Power plant inspection method and system based on intelligent positioning of personnel safety
Patent Information
- Application Number
- CN202510948702.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-10
- Publication Date
- 2025-10-28
Smart Images
Figure CN120851698A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power plant management technology, specifically relating to a power plant inspection method and system based on intelligent positioning for personnel safety. Background Art
[0002] In power plant inspection work, ensuring personnel safety and achieving efficient and accurate personnel task matching are crucial aspects, but current inspection technologies have significant shortcomings in these areas.
[0003] The arrangement of personnel for inspections often adopts a fixed shift or random assignment method. This method does not fully consider the risk differences of each monitoring area in each inspection route, nor does it conduct a comprehensive assessment and reasonable matching of the skills, experience and ability of the inspection personnel, which exposes the inspection personnel to potential dangers.
[0004] In terms of safety matching, the lack of accurate assessment of the work efficiency and inspection quality of inspection personnel leads to unreasonable task allocation; inspection personnel with low work efficiency may be assigned to areas with heavy workloads and high risks, which makes it impossible for them to complete comprehensive and detailed inspection work within a limited time, and may overlook some potential safety hazards; to address this, we propose a power plant inspection method and system based on intelligent positioning of personnel safety. Summary of the Invention
[0005] The purpose of this invention is to provide a power plant inspection method and system based on intelligent positioning for personnel safety, so as to solve the problems mentioned in the background art.
[0006] To achieve the above objectives, the present invention provides the following technical solution: a power plant inspection method based on intelligent positioning for personnel safety, comprising the following steps:
[0007] Step S1: Divide the power plant into inspection areas and then divide the inspection areas into monitoring areas. Construct an equipment anomaly record database. Based on this, analyze the average fault isolation, total single repair capacity consumption, and equipment hidden danger value of the equipment in the monitoring area. And obtain the risk assessment value through comprehensive analysis. Sort the monitoring areas according to the size of the risk assessment value to obtain the risk ranking sequence of the monitoring areas.
[0008] Step S2: Construct a skills database for inspection personnel; quantify and score the work experience, work efficiency, and inspection quality of inspection personnel; analyze the inspection evaluation value based on this; match the task load equivalent value of inspection personnel according to the inspection evaluation value; quantify the task load of the monitoring area; and assign tasks to inspection personnel by combining the risk ranking sequence of the monitoring area, the task load of the monitoring area, the professional skills of the inspection personnel, the inspection evaluation value, and the task load equivalent value.
[0009] Step S3: Obtain the real-time coordinates of the inspection personnel, and combine them with the risk assessment values of each inspection area in the inspection task, the geographical center coordinates of the monitoring area, and the boundary coordinates of the area to plan the inspection route. If the real-time positioning coordinates of the inspection personnel deviate from the planned inspection route by more than the corresponding threshold, an early warning will be triggered.
[0010] Preferably, in step S1, the specific steps for dividing the power plant into inspection areas, dividing the inspection areas into monitoring areas, constructing an equipment anomaly record database, and analyzing fault isolation are as follows:
[0011] Step S101: In the power plant, the power plant area is divided into regions based on the operating functions of each piece of equipment. Equipment that performs the same or related functions is grouped into the same region to obtain the inspection area.
[0012] For each inspection area, the inspection area is divided into several monitoring areas based on the individual electrical equipment within the inspection area;
[0013] Step S102: For each monitoring area, construct an equipment anomaly record library, which covers all anomalies that may occur in the corresponding equipment in that monitoring area;
[0014] When equipment malfunctions within the monitoring area, the anomaly log database will record the time of the malfunction, the repair time, the associated equipment, the number of times the equipment malfunctioned, as well as the equipment's service life and maintenance records.
[0015] Step S103: Obtain the time points of all equipment failures in the anomaly record database, calculate the time interval between adjacent time points to obtain the failure occurrence interval, and calculate the average of all failure occurrence intervals to obtain the average failure interval GZ.
[0016] Preferably, in step S1, the specific steps for analyzing the total consumption of single-repair capacity are as follows:
[0017] Step S104: Obtain the total production capacity generated by the equipment during normal operation within the past month from the equipment operation log within the monitoring area, and at the same time obtain the total operating time of the equipment within the past month. Divide the total production capacity by the total operating time to obtain the hourly production capacity.
[0018] Based on the anomaly log database, the total repair time of each equipment failure is summarized, and the total number of failures is counted. The average repair time per unit is obtained by dividing the total number of failures by the total repair time.
[0019] The lost production capacity of the equipment is obtained by multiplying the hourly production capacity corresponding to the equipment failure in the monitoring area by the average repair time.
[0020] For each fault-related device, the difference in hourly capacity before and after the fault occurs is calculated to obtain the associated impact capacity; by multiplying the associated impact capacity by the average repair time, the single repair impact capacity is obtained.
[0021] The total impact of single-repair on the production capacity of all fault-related equipment is obtained by summing up the impact of single-repair on the production capacity.
[0022] The total capacity loss of equipment is added to the total capacity loss associated with a single repair, GH, to obtain the total capacity loss of a single repair.
[0023] Preferably, in step S1, the specific steps for analyzing equipment hazard values and risk assessment values, and ranking the monitoring areas based on the risk assessment values to obtain a risk ranking sequence for the monitoring areas are as follows:
[0024] Step S105: Obtain the service life and maintenance records of the equipment in the monitoring area from the anomaly record database, and obtain the remaining service life by subtracting the service life of the equipment from the service life.
[0025] Multiple life expectancy intervals are preset, and each life expectancy interval corresponds to a life expectancy risk coefficient. By matching the life expectancy with all life expectancy intervals, the corresponding life expectancy risk coefficient YF is output.
[0026] Obtain the time interval between each maintenance within the equipment's service life to get the maintenance interval; obtain the maintenance cycle time interval marked in the equipment manual to get the maintenance set interval; and obtain the maintenance deviation interval by subtracting the maintenance set interval from the maintenance interval and taking the absolute value.
[0027] A preset safety deviation interval is set, and the maintenance deviation interval is compared with the preset safety deviation interval. If it is greater than the preset safety deviation interval, the maintenance is marked as substandard. The number of substandard maintenance is counted to obtain the substandard maintenance value.
[0028] Multiple maintenance failure value ranges are preset, and each range corresponds to a maintenance risk coefficient. By matching the maintenance failure value with all maintenance failure value ranges, the corresponding maintenance risk coefficient BF is output.
[0029] Using the formula: YH=YF×a1+BF×a2, the equipment hazard value YH is obtained, where a1 and a2 are preset weighting coefficients;
[0030] Step S106: After normalizing the corresponding fault isolation GZ, total single-repair capacity consumption GH, and equipment hidden danger value YH within the monitoring area, the formula is used: The FPZ rating is obtained, where w1, w2, and w3 are preset weighting coefficients;
[0031] For each inspection area, the risk ranking sequence of the monitoring areas is obtained by sorting them from largest to smallest according to the risk assessment value of each monitoring area within the inspection area.
[0032] Preferably, in step S2, the specific steps for constructing the inspection personnel skill base and quantifying the inspection personnel's work experience are as follows:
[0033] Step S201: Construct a skills database for inspection personnel. The skills database records the personal information, professional skills, work experience rating, work efficiency rating, and inspection quality rating of the inspection personnel. Personal information includes: name, ID number, and mobile phone number.
[0034] In the inspection personnel skill library, for each skill of the inspection personnel, a skill certificate level-rating correspondence table, years of service-rating correspondence table, task efficiency index-rating correspondence table, and timely response rate-rating correspondence table are established;
[0035] Step S202: For each inspector, match the inspector's electrician level with the corresponding skill certificate level-scoring correspondence table and output the skill score; at the same time, match the inspector's length of service with the corresponding years of service-scoring correspondence table and output the length of service score.
[0036] Different weighting coefficients are assigned to the skill score and seniority score of the inspection personnel. Then, the skill score and seniority score are multiplied by their respective weighting coefficients and added together to obtain the work experience score YF.
[0037] Preferably, in step S2, the specific steps for quantifying work efficiency and inspection quality are as follows:
[0038] Step S203: For each inspector, record the task assignment duration and corresponding inspection completion duration for each inspection task within the past month;
[0039] For each inspection task, if the inspection completion time is less than or equal to the task's scheduled time, it is recorded as a completed task; otherwise, it is recorded as an incomplete task.
[0040] For completed tasks, the task completion rate is obtained by dividing the task scheduling time by the inspection completion time; for incomplete tasks, the task overtime rate is obtained by dividing the inspection completion time by the task scheduling time.
[0041] The total number of tasks completed by the inspection personnel (WN) and the total number of tasks not completed (BN) in the past month are recorded. The average completion rate of all completed tasks is calculated to obtain the average completion rate (WV). The average timeout rate of all incomplete tasks is calculated to obtain the average timeout rate (BV).
[0042] After normalizing the total number of completed tasks WN, the total number of incomplete tasks BN, the average completion rate WV, and the average timeout rate BV, the task efficiency index ZS is obtained using the formula: ZS=WN×b1+WV×b2-BN×b3-BV×b4.
[0043] The work efficiency score XF is obtained by matching the task efficiency index with the task efficiency index-score correspondence table.
[0044] Step S204: For each piece of equipment that needs to be inspected, record the interval between the actual time of the equipment failure and the time when the inspector discovers and records it. This interval is called the failure response time. A failure handling time limit is preset. If the failure response time is less than the failure handling time limit, the failure is marked as a timely response failure.
[0045] Based on the inspection reports of the inspection personnel, the total number of equipment failures and the number of failures marked as timely responses were counted during the inspection process in the past month; the timely response rate was obtained by dividing the number of timely response failures by the total number of equipment failures.
[0046] The timely response rate of the inspection personnel in the most recent month is matched with the timely response rate-score correspondence table to obtain the inspection quality score ZF.
[0047] Preferably, in step S2, the inspection and evaluation values are analyzed, and the task load equivalent value of the inspection personnel is matched according to the inspection and evaluation values; the specific steps for quantifying the task load of the monitoring area are as follows:
[0048] Step S205: For each inspector, substitute the inspector's work experience score YF, work efficiency score XF, and inspection quality score ZF into the preset formula model: XPZ=YF×c1+XF×c2+ZF×c3 to obtain the inspection evaluation value XPZ, where c1, c2, and c3 are preset weight coefficients.
[0049] Step S206: For each inspection area, establish a monitoring area task quantity value matching library. Each monitoring area in the monitoring area task quantity value matching library has a corresponding task quantity value.
[0050] Substitute the monitoring areas that need to be inspected into the monitoring area task volume scoring database for matching, and output the corresponding task volume value.
[0051] Several inspection value ranges are preset, and each inspection value range corresponds to a task load equivalent value. The inspection value of the inspector is matched with all inspection value ranges, and the corresponding task load equivalent value is output.
[0052] Preferably, in step S2, the specific steps for assigning tasks to inspection personnel, based on the risk ranking sequence of the monitoring area, the workload of the monitoring area, the professional skills of the inspection personnel, the inspection evaluation value, and the task load equivalent value, are as follows:
[0053] Step S207: For each inspection area, based on the professional skills of the inspectors in the inspection personnel skill library, obtain the inspection personnel corresponding to the operation function of the inspection area and mark them as initial screening personnel; from the initial screening personnel, select the inspection personnel to be on duty today and record the assigned personnel.
[0054] For each assigned personnel, calculate the corresponding inspection score, and sort them from largest to smallest according to the inspection score to obtain the priority assignment sequence for assigned personnel;
[0055] The risk ranking sequence of the monitoring area corresponding to the inspection area is matched and assigned with the priority allocation sequence of the assigned personnel. The specific process is as follows:
[0056] Extract the monitoring area with the highest current priority from the risk ranking sequence, obtain its workload value, and start from the first position in the allocation priority ranking sequence, checking the allocation personnel in turn:
[0057] If the task load value of the monitoring area is less than or equal to the remaining task load equivalent value of the assigned personnel, the monitoring area will be assigned to the inspection task of the matched personnel; after each assignment is completed, the remaining task load equivalent value of all assigned personnel will be updated in real time.
[0058] If the workload of the monitoring area is greater than the remaining workload of the assigned personnel, skip the assigned personnel and continue to check the next person in the priority allocation sequence. Repeat the above steps until all monitoring areas in the current inspection area have been allocated, or the personnel in the priority allocation sequence are unable to take on the remaining monitoring area tasks.
[0059] If there are unassigned monitoring areas, patrol personnel will be drawn from those unscheduled personnel to conduct patrols.
[0060] Preferably, the specific implementation process of step S3 is as follows:
[0061] For each inspection personnel, their real-time coordinates within the power plant are obtained through UWB positioning base stations or BeiDou positioning terminals deployed at the power plant. At the same time, the wind assessment values, geographic center coordinates, and boundary coordinates of each monitoring area corresponding to the task assigned to the inspection personnel are obtained from the power plant database.
[0062] The initial power plant coordinates obtained by the inspection personnel are combined with the wind assessment value of the monitoring area, the geographic center coordinates, and the regional boundary coordinates and input into the GIS map system. The GIS map system adopts the Dijkstra geospatial path planning algorithm, using the geographic center coordinates of each monitoring area as path nodes.
[0063] Based on the risk assessment values of the monitored areas, the path nodes corresponding to the monitored areas with the highest risk assessment values are arranged at the beginning of the path planning sequence in descending order.
[0064] If there are monitoring areas with the same risk assessment value, the path nodes corresponding to the monitoring areas with the closest distance will be prioritized according to the straight-line distance between their geographical center coordinates and the initial coordinates of the inspection personnel.
[0065] Based on path traversability as a constraint and with the shortest total journey as the core optimization objective, the optimal connection order between each node is calculated using a path planning algorithm based on the sorted path nodes, generating an inspection route that covers all assigned monitoring areas.
[0066] Patrol personnel conduct patrols of the monitored area according to the patrol route. If the real-time positioning coordinates of the patrol personnel deviate from the planned patrol route by more than a preset distance threshold, an early warning will be triggered, and the patrol personnel will be instructed to continue patrolling according to the planned patrol route.
[0067] Compared with the prior art, the present invention has the following beneficial effects:
[0068] (1) The power plant inspection method and system based on intelligent positioning for personnel safety, by constructing an equipment anomaly record database, records equipment fault information in detail, and deeply analyzes fault isolation, total single-repair capacity consumption and equipment hidden danger value; these quantitative indicators enable the power plant to accurately grasp the equipment operating status and risks, discover potential problems in advance, arrange maintenance in a timely manner, and reduce production interruptions caused by sudden faults; the risk assessment value obtained by comprehensively analyzing these indicators fully reflects the risk level of the monitoring area, and the risk ranking of the monitoring area based on the risk assessment value provides a scientific basis for the subsequent reasonable allocation of inspection tasks, which helps to ensure the safe and stable operation of power plant equipment.
[0069] (2) This power plant inspection method and system based on intelligent positioning for personnel safety constructs a skills database for inspection personnel and uses multiple scoring tables to quantify the work experience, work efficiency, and inspection quality of inspection personnel, measuring the actual ability of each inspection personnel. It matches the task load score according to the inspection suitability and quantifies the task load score of the monitoring area. Combined with the risk ranking of the monitoring area, it accurately matches the risk level and workload of the task with the comprehensive ability of the inspection personnel. This not only gives full play to the advantages of each inspection personnel and improves the quality and efficiency of the inspection work, but also reduces inspection omissions or resource waste caused by unreasonable task allocation. It screens personnel based on their professional skills and then allocates tasks based on the ranking of inspection evaluation values, ensuring that each monitoring area can be assigned inspection personnel with corresponding professional skills and comprehensive ability, which helps to discover and deal with equipment failures in a timely manner and further ensure the safety of power plant equipment.
[0070] (3) This power plant inspection method and system based on intelligent positioning for personnel safety relies on UWB positioning base stations or Beidou positioning terminals to collect personnel coordinates in real time. Combined with information such as the risk assessment value of the monitoring area and geographical coordinates, it uses the Dijkstra algorithm to prioritize the planning of inspection routes in high-risk areas, so that inspection resources are tilted towards areas with high incidence of safety hazards, significantly improving the efficiency of potential risk investigation. On the other hand, by dividing the plant area into inspection sections and constructing a personnel number prediction model based on historical inspection data, it realizes real-time monitoring and prediction of road congestion. Once the congestion threshold is triggered, the route is dynamically adjusted according to the priority of the risk assessment value of the affected area, effectively avoiding efficiency loss caused by overlapping inspection routes. At the same time, the dual protection of real-time positioning and deviation warning mechanism not only ensures that the operation path of inspection personnel is standardized and uniform, but also enables rapid response in case of emergencies, comprehensively ensuring the safety, efficiency and orderliness of power plant inspection work. Attached Figure Description
[0071] Figure 1 This is a flowchart of the present invention. Detailed Implementation
[0072] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0073] Example 1
[0074] Please see Figure 1 This invention provides a power plant inspection method based on intelligent positioning for personnel safety, comprising the following steps:
[0075] Step S1: Divide the power plant into inspection areas, then divide the inspection areas into monitoring areas, and build an equipment anomaly record database. Based on this, analyze the average fault isolation, total single-repair capacity consumption, and equipment hidden danger value of the equipment in the monitoring area, and comprehensively analyze the three to obtain a risk assessment value. Based on the risk assessment value, rank the monitoring areas to obtain a risk ranking sequence of the monitoring areas. The specific process is as follows:
[0076] Step S101: In the power plant, the power plant area is divided into regions based on the operating functions of each piece of equipment. Equipment that performs the same or related functions is grouped into the same region to obtain the inspection area.
[0077] For each inspection area, the inspection area is divided into several monitoring areas based on the individual electrical equipment within the inspection area;
[0078] Step S102: For each monitoring area, construct an equipment anomaly record library. The equipment anomaly record library covers all the abnormal conditions that may occur in the corresponding equipment in the monitoring area, including but not limited to: electrical performance abnormalities, communication failures, and malfunctions of protection devices.
[0079] When equipment malfunctions within the monitoring area, the anomaly log database will record the time of the malfunction, the repair time, the associated equipment, the number of times the equipment malfunctioned, as well as the equipment's service life and maintenance records.
[0080] Step S103: Obtain the time points of all equipment failures in the anomaly record database, calculate the time interval between adjacent time points to obtain the failure occurrence interval, and calculate the average of all failure occurrence intervals to obtain the average failure interval GZ.
[0081] Step S104: Obtain the total production capacity generated by the equipment during normal operation within the past month from the equipment operation log within the monitoring area, and at the same time obtain the total operating time of the equipment within the past month. By dividing the total production capacity by the total operating time, the hourly production capacity is obtained; the hourly production capacity represents the production capacity of the equipment per unit time.
[0082] Based on the anomaly log database, the total repair time of each equipment failure is summarized, and the total number of failures is counted. The average repair time per unit is obtained by dividing the total number of failures by the total repair time.
[0083] The lost production capacity of the equipment is obtained by multiplying the hourly production capacity corresponding to the equipment failure in the monitoring area by the average repair time.
[0084] For each fault-related device, the difference in hourly capacity before and after the fault occurs is calculated. That is, the hourly capacity of the fault-related device is subtracted from the hourly capacity of the related device during the repair period to obtain the associated impact capacity. By multiplying the associated impact capacity by the average repair time, the single repair impact capacity is obtained. That is, the capacity loss of a single related device during a single fault repair.
[0085] The total impact of single-repair on the production capacity of all fault-related equipment is obtained by summing up the impact of single-repair on the production capacity.
[0086] By adding the lost capacity of the equipment to the total associated capacity of a single repair, the total capacity consumption of a single repair is obtained as GH. The total capacity consumption of a single repair covers the direct and indirect capacity loss caused by a single equipment failure, and comprehensively quantifies the degree of impact of the failure on the production system of the monitored area.
[0087] Step S105: Obtain the service life and maintenance records of the equipment in the monitoring area from the anomaly record database, and obtain the remaining service life by subtracting the service life of the equipment from the service life.
[0088] Multiple life expectancy intervals are preset, and each life expectancy interval corresponds to a life expectancy risk coefficient. By matching the life expectancy with all life expectancy intervals, the corresponding life expectancy risk coefficient YF is output.
[0089] Obtain the time interval between each maintenance within the equipment's service life to get the maintenance interval; obtain the maintenance cycle time interval marked in the equipment manual to get the maintenance set interval; and obtain the maintenance deviation interval by subtracting the maintenance set interval from the maintenance interval and taking the absolute value.
[0090] A preset safety deviation interval is set, and the maintenance deviation interval is compared with the preset safety deviation interval. If it is greater than the preset safety deviation interval, the maintenance is marked as substandard. The number of substandard maintenance is counted to obtain the substandard maintenance value.
[0091] Multiple maintenance failure value ranges are preset, and each range corresponds to a maintenance risk coefficient. By matching the maintenance failure value with all maintenance failure value ranges, the corresponding maintenance risk coefficient BF is output.
[0092] Using the formula: YH=YF×a1+BF×a2, the equipment hazard value YH is obtained, where a1 and a2 are preset weighting coefficients; the larger the equipment hazard value, the greater the probability of a safety accident occurring.
[0093] Step S106: After normalizing the corresponding fault isolation GZ, total single-repair capacity consumption GH, and equipment hidden danger value YH within the monitoring area, the formula is used: The risk rating value FPZ is obtained, where w1, w2, and w3 are preset weighting coefficients. The higher the risk rating value for the monitored area, the higher the risk level of the monitored area.
[0094] For each inspection area, the risk assessment values of the monitoring areas that need to be inspected within the inspection area are sorted from largest to smallest to obtain the risk ranking sequence of the monitoring areas.
[0095] It should be noted that by building an equipment anomaly log database and recording various equipment fault information, it is possible to conduct in-depth analysis of fault isolation, total capacity consumption per repair, and equipment hidden danger value. Fault isolation reflects the frequency of equipment failures, total capacity consumption per repair comprehensively quantifies the impact of a single failure on the production system's capacity loss, and equipment hidden danger value comprehensively considers the equipment's remaining life and maintenance status to assess potential safety risks. These quantitative indicators enable power plants to have a more accurate understanding of the equipment's operating status and risks, which helps to identify potential problems in advance, arrange maintenance in a timely manner, and reduce production interruptions caused by sudden failures.
[0096] By comprehensively analyzing fault isolation, total single-repair capacity consumption, and equipment hazard values, a risk assessment value is obtained, which can fully reflect the risk level of the monitored area. Based on the risk assessment value, the risk ranking sequence of the monitored area is used to allocate high-risk areas to experienced and highly skilled inspectors in the subsequent task allocation process, so as to make reasonable use of human resources.
[0097] Step S2: Construct a skills database for inspection personnel. Quantify and score the work experience, efficiency, and quality of inspection personnel. Analyze the evaluation value based on these scores. Simultaneously, analyze the task load score and monitoring area task volume score of the inspection personnel. Combining the risk ranking sequence of the monitoring area, the monitoring area task volume score, the professional skills of the inspection personnel, the evaluation value, and the task load score, assign tasks to the inspection personnel. The specific process is as follows:
[0098] Step S201: Construct a skills database for inspection personnel. The skills database records the personal information, professional skills, work experience rating, work efficiency rating, and inspection quality rating of the inspection personnel. Personal information includes, but is not limited to: name, ID number, and mobile phone number.
[0099] In the inspection personnel skill library, for each skill of the inspection personnel, a skill certificate level-rating correspondence table, years of service-rating correspondence table, task efficiency index-rating correspondence table, and timely response rate-rating correspondence table are established;
[0100] Step S202: For each inspector, match the inspector's electrician level with the corresponding skill certificate level-scoring correspondence table and output the skill score; at the same time, match the inspector's length of service with the corresponding years of service-scoring correspondence table and output the length of service score.
[0101] Different weighting coefficients are assigned to the skill score and seniority score of the inspection personnel. Then, the skill score and seniority score are multiplied by their respective weighting coefficients and added together to obtain the work experience score YF.
[0102] Step S203: For each inspector, record the task assignment duration and corresponding inspection completion duration for each inspection task within the past month;
[0103] For each inspection task, if the inspection completion time is less than or equal to the task's scheduled time, it is recorded as a completed task; otherwise, it is recorded as an incomplete task.
[0104] For completed tasks, the task completion rate is obtained by dividing the task scheduling time by the inspection completion time; for incomplete tasks, the task overtime rate is obtained by dividing the inspection completion time by the task scheduling time.
[0105] The total number of tasks completed by the inspection personnel (WN) and the total number of tasks not completed (BN) in the past month are recorded. The average completion rate of all completed tasks is calculated to obtain the average completion rate (WV). The average timeout rate of all incomplete tasks is calculated to obtain the average timeout rate (BV).
[0106] After normalizing the total number of completed tasks WN, the total number of incomplete tasks BN, the average completion rate WV, and the average timeout rate BV, the task efficiency index ZS is obtained using the formula: ZS=WN×b1+WV×b2-BN×b3-BV×b4; where b1, b2, b3, and b4 are preset weight coefficients.
[0107] The work efficiency score XF is obtained by matching the task efficiency index with the task efficiency index-score correspondence table.
[0108] Step S204: For each piece of equipment that needs to be inspected, record the interval between the actual time of the equipment failure and the time when the inspector discovers and records it. This interval is called the failure response time. A failure handling time limit is preset. If the failure response time is less than the failure handling time limit, the failure is marked as a timely response failure.
[0109] Based on the inspection reports of the inspection personnel, the total number of equipment failures and the number of failures marked as timely responses were counted during the inspection process in the past month; the timely response rate was obtained by dividing the number of timely response failures by the total number of equipment failures.
[0110] The timely response rate of the inspection personnel in the most recent month is matched with the timely response rate-score correspondence table to obtain the inspection quality score ZF;
[0111] Step S205: For each inspector, substitute their work experience score YF, work efficiency score XF, and inspection quality score ZF into the preset formula model: XPZ=YF×c1+XF×c2+ZF×c3 to obtain the inspection evaluation value XPZ, where c1, c2, and c3 are preset weight coefficients. The larger the inspection evaluation value, the higher the overall level of the inspector in performing the inspection task, and the more competent they are in performing the inspection task in monitoring areas with higher risk levels.
[0112] Step S206: For each monitoring area, obtain the corresponding risk assessment value and establish a risk assessment value frequency interval. Each risk assessment value frequency interval corresponds to a monitoring area inspection frequency. The larger the upper and lower bounds of the risk assessment value interval, the more inspections are required for that monitoring area. The corresponding inspection frequency is the number of inspections conducted on that area after each shift change when inspection personnel are reassigned. Match the risk assessment value of the monitoring area with all risk assessment value frequency intervals and output the corresponding inspection frequency XP.
[0113] The inspection duration of each inspection in the monitoring area over the past week was obtained from the power plant database. The average inspection duration was calculated to obtain the average inspection duration XT.
[0114] The specific items that need to be covered during the inspection of the monitoring area are obtained from the power plant database. The number of items that need to be inspected in the monitoring area is counted and recorded as the number of inspections XS.
[0115] After normalizing the inspection frequency XP, average inspection length XT, and number of inspections XS, the task quantity value RWP is obtained using the formula: RWP=XP×d1+XT×d2+XS×d3, where d1, d2, and d3 are preset weighting coefficients; the larger the task quantity value, the heavier the task quantity when inspecting the monitoring area.
[0116] Set multiple task volume value ranges, each task volume value range corresponds to a task volume score. The larger the task volume score, the heavier the task volume. Match the task volume value of the monitored area with all task volume value ranges and output the corresponding task volume score.
[0117] Step S207: For each inspection area, based on the professional skills of the inspectors in the inspection personnel skill library, obtain the inspection personnel corresponding to the operation function of the inspection area and mark them as initial screening personnel; from the initial screening personnel, select the inspection personnel to be on duty today and record the assigned personnel.
[0118] Obtain the risk assessment value FPZ and task quantity value RWP for each monitoring area within the risk ranking sequence of the inspection area;
[0119] For each assigned personnel, the patrol evaluation value XPZ corresponding to the assigned personnel, the risk evaluation value FPZ corresponding to each monitoring area, and the task quantity value RWP are normalized and then processed using the formula: The inspection fit degree SPD is obtained, where f1 and f2 are preset weight coefficients;
[0120] Several inspection suitability ranges are preset, and each inspection suitability range corresponds to a task capacity score. The inspection suitability of the inspector is matched with all inspection suitability ranges, and the corresponding task capacity score is output. The higher the inspection suitability, the larger the corresponding task capacity score, and the greater the inspection capacity of the inspector.
[0121] For each assigned personnel, calculate the corresponding inspection score, and sort them from largest to smallest according to the inspection score to obtain the priority assignment sequence for assigned personnel;
[0122] The risk ranking sequence of the monitoring area corresponding to the inspection area is matched and assigned with the priority allocation sequence of the assigned personnel. The specific process is as follows:
[0123] Extract the monitoring area with the highest current priority from the risk ranking sequence, where the higher the ranking, the higher the priority. Obtain its workload score, starting from the first person in the allocation priority ranking sequence, and check the allocation personnel in turn.
[0124] If the task load score of the monitored area is less than or equal to the remaining task load score of the assigned personnel, the monitored area will be assigned to the inspection task of the matched personnel. After each assignment is completed, the remaining task load scores of all assigned personnel will be updated in real time. In the initial state, the total task load of each person in the queue is zero.
[0125] If the task load score of the monitoring area is greater than the remaining task load score of the assigned personnel, then skip the assigned personnel and continue to check the next personnel in the priority allocation sequence of the assigned personnel; repeat the above steps until all monitoring areas in the current inspection area have been assigned, or the personnel in the priority allocation sequence of the assigned personnel are unable to take on the remaining monitoring area tasks.
[0126] If there are unassigned monitoring areas, patrol personnel will be drawn from those unscheduled personnel to conduct patrols.
[0127] It should be noted that by constructing a skills database for inspection personnel, recording personnel information in detail, and using various scoring tables to quantify work experience, work efficiency, and inspection quality, it is easier to measure the actual ability of each inspection personnel, changing the ambiguity of previous subjective judgment in assessing personnel capabilities, and providing a scientific basis for the subsequent reasonable allocation of tasks.
[0128] The task load score is matched based on the inspection suitability, and the task load score of the monitoring area is quantified. Then, the risk ranking of the monitoring area is combined. This can accurately match the risk level and workload of the task with the comprehensive ability of the inspection personnel. It is convenient to give full play to the strengths of each inspection personnel, improve the quality and efficiency of the inspection work, and reduce inspection omissions or waste of resources caused by unreasonable task allocation.
[0129] Personnel are initially screened based on their professional skills, and then tasks are assigned based on the ranking of inspection values. This ensures that each monitoring area is assigned to an inspector with the corresponding professional skills and comprehensive capabilities. This helps inspectors to better identify potential equipment problems, handle faults in a timely manner, reduce equipment failure rates, and ensure the safe and stable operation of power plant equipment. Moreover, during the task allocation process, the task load score is updated in real time, which can dynamically adjust the task allocation, making the entire inspection work more scientific and reasonable, and further improving the quality and safety of the inspection work.
[0130] Step S3: Obtain the real-time coordinates of the inspection personnel, and combine them with the risk assessment values of each inspection area in the inspection task, the geographic center coordinates of the monitoring area, and the boundary coordinates of the area to plan the inspection route. The real-time coordinates of the inspection personnel are then located. If the real-time location deviates from the planned inspection route by more than a preset distance threshold, an early warning is triggered. The specific process is as follows:
[0131] For each inspection personnel, their real-time coordinates within the power plant are obtained through UWB positioning base stations or BeiDou positioning terminals deployed at the power plant. At the same time, the risk assessment values, monitoring frequency, geographic center coordinates, and boundary coordinates of each monitoring area corresponding to the assigned task of the inspection personnel are obtained from the power plant database.
[0132] The initial power plant coordinates obtained by the inspection personnel are combined with the risk assessment value of the monitoring area, the geographic center coordinates, and the regional boundary coordinates and input into the GIS map system. The GIS map system adopts the Dijkstra geospatial path planning algorithm, using the geographic center coordinates of each monitoring area as path nodes.
[0133] The monitoring areas are sorted from high to low according to the frequency of monitoring and patrol, and the round variable j is set with an initial value of 1;
[0134] First round of inspection: Select all monitoring areas and sort them from high to low according to their risk assessment values. Place the path node corresponding to the monitoring area with the highest risk assessment value at the front of the path planning sequence. If there are monitoring areas with the same risk assessment value, prioritize the path node corresponding to the monitoring area with the closest straight-line distance from its geographical center coordinates to the initial coordinates of the inspectors.
[0135] With path traversability as a constraint and the shortest total distance as the core optimization objective, the optimal connection sequence between nodes is calculated using a path planning algorithm to generate the first round of inspection routes; and the following measures are taken:
[0136] The access roads within the plant area are divided into several inspection sections. Based on the inspection data of the inspection sections at different times (by hour, weekday, weekend, etc.) over the past three months, and with the help of a linear regression algorithm, a personnel number prediction model is constructed. The inspection data includes: the travel time and stay time of inspection personnel in each section, as well as the personnel flow in each section.
[0137] The real-time location coordinates of all inspection personnel are monitored by UWB positioning base stations or Beidou positioning terminals, and a fixed time interval is set for statistical period. At the beginning of each statistical period, the actual number of personnel at that moment is obtained. Combined with the personnel number prediction model, the predicted number of personnel in the inspection section at each moment within the statistical period is obtained.
[0138] For each patrol section, if the actual number of people or the predicted number of people in the patrol section is greater than or equal to the corresponding threshold at any time within the current statistical period, the patrol section is judged to be a congested section.
[0139] If the inspected section is a congested section, obtain the set of inspectors affected by the congestion section, and denot it as the congestion personnel set; for each inspector in the congestion personnel set, obtain the magnitude of the risk assessment value corresponding to the monitoring area to be inspected by the inspector through the section, sort them from smallest to largest, and obtain the priority ranking of the personnel who changed routes; obtain the maximum number of personnel with a number greater than the threshold of the number of people in the congested section within the statistical period, and obtain the number of people who changed routes;
[0140] Based on the number of people whose routes have changed, select the corresponding number of people from the priority list of those whose routes have changed to modify the routes.
[0141] For inspection personnel who need to change their routes, the inspection route is replanned, starting from the current location and using the geographic center coordinates of the remaining uninspected area as the target node. During the replanning process, constraints are added to the original Dijkstra algorithm, and the optimal connection order between nodes is regenerated to obtain the new planned route.
[0142] The specific constraints are: the new planned route cannot pass through road sections that have been identified as congested, and the total distance is minimized based on route accessibility;
[0143] Subsequent rounds of inspection: After each round of inspection is completed, j is incremented by 1. From the remaining monitoring areas, monitoring areas with a monitoring frequency less than or equal to j are removed. For example, in the second round of inspection, monitoring areas that were inspected only once are removed; in the third round of inspection, monitoring areas that were inspected only twice are removed. Then, for the remaining monitoring areas, the steps of planning the path in the first round of inspection are repeated to generate the inspection route for the corresponding round. The inspection personnel continue to inspect according to the newly generated route until all monitoring areas have been inspected.
[0144] During the inspection process, if the real-time positioning coordinates of the inspection personnel deviate from the planned inspection route by more than a preset distance threshold, an early warning will be triggered to remind the inspection personnel to carry out the inspection according to the planned inspection route.
[0145] It should be noted that, on the one hand, by relying on UWB positioning base stations or Beidou positioning terminals to collect personnel coordinates in real time, and combining this with information such as the risk assessment value and geographical coordinates of the monitoring area, the Dijkstra algorithm is used to prioritize the planning of inspection routes in high-risk areas, thus tilting inspection resources towards areas with high incidence of safety hazards and significantly improving the efficiency of potential risk investigation. On the other hand, by dividing the plant area into inspection sections and building a personnel number prediction model based on historical inspection data, real-time monitoring and prediction of road congestion can be achieved. Once the congestion threshold is triggered, the route is dynamically adjusted according to the priority of the risk assessment value of the affected area, effectively avoiding efficiency losses caused by overlapping inspection routes. At the same time, the dual protection of real-time positioning and deviation warning mechanism ensures that the work paths of inspection personnel are standardized and uniform, and can respond quickly in case of emergencies, comprehensively ensuring the safety, efficiency and orderliness of power plant inspection work.
[0146] A power plant inspection system based on intelligent positioning for personnel safety includes:
[0147] Regional Risk Assessment Module: The power plant is divided into inspection areas and monitoring areas. An equipment anomaly record database is built. Based on this, the average fault isolation, total single-repair capacity consumption, and equipment hidden danger value of the equipment in the monitoring area are analyzed. The risk assessment value is obtained through comprehensive analysis. The monitoring areas are ranked according to the size of the risk assessment value to obtain the risk ranking sequence of the monitoring areas.
[0148] Personnel task allocation module: Constructs a skills library for inspection personnel to score and quantify their work experience, efficiency, and inspection quality. Based on the scores of various abilities, it analyzes the inspection evaluation value. At the same time, it analyzes the task load score of the inspection personnel and the task volume score of the monitoring area. Combining the risk ranking sequence of the monitoring area, the task volume score of the monitoring area, the professional skills of the inspection personnel, the inspection evaluation value, and the task load score, it assigns tasks to the inspection personnel.
[0149] Path planning and early warning module: Obtain the real-time coordinates of the inspection personnel, and combine them with the risk assessment value of each inspection area in the inspection task, the geographical center coordinates of the monitoring area, and the boundary coordinates of the area to plan the inspection path. If the real-time positioning coordinates of the inspection personnel deviate from the planned inspection route by more than the corresponding threshold, an early warning will be triggered.
[0150] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.
Claims
1. A power plant inspection method based on intelligent positioning for personnel safety, characterized in that, Includes the following steps: Step S1: Divide the power plant into inspection areas and then divide the inspection areas into monitoring areas. Construct an equipment anomaly record database. Based on this, analyze the average fault isolation, total single repair capacity consumption, and equipment hidden danger value of the equipment in the monitoring area. And obtain the risk assessment value through comprehensive analysis. Sort the monitoring areas according to the size of the risk assessment value to obtain the risk ranking sequence of the monitoring areas. Step S2: Construct a skills database for inspection personnel. Quantify and score the work experience, efficiency, and quality of inspection personnel. Analyze the evaluation value based on the scores of each ability. Simultaneously, analyze the task load score and the task volume score of the monitoring area for each inspection personnel. Tasks are assigned to inspection personnel by combining the risk ranking sequence of the monitoring area, the task load score of the monitoring area, the professional skills of the inspection personnel, the inspection evaluation value, and the task load score. Step S3: Obtain the real-time coordinates of the inspection personnel, and combine them with the risk assessment values of each inspection area in the inspection task, the geographical center coordinates of the monitoring area, and the boundary coordinates of the area to plan the inspection route. If the real-time positioning coordinates of the inspection personnel deviate from the planned inspection route by more than the corresponding threshold, an early warning will be triggered.
2. The power plant inspection method based on intelligent positioning for personnel safety according to claim 1, characterized in that: In step S1, the specific steps for dividing the power plant into inspection areas, dividing the inspection areas into monitoring areas, constructing an equipment anomaly record database, and analyzing the fault isolation and total single-repair capacity consumption are as follows: Step S101: In the power plant, the power plant area is divided into regions based on the operating functions of each piece of equipment. Equipment that performs the same or related functions is grouped into the same region to obtain the inspection area. For each inspection area, the inspection area is divided into several monitoring areas based on the individual electrical equipment within the inspection area; Step S102: For each monitoring area, construct an equipment anomaly record library, which covers all anomalies that may occur in the corresponding equipment in that monitoring area; When equipment malfunctions within the monitoring area, the anomaly log database will record the time of the malfunction, the repair time, the associated equipment, the number of times the equipment malfunctioned, as well as the equipment's service life and maintenance records. Step S103: Obtain the time points of all equipment failures, calculate the time interval between adjacent time points to obtain the failure interval, and calculate the average of all failure intervals to obtain the average failure interval GZ. Step S104: Obtain the total production capacity generated by the equipment during normal operation in the past month, and at the same time obtain the total operating time of the equipment in the past month. Divide the total production capacity by the total operating time to obtain the hourly production capacity. Summarize the total repair time for all equipment failures and count the total number of failures. Divide the total number of failures by the total repair time to obtain the average repair time per failure. The lost production capacity of the equipment is obtained by multiplying the hourly production capacity corresponding to the equipment failure in the monitoring area by the average repair time. For each fault-related device, the difference in hourly capacity before and after the fault occurs is calculated to obtain the associated impact capacity; by multiplying the associated impact capacity by the average repair time, the single repair impact capacity is obtained. The total impact of single-repair on the production capacity of all fault-related equipment is obtained by summing up the impact of single-repair on the production capacity. The total capacity loss of equipment is added to the total capacity affected by single repair, and the total capacity loss of single repair is obtained as GH.
3. The power plant inspection method based on intelligent positioning for personnel safety according to claim 2, characterized in that: In step S1, the specific steps for analyzing equipment hazard values and risk assessment values, and ranking the monitoring areas based on the risk assessment values to obtain the risk ranking sequence of the monitoring areas are as follows: Step S105: Obtain the service life and maintenance records of the equipment in the monitoring area from the anomaly record database, and obtain the remaining service life by subtracting the service life of the equipment from the service life. Multiple life expectancy intervals are preset, and each life expectancy interval corresponds to a life expectancy risk coefficient. By matching the life expectancy with all life expectancy intervals, the corresponding life expectancy risk coefficient YF is output. The time interval between each maintenance within the equipment's service life is recorded as the maintenance interval. The time interval between the maintenance cycles marked in the equipment manual is recorded as the maintenance setting interval. The maintenance deviation interval is obtained by subtracting the maintenance setting interval from the maintenance interval and taking the absolute value. The maintenance deviation interval is compared with the preset safety deviation interval. If it is greater than the preset safety deviation interval, the maintenance is marked as substandard. The number of substandard maintenance is counted to obtain the substandard maintenance value. Multiple maintenance failure value ranges are preset, and each range corresponds to a maintenance risk coefficient. By matching the maintenance failure value with all maintenance failure value ranges, the corresponding maintenance risk coefficient BF is output. The equipment hazard value YH is obtained by using the formula: YH=YF×a1+BF×a2, where a1 and a2 are preset weighting coefficients; Step S106: After normalizing the corresponding fault isolation GZ, total single-repair capacity consumption GH, and equipment hidden danger value YH within the monitoring area, the formula is used: The FPZ rating is obtained, where w1, w2, and w3 are preset weighting coefficients; For each inspection area, the risk ranking sequence of the monitoring areas is obtained by sorting them from largest to smallest according to the risk assessment value of each monitoring area within the inspection area.
4. The power plant inspection method based on intelligent positioning for personnel safety according to claim 3, characterized in that: In step S2, the specific steps for building a skills database for inspection personnel and quantifying their work experience are as follows: Step S201: Construct a skills library for inspection personnel, which records the personal information, professional skills, work experience rating, work efficiency rating, and inspection quality rating of the inspection personnel. In the inspection personnel skill library, for each skill of the inspection personnel, a skill certificate level-rating correspondence table, years of service-rating correspondence table, task efficiency index-rating correspondence table, and timely response rate-rating correspondence table are established; Step S202: For each inspector, match the inspector's electrician level with the corresponding skill certificate level-scoring correspondence table and output the skill score; at the same time, match the inspector's length of service with the corresponding years of service-scoring correspondence table and output the length of service score. Different weighting coefficients are assigned to the skill score and seniority score of the inspection personnel. Then, the skill score and seniority score are multiplied by their respective weighting coefficients and added together to obtain the work experience score YF.
5. The power plant inspection method based on intelligent positioning for personnel safety according to claim 4, characterized in that: In step S2, the specific steps for quantifying work efficiency and inspection quality are as follows: Step S203: For each inspector, record the task assignment duration and corresponding inspection completion duration for each inspection task within the past month; For each inspection task, if the inspection completion time is less than or equal to the task's scheduled time, it is recorded as a completed task; otherwise, it is recorded as an incomplete task. For completed tasks, the task completion rate is obtained by dividing the task scheduling time by the inspection completion time; for incomplete tasks, the task overtime rate is obtained by dividing the inspection completion time by the task scheduling time. The total number of tasks completed by the inspection personnel (WN) and the total number of tasks not completed (BN) in the past month are recorded. The average completion rate of all completed tasks is calculated to obtain the average completion rate (WV). The average timeout rate of all incomplete tasks is calculated to obtain the average timeout rate (BV). After normalizing the total number of completed tasks WN, the total number of incomplete tasks BN, the average completion rate WV, and the average timeout rate BV, the task efficiency index ZS is obtained using the formula: ZS=WN×b1+WV×b2-BN×b3-BV×b4; where b1, b2, b3, and b4 are preset weight coefficients. The work efficiency score XF is obtained by matching the task efficiency index with the task efficiency index-score correspondence table. Step S204: For each piece of equipment that needs to be inspected, record the interval between the actual time of the equipment failure and the time when the inspector discovers and records it. This interval is called the failure response time. A failure handling time limit is preset. If the failure response time is less than the failure handling time limit, the failure is marked as a timely response failure. Based on the inspection reports of the inspection personnel, the total number of equipment failures that occurred during the inspection process in the past month and the number of failures marked as timely responses were calculated. The timely response rate is calculated by dividing the number of timely response failures by the total number of equipment failures. The timely response rate of the inspection personnel in the most recent month is matched with the timely response rate-score correspondence table to obtain the inspection quality score ZF.
6. The power plant inspection method based on intelligent positioning for personnel safety according to claim 5, characterized in that: In step S2, the specific steps for simultaneously analyzing the inspection personnel's task load score and the monitoring area's task volume score are as follows: Step S206: For each monitoring area, obtain the corresponding risk assessment value, establish a risk assessment value frequency interval, and each risk assessment value frequency interval corresponds to a monitoring area inspection frequency; match the risk assessment value of the monitoring area with all risk assessment value frequency intervals, and output the corresponding inspection frequency; The inspection duration of each inspection in the monitoring area over the past week was obtained from the power plant database. The average inspection duration of all inspections was calculated to obtain the average inspection duration. The specific items that need to be covered during the inspection of the monitoring area are obtained from the power plant database. The number of items that need to be inspected in the monitoring area is counted and recorded as the number of inspections. By comprehensively analyzing the frequency of inspections, the average length of each inspection, and the number of inspections, the task quantity value (RWP) is obtained. Set multiple task volume value ranges, with each task volume value range corresponding to a task volume score; match the task volume value of the monitored area with all task volume value ranges, and output the corresponding task volume score; Step S207: For each inspection area, based on the professional skills of the inspectors in the inspection personnel skill library, obtain the inspection personnel corresponding to the operation function of the inspection area, and mark them as the initial screening personnel; From the initial screening, select the patrol personnel for today's duty and assign them to personnel; Obtain the risk assessment value FPZ and task quantity value RWP for each monitoring area within the risk ranking sequence of the inspection area; For each assigned personnel, the patrol evaluation value XPZ corresponding to the assigned personnel, the risk evaluation value FPZ corresponding to each monitoring area, and the task quantity value RWP are normalized and then processed using the formula: The inspection fit degree SPD is obtained, where f1 and f2 are preset weight coefficients; Several inspection fit intervals are preset, and each inspection fit interval corresponds to a task capacity score. The inspection personnel's inspection fit is matched with all inspection fit intervals, and the corresponding task capacity score is output. For each assigned personnel, calculate the corresponding inspection value and sort them from largest to smallest according to the inspection value to obtain the priority assignment sequence of the assigned personnel.
7. The power plant inspection method based on intelligent positioning for personnel safety according to claim 6, characterized in that: In step S2, the specific steps for assigning tasks to inspection personnel are as follows: Extract the monitoring area with the highest current priority from the risk ranking sequence; obtain its workload score, starting from the first position in the allocation priority ranking sequence, and check the allocation personnel in turn: If the task load score of the monitored area is less than or equal to the remaining task load score of the assigned personnel, then the monitored area will be assigned to the inspection task of the matched personnel. After each assignment is completed, the remaining task load rating of all assignees is updated in real time. If the task load score of the monitoring area is greater than the remaining task load score of the assigned personnel, then skip the assigned personnel and continue to check the next personnel in the priority allocation sequence of the assigned personnel; repeat the above steps until all monitoring areas in the current inspection area have been assigned, or the personnel in the priority allocation sequence of the assigned personnel are unable to take on the remaining monitoring area tasks. If there are unassigned monitoring areas, patrol personnel will be drawn from those unscheduled personnel to conduct patrols.
8. The power plant inspection method based on intelligent positioning for personnel safety according to claim 7, characterized in that: The specific implementation process of step S3 is as follows: For each inspection personnel, obtain their real-time coordinates in the power plant; the risk assessment value, inspection frequency, geographic center coordinates, and boundary coordinates of each monitoring area corresponding to the assigned tasks of the inspection personnel; The initial power plant coordinates obtained by the inspection personnel, combined with the risk assessment value of the monitoring area, the geographic center coordinates, and the regional boundary coordinates, are completely input into the GIS map system; the geographic center coordinates of each monitoring area are used as path nodes; The monitoring areas are sorted from high to low according to the frequency of monitoring and patrol, and the round variable j is set with an initial value of 1; First round of inspection: Select all monitoring areas and sort them from high to low according to their risk assessment values. Place the path node corresponding to the monitoring area with the highest risk assessment value at the front of the path planning sequence. If there are monitoring areas with the same risk assessment value, prioritize the path node corresponding to the monitoring area with the closest straight-line distance from its geographical center coordinates to the initial coordinates of the inspectors. The first round of inspection routes is generated with path traversability as a constraint and the shortest total distance as the optimization objective. And the following measures will be taken: The access roads within the plant area are divided into several inspection sections. Based on the inspection data of the inspection sections in the power plant's history at various time periods, and with the help of a linear regression algorithm, a personnel number prediction model is constructed. The inspection data includes: the travel time and dwell time of inspection personnel in each road section, as well as the flow of people in each road section; Obtain the real-time location coordinates of all inspection personnel, and at the beginning of each statistical period, obtain the actual number of personnel at that moment; combine the personnel number prediction model to obtain the predicted number of personnel for each inspection section at each moment within the statistical period. Determine whether the inspected section of road is congested based on the actual or predicted number of personnel.
9. The power plant inspection method based on intelligent positioning for personnel safety according to claim 8, characterized in that: The specific implementation details for congested sections are as follows: If it is a congested road section, obtain the set of inspection personnel affected by the congested road section, and denot it as the congested personnel set; for each inspection personnel in the congested personnel set, obtain the magnitude of the risk assessment value corresponding to the monitoring area that the inspection personnel is about to inspect through the road section, sort them from smallest to largest, and obtain the priority ranking of the personnel who changed routes; obtain the maximum number of personnel greater than the threshold of the number of people in the congested road section within the statistical period, and obtain the number of people who changed routes. Based on the number of people whose routes have changed, select the corresponding number of people from the priority list of those whose routes have changed to modify the routes. For inspection personnel who need to change their routes, the inspection route is replanned with the current location as the starting point and the geographic center coordinates of the remaining uninspected area as the target node. During the replanning, constraints are added to the original Dijkstra algorithm, and the optimal connection order between each node is regenerated to obtain the new planned route. The specific constraints are: the new planned route cannot pass through road sections that have been identified as congested, and the total distance is minimized based on route accessibility; Subsequent rounds of inspection: After each round of inspection is completed, j is incremented by 1. From the remaining monitoring areas, the monitoring areas with a monitoring frequency less than or equal to j are removed. For the remaining monitoring areas, the steps of planning the path in the first round of inspection are repeated to generate the inspection route for the corresponding round. During the inspection, the decision to issue an early warning is based on the distance along the inspection route.
10. A power plant inspection system based on intelligent positioning for personnel safety, applied to the power plant inspection method based on intelligent positioning for personnel safety proposed in any one of claims 1-9, characterized in that: include: Regional Risk Assessment Module: The power plant is divided into inspection areas and monitoring areas. An equipment anomaly record database is built. Based on this, the average fault isolation, total single-repair capacity consumption, and equipment hidden danger value of the equipment in the monitoring area are analyzed. The risk assessment value is obtained through comprehensive analysis. The monitoring areas are ranked according to the size of the risk assessment value to obtain the risk ranking sequence of the monitoring areas. Personnel task allocation module: Constructs a skills library for inspection personnel, scores and quantifies their work experience, efficiency, and inspection quality, and analyzes the inspection evaluation value based on the scores of various abilities; at the same time, it analyzes the task load score of inspection personnel and the task volume score of the monitoring area. Tasks are assigned to inspection personnel by combining the risk ranking sequence of the monitoring area, the task load score of the monitoring area, the professional skills of the inspection personnel, the inspection evaluation value, and the task load score. Path planning and early warning module: Obtain the real-time coordinates of the inspection personnel, and combine them with the risk assessment value of each inspection area in the inspection task, the geographical center coordinates of the monitoring area, and the boundary coordinates of the area to plan the inspection path. If the real-time positioning coordinates of the inspection personnel deviate from the planned inspection route by more than the corresponding threshold, an early warning will be triggered.