integrates a logically linked solution of 3D scanning modeling, gas diffusion simulation, and ventilation optimization. This application constructs a digital spatial model, combines real-time wind speed and direction data, uses finite element analysis to simulate airflow distribution, accurately identifies low-ventilation areas, and predicts the diffusion trajectory and accumulation areas of harmful gases using a particle tracking algorithm. Subsequently, a risk heat map is generated for high-concentration areas to determine priority intervention points, and ventilation configuration is iteratively optimized to significantly reduce the risk of gas accumulation, ultimately generating a safe entry permit assessment report.

[0074] The core innovation of this application lies in the seamless integration of spatial modeling, dynamic simulation and risk assessment, enabling accurate prediction and intelligent control of gas risks in confined spaces, and comprehensively improving the safety of on-site operations in coal-fired power plants. Attached Figure Description

[0075] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. The description in this part is only exemplary and explanatory, and should not limit the scope of protection of this application in any way.

[0076] Figure 1 This is a schematic flowchart of one embodiment of the intelligent risk management method for on-site operations in coal-fired power plants proposed in this application.

[0077] Figure 2 This is a block diagram of the module composition of the intelligent risk management and control system for on-site operations in coal-fired power plants, as described in this application. Detailed Implementation

[0078] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this application will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this application.

[0079] The terms “a,” “one,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended inclusion and that other elements / components / etc. may exist in addition to those listed. The terms “first” and “second” are used only as markers and are not a limitation on the number of objects.

[0080] like Figure 1 As shown in the figure, a specific method for intelligent risk management and control of on-site operations in a coal-fired power plant in this embodiment may include:

[0081] S1) Obtain the geometric structure data of the confined space through a pre-established three-dimensional scanning method, and combine it with the initial gas concentration distribution data collected by the field sensors to construct an initial spatial model that includes boundary conditions and grid distribution;

[0082] S2) Import the real-time monitored wind speed and wind direction data according to the initial spatial model, and use finite element analysis to parameterize the ventilation conditions to obtain ventilation parameters;

[0083] S3) Extract the dominant airflow vector data from the ventilation parameters. If the dominant airflow vector data is lower than the preset airflow threshold, it is marked as a low ventilation area. The diffusion trajectory of harmful gas after being released from the source is simulated by the particle tracking algorithm to determine the location of potential harmful gas accumulation areas and generate a harmful gas concentration distribution map.

[0084] S4) Obtain the concentration distribution map of harmful gases in the simulation results of harmful gas diffusion, perform cluster analysis on high concentration nodes, and obtain the boundary contour and peak concentration value of the harmful gas accumulation area;

[0085] S5) Based on the boundary contour of the harmful gas accumulation area and the initial spatial model, if the peak concentration value exceeds the safe gas concentration threshold, a risk heat map is generated.

[0086] S6) Extract high-risk coordinates from the risk heat map, and use iterative optimization to adjust and simulate ventilation conditions to obtain an optimized harmful gas diffusion mode and a ventilation configuration scheme to reduce accumulation;

[0087] S7) By comparing the optimized ventilation configuration with the initial space model, the degree of reduction in the area where harmful gases accumulate is determined, and the final safe entry permit assessment report is obtained.

[0088] Step S1) Obtain the geometric structure data of the confined space using a pre-established 3D scanning method, and combine it with the initial gas concentration distribution data collected by on-site sensors to construct an initial spatial model that includes boundary conditions and grid distribution. One possible implementation is as follows:

[0089] The pre-established 3D scanning method involves determining the scanning coordinate system, scanning point layout rules, point cloud sampling density, scanning accuracy, data output format, and coordinate registration rules before scanning in a confined space, and using these rules as the execution parameters for this scanning acquisition.

[0090] A comprehensive data collection was conducted in the confined space using a pre-established 3D scanning method. The collected geometric data was then formatted to obtain a preliminary spatial geometric dataset.

[0091] Based on the preliminary spatial geometry dataset, the restricted space is structurally segmented using a grid partitioning method to generate a refined spatial framework that includes boundary conditions and grid distribution.

[0092] By using on-site sensor equipment, the initial distribution data of gas concentration in the confined space is obtained. The collected concentration data is standardized to determine a unified concentration distribution matrix. If there are missing data or outliers in the concentration distribution matrix, the missing areas are filled in by interpolation to obtain a complete concentration distribution dataset.

[0093] Based on the refined spatial framework and the complete concentration distribution dataset, data fusion processing is performed to spatially map the geometric structure with gas concentration information, generating an initial spatial model with concentration distribution.

[0094] Using the initial spatial model with concentration distribution, the finite element analysis method is used to conduct a preliminary simulation of the gas diffusion trend in the confined space and determine the distribution of potential high-concentration areas.

[0095] Based on the high-concentration region distribution obtained from the simulation, the initial spatial model with concentration distribution is subjected to local mesh refinement processing for subsequent in-depth analysis.

[0096] To further describe the method of step S1) of this application, a specific embodiment is described below. Those skilled in the art will understand that this embodiment may not include all possible situations covered by the method of this application and does not constitute a limitation on the method of this application.

[0097] Comprehensive data collection is conducted in a confined space using a pre-established 3D scanning method. The collected geometric information is then formatted to obtain a preliminary spatial geometric dataset. For example, a lidar device is used to scan the interior of an underground storage tank measuring 5 meters long, 3 meters wide, and 2 meters high, generating point cloud data with a density of 100 points per cubic centimeter and a scanning accuracy controlled within ±1 millimeter. Data processing employs point cloud filtering algorithms, such as voxel mesh filtering, to downsample the point cloud data to 50 points per cubic centimeter to reduce computational load. Subsequently, a 3D surface model is constructed using triangulation algorithms (such as Delaunay triangulation), generating a spatial geometric dataset containing 100,000 triangular facets.

[0098] Based on the preliminary spatial geometry dataset, a mesh generation method is used to structurally segment the confined space, generating a refined spatial framework that includes boundary conditions and mesh distribution. This refined spatial framework represents the geometric boundaries, mesh cells, and boundary conditions of the confined space, serving as the geometric and mesh foundation for generating the initial spatial model. For example, by setting the tank wall as a no-slip boundary condition and the inlet wind speed as 0.5 m / s, a structured mesh generation method is used to divide the space into 1 million volumetric cells. The mesh size gradually changes from 0.01 meters at the boundary to 0.05 meters in the central region, ensuring the accuracy of the boundary layer flow field calculation. The mesh quality is verified by skewness testing, with the average skewness value controlled below 0.3, thus generating a refined spatial framework that includes boundary conditions and mesh distribution.

[0099] By deploying sensor equipment on-site, initial gas concentration distribution data within the confined space is acquired. The collected concentration information is then standardized to determine a unified concentration distribution matrix. If data gaps or outliers exist in the concentration distribution matrix, interpolation methods are used to fill in the missing data, resulting in a complete concentration distribution dataset. For example, firstly, initial gas concentration distribution information is collected from on-site sensors. Assuming five gas sensors are deployed inside the storage tank, located at the center and four corners, measuring methane concentrations of 2.5%, 1.8%, 2.0%, 1.9%, and 2.2%, respectively, Kriging interpolation is used to spatially interpolate the concentration data, generating a concentration field distribution map with a resolution of 0.1 m × 0.1 m × 0.1 m, with an error controlled within ±0.2%, thus obtaining a complete concentration distribution dataset.

[0100] Finally, the spatial geometry dataset, concentration distribution dataset, and refined spatial framework including boundary conditions and mesh distribution obtained above are integrated and fused to spatially map the geometric structure with gas concentration information, generating an initial spatial model with concentration distribution. The finite element method is then used to perform a preliminary simulation of the gas diffusion trend within the confined space, identifying potential high-concentration regions. Based on the simulated high-concentration region distribution, the initial spatial model with concentration distribution is locally meshed for subsequent in-depth analysis.

[0101] Through the above steps, a complete logical chain is formed from data acquisition to the construction of the initial spatial model. The spatial geometry dataset provides the spatial basis for the concentration distribution dataset, and the concentration distribution dataset combined with the boundary conditions provides constraints for the grid distribution, thereby constructing an initial spatial model that includes boundary conditions and grid distribution.

[0102] Further, in step S2), the real-time monitored wind speed and direction data are imported based on the initial spatial model, and the ventilation conditions are parameterized using finite element analysis to obtain ventilation parameters. One possible implementation is as follows:

[0103] By using a real-time monitoring system, wind speed and direction data are acquired from environmental sensors to construct an initial airflow input dataset. After data preprocessing, standardized input parameters are obtained. The ventilation parameters include one or more of the following: wind speed, wind direction, vent location, vent area, outlet boundary conditions, fan power, and air exchange rate. These parameters are used to describe the impact of ventilation conditions on the gas flow state within a confined space.

[0104] Standardized input parameters are imported into a pre-established initial spatial model, and the airflow distribution under different ventilation scenarios is simulated using the finite element analysis method to calculate the corresponding fluid dynamic parameters.

[0105] Based on the fluid dynamics parameters, the gas flow path under different ventilation scenarios is analyzed, the key nodes of gas flow path changes are extracted, and the main influencing factors of gas flow path changes are determined.

[0106] To address the main influencing factors of gas flow path changes, the simulation boundary conditions of the ventilation scenario were adjusted, the dynamic changes in airflow distribution were recalculated, and updated fluid dynamics data were obtained.

[0107] By comparing the differences in gas flow paths under different ventilation scenarios using updated fluid dynamics data, the specific effects of adjusting ventilation parameters are determined. If the differences in gas flow paths exceed the preset threshold, a secondary parametric simulation of the ventilation scenario is performed to iteratively optimize the gas flow distribution model and obtain the final gas flow path impact analysis results.

[0108] To further describe the method of step S2 of this application, a specific embodiment is described below. Those skilled in the art will understand that this embodiment may not include all possible situations covered by the method of this application and does not constitute a limitation on the method of this application.

[0109] Specifically, based on the initial spatial model, wind speed and wind direction data are first obtained in real time from meteorological monitoring equipment through a data interface. For example, if a monitoring point records a wind speed of 5.2 m / s and a wind direction of 45 degrees northeast at a specific time, these data are automatically transmitted to the system database through an API interface and stored using timestamps as indexes.

[0110] Next, the obtained data was imported into the finite element analysis software. The structured mesh generation method was used to divide the spatial model into 1 million micro-units. The mesh size was controlled within 0.5 meters to ensure calculation accuracy. Then, the Navier-Stokes equations were used as the core algorithm to iteratively solve the velocity and pressure distribution of the airflow field. The boundary conditions were set as an inlet wind speed of 5.2 m / s and a free flow outlet. The calculated airflow velocity distribution error was controlled within ±0.1 m / s.

[0111] Furthermore, parameterization is performed for different ventilation scenarios. For example, two modes, natural ventilation and mechanical ventilation, are simulated, with ventilation opening areas of 2 square meters and 4 square meters, and fan power of 500 watts and 1000 watts, respectively. Multiple simulations can be performed here to obtain the fluid dynamic parameters for each scenario.

[0112] Finally, the influence of ventilation parameters on the gas flow path is analyzed. A sensitivity analysis algorithm is used to quantify that the airflow path offset angle is 5 degrees when the ventilation opening area changes by 10%, and the path offset distance is 1.2 meters when the wind speed changes by 0.5 m / s. Here, a visualization tool can be used to generate an airflow path diagram to more clearly show the impact of different parameters on the flow trajectory. The results are stored as report data and can be set to automatically link to subsequent ventilation optimization business modules, forming a complete chain logic from data collection to analysis to application.

[0113] Further, in step S3), the dominant airflow vector data is extracted from the ventilation parameters. If the dominant airflow vector data is lower than a preset airflow threshold, it is marked as a low ventilation area. The diffusion trajectory of harmful gas after being released from the source is simulated by a particle tracking algorithm to determine the location of potential harmful gas accumulation areas and generate a harmful gas concentration distribution map. One possible implementation is as follows: By extracting the dominant airflow vector data from the ventilation parameters, the velocity modulus value in the dominant airflow vector data is compared using a pre-established airflow vector analysis model. If the dominant airflow vector data is lower than a preset airflow threshold, the area is marked as a low ventilation area.

[0114] Data on the source release location of harmful gases in low-ventilation areas are obtained, and particle tracking algorithms are used to simulate the diffusion path of harmful gases to determine the distribution range of the diffusion trajectory of harmful gases.

[0115] Based on the distribution range of the harmful gas diffusion trajectory, the residence time of the harmful gas in each grid cell through which the harmful gas diffusion trajectory passes is obtained, the cumulative amount of harmful gas in each grid cell is calculated, and the specific location of the potential harmful gas accumulation area is determined.

[0116] From the specific location of potential hazardous gas accumulation areas, extract grid cells with high accumulation of hazardous gases, analyze the changes in hazardous gas concentration between adjacent grid cells, and determine the hazardous gas concentration distribution map;

[0117] Based on the distribution map of harmful gas concentrations, the overlapping area between high-concentration areas and low-ventilation areas is obtained. The changes in airflow vectors in the overlapping area are analyzed to determine whether there is a risk of continuous accumulation.

[0118] Based on the assessment results of continuous risk accumulation, the locations of high-risk areas are extracted. Combined with real-time updated data of ventilation parameters, the changes in diffusion paths after the adjustment of airflow vectors are simulated to obtain the final location of the harmful gas accumulation area.

[0119] For the final location of the hazardous gas accumulation area, the interaction between the hazardous gas concentration gradient and the airflow vector is analyzed to determine the priority ranking of high-risk areas and complete the risk assessment output.

[0120] To further describe the method of step S3 of this application, a specific embodiment is described below. Those skilled in the art will understand that this embodiment may not include all possible situations covered by the method of this application and does not constitute a limitation on the method of this application.

[0121] Specifically, when processing ventilation parameters to extract the dominant airflow vector, vector decomposition can be performed on the wind speed data collected by the ventilation system. For example, assuming the wind speed measured in a certain area is 3.5 meters per second in a northeast 45-degree direction, it can be decomposed into components of 2.47 meters per second in the x-axis direction and 2.47 meters per second in the y-axis direction through vector calculation. Then, the vectors of multiple measurement points are weighted and averaged to obtain the dominant airflow vector data. The dominant airflow vector data includes velocity magnitude and direction information, such as a velocity magnitude of 2.8 meters per second and a direction 40 degrees east.

[0122] Next, the vector is compared with the preset airflow threshold of 3.0 meters per second. It is found that the vector is lower than the preset airflow threshold. Therefore, the area is marked as a low ventilation area, and the relevant data is automatically entered into the system database and a warning label is generated.

[0123] Subsequently, for low-ventilation areas, a particle tracking algorithm was used to simulate the diffusion trajectory of harmful gases. Source release location data included the three-dimensional coordinates of the release point in the initial spatial model and its corresponding grid cell number. The source release rate of the harmful gas was used as a boundary condition for particle release in the diffusion simulation. Assuming a source release rate of 0.5 cubic meters per hour, a Lagrange particle tracking model was adopted, with 10,000 particles and a simulation time step of 0.1 seconds. By calculating the particle trajectories in the airflow field and combining this with a turbulent diffusion coefficient of 0.02, the diffusion path of harmful gases within a 5-meter radius in the low-ventilation area was determined. When the particle residence time and simulated concentration at a distance of 3 meters from the source both exceeded the corresponding threshold, a potential harmful gas accumulation area was identified at that location. Based on the specific location of the potential harmful gas accumulation area and the harmful gas... The system extracts grid cells with high accumulation of harmful gases based on their diffusion trajectories. It analyzes concentration changes between adjacent cells, such as a concentration gradient decreasing from 500 mg / m³ at the source to 300 mg / m³ in the accumulation zone, to obtain the distribution characteristics of the concentration gradient and determine the harmful gas concentration distribution map. Based on this map, it identifies the overlap between high-concentration and low-ventilation areas, analyzes airflow vector changes in the overlapping areas, and assesses the risk of continued accumulation. If a risk of continued accumulation exists, it is marked as a high-risk area. Combining this with real-time updated ventilation parameters, it simulates the diffusion path changes after airflow adjustment to obtain the final location of the harmful gas accumulation area. Finally, it analyzes the interaction between the harmful gas concentration gradient and airflow vector to determine the priority ranking of high-risk areas, completing the risk assessment output.

[0124] Further, in step S4), the concentration distribution map of harmful gases in the simulation results of harmful gas diffusion is obtained, and cluster analysis is performed on high-concentration nodes to obtain the boundary contours and peak concentration values ​​of the harmful gas accumulation regions. One possible implementation is as follows:

[0125] The concentration distribution map of harmful gases is extracted from the simulation results of harmful gas diffusion and preliminarily cleaned to remove outliers, resulting in a standardized dataset of harmful gas concentration distribution.

[0126] For a standardized dataset of hazardous gas concentration distribution, if the hazardous gas concentration is higher than a preset hazardous gas concentration threshold, it is selected as a node with high hazardous gas concentration.

[0127] For the nodes with high concentrations of harmful gases, the K-means clustering method is used to group adjacent nodes with high concentrations of harmful gases, and the preliminary division results of multiple harmful gas accumulation regions are obtained.

[0128] Based on the preliminary division of harmful gas accumulation areas, the spatial distribution characteristics of the concentration values ​​within each harmful gas accumulation area are calculated, and the boundary contour of each harmful gas accumulation area is obtained.

[0129] By sorting the concentration values ​​of each harmful gas accumulation region, the highest concentration value in each harmful gas accumulation region is extracted to obtain the corresponding peak concentration value.

[0130] Based on the boundary contour data and peak concentration values, an overlay analysis is performed according to the initial spatial model to generate a comprehensive mapping map showing the concentration distribution and the accumulation area of ​​harmful gases.

[0131] If there is overlapping boundary contours in the comprehensive mapping map, the concentration values ​​of the overlapping areas are weighted and averaged, and the boundary contours are redefined to obtain the final distribution results of the harmful gas accumulation areas.

[0132] To further describe the method of step S4 of this application, a specific embodiment is described below. Those skilled in the art will understand that this embodiment may not include all possible situations covered by the method of this application and does not constitute a limitation on the method of this application.

[0133] Specifically, after obtaining the hazardous gas concentration distribution map in the hazardous gas diffusion simulation results, the spatial coordinates (x, y, z) of each grid node and its corresponding hazardous gas concentration value C are read through the data extraction interface. For example, a dataset containing 10,000 nodes, where the hazardous gas concentration value ranges from 0 to 500 mg / m³.

[0134] Next, a preset hazardous gas concentration threshold is set, such as 200 mg / m³. The preset hazardous gas concentration threshold is determined based on confined space operation safety standards, enterprise safety control limits, or historical monitoring data. All nodes with hazardous gas concentrations higher than the preset hazardous gas concentration threshold are screened out, assuming 350 nodes with high hazardous gas concentrations are obtained.

[0135] Subsequently, K-means clustering analysis was used to perform spatial clustering analysis on these high-concentration nodes of harmful gases. Cluster samples were constructed using the three-dimensional coordinates of the nodes and the concentration values ​​of harmful gases. The number of cluster centers K was determined according to the inflection point method of the spatial distribution of high-concentration nodes or the number of preset regions. After iteratively updating the cluster centers, the high-concentration nodes of harmful gases that are closest to the same cluster center and are spatially continuous were divided into the same group, resulting in the preliminary division results of multiple harmful gas agglomeration regions.

[0136] For the preliminary division results of each harmful gas accumulation region, the spatial distribution characteristics of the high-concentration nodes of harmful gases in the region are calculated. The spatial distribution characteristics include the three-dimensional coordinate range of the nodes, the location of the cluster center, the distance between nodes and the concentration gradient distribution. Based on the spatial distribution characteristics, the boundary contour is extracted using the AlphaShape algorithm or the convex hull calculation method. The boundary contour is defined by a set of ordered three-dimensional coordinate points and is used to represent the outer boundary of the harmful gas accumulation region in the initial spatial model.

[0137] Meanwhile, within each harmful gas accumulation region, the maximum concentration of harmful gas is found by traversing the nodes, which is used as the peak concentration value of that harmful gas accumulation region. For example, the peak concentration of the first harmful gas accumulation region is 478.3 mg / m³, the peak concentration of the second harmful gas accumulation region is 432.1 mg / m³, and the peak concentration of the third harmful gas accumulation region is 398.7 mg / m³. These boundary contours and peak concentration values ​​can be stored as structured JSON data.

[0138] For the boundary contour data and peak concentration data of each harmful gas accumulation area, an overlay analysis is performed based on the initial spatial model to generate a comprehensive mapping map of concentration distribution and harmful gas accumulation area, and to determine the key locations of high concentration accumulation.

[0139] If there is overlapping boundary contours in the comprehensive mapping map, the concentration values ​​of harmful gases in the overlapping areas are weighted and averaged, and the boundary contours are redefined to obtain the final distribution results of the harmful gas accumulation areas.

[0140] Further, in step S5), based on the superposition of the boundary contour of the harmful gas accumulation area and the initial spatial model, if the peak concentration value exceeds the safe gas concentration threshold, a risk heat map is generated. One possible embodiment is as follows:

[0141] By overlaying the boundary contour data of the harmful gas accumulation area with the initial spatial model, the distribution characteristics of the overlapping area are obtained. The purpose of this step is to determine the potential risk range.

[0142] Based on the distribution characteristics obtained from the overlay analysis, the concentration data in the overlapping area is collected in real time to obtain the specific value of the peak concentration. If the detected peak concentration exceeds the preset safe gas concentration threshold, the risk assessment mechanism is triggered to generate the corresponding risk heat map.

[0143] By analyzing the distribution of risk heatmaps, high-risk areas can be identified, and priority intervention points can be pinpointed.

[0144] For the locked work site, acquire the surrounding environmental data of the work site in the initial spatial model to determine the boundary contour of the intervention range;

[0145] By using a pre-established decision tree model, combined with risk heat maps and boundary contour data of the work sites, specific intervention sequences and resource allocation plans are generated.

[0146] By adjusting the intervention sequence and resource allocation plan, the risk assessment data in the initial spatial model is updated, and key areas for subsequent monitoring are identified.

[0147] To further describe the method of step S5 of this application, a specific embodiment is described below. Those skilled in the art will understand that this embodiment may not include all possible situations covered by the method of this application and does not constitute a limitation on the method of this application.

[0148] Specifically, in processing the overlay of the boundary contour of the hazardous gas accumulation area with the initial spatial model, firstly, 3D spatial modeling software is used to overlay the boundary contour data of the hazardous gas accumulation area with the pre-constructed initial spatial model. It is assumed that the boundary contour of the hazardous gas accumulation area consists of a series of coordinate points, for example, a point set of {(10.5, 20.3, 5.2), (15.7, 22.1, 6.0), ...}. The initial spatial model is stored in a grid format, with each grid cell being 1 meter × 1 meter × 1 meter. The overlay algorithm uses spatial intersection calculation, specifically determining whether points within the boundary contour fall within the grid cells of the initial spatial model to calculate the volume of the overlapping area. Assuming the overlapping volume is 500 cubic meters, accounting for 10% of the total model volume, then real-time collection of concentration data within the overlapping area is performed to obtain the specific value of the peak concentration. If a peak concentration of 80 ppm is detected, and the safe gas concentration threshold is 50 ppm, the system automatically triggers a risk assessment mechanism, calculating the concentration value of each grid cell using the concentration distribution formula. The concentration distribution formula is:

[0149]

[0150] In the formula, Coordinates The concentration of harmful gases at the location coordinate point The Euclidean distance between the peak point and the peak point.

[0151] Based on the concentration values ​​of each grid cell, areas with concentrations exceeding 50 ppm are selected. Assuming the selection result is 200 grid cells, a risk heat map is generated accordingly. The resulting risk heat map can use color gradients to represent gas concentration levels, with 80 ppm represented by red and 50 ppm by yellow. The heat map data is smoothed using an interpolation algorithm (such as Kriging interpolation) to ensure continuous visualization. The heat map resolution is 0.5 meters per pixel.

[0152] Next, it is necessary to determine the priority intervention points. The system calculates the points with the highest concentration based on the grid cell with the highest concentration in the heat map, combined with the operation point database (including point coordinates and operation type). For example, if point A is only 2 meters away from the peak point and has a concentration of 75 ppm, it has the highest priority and is automatically marked as the primary intervention target. This process is repeated to lock in all operation points that need intervention.

[0153] For the identified work sites, acquire the surrounding environmental data of the work sites in the initial spatial model to determine the boundary contour of the intervention range.

[0154] By using a pre-established decision tree model, combined with risk heat maps and boundary contour data of the locked work sites, specific intervention sequences and resource allocation plans are generated.

[0155] By adjusting the intervention sequence and resource allocation plan, the risk assessment data in the initial spatial model is updated, and key areas for subsequent monitoring are identified.

[0156] Furthermore, for the locked work sites, the system can be linked to other relevant business data, such as the equipment maintenance records of site A, to find that its recent maintenance frequency is lower than the standard value (once a month, but actually once every two months). This further infers that it may be a source of risk, logically forming a complete chain from spatial overlay to risk identification and then to the determination of intervention sites, ensuring seamless connection of technical processing.

[0157] Further, in step S6), high-risk coordinates are extracted from the risk heat map, and iterative optimization is used to adjust and simulate ventilation conditions to obtain an optimized harmful gas diffusion pattern and a ventilation configuration scheme to reduce accumulation. One possible embodiment is as follows:

[0158] By using risk heatmap data, the distribution of high-risk coordinates can be obtained, thereby identifying key monitoring areas;

[0159] Based on the location information of high-risk coordinates, and using preset ventilation condition parameters, the initial diffusion state of harmful gases is simulated to obtain preliminary diffusion pattern data.

[0160] If the accumulation risk is found to be higher than the preset accumulation risk threshold based on the preliminary diffusion model data, the ventilation condition parameters are adjusted through an iterative method to obtain updated diffusion simulation results. From the updated diffusion simulation results, the distribution changes of harmful gas diffusion are extracted to determine whether the accumulation risk has been reduced.

[0161] If the risk of accumulation is reduced, the corresponding ventilation configuration scheme data is generated using optimized ventilation condition parameters, and the final risk-reducing configuration scheme is determined by combining the output of simulation analysis.

[0162] Obtain simulation data of the final configuration scheme, determine whether it meets the preset risk accumulation control standards, and output a complete solution.

[0163] To further describe the method of step S6) of this application, a specific embodiment is described below. Those skilled in the art will understand that this embodiment may not include all possible situations covered by the method of this invention, and does not constitute a limitation on the method of this application. Specifically, the process of extracting high-risk coordinates from a risk heat map and performing ventilation condition optimization simulation can be achieved through a series of information technology means.

[0164] First, the risk heatmap is analyzed using image processing algorithms. Assuming that the heatmap uses pixel values ​​to indicate risk levels (ranging from 0 to 255), a threshold of 200 or higher is set as a high-risk area. By traversing the image matrix, coordinates of pixel values ​​exceeding 200 are extracted. For example, the high-risk coordinates are (x=50, y=30, z=5) and (x=80, y=60, z=5), and their risk values ​​are recorded as 230 and 210, respectively.

[0165] Next, based on these coordinates, a computational fluid dynamics simulation algorithm was used to model the diffusion mode of harmful gases. The initial conditions were set as a wind speed of 2.5 m / s and the location of the ventilation opening at coordinates (x=10, y=10, z=3). The Navier-Stokes equations were solved using the finite element method to calculate the concentration distribution of harmful gases. It was found that the concentration at the high-risk point (x=50, y=30, z=5) was 0.8 ppm, which exceeded the safe gas concentration threshold of 0.5 ppm.

[0166] Subsequently, the iterative optimization phase was initiated, using a genetic algorithm to adjust ventilation parameters. The population size was set to 50, and the number of iterations was set to 100. In each iteration, the wind speed and the location of the ventilation openings were randomly adjusted. The wind speed ranged from 1.0 m / s to 5.0 m / s, and the location of the ventilation openings was the three-dimensional coordinate point in the initial spatial model. The objective function was to reduce the concentration of high-risk points. The fitness value was calculated as the sum of squares of the difference between the concentration and the threshold. After 80 iterations, the optimal solution was obtained with a wind speed of 3.8 m / s and a ventilation opening location (x=40, y=20, z=3). At this point, the concentration of high-risk points dropped to 0.3 ppm.

[0167] Finally, the optimized harmful gas diffusion pattern diagram is output, and a ventilation configuration scheme is generated, which includes specific values ​​for wind speed and ventilation outlet location.

[0168] This process not only enables accurate identification of risk points, but also effectively reduces the risk of gas accumulation through algorithm optimization. At the same time, the results can be integrated with the factory environmental monitoring system to automatically adjust the parameters of ventilation equipment and form a closed-loop control logic.

[0169] Further, in step S7), by comparing the optimized ventilation configuration with the initial space model, the degree of reduction in the area where harmful gases accumulate is determined, and a final safety access permit assessment report is obtained. One possible implementation is as follows:

[0170] By acquiring the optimized ventilation configuration scheme data and the initial space model data, preliminary data format unification processing is performed to obtain the first standardized dataset.

[0171] Based on the standardized first dataset, a pre-set comparative analysis module is used to compare the distribution range of the harmful gas accumulation area layer by layer to determine the specific range of reduction of the harmful gas accumulation area.

[0172] By reducing the specific range of the harmful gas accumulation area, and combining it with the preset safety permit standards, if the range value meets the preset safety permit standard threshold, it is determined that the safe entry conditions are met, and the first judgment result is generated.

[0173] Based on the first judgment result, the reduction range value, peak concentration value and the proportion of areas with excessive concentration of harmful gas accumulation area are obtained. The support vector machine algorithm is used to classify the harmful gas accumulation situation to obtain the classified second dataset.

[0174] Using the second dataset after classification, a second verification is performed on the matching degree between the classification results and the security permission standards. If the matching degree reaches the preset matching degree standard, the final evaluation state is confirmed.

[0175] The revised version is: "Based on the final assessment status and a pre-set assessment report template, automatically fill in the reduction range of hazardous gas accumulation areas, peak concentration values, percentage of areas exceeding concentration limits, and safe entry determination results to generate a safe entry permit assessment report."

[0176] The security access permission assessment report is stored in a preset database, and the real-time status data of the initial space model is updated synchronously to complete the entire assessment process.

[0177] To further describe the method of step S7) of this application, a specific embodiment is described below. Those skilled in the art will understand that this embodiment may not include all possible situations covered by the method of this application and does not constitute a limitation on the method of this application.

[0178] Specifically, by comparing the optimized ventilation configuration with the initial space model, the extent to which the area where harmful gases accumulate is determined, and a final safety access permit assessment report is generated. This can be implemented in the following ways.

[0179] First, the initial spatial model is digitized using 3D modeling software. Assuming the volume of the harmful gas accumulation area in the initial spatial model is 500 cubic meters, the proportion of the area where the harmful gas concentration exceeds the safe gas concentration threshold is recorded as 30% through data collection by gas concentration sensors. The safe gas concentration threshold is set to 0.5%.

[0180] Subsequently, based on the optimized ventilation configuration, a computational fluid dynamics algorithm was used to simulate the ventilation effect. The ventilation speed was set to 2.5 m / s, and the number of ventilation openings was increased to 5. After running the simulation, the volume of the area where harmful gas accumulates was reduced to 300 cubic meters, and the proportion of the area with excessive concentration was reduced to 15%. By comparing the data before and after, the reduction was calculated to be (500-300) / 500×100%=40%.

[0181] Next, combining the reduction degree and concentration distribution changes, a risk assessment algorithm is used to compare the reduction degree of the harmful gas accumulation area with the preset safety permit threshold. Assuming the preset safety permit threshold is 20%, the risk reduction index is 0.6, where the risk reduction index = 1 - current exceedance percentage / initial exceedance percentage.

[0182] Based on the risk reduction index, a safe entry permit assessment report is generated. The report automatically marks the risk level as "moderate to low" and includes a three-dimensional visualization chart of the hazardous gas accumulation area and a heat map of the hazardous gas concentration distribution.

[0183] Finally, the security access permit assessment report is correlated with historical data. By comparing the risk change trends of similar scenarios in the past three months through database comparison, a security access permit assessment logic chain is formed, and the assessment results are used as the basis for personnel to enter the restricted space.

[0184] Embodiments of this application also provide an intelligent risk management and control system for on-site operations in coal-fired power plants, the system comprising:

[0185] The spatial modeling module is used to acquire confined space geometry data through a pre-established 3D scanning method, and combine it with initial gas concentration distribution data collected by on-site sensors to construct an initial spatial model that includes boundary conditions and grid distribution.

[0186] The airflow field simulation module is used to import real-time monitored wind speed and wind direction data based on the initial spatial model, and to perform parameterization processing of ventilation conditions using finite element analysis to obtain ventilation parameters.

[0187] The accumulation area prediction module is used to extract the dominant airflow vector data from the ventilation parameters. If the dominant airflow vector data is lower than the preset airflow threshold, it is marked as a low ventilation area. The particle tracking algorithm is used to simulate the diffusion trajectory of harmful gases after they are released from the source, determine the location of potential harmful gas accumulation areas, and generate a harmful gas concentration distribution map.

[0188] The accumulation region analysis module is used to obtain the concentration distribution map of harmful gases in the simulation results of harmful gas diffusion, and to perform cluster analysis on high concentration nodes to obtain the boundary contour and peak concentration value of the harmful gas accumulation region.

[0189] The risk location module is used to generate a risk heat map by superimposing the boundary contour of the harmful gas accumulation area with the initial spatial model. If the peak concentration value exceeds the safe gas concentration threshold, the heat map is generated.

[0190] The ventilation optimization module is used to extract high-risk coordinates from the risk heat map, and to perform iterative optimization to adjust and simulate ventilation conditions, thereby obtaining an optimized harmful gas diffusion mode and a ventilation configuration scheme to reduce accumulation.

[0191] The safety assessment module is used to compare the optimized ventilation configuration with the initial space model to determine the extent to which the area where harmful gases accumulate is reduced, and to obtain the final safety access permit assessment report.

[0192] This application also provides an electronic device, which includes one or more processors and memory resources represented by a memory for storing instructions executable by the processor, such as application programs. The application programs stored in the memory may include one or more modules, each corresponding to a set of instructions. Furthermore, the processor is configured to execute instructions to perform the aforementioned intelligent risk management method for on-site operations in coal-fired power plants.

[0193] The aforementioned electronic device may also include a power supply component configured to perform power management of the electronic device, a wired or wireless network interface configured to connect the electronic device to a network, and an input / output (I / O) interface. The electronic device can be operated based on operating devices stored in memory, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or similar.

[0194] In one embodiment, a computer device, which may be a server, is also provided. The computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device stores data. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for intelligent risk management and control of on-site operations in a coal-fired power plant.

[0195] In one embodiment, a computer device is provided, which may be a terminal. The computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for intelligent risk management and control of on-site operations in coal-fired power plants. The display unit of the computer device is used to form a visually visible image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0196] This application also provides a non-transitory computer-readable storage medium, which, when the instructions in the storage medium are executed by the processor of the aforementioned electronic device, enables the aforementioned electronic device to execute a method for intelligent management and control of on-site operational risks in coal-fired power plants.

[0197] This application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information may be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to: phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0198] It should be noted that, in this document, the terms "comprising," "including," and any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Specific examples have been used in this document to illustrate the principles and implementation methods of the technical solutions of this application. The above examples are only for the purpose of helping to understand the methods and core ideas of this application. The above descriptions are merely preferred embodiments of this application. It should be pointed out that, due to the limitations of written expression and the objective existence of infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes, or combinations, or the direct application of the concept and technical solutions of this application to other situations without modification, should all be considered within the scope of protection of this application.

Claims

1. A method for intelligent risk management and control of on-site operations in coal-fired power plants, characterized in that, include: S1) Obtain the geometric structure data of the confined space through a pre-established three-dimensional scanning method, and combine it with the initial gas concentration distribution data collected by the field sensors to construct an initial spatial model that includes boundary conditions and grid distribution; S2) Import the real-time monitored wind speed and wind direction data according to the initial spatial model, and use finite element analysis to parameterize the ventilation conditions to obtain ventilation parameters; S3) Extract the dominant airflow vector data from the ventilation parameters. If the dominant airflow vector data is lower than the preset airflow threshold, it is marked as a low ventilation area. The diffusion trajectory of harmful gas after being released from the source is simulated by the particle tracking algorithm to determine the location of potential harmful gas accumulation areas and generate a harmful gas concentration distribution map. S4) Obtain the concentration distribution map of harmful gases in the simulation results of harmful gas diffusion, perform cluster analysis on high concentration nodes, and obtain the boundary contour and peak concentration value of the harmful gas accumulation area; S5) Based on the boundary contour of the harmful gas accumulation area and the initial spatial model, if the peak concentration value exceeds the safe gas concentration threshold, a risk heat map is generated. S6) Extract high-risk coordinates from the risk heat map, and use iterative optimization to adjust and simulate ventilation conditions to obtain an optimized harmful gas diffusion mode and a ventilation configuration scheme to reduce accumulation; S7) By comparing the optimized ventilation configuration with the initial space model, the degree of reduction in the area where harmful gases accumulate is determined, and the final safe entry permit assessment report is obtained.

2. The intelligent risk management method for on-site operations in coal-fired power plants according to claim 1, characterized in that, Step S1) specifically includes: A comprehensive data collection was conducted in the confined space using a pre-established 3D scanning method. The collected geometric data was then formatted to obtain a preliminary spatial geometric dataset. Based on the preliminary spatial geometry dataset, the restricted space is structurally segmented using a grid partitioning method to generate a refined spatial framework that includes boundary conditions and grid distribution. By using on-site sensor equipment, the initial distribution data of gas concentration in the confined space is obtained. The collected concentration data is standardized to determine a unified concentration distribution matrix. If there are missing data or outliers in the concentration distribution matrix, the missing areas are filled in by interpolation to obtain a complete concentration distribution dataset. Based on the refined spatial framework and the complete concentration distribution dataset, data fusion processing is performed to spatially map the geometric structure with gas concentration information, generating an initial spatial model with concentration distribution. Using the initial spatial model with concentration distribution, the finite element analysis method is used to conduct a preliminary simulation of the gas diffusion trend in the confined space and determine the distribution of potential high-concentration areas. Based on the high-concentration region distribution obtained from the simulation, the initial spatial model with concentration distribution is subjected to local mesh refinement processing for subsequent in-depth analysis.

3. The intelligent risk management method for on-site operations in coal-fired power plants according to claim 2, characterized in that, Step S2) specifically includes: By using a real-time monitoring system, wind speed and direction data are acquired from environmental sensors to construct an initial airflow input dataset. After data preprocessing, standardized input parameters are obtained. Standardized input parameters are imported into a pre-established initial spatial model, and the airflow distribution under different ventilation scenarios is simulated using the finite element analysis method to calculate the corresponding fluid dynamic parameters. Based on the fluid dynamics parameters, the gas flow path under different ventilation scenarios is analyzed, the key nodes of gas flow path changes are extracted, and the main influencing factors of gas flow path changes are determined. To address the main influencing factors of gas flow path changes, the simulation boundary conditions of the ventilation scenario were adjusted, the dynamic changes in airflow distribution were recalculated, and updated fluid dynamics data were obtained. By comparing the differences in gas flow paths under different ventilation scenarios using updated fluid dynamics data, the specific effects of adjusting ventilation parameters are determined. If the differences in gas flow paths exceed the preset threshold, a secondary parametric simulation of the ventilation scenario is performed to iteratively optimize the gas flow distribution model and obtain the final gas flow path impact analysis results.

4. The intelligent risk management method for on-site operations in coal-fired power plants according to claim 3, characterized in that, Step S3) specifically includes: By extracting the dominant airflow vector data from the ventilation parameters, and comparing the velocity magnitude in the dominant airflow vector data with a pre-established airflow vector analysis model, if the dominant airflow vector data is lower than a preset airflow threshold, the area is marked as a low ventilation area. Data on the source release location of harmful gases in low-ventilation areas are obtained, and particle tracking algorithms are used to simulate the diffusion path of harmful gases to determine the distribution range of the diffusion trajectory of harmful gases. Based on the distribution range of the harmful gas diffusion trajectory, the residence time of the harmful gas in each grid cell through which the harmful gas diffusion trajectory passes is obtained, the cumulative amount of harmful gas in each grid cell is calculated, and the specific location of the potential harmful gas accumulation area is determined. From the specific location of potential hazardous gas accumulation areas, extract grid cells with high accumulation of hazardous gases, analyze the changes in hazardous gas concentration between adjacent grid cells, and determine the hazardous gas concentration distribution map; Based on the distribution map of harmful gas concentrations, the overlapping area between high-concentration areas and low-ventilation areas is obtained. The changes in airflow vectors in the overlapping area are analyzed to determine whether there is a risk of continuous accumulation. Based on the assessment results of continuous risk accumulation, the locations of high-risk areas are extracted. Combined with real-time updated data of ventilation parameters, the changes in diffusion paths after the adjustment of airflow vectors are simulated to obtain the final location of the harmful gas accumulation area. For the final location of the hazardous gas accumulation area, the interaction between the hazardous gas concentration gradient and the airflow vector is analyzed to determine the priority ranking of high-risk areas and complete the risk assessment output.

5. The intelligent risk management method for on-site operations in coal-fired power plants according to claim 4, characterized in that, Step S4) specifically includes: The concentration distribution map of harmful gases is extracted from the simulation results of harmful gas diffusion and preliminarily cleaned to remove outliers, resulting in a standardized dataset of harmful gas concentration distribution. For a standardized dataset of hazardous gas concentration distribution, if the hazardous gas concentration is higher than a preset hazardous gas concentration threshold, it is selected as a node with high hazardous gas concentration. For the nodes with high concentrations of harmful gases, the K-means clustering method is used to group adjacent nodes with high concentrations of harmful gases, and the preliminary division results of multiple harmful gas accumulation regions are obtained. Based on the preliminary division of harmful gas accumulation areas, the spatial distribution characteristics of the concentration values ​​within each harmful gas accumulation area are calculated, and the boundary contour of each harmful gas accumulation area is obtained. By sorting the concentration values ​​of each harmful gas accumulation region, the highest concentration value in each harmful gas accumulation region is extracted to obtain the corresponding peak concentration value. Based on the boundary contour data and peak concentration values, an overlay analysis is performed according to the initial spatial model to generate a comprehensive mapping map showing the concentration distribution and the accumulation area of ​​harmful gases. If there is overlapping boundary contours in the comprehensive mapping map, the concentration values ​​of the overlapping areas are weighted and averaged, and the boundary contours are redefined to obtain the final distribution results of the harmful gas accumulation areas.

6. The intelligent risk management method for on-site operations in coal-fired power plants according to claim 5, characterized in that, Step S5) specifically includes: By superimposing the boundary contour of the harmful gas accumulation region with the initial spatial model, the distribution characteristics of the overlapping region are obtained. Based on the distribution characteristics of the overlapping area, the concentration values ​​in the overlapping area are collected in real time to obtain the peak concentration value. If the detected peak concentration value exceeds the preset safe gas concentration threshold, the risk assessment mechanism is triggered to generate the corresponding risk heat map. By using risk heat maps, high-risk areas can be identified, and priority intervention points can be identified. For the locked work site, acquire the surrounding environmental data of the work site in the initial spatial model to determine the boundary contour of the intervention range; By using a pre-established decision tree model, combined with the risk heat map and boundary contour of the locked operation points, specific intervention sequences and resource allocation plans are generated. By adjusting the intervention sequence and resource allocation plan, the risk assessment data in the initial spatial model is updated, and key areas for subsequent monitoring are identified.

7. The intelligent risk management method for on-site operations in coal-fired power plants according to claim 6, characterized in that, Step S6) specifically includes: The distribution of high-risk coordinates can be obtained through risk heatmaps; Based on the location of high-risk coordinates, the initial diffusion state of harmful gases is simulated using preset ventilation scenario parameters to obtain preliminary diffusion pattern data. If the accumulation risk is found to be higher than the preset accumulation risk threshold based on the preliminary diffusion pattern data, the ventilation scenario parameters are adjusted through an iterative method to obtain updated diffusion simulation results, and the distribution changes of harmful gas diffusion are extracted from them to determine whether the accumulation risk has been reduced. If the risk of accumulation is reduced, the optimized ventilation scenario parameters are used, and the output of the simulation analysis is combined to generate the final configuration scheme to reduce the risk. Obtain simulation results data of the final configuration scheme, determine whether it meets the preset risk accumulation control standards, and output a complete solution.

8. The intelligent risk management method for on-site operations in coal-fired power plants according to claim 7, characterized in that, Step S7) specifically includes: By obtaining the optimized ventilation configuration scheme and the initial space model, preliminary data format unification processing is performed to obtain the first standardized dataset. Based on the standardized first dataset, a pre-set comparative analysis module is used to compare the distribution range of the harmful gas accumulation area layer by layer to determine the specific range of reduction of the harmful gas accumulation area. If the range value of the reduction of the harmful gas accumulation area meets the preset safety permit standard threshold, it is determined that the safe entry conditions are met, and the first judgment result is generated. Based on the first determination result, the relevant gas accumulation situation is obtained, and the accumulation situation is classified by the support vector machine algorithm to obtain the classified second dataset; Using the second dataset after classification, a second verification is performed on the matching degree between the classification results and the security permission standards. If the matching degree reaches the preset matching degree standard, the final evaluation state is confirmed. Based on the final status of the assessment, and in conjunction with the preset assessment report template, the relevant data fields are automatically filled in to generate the final conclusion of secure access permission; The final conclusions are stored in a preset database, and the real-time status data of the initial spatial model is updated synchronously to complete the entire evaluation process.

9. A smart risk management and control system for on-site operations in a coal-fired power plant, characterized in that, The system is used to implement the intelligent risk management method for on-site operations in coal-fired power plants according to any one of claims 1-8, and the system includes: The spatial modeling module is used to acquire confined space geometry data through a pre-established 3D scanning method, and combine it with initial gas concentration distribution data collected by on-site sensors to construct an initial spatial model that includes boundary conditions and grid distribution. The airflow field simulation module is used to import real-time monitored wind speed and wind direction data based on the initial spatial model, and to perform parameterization processing of ventilation conditions using finite element analysis to obtain ventilation parameters. The accumulation area prediction module is used to extract the dominant airflow vector data from the ventilation parameters. If the dominant airflow vector data is lower than the preset airflow threshold, it is marked as a low ventilation area. The particle tracking algorithm is used to simulate the diffusion trajectory of harmful gases after they are released from the source, determine the location of potential harmful gas accumulation areas, and generate a harmful gas concentration distribution map. The accumulation region analysis module is used to obtain the concentration distribution map of harmful gases in the simulation results of harmful gas diffusion, and to perform cluster analysis on high concentration nodes to obtain the boundary contour and peak concentration value of the harmful gas accumulation region. The risk location module is used to generate a risk heat map by superimposing the boundary contour of the harmful gas accumulation area with the initial spatial model. If the peak concentration value exceeds the safe gas concentration threshold, the heat map is generated. The ventilation optimization module is used to extract high-risk coordinates from the risk heat map, and to perform iterative optimization to adjust and simulate ventilation conditions, thereby obtaining an optimized harmful gas diffusion mode and a ventilation configuration scheme to reduce accumulation. The safety assessment module is used to compare the optimized ventilation configuration with the initial space model to determine the extent to which the area where harmful gases accumulate is reduced, and to obtain the final safety access permit assessment report.

Images