Dynamic early warning method and system for personnel safety in distribution network line operations

Abstract

The present invention provides a dynamic safety warning method and system for personnel in distribution network line operations, which relates to the field of smart grid management technology, including: calling an operation work ticket, which includes work tasks, operator information, safety measures requirements, grounding wire installation information, and control instructions; performing text semantic structured preprocessing to extract three-dimensional safety features; extracting safety statements and establishing a structured task template; activating the target area when the location information of the target operator is identified to be consistent with the area range of the target area; establishing a time-series work data set based on intelligent perception of grounding wires and intelligent perception of operating tools; performing safety verification and generating dynamic safety warnings. The present invention solves the technical problem that the existing distribution network line operations lack intelligent information extraction and task verification mechanisms, resulting in inaccurate safety assessments of the operation tasks, thereby reducing the response speed and accuracy of safety warnings.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grid management, and in particular to a dynamic early warning method and system for personnel safety in distribution network line operations. Background Art

[0002] In the power distribution network system, line operations involve multiple links such as maintenance, repair, and installation. Operators need to work in a high-voltage and dangerous environment. With the continuous development of the power system, the scale and complexity of distribution network line operations are also increasing. How to ensure the safety of operators has become a key issue.

[0003] Currently, most distribution network line operations rely on traditional manual safety management and operation processes. In many cases, operators need to obtain operation instructions through paper work tickets and manually perform safety measures. There is a lack of intelligent information extraction and task verification mechanisms, and they fail to deeply explore the safety features of the tasks, resulting in inaccurate safety assessments of the operation tasks, which may in turn miss certain key safety hazards and reduce the response speed and accuracy of the safety warning system. Summary of the Invention

[0004] This application provides a dynamic warning method and system for personnel safety in distribution network line operations, aiming to solve the technical problem that the existing distribution network line operations lack intelligent information extraction and task verification mechanisms, resulting in inaccurate safety assessments of operation tasks, thereby reducing the response speed and accuracy of safety warnings.

[0005] The first aspect disclosed in the present application provides a dynamic early warning method for personnel safety in distribution network line operations, the method comprising: calling an operation work ticket, the operation work ticket including work tasks, operator information, safety measures requirements, grounding wire installation information, and control instructions; performing text semantic structured preprocessing of the operation work ticket to extract three-dimensional safety features, the three-dimensional safety features including control center execution features, operator execution safety features, and grounding wire installation and removal task features; utilizing an electric power NLP model to extract safety statements of the three-dimensional safety features and establish a structured task template; configuring a target area to be activated according to the operation work ticket, and activating the target area when the positioning information of the target operator is identified to be consistent with the area range of the target area; after the target area is activated, establishing a time series work data set based on intelligent perception of the grounding wire and intelligent perception of the operation tool; utilizing the structured task template to perform safety verification of the time series work data set to generate a dynamic early warning for safety.

[0006] The second aspect disclosed in the present application provides a dynamic early warning system for personnel safety in distribution network line operations, the system being used for the above-mentioned dynamic early warning method for personnel safety in distribution network line operations, the system comprising: a work ticket calling module for calling a work ticket, the work ticket containing work tasks, operator information, safety measures requirements, grounding wire installation information, and control instructions; a structured preprocessing module for performing text semantic structured preprocessing of the work ticket and extracting three-dimensional safety features, the three-dimensional safety features including control center execution features, operator execution safety features, and grounding wire installation and removal task features; a safety statement extraction module for extracting safety statements from the three-dimensional safety features using an electric power NLP model and establishing a structured task template; a target area activation module for configuring a target area to a pending activation state based on the work ticket, and activating the target area when the location information of the target operator is identified to be consistent with the area range of the target area; a work data set establishment module for establishing a time-series work data set based on intelligent grounding wire perception and intelligent work tool perception after the target area is activated; and a safety verification module for performing safety verification of the time-series work data set using the structured task template to generate a dynamic early warning for safety.

[0007] One or more technical solutions provided in this application have at least the following beneficial effects:

[0008] By calling the work ticket, key information such as work tasks, operator information, safety measures requirements, etc. can be accurately obtained, and combined with grounding wire installation information, control instructions, etc., to ensure that operators have sufficient safety measures when performing tasks, which helps to avoid safety accidents caused by human negligence or information omissions; by performing text semantic structured preprocessing of the work ticket, three-dimensional safety features are extracted, including the control center execution features in the work task, the operator execution safety features, the grounding wire installation and removal task features, etc. The extraction and structured processing of these features significantly improves the readability and safety verification capabilities of the work task information; the power NLP model is used to extract safety statements from the extracted three-dimensional safety features, thereby establishing a structured task template. This process allows the entire work task to be automatically evaluated for safety through intelligent means, reducing errors and omissions in human intervention; when the operator's positioning information When the information meets the scope of the designated area, the target area is activated, which means that the working area can be dynamically managed in real time according to the location of the personnel. This precise area division and real-time location monitoring effectively reduce the risk of personnel entering dangerous areas during the operation; after the target area is activated, a time-series working data set is established based on the intelligent perception of the grounding wire and the intelligent perception of the working tool. By real-time monitoring of the status of the working tools and the grounding wire, a timely response can be made to potential safety hazards in the operation process. The construction of this time-series data set provides a more accurate basis for dynamic monitoring during the operation process; using structured task templates to perform safety verification on the time-series working data set can automatically detect potential safety hazards in the operation process and generate dynamic safety warnings, which can issue safety warnings in time during the operation to avoid accidents caused by safety hazards not being discovered in time. The automated warning mechanism improves the safety of the operating personnel.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flowchart of a dynamic early warning method for personnel safety in distribution network line operations provided in an embodiment of the present application.

[0011] Figure 2 A schematic diagram of the structure of a dynamic warning system for personnel safety in distribution network line operations provided in an embodiment of the present application.

[0012] Explanation of the reference numerals: work ticket calling module 10, structured preprocessing module 20, security statement extraction module 30, target area activation module 40, work data set establishment module 50, security verification module. DETAILED DESCRIPTION

[0013] The embodiments of the present application provide a dynamic personnel safety warning method and system for distribution network line operations, thereby solving the technical problem that the existing distribution network line operations lack intelligent information extraction and task verification mechanisms, resulting in inaccurate safety assessments of operation tasks, and thus reducing the response speed and accuracy of safety warnings.

[0014] After introducing the basic principles of this application, various non-limiting embodiments of this application will be specifically described below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0015] Example 1, as Figure 1 As shown, an embodiment of the present application provides a dynamic early warning method for personnel safety in distribution network line operations, the method comprising:

[0016] The operation work ticket is called, and the operation work ticket includes work tasks, operator information, safety measures requirements, grounding wire installation information, and control instructions.

[0017] Obtain a work ticket from a database or job management system. A work ticket is a document containing the work task, operator information, safety measures, grounding wire installation information, and control instructions. The work task describes the specific objectives and operational steps of the work, such as line inspection and equipment maintenance. Operator information includes the operator's identity, qualifications, and job position. Safety measures include safety rules and precautions to be followed during the work. Grounding wire installation information includes the installation and removal methods of the grounding wire, as well as whether the equipment is grounded. Control instructions are related to power system control operations, such as disconnecting a line and adjusting the load. Work ticket data may exist in various formats, such as PDF, paper documents, and Word files, and this information needs to be extracted and structured.

[0018] The text semantic structured preprocessing of the operation work ticket is performed to extract three-dimensional safety features, which include control center execution features, operator execution safety features, and ground wire installation and removal task features.

[0019] Using semantic analysis and entity recognition technology in natural language processing, semantic analysis is performed on the text extracted from the work ticket, and unstructured data (such as natural language descriptions) is converted into structured data (such as actionable features). Three-dimensional safety features are extracted from it. The three-dimensional safety features are represented in a structured data format, such as vector representation, where each feature corresponds to a dimension. These features will serve as the basis for subsequent safety assessments to determine whether there are safety hazards and provide data support for the dynamic early warning system.

[0020] The extracted three-dimensional safety features include control center execution features, operator execution safety features, and grounding wire installation and removal task features. Among them, the control center execution features are execution features related to the control center, including dispatching instructions, load adjustment, system monitoring, etc. It reflects the role and tasks undertaken by the control center during the operation process, affecting the safety of operators and the execution of operation tasks; operator execution safety features include operator safety execution requirements, training records, task performance, etc. For example, the safety behavior characteristics of operators, such as whether they wear protective equipment and whether they operate in accordance with regulations, all of which affect the safety of the operation process; grounding wire installation and removal task features include the correct installation of grounding wires, monitoring of the removal process, grounding resistance measurement, etc. Grounding is a key link to ensure the safety of operators. Incorrect grounding operations may lead to serious electrical accidents.

[0021] The electric power NLP model is used to extract safety statements with three-dimensional safety features and establish a structured task template.

[0022] The power NLP model is a pre-trained language model based on deep learning. It is used to process the work ticket text containing work tasks, safety measures, grounding wire information, etc. Through the model, it can identify safety-related statements, operating requirements, and execution specifications in the text.

[0023] Safety statement extraction is to extract key safety-related statements or commands from three-dimensional safety features, such as "all workers must wear safety helmets" and "the grounding wire must be confirmed to be installed before operation" in the operational safety requirements, and "electrical equipment must be powered off before maintenance" in the risk warning. Through the named entity recognition technology of the NLP model, safety statement information in the operation process is extracted and converted into a structured format. Based on the extracted safety statements, a structured task template is created. The template contains detailed steps for task execution, safety requirements for each step, equipment requirements, personnel requirements, etc. The structured task template is a graphical structure based on task nodes. Each node represents an operation task or operation step, and the relationship between nodes describes the operation process.

[0024] The target area is configured as a waiting-for-activation state according to the operation work ticket, and the target area is activated when the positioning information of the target operator is identified to be consistent with the area range of the target area.

[0025] The target area is a spatial range that includes specific work locations, facilities, and equipment. It can be a two-dimensional plane area, such as a section of a distribution network line, or a three-dimensional area, such as the equipment area of ​​a substation. In the work ticket, the scope of the target area is determined according to the requirements of the work task. Based on this information, the target area is configured to be activated, which means that the target area is ready to receive subsequent activation instructions. During the operation process, the location information of the operator is tracked in real time. The operator is usually equipped with positioning equipment such as GPS, indoor positioning systems, RFID, etc., which is used to collect the location information of the operator in real time. When the location information of the target operator is identified as being within the scope of the target area, the target area is activated. After activation, the target area enters the real-time monitoring state, and all the operator's operational behaviors will be recorded and analyzed.

[0026] After the target area is activated, a time-series working data set is established based on the intelligent perception of the grounding wire and the intelligent perception of the working tool.

[0027] Grounding cables are crucial for ensuring worker safety. Therefore, their installation, removal, and location require real-time monitoring. Intelligent grounding cable sensing includes location monitoring, height monitoring, conductor identification monitoring, grounding depth monitoring, and connection loop resistance calculation. This data is collected in real time and transmitted to the system, forming part of the time-series work data set.

[0028] The usage status of work tools is crucial to ensuring operational safety. Intelligent perception of work tools includes tool identification and positioning, tool status monitoring, and tool usage monitoring. Tool identification and positioning uses sensors to identify the work tool in use and record the location and time of use. Tool status monitoring monitors the tool's status in real time, such as whether it is intact, operating normally, or has any faults. Tool usage monitoring records the specific time and method of tool use to ensure that workers use tools in accordance with regulations. Intelligent perception of work tools provides dynamic data on work tools, which constitutes another part of the time-series work dataset and is used for subsequent analysis and early warning.

[0029] The structured task template is used to perform security verification of the time series work data set and generate a dynamic security warning.

[0030] The structured task template is matched with the sequential work dataset. The structured task template provides the standardized process and safety requirements for the task, while the sequential work dataset reflects the real-time execution status of the task. By comparing each task node in the structured task template with the real-time data in the sequential work dataset, task execution is verified. This includes: task node verification, such as whether tasks are executed in sequence, whether grounding wires are correctly installed, and whether work tools are functioning properly; and safety requirement verification, checking whether workers comply with safety regulations, such as whether they wear protective equipment and perform necessary inspections.

[0031] If any behavior that does not meet safety requirements is found during the verification process, such as the operator not following safety procedures, improper use of tools, or non-standard installation of grounding wires, a dynamic safety warning will be triggered. In this case, a warning message will be sent to the operation management personnel or dispatch center. Based on the level and type of the warning, relevant personnel will be reminded to take appropriate response measures, such as stopping operations, adjusting operation processes, or conducting on-site inspections, to provide timely safety protection for the operators.

[0032] Furthermore, the use of the structured task template to perform security verification of the time series work data set and generate a dynamic security warning includes:

[0033] A directed task graph is constructed based on the structured task template, and state transition edges are set; the time-series work data set is converted into a time-annotated graph, and a behavior path graph is constructed; task node pairing analysis is performed based on the directed task graph and the behavior path graph; a multimodal perception score is established using the task node pairing analysis results, and a dynamic safety warning is generated using the multimodal perception score.

[0034] A directed task graph is a graph structure consisting of task nodes and directed edges between connecting nodes. Each node represents a task step, and the edges represent the dependencies between task steps. When constructing a directed task graph, first determine the various task steps in the structured task template, such as cutting off the power supply, installing the grounding wire, checking the grounding resistance, and restoring the power supply. Each task step will be placed in the task graph as a node, and the dependencies between task nodes (such as some tasks must be executed after other tasks are completed) are represented by directed edges.

[0035] In a directed task graph, in addition to describing the sequential relationship between tasks, it is also necessary to set a state transition edge for each task node. The state transition edge reflects the possible state changes during the task execution process. The definition of the state transition edge is usually based on the execution status of the job task, such as task in progress, task completed, task failed, and task paused. The execution state change of each task can trigger different state transitions. For example, after the "cut off the power" task is completed, the task status changes from in progress to completed. At this time, the edge in the directed graph points to the next task "installing the grounding wire". If the new task "installing the grounding wire" fails, the state transition points to the task failure state, indicating that the task needs to be re-executed or fault handling needs to be performed.

[0036] The execution flow of a directed task graph represents the sequence from one task to the next, as well as the status updates after task completion. This structured approach can clearly describe the execution process of job tasks, facilitating real-time monitoring and management.

[0037] A time-annotated graph is a graphical structure that combines the execution of job tasks with a timeline. In this graph, task nodes are not just static task steps, but nodes with time information. Each node represents the execution status of a task and is accompanied by a time annotation, indicating the start time, end time, and duration of the task. The time-annotated graph can help determine whether tasks are executed on schedule and whether the time intervals between tasks meet the requirements.

[0038] The behavior path graph is built based on the time-annotated graph and is used to display each behavior path in the operation process. In the behavior path graph, nodes represent key behaviors in the operation, such as cutting off the power supply, installing the grounding wire, and checking the equipment. Edges represent the path relationship between these behaviors. For example, a certain behavior must be performed after another behavior, and these paths contain the time information of the behavior execution.

[0039] The role of the behavior path diagram is to show the overall picture of the execution of the job task, as well as the time sequence, dependencies and time differences of each task. Through the behavior path diagram, the task execution process can be monitored globally to ensure that the job process proceeds according to the predetermined plan.

[0040] Task node pairing refers to pairing each task node in the directed task graph with the behavior node in the behavior path graph, and performing temporal, spatial and semantic condition matching to ensure that the task execution sequence and behavior path in the operation process meet the predetermined requirements. The specific methods include: checking whether the time between the task node and the behavior node is consistent. For example, the "cut off the power" task should be completed within a specific time period, and the actual execution time recorded in the behavior path graph should match it; by analyzing the semantic content of the task node and the behavior node, ensure that the actual behavior is consistent with the task requirements. For example, the requirement of the task node "install the grounding wire" is to install the grounding wire, and the behavior node in the behavior path graph should reflect this action.

[0041] During the pairing analysis process, the task nodes and the behavioral path graph are checked for anomalies. For example, if the execution time of a task node exceeds the scheduled time, it may cause task delays. For example, if the task "Restore Power" should be executed after the completion of "Check Ground Resistance," if the behavioral path graph shows that the "Restore Power" task has an unnecessary delay, it indicates that there may be a problem with the task execution process. After the pairing analysis is completed, the successfully matched task node pairs and the abnormal node pairs are recorded. This result provides a basis for subsequent safety verification and early warning.

[0042] The multimodal perception score is a comprehensive score that evaluates the execution of task nodes based on multiple perception dimensions, including temporal consistency score, behavioral consistency score, state consistency score, etc. The score of each dimension can be combined through the weighted average method to obtain a comprehensive score for each task node.

[0043] Multimodal perception scores are dynamic, updating in real time as tasks are executed and behaviors occur. For example, if a task is delayed or doesn't meet expectations, the score will automatically adjust, reflecting the increased risk of task execution. Dynamic safety alerts are generated based on the multimodal perception score. Specifically, alert levels are assigned based on the multimodal perception score: low scores (indicating high risk) generate high-priority alerts, while high scores (indicating low risk) generate low-priority alerts. Comprehensive dynamic safety alerts are generated based on the alert level and the cause of the alert. Depending on the severity of the alert, relevant personnel can take immediate response measures, such as adjusting work plans, strengthening on-site monitoring, and dispatching inspectors, to ensure operational safety.

[0044] Furthermore, performing task node pairing analysis based on the directed task graph and the behavior path graph includes:

[0045] Based on the directed task graph, safety steps are extracted as task nodes, and the task nodes include operation name, executor, tool equipment, location, and execution time period attributes; with the task nodes as matching nodes, candidate nodes that meet the spatiotemporal and semantic conditions in the behavior path graph are searched to construct an initial matching set; the node pairing channel is used to perform node pair matching analysis of the initial matching set under four-dimensional features to establish the task node pairing analysis results.

[0046] A task node is the specific representation of each safety step in a directed task graph. Each node represents a specific safety step in a task. Each task node includes the attributes of operation name, executor, tools and equipment, location, and execution time period. The operation name is the specific operation or step name of the task, such as cutting off the power supply and installing the grounding wire; the executor is the operator who performs the task step, such as electrician A; the tools and equipment are the tools and equipment used during the execution of the task, such as insulating gloves and grounding wire installation tools; the location is the specific location where the task is performed, which is usually related to the work area or facility, such as distribution line X; and the execution time period is the time range for task execution.

[0047] In the behavior path diagram, each behavior node represents the actual behavior in the operation process. The task node, as a predefined operation step, must be matched with the relevant behavior nodes in the behavior path diagram, including spatiotemporal matching and semantic matching. Specifically, filtering is performed through time windows and spatial ranges to ensure that the behavior nodes in the behavior path diagram and the task nodes have the same execution time period and spatial range. For example, if the scheduled time of the "cut off the power" task in the task node is "08:00-08:05", then the matching behavior nodes should occur in the same or overlapping time period and the location of occurrence should be consistent; matching is performed according to the semantics of the task node. For example, the task node "install the grounding wire" should match the behavior node that performs the operation in the behavior path diagram to ensure that the actually executed task is consistent with the predefined task steps.

[0048] Based on the spatiotemporal and semantic conditions, an initial matching set is generated. This set contains all eligible behavior nodes, which are candidate nodes for matching with the task node. In this way, multiple candidate behavior nodes are found for each task node, providing options for subsequent node pairing analysis.

[0049] A node pairing channel is introduced to perform node pair matching analysis, including node pair area matching analysis, time window overlap matching analysis, operation type matching analysis, and node pair verification matching. Through four independent channels, each feature is analyzed through an independent channel to determine whether the task nodes and candidate behavior nodes in the initial matching set meet the requirements. Through these pairing analyses, the task node pairing analysis results are established. These results show which task nodes and behavior nodes are successfully matched and which ones are inconsistent or abnormal.

[0050] Furthermore, the method of establishing a multimodal perception score using the task node pairing analysis results and generating a dynamic security warning using the multimodal perception score includes:

[0051] The task node pairing analysis results are used to compare the execution path sequence with the dependency order in the original directed task graph to establish a dependency order comparison result; risk nodes are marked based on the matching anomaly score of the task node pairing analysis results and the sequence anomaly score of the dependency order comparison results, and a dynamic warning level is output.

[0052] Based on the results of the task node pairing analysis, an actual execution path sequence is generated. This execution path sequence represents the actual execution order of the job tasks and is generated based on factors such as the chronological order of the nodes in the behavioral path graph and the order of operations. The original directed task graph defines the expected execution order of the job tasks, that is, the dependencies between tasks. For example, the grounding wire can only be installed after the power is cut off. After the grounding wire is installed, the grounding resistance can be checked, and finally the power can be restored.

[0053] Compare the actual execution path sequence with the dependency order in the original directed task graph to determine whether there are any task sequence errors or delayed execution problems, whether there are any task skipping situations, or whether there are tasks executed in advance. The key to the comparison is to determine whether the execution of each task conforms to the original dependency relationship. Based on the comparison results, the dependency order comparison results are established.

[0054] Based on the results of the task node pairing analysis, a matching anomaly score is assigned to each task node. This score reflects the degree of matching between the task node and the behavior node. The high or low score indicates the accuracy of the task node execution. Among them, a high score indicates that the task node and the behavior node are completely consistent and the task execution is normal; a low score indicates that there is a significant mismatch between the task node and the behavior node, which may be due to incorrect operation, delay, missing necessary steps, etc.

[0055] Based on the dependency order comparison results, a sequence anomaly score is generated for each task node. This score reflects whether the task execution order conforms to the original dependency order. A high score indicates that the tasks are executed in the correct order and the dependencies between tasks are satisfied; a low score indicates that the task order does not meet expectations, and there are cases where tasks are skipped, executed early or delayed, which may lead to conflicts between tasks or safety hazards.

[0056] Based on matching anomaly scores and sequential anomaly scores, each task node in the operation process is risk-assessed. Nodes with low scores are marked as risky, indicating that these task steps pose a high safety risk. Dynamic warning levels are then generated based on the risk score of each task node. The marking of risky nodes and the output of dynamic warning levels provide a basis for safety management during the operation. Operation managers can take timely preventive or corrective measures based on the warning level to ensure the smooth and safe completion of the operation.

[0057] Furthermore, the node pair matching analysis of the initial matching set under the four-dimensional features is performed using the node pairing channel to establish the task node pairing analysis result, including:

[0058] Activate the regional consistency evaluation sub-channel, use the regional consistency evaluation sub-channel to perform node-to-region matching analysis in the initial matching set, and establish a first dimension matching result; activate the time window evaluation sub-channel, use the time window evaluation sub-channel to perform node-to-time window overlap matching analysis in the initial matching set, and establish a second dimension matching result; activate the operation type matching sub-channel, use the operation type matching sub-channel to perform node-to-operation type matching analysis in the initial matching set, and establish a third dimension matching result; activate the sensor verification sub-channel, use the sensor verification sub-channel to perform node-to-node verification matching in the initial matching set, and establish a fourth dimension matching result, wherein the regional consistency evaluation sub-channel, time window evaluation sub-channel, operation type matching sub-channel, and sensor verification sub-channel are all processing sub-channels in the node pairing channel; establish a task node pairing analysis result based on the first dimension matching result, the second dimension matching result, the third dimension matching result, and the fourth dimension matching result.

[0059] The regional consistency evaluation subchannel is an analytical tool used to evaluate the positional consistency between task nodes and behavior nodes. Its function is to ensure that the operation location specified in the task node is consistent with the execution location recorded in the behavior path diagram. During the pairing analysis process, each task node contains location information, and the behavior node in the behavior path diagram also contains the location information for executing the task. The regional consistency evaluation subchannel matches the spatial locations of task nodes and behavior nodes to evaluate whether they are located in the same or overlapping regions. Specific operations include: extracting the location attributes from the task nodes and behavior nodes, and comparing the operation location specified by the task node with the actual execution location of the behavior node. If the locations do not match, it indicates that the match has failed and there may be problems with task execution. Based on the results of the regional matching analysis, a first-dimensional matching result is generated, reflecting the positional consistency of the task node and the behavior node.

[0060] The time window evaluation subchannel is an analytical tool used to assess the temporal consistency between task nodes and behavior nodes. Its purpose is to ensure that the execution time range of a task matches the time nodes in the behavior path graph. Task nodes typically specify a time period, and behavior nodes in the behavior path graph also record the execution time period. By comparing this time information, it is possible to check whether the task is executed on time. If the execution times of the task node and behavior node overlap, it indicates that their time windows match and the task is executed on time. If the time windows do not overlap, the match is marked as failed, indicating that the task execution time does not meet the requirements and the workflow needs to be rescheduled or revised. The specific operation involves extracting the time information from the task node and behavior node and determining whether the time periods of the task node and behavior node overlap or meet the predetermined time requirements. If the time periods overlap, the match is successful; otherwise, the match fails. Based on the time window overlap matching analysis results, a second dimension matching result is generated, indicating the temporal consistency of the task node and behavior node.

[0061] The operation type matching subchannel is an analysis tool used to evaluate the consistency of operation types between task nodes and behavior nodes. Its function is to ensure that the operation specified in the task node matches the operation type performed in the behavior node. Each task node usually defines a specific operation type, and the behavior node in the behavior path diagram records the actual operation type performed. By comparing these operation types, it can be determined whether the task is executed correctly. If the operation type specified by the task node is exactly the same as the operation type performed by the behavior node, it means that the task is executed correctly; if the operation types are inconsistent, it is marked as a match failure, indicating that the actual operation does not meet the task requirements and needs to be corrected. Based on the results of the operation type matching analysis, a third dimension matching result is generated, indicating the consistency of the task node and the behavior node in terms of operation type.

[0062] The sensor verification subchannel is an analysis tool used to verify whether the physical or perception data supported by the sensor data is consistent between the task node and the behavior node. It is used to ensure that the execution of the task complies with the monitoring data or conditions of the sensor. In some power operations, task execution is usually accompanied by some physical or environmental changes, such as temperature, pressure, current, voltage changes, etc. These data can be monitored in real time through sensors and used to verify the execution of the task. The purpose of sensor verification is to ensure that the actual operations recorded in the behavior path diagram meet the physical or environmental requirements, and that the operations required by the task node can be implemented in the actual perception environment.

[0063] Sensor validation analysis compares sensor data for each node pair in the initial matching set. Specifically, the environmental data provided by the sensors, such as current, voltage, and ground resistance, is checked to see if they are consistent with the standard or expected values ​​defined in the task node. This matching analysis is performed on each node pair to confirm whether the physical conditions of the task node and the behavior node are consistent. Based on the results of sensor validation analysis, a fourth dimension matching result is generated, reflecting the matching between the task node and the behavior node at the physical or sensor device level.

[0064] Among them, the regional consistency evaluation sub-channel, time window evaluation sub-channel, operation type matching sub-channel, and sensor verification sub-channel are all processing sub-channels in the node pairing channel. Each sub-channel runs independently. Through four independent channels, each feature can be analyzed separately to determine whether the task nodes and candidate behavior nodes in the initial matching set meet the requirements. Through these pairing analyses, task node pairing analysis results are established. These results indicate which task nodes and behavior nodes are successfully matched and which are inconsistent or abnormal.

[0065] The first dimension matching results, the second dimension matching results, the third dimension matching results, and the fourth dimension matching results are integrated to form a complete task node pairing analysis result. The task node pairing analysis result provides the execution status of each task node, helping the system to accurately identify potential problems in task execution.

[0066] Furthermore, the method of establishing a multimodal perception score using the task node pairing analysis results and generating a dynamic security warning using the multimodal perception score includes:

[0067] Establish a trend factor set, which includes a spatial drift factor, a time delay factor, a repeated operation factor, a tool state deviation factor, and a sequential fallback factor; call the trend factor set to perform the identification of the operational behavior risk evolution tendency based on the task node pairing analysis results; use the operational behavior risk evolution tendency identification results to establish a risk evolution trend score, construct an additional warning based on the risk evolution trend score, and integrate the additional warning into the safety dynamic warning.

[0068] A set of trend factors is established to describe the different risk dynamics that occur during task execution. Each factor will affect the safety of the operation process and may lead to safety hazards. Among them, the spatial drift factor describes whether the operation task deviates from the predetermined operation area, for example, whether the operator or equipment performs the task in the wrong location, or whether it is active outside the boundary; the time delay factor measures whether the operation task is delayed, whether the task execution time exceeds the predetermined time range, or whether it is not completed on time, which will affect the execution of subsequent tasks; the repeated operation factor is used to check whether there are repeated operations in the task execution, for example, whether the operator repeats the same task without completing the previous operation, or whether the task is repeated for some reason; the tool state deviation factor measures whether the operation tool is in the predetermined working state, for example, the tool is damaged, improperly operated, or used without inspection, which may affect the normal progress of the task; the sequential fallback factor measures whether the execution order of the task is reversed, or whether the execution of the task node returns to the previous step, violating the original dependency order.

[0069] The trend factor set is used to analyze the results of task node pairing analysis to identify the risk evolution tendency of operational behavior. Risk evolution tendency identification dynamically evaluates operational behavior based on the trend factor set and identifies the potential risk evolution during task execution. Specifically, the task node pairing analysis results are analyzed and combined with the dynamic changes of factors such as spatial drift, time delay, and repeated operations in the trend factor set to identify the risk evolution tendency during the operation. For example, if the time delay factor gradually increases and the operation task enters a state of multiple repeated operations, the operation behavior may be at risk of continuous delay or interruption. If the spatial drift factor gradually increases, it may lead to the expansion of the operation area, thereby affecting the order and execution environment of the operation tasks. Using the feedback of the trend factor set, potential safety hazards in the operation process are predicted and the corresponding risk evolution tendency is generated.

[0070] Based on the results of identifying the risk evolution trends of operational behaviors, a risk evolution trend score is calculated for each task node. This score reflects the changing trend of potential risks during the execution of the task. A high score indicates that the risk evolution of the task is tending to worsen, posing a significant safety hazard and requiring urgent intervention; a low score indicates that the risk evolution of the task is within a controllable range and the risk of task execution is low. The risk evolution trend score is constructed based on the comprehensive calculation results of multiple factors. If multiple factors show negative trends, the score is higher.

[0071] Building on the existing dynamic safety warnings, additional warnings are generated based on risk evolution trend scores. These additional warnings reflect potential trends during task execution and provide operational managers with early warnings of future risks. For example, if the risk evolution trend score is high, the additional warning informs the operator or monitoring center that the task may present safety issues in the near future and recommends additional inspections or adjustments. By integrating additional warnings into dynamic safety warnings, dynamic safety warnings not only reflect the risks of the current task execution but also predict potential future risks based on risk evolution trends, helping operators take timely measures.

[0072] Furthermore, the method of using the structured task template to perform security verification of the time series work data set and generate a dynamic security warning further includes:

[0073] A multi-job interaction graph is constructed based on the structured task template, and the multi-job interaction graph includes multiple job task nodes and spatial conflict, time conflict, and resource conflict edge relationships; the multi-job interaction graph is used to identify cross-job conflicts in the time-series work data set to determine whether there are area overlap conflicts, operation conflicts, timing logic conflicts, and electrical interlocking conflicts; and a dynamic safety warning is generated based on the judgment results.

[0074] A multi-job interaction graph is constructed based on a structured task template to represent the relationships between multiple job tasks. Each job task is regarded as a node, and the edges between nodes represent different types of conflict relationships, including spatial conflict, temporal conflict, and resource conflict edge relationships. Among them, spatial conflict refers to the location conflict caused by the requirement that two tasks be executed in the same physical space. For example, two tasks need to be operated simultaneously in the same work area, resulting in spatial overlap; temporal conflict refers to the conflict caused by the overlapping execution time of two tasks. For example, task A and task B need to be executed at the same time, resulting in a time conflict; resource conflict refers to the resource allocation conflict caused by the need for two tasks to use the same resources, such as equipment, tools, personnel, etc.

[0075] The time series work dataset contains the actual execution order and time information of task nodes. By cross-validating the conflict edges in the multi-task interaction graph with the actual execution status in the time series dataset, we check whether conflicts occur between tasks, including checking for area overlap conflicts, operation conflicts, timing logic conflicts, and electrical interlock conflicts. Specifically, we analyze the execution space of tasks to check whether two tasks attempt to execute in the same area. For example, if two workers work in the same area at the same time, this may lead to safety risks, and thus an area overlap conflict exists. We also check whether there are repeated operation requirements in the tasks, especially on the same equipment or tools. For example, if two tasks require the use of the same power tool or equipment, then an operation conflict exists. We compare the execution order of task nodes with the task dependencies to determine whether there are any order violations. For example, if task B is supposed to be executed after task A is completed, but task B is actually started earlier during execution, then a timing logic conflict exists. We also check whether the interlocking mechanism of electrical equipment is violated. For example, some equipment requires power to be disconnected under safe conditions before maintenance. If multiple tasks require the operation of the same equipment at the same time, this may lead to an electrical interlock conflict, and thus an electrical interlock conflict exists.

[0076] According to the judgment results of the conflict, a corresponding dynamic security warning is generated. That is, if any of the above conflicts is found, a corresponding warning will be triggered.

[0077] Furthermore, generating a dynamic security warning based on the judgment result includes:

[0078] The historical behavior risk profile of the operator is called; an individual risk sensitivity coefficient is generated according to the historical behavior risk profile, and a conflict level rise analysis is performed on the judgment result based on the individual risk sensitivity coefficient, and a dynamic safety warning is generated based on the conflict level rise analysis result.

[0079] The historical behavioral risk profiles of operators are called up. These historical behavioral risk profiles are constructed based on information such as the operators' historical task execution, safety violation records, historical accident data, and operational behavior assessments. They are used to assess the operators' safety behavior patterns and potential risks. For example, if an operator has often caused risk events due to time delays or operational errors in the past, then his historical behavioral risk profile will show a higher risk value.

[0080] An individual risk sensitivity coefficient is generated based on the operator's historical behavioral risk profile. The individual risk sensitivity coefficient is a numerical value that reflects the operator's risk propensity. This coefficient measures the operator's sensitivity to potential risks during the operation. For example, the higher the frequency of risk events in the operator's past, the higher the individual risk sensitivity coefficient; the greater the severity of the operator's safety violations (such as the consequences of accidents), the higher the individual risk sensitivity coefficient; and if the operator has strictly adhered to safety measures and procedures in the past, their risk sensitivity coefficient will be relatively low. The data in the historical behavioral risk profile is mapped into a quantified individual risk sensitivity coefficient, which reflects the operator's risk tolerance and responsiveness when performing tasks.

[0081] Conflict Rising Analysis dynamically assesses the conflict level in the current task based on individual risk sensitivity coefficients. When an operator's historical behavioral risk is high, the conflict score in the current task is weighted, raising the warning level to reflect the potential risk changes posed by the operator. For example, an operator with a high risk sensitivity coefficient may lead to an increase in the conflict level during task execution, predicting that these conflicts may cause major safety issues. An operator with a low risk sensitivity coefficient indicates that the operator is relatively safe, with a lower likelihood of an escalation in the conflict level. Based on the results of the conflict rising analysis, a dynamic safety warning is generated, reflecting potential safety issues that may exist in the current task. This process allows the task risk level to be dynamically adjusted based on the operator's historical behavioral characteristics, providing more personalized safety measures through refined personnel risk assessments.

[0082] Furthermore, the intelligent sensing of the ground wire includes positioning monitoring, height monitoring, conductor identification monitoring, ground insertion depth monitoring, and connection loop resistance calculation.

[0083] The function of the grounding wire is to provide safety for equipment and personnel, so its installation and monitoring process is crucial. Intelligent sensing of the grounding wire involves multiple key monitoring functions, aiming to ensure the safety, effectiveness and compliance of the grounding wire with standard requirements. Among them, positioning monitoring refers to the real-time tracking of the position of the grounding wire through positioning technology to ensure that the grounding wire is installed in the correct working area and does not deviate from the predetermined position. Common technologies include GPS positioning, RFID tags or other geographic positioning systems. For example, in the grounding requirements of key facilities such as substations and distribution lines, incorrectly positioned grounding wires may not be able to effectively prevent electrical accidents; height monitoring refers to monitoring the installation height of the grounding wire. Usually, the grounding wire needs to be installed within a specific height range to ensure that the current will not be affected by physical obstacles when flowing. Abnormal height may lead to poor grounding and affect the safety of the electrical system; conductor identification monitoring is used to confirm whether the grounding wire is made of suitable conductor material and that the material is not damaged. Current detection is usually used. Technology is used to judge the conductive capacity of the grounding wire. If the grounding wire is made of low-quality materials or is damaged, poor grounding will result, which cannot effectively prevent electrical equipment or personnel from electric shock. Ground depth monitoring is a real-time monitoring of the depth of the grounding wire buried in the ground. The grounding wire needs to be buried deep enough in the soil to ensure the stability and effectiveness of the grounding. Insufficient grounding depth may make it impossible to effectively guide the current into the earth, resulting in electrical equipment unable to correctly release current when a fault occurs, causing safety hazards. Hanging loop resistance measurement refers to measuring the resistance value of the grounding loop to ensure that it meets safety requirements. The smaller the grounding loop resistance, the higher the current conduction efficiency and the better the grounding effect. Usually, the resistance value should be lower than the standard value, such as 1Ω, otherwise the grounding effect may be poor and electrical accidents cannot be effectively prevented.

[0084] In summary, the dynamic early warning method for personnel safety during distribution network line operations provided by the embodiments of the present application has the following technical effects:

[0085] By calling the work ticket, key information such as work tasks, operator information, safety measures requirements, etc. can be accurately obtained, and combined with grounding wire installation information, control instructions, etc., to ensure that operators have sufficient safety measures when performing tasks, which helps to avoid safety accidents caused by human negligence or information omissions; by performing text semantic structured preprocessing of the work ticket, three-dimensional safety features are extracted, including the control center execution features in the work task, the operator execution safety features, the grounding wire installation and removal task features, etc. The extraction and structured processing of these features significantly improves the readability and safety verification capabilities of the work task information; the power NLP model is used to extract safety statements from the extracted three-dimensional safety features, thereby establishing a structured task template. This process allows the entire work task to be automatically evaluated for safety through intelligent means, reducing errors and omissions in human intervention; when the operator's positioning information When the information meets the scope of the designated area, the target area is activated, which means that the working area can be dynamically managed in real time according to the location of the personnel. This precise area division and real-time location monitoring effectively reduce the risk of personnel entering dangerous areas during the operation; after the target area is activated, a time-series working data set is established based on the intelligent perception of the grounding wire and the intelligent perception of the working tool. By real-time monitoring of the status of the working tools and the grounding wire, a timely response can be made to potential safety hazards in the operation process. The construction of this time-series data set provides a more accurate basis for dynamic monitoring during the operation process; using structured task templates to perform safety verification on the time-series working data set can automatically detect potential safety hazards in the operation process and generate dynamic safety warnings, which can issue safety warnings in time during the operation to avoid accidents caused by safety hazards not being discovered in time. The automated warning mechanism improves the safety of the operating personnel.

[0086] Example 2, based on the same inventive concept as the method for dynamic early warning of personnel safety in distribution network line operations in the above embodiment, as shown in FIG. Figure 2 As shown, an embodiment of the present application provides a dynamic early warning system for personnel safety in distribution network line operations, the system comprising:

[0087] The work ticket calling module 10 is used to call the operation work ticket, which contains work tasks, operator information, safety measures requirements, grounding wire installation information, and control instructions.

[0088] The structured preprocessing module 20 is used to perform text semantic structured preprocessing of the work order and extract three-dimensional safety features, which include control center execution features, operator execution safety features, and grounding wire installation and removal task features.

[0089] The safety statement extraction module 30 is used to extract safety statements with three-dimensional safety features using the power NLP model and establish a structured task template.

[0090] The target area activation module 40 is used to configure the target area to be activated according to the operation work ticket, and activate the target area when the positioning information of the target operator is identified to be consistent with the area range of the target area.

[0091] The working data set establishing module 50 is used to establish a time-series working data set based on the intelligent perception of the ground wire and the intelligent perception of the working tool after the target area is activated.

[0092] The safety verification module 60 is used to perform safety verification of the time series work data set using the structured task template and generate a safety dynamic warning.

[0093] Furthermore, the security verification module 60 is configured to perform the following steps:

[0094] A directed task graph is constructed based on the structured task template, and state transition edges are set; the time-series work data set is converted into a time-annotated graph, and a behavior path graph is constructed; task node pairing analysis is performed based on the directed task graph and the behavior path graph; a multimodal perception score is established using the task node pairing analysis results, and a dynamic safety warning is generated using the multimodal perception score.

[0095] Furthermore, the security verification module 60 is configured to perform the following steps:

[0096] Based on the directed task graph, safety steps are extracted as task nodes, and the task nodes include operation name, executor, tool equipment, location, and execution time period attributes; with the task nodes as matching nodes, candidate nodes that meet the spatiotemporal and semantic conditions in the behavior path graph are searched to construct an initial matching set; the node pairing channel is used to perform node pair matching analysis of the initial matching set under four-dimensional features to establish the task node pairing analysis results.

[0097] Furthermore, the security verification module 60 is configured to perform the following steps:

[0098] The task node pairing analysis results are used to compare the execution path sequence with the dependency order in the original directed task graph to establish a dependency order comparison result; risk nodes are marked based on the matching anomaly score of the task node pairing analysis results and the sequence anomaly score of the dependency order comparison results, and a dynamic warning level is output.

[0099] Furthermore, the security verification module 60 is configured to perform the following steps:

[0100] Activate the regional consistency evaluation sub-channel, use the regional consistency evaluation sub-channel to perform node-to-region matching analysis in the initial matching set, and establish a first dimension matching result; activate the time window evaluation sub-channel, use the time window evaluation sub-channel to perform node-to-time window overlap matching analysis in the initial matching set, and establish a second dimension matching result; activate the operation type matching sub-channel, use the operation type matching sub-channel to perform node-to-operation type matching analysis in the initial matching set, and establish a third dimension matching result; activate the sensor verification sub-channel, use the sensor verification sub-channel to perform node-to-node verification matching in the initial matching set, and establish a fourth dimension matching result, wherein the regional consistency evaluation sub-channel, time window evaluation sub-channel, operation type matching sub-channel, and sensor verification sub-channel are all processing sub-channels in the node pairing channel; establish a task node pairing analysis result based on the first dimension matching result, the second dimension matching result, the third dimension matching result, and the fourth dimension matching result.

[0101] Furthermore, the security verification module 60 is configured to perform the following steps:

[0102] Establish a trend factor set, which includes a spatial drift factor, a time delay factor, a repeated operation factor, a tool state deviation factor, and a sequential fallback factor; call the trend factor set to perform the identification of the operational behavior risk evolution tendency based on the task node pairing analysis results; use the operational behavior risk evolution tendency identification results to establish a risk evolution trend score, construct an additional warning based on the risk evolution trend score, and integrate the additional warning into the safety dynamic warning.

[0103] Furthermore, the security verification module 60 is configured to perform the following steps:

[0104] A multi-job interaction graph is constructed based on the structured task template, and the multi-job interaction graph includes multiple job task nodes and spatial conflict, time conflict, and resource conflict edge relationships; the multi-job interaction graph is used to identify cross-job conflicts in the time-series work data set to determine whether there are area overlap conflicts, operation conflicts, timing logic conflicts, and electrical interlocking conflicts; and a dynamic safety warning is generated based on the judgment results.

[0105] Furthermore, the security verification module 60 is configured to perform the following steps:

[0106] The historical behavior risk profile of the operator is called; an individual risk sensitivity coefficient is generated according to the historical behavior risk profile, and a conflict level rise analysis is performed on the judgment result based on the individual risk sensitivity coefficient, and a dynamic safety warning is generated based on the conflict level rise analysis result.

[0107] Furthermore, the intelligent sensing of the ground wire includes positioning monitoring, height monitoring, conductor identification monitoring, ground insertion depth monitoring, and connection loop resistance calculation.

[0108] Through the above detailed description of the personnel safety dynamic warning method for distribution network line operations in this specification, those skilled in the art can clearly understand the personnel safety dynamic warning system for distribution network line operations in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For relevant matters, please refer to the method section.

[0109] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A dynamic early warning method for personnel safety during distribution network line operations, characterized in that: The method comprises: Calling an operation work ticket, wherein the operation work ticket includes work tasks, operator information, safety measures requirements, grounding wire installation information, and control instructions; Performing text semantic structured preprocessing of the work order to extract three-dimensional safety features, including control center execution features, operator execution safety features, and ground wire installation and removal task features; Use the power NLP model to extract safety statements with three-dimensional safety features and establish a structured task template; The target area is configured as a waiting state according to the operation work ticket, and the target area is activated when the positioning information of the target operator is identified to be consistent with the area range of the target area; After the target area is activated, a time series working data set is established based on intelligent perception of the grounding wire and the working tool; Using the structured task template to perform security verification of the time series work data set and generate a dynamic security warning; Among them, generating dynamic security warnings includes: Constructing a directed task graph based on the structured task template and setting state transition edges; Convert the time series work data set into a time-annotated graph and construct a behavior path graph; Performing task node pairing analysis based on the directed task graph and the behavior path graph; Establishing a multimodal perception score using the task node pairing analysis results, and generating a dynamic safety warning using the multimodal perception score; Among them, task node pairing analysis is performed, including: Extracting safety steps as task nodes based on the directed task graph, wherein the task nodes include attributes such as operation name, performer, tool equipment, location, and execution time period; Taking the task node as a matching node, searching for candidate nodes that meet spatiotemporal and semantic conditions in the behavior path graph, and constructing an initial matching set; Use the node pairing channel to perform node pair matching analysis of the initial matching set under four-dimensional features and establish the task node pairing analysis results; Among them, establishing task node pairing analysis results includes: Activate the regional consistency evaluation sub-channel, use the regional consistency evaluation sub-channel to perform node-to-region matching analysis in the initial matching set, and establish a first dimension matching result; Activate the time window evaluation subchannel, use the time window evaluation subchannel to perform time window coincidence matching analysis of nodes in the initial matching set, and establish a second dimension matching result; activating an operation type matching subchannel, and using the operation type matching subchannel to perform operation type matching analysis on the nodes in the initial matching set, and establishing a third dimension matching result; Activate the sensor verification subchannel, use the sensor verification subchannel to perform node pair verification matching in the initial matching set, and establish a fourth-dimensional matching result. The regional consistency evaluation subchannel, time window evaluation subchannel, operation type matching subchannel, and sensor verification subchannel are all processing subchannels in the node pairing channel. A task node pairing analysis result is established according to the first dimension matching result, the second dimension matching result, the third dimension matching result, and the fourth dimension matching result.

2. The dynamic early warning method for personnel safety in distribution network line operations according to claim 1, characterized in that: The method of establishing a multimodal perception score using the task node pairing analysis results and generating a dynamic security warning using the multimodal perception score includes: Using the task node pairing analysis results, the execution path sequence is compared with the dependency order in the original directed task graph to establish a dependency order comparison result; Based on the matching anomaly score of the task node pairing analysis result and the sequence anomaly score of the dependency sequence comparison result, risk nodes are marked and a dynamic warning level is output.

3. The dynamic early warning method for personnel safety in distribution network line operation according to claim 1, characterized in that: The method of establishing a multimodal perception score using the task node pairing analysis results and generating a dynamic security warning using the multimodal perception score includes: Establishing a trend factor set, wherein the trend factor set includes a spatial drift factor, a time delay factor, a repeated operation factor, a tool state deviation factor, and a sequential fallback factor; Calling the trend factor set to perform task node pairing analysis results to identify operational behavior risk evolution tendencies; A risk evolution trend score is established using the results of the risk evolution tendency identification of the operational behavior, an additional warning is constructed based on the risk evolution trend score, and the additional warning is integrated into the safety dynamic warning.

4. The dynamic early warning method for personnel safety in distribution network line operation according to claim 1, characterized in that: The method of using the structured task template to perform security verification of the time series work data set and generate a dynamic security warning further includes: Constructing a multi-job interaction graph based on the structured task template, wherein the multi-job interaction graph includes multiple job task nodes and spatial conflict, time conflict, and resource conflict edge relationships; Using the multi-operation interaction graph to identify cross-operation conflicts in a sequential work data set, and determining whether there are area overlap conflicts, operation conflicts, sequential logic conflicts, and electrical interlocking conflicts; Generate dynamic security warnings based on the judgment results.

5. The dynamic early warning method for personnel safety in distribution network line operation according to claim 4, characterized in that: Generating a dynamic security warning based on the judgment result includes: Call the historical behavioral risk profile of the operator; An individual risk sensitivity coefficient is generated according to the historical behavior risk profile, and a conflict level rise analysis of the judgment result is performed based on the individual risk sensitivity coefficient, and a security dynamic warning is generated based on the conflict level rise analysis result.

6. The dynamic early warning method for personnel safety in distribution network line operation according to claim 1, characterized in that: The intelligent sensing of the grounding wire includes positioning monitoring, height monitoring, conductor identification monitoring, ground insertion depth monitoring, and connection loop resistance calculation.

7. A dynamic early warning system for personnel safety during distribution network line operations, characterized in that: A system for implementing the dynamic early warning method for personnel safety in distribution network line operations according to any one of claims 1 to 6, comprising: A work ticket calling module is used to call an operation work ticket, which contains work tasks, operator information, safety measures requirements, grounding wire installation information, and control instructions; a structured preprocessing module for performing text semantic structured preprocessing of the work order and extracting three-dimensional safety features, the three-dimensional safety features including control center execution features, operator execution safety features, and ground wire installation and removal task features; The safety statement extraction module is used to extract safety statements with three-dimensional safety features using the power NLP model and establish a structured task template; A target area activation module is used to configure the target area to a pending activation state according to the operation work ticket, and activate the target area when the positioning information of the target operator is identified to be consistent with the area range of the target area; A working data set establishment module is used to establish a time-series working data set based on intelligent perception of the ground wire and the working tool after the target area is activated; The safety verification module is used to use the structured task template to perform safety verification of the time series work data set and generate a dynamic safety warning.

Images