A remote security coordination platform for power construction sites

Abstract

This invention belongs to the field of engineering safety and discloses a remote safety coordination platform for power construction sites, including an acquisition module, a first generation module, an adjustment module, an analysis module, and a second generation module. The acquisition module is used to access dynamic data from the entire "two-ticket, three-system" process and personnel behavior data at the construction site; the first generation module dynamically generates a personnel behavior monitoring benchmark model adapted to the current working conditions; the adjustment module adaptively adjusts the full-dimensional monitoring rules based on this model; the analysis module analyzes the behavior data in real time based on the adjusted rules and determines the risk level; the second generation module generates tiered emergency control instructions based on the analysis results and sends them back to the "two-ticket, three-system" management system, forming a closed-loop control system. This invention solves the problem of invalid alarms caused by static rules by deeply binding the monitoring system with the "two-ticket, three-system" statutory operation management, achieving a leap from post-event alarms to in-event intervention, and eliminating the management drawbacks of the disconnect between ticket approval and on-site execution.

Description

Technical Field

[0001] This invention relates to the field of engineering safety, and in particular to a remote safety coordination platform for power construction sites. Background Technology

[0002] Power engineering construction is characterized by high voltage, high risk, and complex and ever-changing work scenarios. Personnel violations are a core cause of personal injury and death accidents in power construction. Relevant regulations explicitly list the "two-ticket, three-system" (work permit, operation permit, shift handover system, patrol inspection system, and equipment periodic testing and rotation system) as a mandatory core legal system for power safety production. All power construction operations must be based on compliantly approved work permits and operation permits to achieve compliant control throughout the entire operation process. Therefore, personnel behavior safety monitoring must be deeply integrated with the legal requirements of the "two-ticket, three-system" to fundamentally ensure the compliance and safety of power construction.

[0003] With the development of the Internet of Things in the power industry and intelligent monitoring technology, the safety monitoring of construction workers in power engineering has gradually upgraded from the traditional manual inspection and on-site supervision mode to an automated monitoring mode based on sensor networks and AI video recognition. The applicant's earlier invention patent application, publication number CN119831275A, provides a method for safety management in power engineering construction. It constructs a multi-dimensional safety monitoring system covering the construction environment, equipment operation, and worker behavior, achieving automated identification and early warning for four core scenarios: unauthorized entry, falls from heights, equipment collisions, and electric shock risks. This solves the problems of high reliance on manual labor, incomplete monitoring coverage, and delayed risk identification in traditional safety management solutions, providing a fundamental technical solution for automated safety monitoring in power construction.

[0004] However, during the large-scale engineering implementation of the above-mentioned technical solutions, the applicant discovered that the existing technology still has fundamental application limitations. These limitations are also common technical problems faced by similar personnel behavior monitoring solutions in the current field: the existing personnel behavior safety monitoring system is completely disconnected from the "two-ticket, three-system" full-process operation management system that is legally mandated by the power industry. Its monitoring solution adopts static and fixed rule design logic, which is difficult to adapt to the dynamic operation compliance requirements of the power construction site that are accompanied by the full-process changes of the "two-ticket, three-system". This leads to serious problems such as misjudgment and omission in actual application of the monitoring system, failure of in-process control, and insufficient compliance support capabilities, making it difficult to fundamentally eliminate illegal operation behavior.

[0005] Specifically, the aforementioned core technical issues lead to the following direct defects in engineering applications: Firstly, the existing monitoring solutions rely on static content such as work permission zoning maps, risk assessment thresholds, and violation identification rules, all of which are preset before construction. However, the permitted work scope, equipment power outage and restoration status, operator permissions, and on-site risk levels at power construction sites are all dynamically changing throughout the entire process of work permit and operation ticket approval, execution, extension, and termination. The significant mismatch between the static and fixed monitoring rules and the dynamic compliance requirements on-site leads to a large number of invalid alarms. For example, compliant operators with valid work permits may still trigger boundary crossing warnings when entering the permitted work area, causing on-site operators to resist the monitoring system and even actively shut it down, resulting in a substantial failure of the monitoring function.

[0006] Secondly, existing solutions can only identify and alert on violations after the fact, and cannot be linked to the entire lifecycle of the "two-ticket, three-system" process to achieve dynamic control during the process. Although the current electronic two-ticket system can complete the online approval of tickets in advance, it cannot synchronize the real-time execution status of tickets to the monitoring system. This makes it impossible for the system to prevent and intercept core violations such as working without a ticket, working beyond the scope of permission, and operating beyond authorization. This has created a long-standing management drawback in the industry of "two separate systems for ticket approval and on-site execution," making it difficult to eliminate violations through technical means.

[0007] Third, the existing monitoring scheme's basis for judging violations is only general safety standards, and it does not take the legally mandated "two-ticket, three-system" compliance requirements as the core judgment benchmark. This results in a disconnect between the monitoring and early warning results and the compliance traceability requirements, making it difficult to have the legally mandated compliance support effect and failing to meet the National Energy Administration's regulatory requirements for full-process compliance traceability of power construction safety.

[0008] Regarding the aforementioned core technical issues, while attempts have been made in this field to achieve simple interface integration between the electronic work permit system and the monitoring system, these efforts are limited to a single permission verification mechanism: "entry into the corresponding work area is permitted only with a valid work permit." This fails to fundamentally resolve the systemic problems mentioned above. The root cause lies in a long-standing and prevalent technical bias within the industry: personnel behavior safety monitoring and work permit management are generally considered two independent technical systems. The core of the monitoring system is identifying unsafe personnel behavior, while the core of the work permit system is work approval management. It is believed that only basic permission data exchange is needed, without recognizing that the dynamic data of the entire "two-permit, three-system" process should be used as the core driver for comprehensive personnel behavior monitoring, enabling adaptive adjustments to monitoring rules, judgment models, and risk thresholds across all scenarios. Therefore, existing technologies have consistently failed to break through the static monitoring design logic and cannot address the fundamental problem of the disconnect between the monitoring system and legally mandated work management regulations. Summary of the Invention

[0009] The purpose of this invention is to disclose a remote safety coordination platform for power construction sites, in order to address the shortcomings of existing map-level task methods.

[0010] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a remote safety coordination platform for power construction sites, comprising: The acquisition module is used to acquire dynamic data of the entire process of the two-ticket and three-system management system output by the two-ticket and three-system management system, as well as the construction site personnel behavior data collected by the intelligent monitoring network; The first generation module is used to dynamically generate a benchmark model for dynamic monitoring of personnel behavior that is adapted to the current working conditions based on the dynamic data of the entire process of the two-ticket three-system. The adjustment module is used to adaptively adjust the full-dimensional monitoring rules for personnel behavior based on the aforementioned dynamic monitoring benchmark model for personnel behavior. The analysis module is used to perform real-time monitoring and analysis of the personnel behavior data at the construction site based on the adjusted full-dimensional monitoring rules for personnel behavior, and to determine the safety monitoring and analysis results. The second generation module is used to generate corresponding safety emergency control instructions based on the safety monitoring and analysis results and send them to the target terminal, and synchronously transmit the safety monitoring and analysis results back to the two-ticket three-system management system to form a closed loop for the entire operation process control.

[0011] Preferably, the dynamic data of the two-ticket, three-system process includes the full life cycle status data of work tickets and operation tickets, basic data of work permits, and work safety management data; wherein, the full life cycle status data includes the status of the ticket as pending approval, effective, in execution, extended, terminated, and invalid; the basic data of work permits includes the scope of permitted work, list of operators, type of work, work time, and voltage level of the work equipment; the work safety management data includes the power outage and restoration status of equipment, the layout of safety measures, and the distribution information of energized equipment.

[0012] Preferably, the step of dynamically generating a dynamic monitoring benchmark model for personnel behavior adapted to the current working conditions based on the dynamic data of the entire process of the two-ticket, three-system system includes: Based on the full lifecycle status data in the dynamic data of the two-ticket three-system process, the execution status of the current ticket is determined. When the ticket is in the effective or in execution state, the benchmark model generation action is triggered. Extract the basic data of work permits and the data of work safety management from the dynamic data of the two-ticket three-system process to determine the compliance authority boundaries, risk level and safety management requirements of the current work. Based on the aforementioned compliance authority boundaries, risk levels, and security control requirements, a dynamic monitoring benchmark model for personnel behavior adapted to the current work conditions is generated.

[0013] Preferably, the multi-dimensional monitoring rules for personnel behavior include: rules for monitoring personnel violations of permissions, rules for monitoring personnel fall risks, rules for monitoring personnel-equipment collisions, and rules for monitoring personnel electric shock risks.

[0014] Preferably, the step of adaptively adjusting the multi-dimensional monitoring rules for personnel behavior based on the dynamic monitoring benchmark model includes: Based on the compliance permission boundaries in the aforementioned personnel behavior dynamic monitoring benchmark model, the power construction permission zoning map and the corresponding compliance standards for each zoning are updated in real time, and the permission verification logic and violation judgment threshold in the personnel violation permission monitoring rules are adaptively adjusted.

[0015] Preferably, the step of adaptively adjusting the multi-dimensional monitoring rules for personnel behavior based on the dynamic monitoring benchmark model for personnel behavior further includes: Based on the risk level and safety control requirements in the dynamic monitoring benchmark model of personnel behavior, the posture recognition model and fall risk threshold in the personnel fall risk monitoring rule, the safety distance threshold in the personnel and equipment collision monitoring rule, and the boundary of the live danger zone and safety protection threshold in the personnel electric shock risk monitoring rule are adaptively adjusted respectively.

[0016] Preferably, the step of performing real-time monitoring and analysis of the construction site personnel behavior data according to the adjusted full-dimensional monitoring rules for personnel behavior, and determining the safety monitoring and analysis results, includes: The on-site personnel behavior data is compared in real time with the adjusted all-dimensional monitoring rules for personnel behavior. Based on the classification standards for violation risk levels, the risk level of the safety monitoring analysis results is determined. The risk levels include normal compliance, general risk, significant risk, and major risk.

[0017] Preferably, the step of generating corresponding security emergency control instructions based on the security monitoring and analysis results and sending them to the target terminal includes: When the risk level of the safety monitoring and analysis results is general risk, a voice warning message is generated on-site and sent to the smart terminal worn by the operator and the on-site safety officer's terminal. When the risk level of the safety monitoring and analysis results is relatively high, a violation prevention instruction and on-site audible and visual alarm information are generated and sent to the on-site control terminal, safety officer terminal and work supervisor terminal. When the risk level of the safety monitoring and analysis results is major risk, an emergency equipment lockout command, a site work stoppage and evacuation command, and the highest level alarm information are generated and sent to the control terminal of the corresponding work equipment, the site emergency broadcast terminal, the project manager's terminal, and the enterprise safety management department's terminal.

[0018] Preferably, the step of synchronously transmitting the safety monitoring and analysis results back to the two-ticket, three-system management system to form a closed loop for the entire operation process control includes: The safety monitoring and analysis results, the corresponding details of violations, and the data on the handling process are synchronously transmitted back to the two-ticket three-system management system and bound and archived with the corresponding work ticket or operation ticket file; When the security monitoring analysis results indicate a significant or major risk, a compliance verification warning is triggered for the corresponding ticket in the two-ticket three-system management system. After the risk is handled and reviewed, the ticket is allowed to enter the final process. Based on the archived safety monitoring and analysis results, a compliance traceability report for the corresponding operation is generated, completing the closed loop of operation process control.

[0019] Preferred options also include: The update module is used to re-trigger the first generation module, adjustment module, and analysis module to perform corresponding operations when changes are detected in the dynamic data of the two-ticket three-system process, so as to update the dynamic monitoring benchmark model of personnel behavior and the full-dimensional monitoring rules of personnel behavior in real time.

[0020] Beneficial effects: This invention breaks down the information silos between the monitoring system and the work permit management system by acquiring real-time dynamic data from the entire "two tickets, three systems" process through an acquisition module. Based on this, the first generation module and the adjustment module dynamically generate and adjust monitoring benchmarks and rules, ensuring that the monitoring logic closely follows the real-time changes in the approval and execution status of permits, thus achieving a deep integration between the safety monitoring system and the statutory work permit management system.

[0021] In addition, this invention also updates the monitoring benchmark (such as the permission zoning map and violation judgment threshold) in real time based on the dynamically changing "two tickets and three systems" data (such as the permitted scope of work and personnel permissions), so that the monitoring rules can accurately match the actual compliance requirements of the current site. This solves the problem of "invalid alarms" caused by the rigidity of rules in the existing static solution, improves the accuracy of monitoring and the acceptance of operators, and ensures that the monitoring function is substantially effective.

[0022] Secondly, this invention uses adjusted monitoring rules to monitor and analyze real-time behavioral data, enabling the identification of violations during the event stage (such as unlicensed operation or operation beyond the scope of the operation). The second generation module immediately generates and sends control commands such as voice warnings and emergency lockouts, realizing a leap from "post-event alarm" to "in-event intervention and interception". This technically eliminates the management drawbacks of "two separate systems of ticket approval and on-site execution". Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of a remote safety coordination platform for power construction sites according to the present invention.

[0025] Figure 2 This is a schematic diagram illustrating the process by which the present invention dynamically generates a benchmark model for monitoring personnel behavior that is adapted to the current working conditions. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] refer to Figure 1 This invention provides a remote safety coordination platform for power construction sites, comprising: The acquisition module is used to acquire dynamic data of the entire process of the two-ticket, three-system management system output by the two-ticket, three-system management system, as well as the on-site personnel behavior data collected by the intelligent monitoring network.

[0028] Specifically, the "Two-Ticket, Three-System" management system is a professional information platform established by power companies in accordance with relevant regulations to achieve standardization, digitalization, and process-oriented operation safety management. Its core function is to conduct online closed-loop management of the entire lifecycle of the two tickets (work tickets and operation tickets), including generation, issuance, authorization, execution, extension, and termination. Simultaneously, it electronically records and verifies the implementation of the three systems (shift handover system, patrol inspection system, and equipment periodic testing and rotation system). The system stores and outputs dynamic data throughout the entire process, including operator permissions, authorized work scope, equipment status, and safety measure deployment, serving as a legally mandated data source for determining the compliance of power construction operations.

[0029] Specifically, the intelligent monitoring network is an IoT monitoring system deployed at power construction sites, consisting of interconnected sensing terminals. It primarily collects real-time data reflecting the behavior of on-site personnel and the state of the environment. Its components typically include: video surveillance units (such as high-definition cameras and AI edge computing cameras) to capture personnel movement trajectories, postures, and intrusion situations; positioning and sensing units (such as UWB positioning tags, smart safety helmets, and wearable devices) to obtain precise personnel location, proximity status, and physiological state; and environmental and equipment monitoring units to sense the electrical status of equipment, environmental parameters, and the status of safety protection facilities. This network can transform unstructured physical scenes of the construction site into structured data on personnel behavior, providing real-time input for safety monitoring and analysis.

[0030] Preferably, the dynamic data of the two-ticket, three-system process includes the full life cycle status data of work tickets and operation tickets, basic data of work permits, and work safety management data; wherein, the full life cycle status data includes the status of the ticket as pending approval, effective, in execution, extended, terminated, and invalid; the basic data of work permits includes the scope of permitted work, list of operators, type of work, work time, and voltage level of the work equipment; the work safety management data includes the power outage and restoration status of equipment, the layout of safety measures, and the distribution information of energized equipment.

[0031] The first generation module is used to dynamically generate a benchmark model for dynamic monitoring of personnel behavior that is adapted to the current working conditions based on the dynamic data of the entire process of the two-ticket three-system.

[0032] Specifically, after obtaining dynamic data for the entire process of the "two-ticket, three-system" system, a dynamic monitoring benchmark model for personnel behavior that is adapted to the current working conditions can be adaptively generated based on the dynamic data of the entire process of the "two-ticket, three-system" system.

[0033] Preferred, such as Figure 2 As shown, the step of dynamically generating a dynamic monitoring benchmark model for personnel behavior adapted to the current working conditions based on the dynamic data of the entire process of the two-ticket, three-system system includes: Based on the full lifecycle status data in the dynamic data of the two-ticket three-system process, the execution status of the current ticket is determined. When the ticket is in the effective or in execution state, the benchmark model generation action is triggered. Extract the basic data of work permits and the data of work safety management from the dynamic data of the two-ticket three-system process to determine the compliance authority boundaries, risk level and safety management requirements of the current work. Based on the aforementioned compliance authority boundaries, risk levels, and security control requirements, a dynamic monitoring benchmark model for personnel behavior adapted to the current work conditions is generated.

[0034] Specifically, after triggering the baseline model generation action, the platform parses and semantically maps the acquired dynamic data of the entire two-ticket, three-system process, extracts the structured fields, and transforms them into parameters recognizable by the monitoring model. The specific implementation logic is as follows: Defining the boundaries of compliance authority: The platform extracts the permitted work area (such as latitude and longitude coordinates, fence, floor, and equipment number), the list of workers (including name, job type, and permission level), and the work time (planned start time and planned end time) from the basic work permit data. By combining this data, the platform forms the spatiotemporal permission boundary of the current work: that is, who can enter which area or equipment at what time. For example, for a valid substation maintenance work permit, the platform will extract the list of maintenance team members specified on the permit as the legal personnel set, extract the No. 3 main transformer and the surrounding 5-meter area as the legal work area, and extract 09:00 to 17:00 on May 20, 2024 as the legal work time period, thereby constructing a unique compliant permission boundary for the work.

[0035] Determining the risk level: The platform automatically assesses risks based on preset risk classification standards, combining the type of work (e.g., working at height, working on live lines, hot work) and the voltage level of the equipment (e.g., 10kV, 110kV, 220kV). For example, when the work type is identified as working at height and the voltage level is 220kV, the platform automatically classifies it as high risk according to the relevant regulations. If it is only ground inspection and the voltage level is 10kV, it is classified as moderate risk. Simultaneously, the platform also extracts information on the distribution of live equipment to identify potential risk sources in the vicinity of live conductors in the work environment.

[0036] Determining safety control requirements: The platform extracts information such as equipment power status (e.g., switched to maintenance or operation status) and safety measure deployment (e.g., grounding wires are connected, barriers are in place, and signs are posted) from operational safety management data. This data is then converted into safety constraints that the monitoring system must follow. For example, if it is found that busbar No. 3 has been switched to cold standby status, the monitoring system updates the corresponding boundary of the energized hazardous area; if it is found that grounding wires are connected, the grounding point is identified as an effective safety protection facility.

[0037] Specifically, after extracting and determining the aforementioned parameters, the platform structures and encapsulates this dynamic data to generate a real-time monitoring baseline model tailored to the current specific task. This model is essentially a task compliance image file containing multi-dimensional dynamic parameters, used to guide subsequent adjustments to monitoring rules. The specific implementation is as follows: Model data structure and instantiation: The platform uses key-value pairs or JSON data format to build a baseline model. The model contains multiple core field clusters. For example: Permission baseline cluster: contains {list of legal user IDs: [user01, user02, ...], legal job area GeoJSON: {...}, job time window: [startTime, endTime]}.

[0038] Risk baseline cluster: contains {Operational risk level: High risk, Voltage level: 220kV, List of coordinates of nearby energized bodies: [x1, y1, z1, ...]}.

[0039] Control baseline cluster: includes {equipment status: under maintenance, energized area: [a1, b1, c1], list of safety measure IDs: [M001, M002, ...]}.

[0040] Dynamic mapping and association: The process of generating models is not simply a matter of piling up data, but rather establishing logical relationships between data points. For example, by overlaying legal work areas with information on the distribution of live equipment, high-risk zones within the work area (i.e., the adjacent boundaries between legal and live areas) are automatically generated. Furthermore, by associating the list of workers with the type of work, standards for verifying the certification of special operations personnel are generated.

[0041] Uniqueness and real-time characteristics of the model: The platform assigns a unique ID to this benchmark model (such as the corresponding work order number) and adds a timestamp. When the work order is in progress, this model is the only effective monitoring benchmark at that moment. Once the work order status changes (such as extension or termination), the platform will trigger a model update or destruction mechanism to ensure that the monitoring benchmark remains highly consistent with the on-site work conditions.

[0042] Through the above methods, the generated dynamic monitoring benchmark model is no longer a static and fixed set of rules, but a digital dynamic image that can accurately reflect all compliance requirements of a specific work order at the current moment, providing a precise and authoritative basis for the adaptive adjustment of subsequent full-dimensional monitoring rules.

[0043] The adjustment module is used to adaptively adjust the full-dimensional monitoring rules for personnel behavior based on the aforementioned dynamic monitoring benchmark model for personnel behavior.

[0044] After obtaining the baseline model for dynamic monitoring of personnel behavior, the rules for full-dimensional monitoring of personnel behavior can be adaptively adjusted according to actual needs.

[0045] Preferably, the multi-dimensional monitoring rules for personnel behavior include: rules for monitoring personnel violations of permissions, rules for monitoring personnel fall risks, rules for monitoring personnel-equipment collisions, and rules for monitoring personnel electric shock risks.

[0046] Furthermore, the personnel violation monitoring rules are used to verify the compliance of on-site personnel's work permissions and real-time location. Specifically, the rules are configured to: determine in real time whether various personnel entering the construction site have the legal authorization to enter the current area at the current time, based on the compliance permission boundaries in the dynamic monitoring benchmark model. The monitoring logic includes, but is not limited to: personnel entering the work area without a ticket, exceeding the work area specified on the ticket, and lingering on site during non-permitted hours, among other violations.

[0047] The personnel fall risk monitoring rule is used to identify the safety status of personnel in high-altitude work scenarios. This rule is configured to: acquire personnel posture and height data through an intelligent monitoring network, and analyze in real time the personnel's edge-to-edge work, climbing behavior, and safety belt suspension status based on the current work risk level in the dynamic monitoring benchmark model. When a personnel is identified as being in a dangerous, high-altitude position and not wearing protective equipment as required, a fall risk warning is triggered.

[0048] Personnel and equipment collision monitoring rules are used to prevent the risk of mechanical injury between mobile machinery and workers. These rules include: real-time tracking of the positional relationship between mobile equipment (such as cranes and excavators) and workers at the construction site, and dynamically calculating the safe distance between them based on the operation type and equipment status in the dynamic monitoring baseline model. When the distance between personnel and equipment is less than a preset safety threshold, a collision risk is determined to exist.

[0049] The personnel electric shock risk monitoring rule is used to prevent electrical safety risks associated with workers approaching or accidentally touching live equipment. Based on the distribution information of live equipment and the power outage / on status of the equipment in the dynamic monitoring benchmark model, the rule delineates the boundaries of live hazardous areas in real time. When the distance between a person, tool, or machinery and a live conductor is less than the prescribed safety protection threshold, an electric shock risk is determined to exist.

[0050] Preferably, the step of adaptively adjusting the multi-dimensional monitoring rules for personnel behavior based on the dynamic monitoring benchmark model includes: Based on the compliance permission boundaries in the aforementioned personnel behavior dynamic monitoring benchmark model, the power construction permission zoning map and the corresponding compliance standards for each zoning are updated in real time, and the permission verification logic and violation judgment threshold in the personnel violation permission monitoring rules are adaptively adjusted.

[0051] Specifically, the power construction permission zoning map is an electronic fence layer dynamically generated based on real-time data in the monitoring system's backend. It divides the physical power construction site (such as substations and pole work areas) into virtual areas with different attributes according to current work permit requirements. For example, it may include permitted work areas, hazardous areas for live equipment, and equipment parking areas. This map is not a static construction drawing, but a digital spatial benchmark used for permission determination, which is updated in real time with the dynamic data of the two-ticket, three-system system.

[0052] Compliance standards are a set of rules that personnel must follow for different areas in the aforementioned zoning map. They define when, who, and how are permitted to enter or remain in which areas. For example, for a permitted work area, the compliance standard might be that only personnel listed on the work permit are allowed to enter during working hours; for a hazardous area with live electrical equipment, the compliance standard would be that entry by any personnel without special permission is strictly prohibited.

[0053] Furthermore, the platform adaptively adjusts the permission verification logic and violation judgment threshold in the personnel violation permission monitoring rules, including: parsing the baseline model and mapping it to a partitioned map and compliance standards. The platform obtains the dynamic monitoring baseline model of personnel behavior created by the first generation module in real time. This model contains key parameters such as the list of legal personnel defined by the current work order, the permitted work scope, and the distribution of live equipment. The platform maps these parameters to a pre-built 3D construction site digital twin scene and performs the following map update operations: Permitted Work Area Activation: The physical coordinate area corresponding to the permitted work scope in the baseline model is rendered as a green permitted work area on the permission partition map. The activation status of this area is bound to the execution status of the ticket; once the ticket is terminated, the area is removed from the map.

[0054] Hazard zone delineation: By combining the distribution of energized equipment in the baseline model with the layout of safety measures, red energized hazard zones and yellow warning zones are automatically calculated and rendered on the zoning map. For example, for equipment under maintenance and with its grounding wire connected, its corresponding energized hazard zone may shrink or disappear; while for adjacent energized operating equipment, its hazard zone boundary is strictly generated according to the safety distance corresponding to the voltage level.

[0055] Compliance Standard Generation: Based on the work time and personnel list in the baseline model, dynamic compliance standards are assigned to each of the generated areas. For example, the compliance standard for a green permit work area is defined as: only personnel listed in the baseline model's list of legitimate personnel IDs are allowed to enter within the work time window specified in the baseline model. Other areas correspond to standards such as prohibiting entry or requiring secondary authorization for entry.

[0056] Secondly, based on the updated map and standards, the permission verification logic is adaptively adjusted.

[0057] After updating the zoning map and compliance standards, the platform uses this dynamic data to adaptively adjust the execution engine of the rules for monitoring personnel's violation of permissions. Dynamic binding of verification logic: The permission verification logic is bound to the dynamically generated area attributes mentioned above. This means that the monitoring system no longer uses a simple logic of fixed "entry prohibited" or "entry permitted" across the entire plant, but instead adopts a four-dimensional composite verification logic of personnel identity + real-time location + current time + dynamic area attributes. For example, when the location tag on the safety helmet indicates that a person has entered a green-permitted work area, the verification logic will automatically extract the compliance standards for that area, determine whether the current time is within the work time window, and whether the person's ID is in the list of legitimate personnel.

[0058] Real-time adjustment of violation judgment thresholds: Based on the risk level and operation type in the dynamic benchmark model, the sensitivity threshold for violation judgment is adaptively fine-tuned. For example, for low-voltage substation operations with a general risk level, a small tolerance time threshold may be set for brief, minor boundary crossings (such as stepping on a line) by personnel (e.g., a violation is judged only after 2 seconds); while for cross-live busbar operations with a major risk level, the platform will automatically adjust this threshold to zero tolerance. Once the positioning signal shows that the physical location of the personnel exceeds the boundary of the permitted operation area, it is immediately judged as a violation, thus achieving adaptive matching between risk level and control precision.

[0059] Through the above process, the personnel violation monitoring rules have been upgraded from static configuration to dynamic adaptation. Every change to a work order and every switch in device status will drive the real-time adjustment of the permission partition map, compliance standards, verification logic, and judgment thresholds, ensuring that the monitoring system always operates on a basis that is completely consistent with the legally required on-site operations.

[0060] Preferably, the step of adaptively adjusting the multi-dimensional monitoring rules for personnel behavior based on the dynamic monitoring benchmark model for personnel behavior further includes: Based on the risk level and safety control requirements in the dynamic monitoring benchmark model of personnel behavior, the posture recognition model and fall risk threshold in the personnel fall risk monitoring rule, the safety distance threshold in the personnel and equipment collision monitoring rule, and the boundary of the live danger zone and safety protection threshold in the personnel electric shock risk monitoring rule are adaptively adjusted respectively.

[0061] Specifically, to address the fall risk in high-altitude work scenarios, the platform implements the following dynamic adjustment logic for the posture recognition model and fall risk threshold: The platform dynamically switches or loads corresponding posture recognition algorithm models based on the risk level and job type in the baseline model. For example, when the baseline model indicates that the current job is a ground inspection job with general risk, the platform loads a basic human standing / squatting / walking posture recognition model; when the baseline model determines that the current job is a high-risk climbing job (such as pole tower work), the platform automatically loads specialized posture recognition sub-models trained specifically for climbing scenarios, such as safety belt hook status recognition, double hook alternating displacement recognition, and fall protection device wearing recognition, thereby improving the targeting of recognition in high-risk scenarios.

[0062] In addition, the platform dynamically adjusts the sensitivity threshold for triggering alarms based on the work height data and risk level in the benchmark model. Specifically, for low-height operations (such as below 2 meters), the platform sets a longer fall detection time window (such as triggering an alarm only if the person does not recover within 3 seconds after losing stability) to reduce false alarms; for high-risk high-altitude operations (such as above 15 meters), the platform automatically shortens the detection time window (such as triggering an alarm 1 second after losing stability) and adjusts the body tilt angle threshold from 45 degrees to 30 degrees, achieving adaptive matching between risk level and control precision.

[0063] Specifically, the platform dynamically adjusts the safe distance threshold based on the risk of mechanical injury between mobile devices and personnel using the following logic: Based on the risk level in the baseline model, and combined with equipment type and working environment, the platform dynamically generates multi-dimensional safety distance thresholds. For example, for manual tool operations in general risk areas, the safety distance threshold can be set to a smaller value; for high-risk operations with large lifting equipment, the platform automatically expands the safety distance threshold to a larger range.

[0064] The system continuously monitors the power supply and movement status of equipment in the baseline model, dynamically adjusting the safety distance threshold. For example, when a crane in the baseline model is detected as stationary, the safety distance threshold remains at its base setting. When the crane enters a dynamic operation state such as rotation, luffing, or movement, the platform automatically amplifies the safety distance threshold and activates a fan-shaped warning zone calculation model for the area in front of the crane's trajectory, achieving real-time linkage between equipment status and safety threshold. Based on the operation type in the baseline model, the platform adaptively activates differentiated collision avoidance strategies. For example, for routine cross-operations between equipment and personnel, a general approach warning strategy is used. For collaborative operations involving equipment debugging or maintenance as indicated by the baseline model, the platform automatically switches to collaborative mode, temporarily suppressing alarms within a certain distance to avoid frequent false alarms interfering with normal collaborative operations. Standard monitoring is restored once personnel exceed the collaborative range.

[0065] Specifically, regarding electrical safety risks, the following dynamic adjustment logic is implemented for the boundaries of energized hazardous areas and safety protection thresholds: Based on the distribution information of energized equipment and the power outage / restoration status of the equipment in the baseline model, the boundary of the energized hazard area in the electronic fence system is updated in real time. Specifically, when the baseline model indicates that a line or busbar has changed from operation to maintenance (i.e., de-energized), the platform automatically removes the corresponding energized hazard area from the monitoring map or marks it as non-energized; when the equipment changes from maintenance to operation (i.e., energized), the platform immediately calculates and generates a new energized hazard area boundary based on the voltage level of the equipment and the prescribed safety distance requirements. For example, for 220kV energized equipment, the platform automatically generates a circular hazard area with a radius of 3 meters or a strip-shaped hazard area along the direction of the equipment.

[0066] Based on the voltage and risk levels in the benchmark model, different safety protection thresholds for approaching live parts are dynamically set. For example, for low-voltage live equipment, the safety protection threshold (i.e., the distance at which an alarm is triggered) is set to a smaller value; for ultra-high-voltage live equipment, the platform automatically increases the safety protection threshold to ensure sufficient safety margin. Simultaneously, the platform makes differentiated adjustments based on the qualifications of the operators in the benchmark model: a stricter safety protection threshold is used when non-electrical professionals approach live areas; for qualified electrical testing personnel, when performing their permitted testing tasks, the platform can appropriately adjust the threshold or switch the monitoring logic according to the safety control requirements in the benchmark model, allowing them to approach the equipment under specific protective measures, thereby achieving refined electric shock protection based on identity, task, and risk.

[0067] The analysis module is used to perform real-time monitoring and analysis of the personnel behavior data at the construction site based on the adjusted full-dimensional monitoring rules for personnel behavior, and to determine the safety monitoring and analysis results.

[0068] Specifically, after obtaining the rules for monitoring personnel behavior across all dimensions, the results of security monitoring analysis can be obtained based on these rules.

[0069] Preferably, the step of performing real-time monitoring and analysis of the construction site personnel behavior data according to the adjusted full-dimensional monitoring rules for personnel behavior, and determining the safety monitoring and analysis results, includes: The on-site personnel behavior data is compared in real time with the adjusted all-dimensional monitoring rules for personnel behavior. Based on the classification standards for violation risk levels, the risk level of the safety monitoring analysis results is determined. The risk levels include normal compliance, general risk, significant risk, and major risk.

[0070] Specifically, in this platform, once the rules for comprehensive monitoring of personnel behavior have completed adaptive adjustments, the analysis module enters real-time monitoring and analysis mode. This process is not simply a threshold-triggered alarm, but a progressive processing flow that includes four core stages: data preprocessing, rule comparison, violation identification, and risk assessment.

[0071] Step 1: Real-time access and preprocessing of personnel behavior data at the construction site: The platform continuously acquires multi-source, heterogeneous personnel behavior data at construction sites through an intelligent monitoring network, including: Location data: Real-time coordinates of personnel from smart safety helmets, UWB positioning tags, or video positioning systems.

[0072] Posture data: Key skeletal data of people from AI video analysis, such as standing, squatting, climbing, lying down and other postures.

[0073] Interactive data includes: the relative distance between personnel and equipment, equipment operating status, and the wearing of safety protective equipment.

[0074] Identity data: Personal identification information obtained through facial recognition or RFID.

[0075] The analysis module first performs spatiotemporal alignment and cleaning filtering on these streaming data, removing invalid data caused by signal drift or obstruction, and forming a standardized single-person behavior snapshot data structure, which includes core fields such as [person ID, timestamp, coordinates, posture, identity].

[0076] Step Two: Real-time comparison with dynamic monitoring rules across multiple dimensions: The platform will perform parallel comparisons between the processed real-time personnel behavior data and the four categories of full-dimensional monitoring rules that have been adjusted according to the dynamic benchmark model. The comparison logic adopts a layered triggering mechanism: First layer: Permission compliance comparison. The [identity + time + coordinates] of personnel are compared with the currently effective compliant permission boundaries in the personnel violation permission monitoring rules to determine whether there is unauthorized entry, overstaying, or operation beyond the scope.

[0077] The second layer: behavioral safety comparison. If the permissions are compliant, the personnel's [posture + coordinates] are further compared with the posture recognition model and risk threshold in the fall risk monitoring rules; the distance between the personnel's [coordinates] and the equipment's [coordinates] is compared with the dynamic safety distance threshold in the collision risk monitoring rules; and the personnel's [coordinates] are compared with the boundary of the live dangerous area in the electric shock risk monitoring rules.

[0078] The third layer: composite risk comparison. For situations where multiple types of risks are triggered simultaneously (such as working at heights at the edge of a live area), the platform initiates composite risk analysis logic to comprehensively assess the cumulative effect of risks.

[0079] Step 3: Risk Identification and Classification Based on Violation Events: During the comparison process, once a person's behavior deviates from the compliance scope or safety threshold stipulated by the monitoring rules, the platform determines that a violation event has occurred. The content of the violation event record includes: the person who violated the rules, the type of violation (permission / fall / collision / electric shock), the time of the violation, and a specific description of the violation (such as entering a prohibited area, not wearing a seat belt, or being too close to a live object).

[0080] Subsequently, the platform classified the violation according to its preset risk level classification standards. The classification logic comprehensively considers the following four dimensions: Degree of behavioral deviation: the distance or extent of deviation from the rule boundary.

[0081] Severity of potential consequences: based on the voltage level, working height, and equipment operating dynamics of the current working environment.

[0082] Personnel Status: Whether personnel have emergency response capabilities and whether they are out of control.

[0083] Cumulative effect: Whether the same person triggers the same or different types of warnings multiple times within a short period of time.

[0084] Based on the above dimensions, the platform classifies the security monitoring and analysis results into the following four risk levels: Normal compliance: Judgment criteria: The personnel's behavior was entirely within the scope permitted by the dynamic monitoring rules and did not trigger any violation conditions.

[0085] Handling method: The system only records background data and does not generate any external output or intervention.

[0086] General risks: Judgment criteria: Personnel's behavior slightly deviates from the rule boundaries, but does not yet pose a direct threat to personal or equipment safety, or is in the early stages of violation. For example: personnel approaching but not yet entering the warning line of a live hazardous area (such as lingering outside the distance threshold); workers performing work at heights briefly exhibiting non-edge-adjacent movements but quickly correcting themselves; personnel briefly stepping on the boundary line of the permitted work area.

[0087] The classification logic is as follows: the deviation is slight (e.g., the deviation from the threshold is less than 10%), and the potential consequences are low voltage level or no direct risk of personal injury.

[0088] Analysis results output: This level serves as a warning, intended to alert the individual.

[0089] Significant risk: Judgment criteria: Personnel's behavior clearly violates the rules and has created a definite safety hazard. If not stopped in time, it is highly likely to cause personal injury or equipment accident. For example: personnel enter the permitted work area or the live electrical hazard area without authorization; workers at heights do not wear safety belts in the vicinity of the edge; the safe distance between personnel and operating mobile equipment is insufficient.

[0090] The classification logic is to clearly define the threshold for triggering violations of permissions or safety rules, and the potential consequences involve clear risks of personal injury (such as falls from heights or mechanical collisions).

[0091] Analysis results output: This level serves as the trigger condition for a mandatory stop order, requiring immediate on-site intervention.

[0092] Major risks: Judgment criteria: The person's behavior has caused or is about to cause irreversible serious consequences, and the person is in a critical state of accident occurrence or an accident has already occurred. For example: a person enters an ultra-high voltage live area and is extremely close; a person falls from a height and loses stability (attitude data shows characteristics of weightlessness and free fall); a mobile device is about to hit a person and shows no signs of deceleration.

[0093] The grading logic is based on AI video analysis or sensor data to identify critical scenarios that meet preset accident thresholds or accident characteristics. This type of determination typically uses a zero-tolerance threshold; once identified, the highest level of emergency response is immediately triggered.

[0094] Analysis results output: This level serves as a trigger condition for emergency locking and emergency evacuation, requiring the system to automatically execute emergency intervention measures.

[0095] Step 4: Generation and encapsulation of security monitoring and analysis results: After the classification is completed, the platform encapsulates the above analysis process into a standardized security monitoring analysis result data package. An example of the data package structure is as follows: Basic information: [Timestamp], [Location], [Personnel ID], [Associated Ticket Number].

[0096] Analysis conclusion: [Risk level] (Normal compliance / General risk / Significant risk / Major risk).

[0097] Violation details: [Violation type], [Specific violation description] (e.g., Person Zhang San entered the prohibited area at 09:30:15, exceeding the boundary by 1.5 meters), [Trigger rule ID].

[0098] Through the above implementation methods, this platform can not only detect violations in real time, but also achieve precise risk stratification based on dynamic monitoring benchmarks and scientific grading standards, providing accurate decision-making basis for subsequent graded emergency control, thereby solving the problem of extensive monitoring in existing technologies that only alarms, do not grade, and are difficult to respond to.

[0099] The second generation module is used to generate corresponding safety emergency control instructions based on the safety monitoring and analysis results and send them to the target terminal, and synchronously transmit the safety monitoring and analysis results back to the two-ticket three-system management system to form a closed loop for the entire operation process control.

[0100] Preferably, the step of generating corresponding security emergency control instructions based on the security monitoring and analysis results and sending them to the target terminal includes: When the risk level of the safety monitoring and analysis results is general risk, a voice warning message is generated on-site and sent to the smart terminal worn by the operator and the on-site safety officer's terminal. On-site voice warnings are alerts issued for general risks. They are delivered via pre-set voice prompts broadcast through smart terminals, such as "You are approaching a live area, please be careful" or "Do not linger in the boundary area." The core function of this information is to immediately remind the person involved to be aware of any behavioral deviations, allowing them to correct the risk themselves and prevent escalation. The messages are sent to the worker wearing a smart safety helmet or handheld terminal and the on-site safety officer, aiming to achieve early intervention without interfering with normal operations.

[0101] When the risk level of the safety monitoring and analysis results is relatively high, a violation prevention instruction and on-site audible and visual alarm information are generated and sent to the on-site control terminal, safety officer terminal and work supervisor terminal. Violation cessation instructions are mandatory intervention commands issued for significant risks. Their core characteristic is a clear intent to stop the violation; the instructions typically state to cease current work, immediately evacuate the danger zone, or immediately fasten safety belts. These instructions are not only broadcast verbally but may also be continuously displayed as a prominent pop-up on the receiving terminal until the violation is rectified. The recipients of these instructions extend to on-site safety officers and work supervisors, requiring on-site management personnel to intervene and supervise the immediate correction of violations, achieving mandatory control during the process. On-site audible and visual alarms are usually sent simultaneously with the cessation instructions, serving as an enhanced environmental warning method. They manifest as strong flashing red lights and high-frequency sirens emitted by sound columns, alarm lights, or warning posts installed at the work site. The core function of this information is to visually and audibly cover the entire work area, alerting not only the violator but also other workers in the violator to the danger, while simultaneously providing guidance for on-site management personnel to quickly locate the violation, creating an urgent atmosphere that indicates a high risk has been activated.

[0102] When the risk level of the safety monitoring and analysis results is major risk, an emergency equipment lockout command, a site work stoppage and evacuation command, and the highest level alarm information are generated and sent to the control terminal of the corresponding work equipment, the site emergency broadcast terminal, the project manager's terminal, and the enterprise safety management department's terminal.

[0103] The equipment emergency lockout command is the highest-level technical intervention command issued in response to major risks. Its core lies in direct linkage with the control systems of on-site intelligent equipment (such as cranes, construction hoists, power cabinets, etc.) through a system interface. When the platform determines that a major risk that will lead to personal injury or death has occurred (such as personnel entering a mechanical blind spot and being unable to evacuate), it automatically sends a stop or braking signal to the equipment's control terminal, forcing the equipment to stop operating or enter a safety lockout state. This command achieves a leap from human-based prevention to technical prevention, cutting off the risk source through physical means and preventing accidents. The on-site work stoppage and evacuation command is an organized emergency command issued in response to major risks. It is manifested as a standardized emergency evacuation notice simultaneously released through the on-site emergency broadcast system, large-screen display system, and all personnel terminals. The command content typically includes a message to all personnel: "Attention, a major risk has occurred on-site. Please evacuate immediately along the green channel to the assembly point." The core function of this command is to quickly organize all non-essential personnel to evacuate in an orderly manner, avoiding secondary injuries caused by chaos and creating a safe environment for emergency response. The highest-level alarm information is a management escalation alarm issued in response to major risks. The alerts are delivered via encrypted messages to remote management personnel through various channels, including SMS, app push notifications, and pop-ups on monitoring platforms. The messages are highly condensed, including the incident type, risk level, precise location, and preliminary video evidence. These messages are sent to project managers and the company's safety management department to ensure that management is informed of the emergency situation immediately, promptly activates the company-wide emergency response plan, meets regulatory requirements for immediate reporting of major incidents, and preserves complete evidence for subsequent incident investigations.

[0104] Preferably, the step of synchronously transmitting the safety monitoring and analysis results back to the two-ticket, three-system management system to form a closed loop for the entire operation process control includes: The safety monitoring and analysis results, the corresponding details of violations, and the data on the handling process are synchronously transmitted back to the two-ticket three-system management system and bound to the corresponding work ticket or operation ticket file for archiving.

[0105] The platform establishes a standard two-way data interface with the two-ticket, three-system management system, and uses a web service or message queue mechanism for real-time data synchronization. The specific implementation steps are as follows: After a monitoring and analysis process is completed (regardless of whether a violation occurred), the platform will structure and encapsulate the security monitoring and analysis results (including risk level), details of the violation (such as violation type, time, location, on-site screenshots or short video evidence), and data on the handling process (such as sent control instructions, received confirmation information, and violation elimination time) to form a complete security monitoring record package.

[0106] The record package contains a work order or operation ticket number that uniquely corresponds to the current operation. Upon receiving the data packet, the two-ticket, three-system management system automatically retrieves the corresponding electronic ticket file based on this number and permanently attaches the safety monitoring record package as a copy or attachment to the ticket file. Thus, each work order or operation ticket not only contains traditional approval process records but also links to all real-time monitoring evidence generated during its execution, forming a complete data chain of ticket + action.

[0107] When the security monitoring and analysis results indicate a significant or major risk, a compliance verification warning is triggered for the corresponding ticket in the two-ticket, three-system management system. After the risk is handled and reviewed, the ticket is allowed to enter the final process.

[0108] When the risk level of the security monitoring and analysis results reaches a relatively high risk or a major risk, the platform triggers a higher level of management intervention, namely, controlling the lifecycle status of the ticket itself. The specific implementation steps are as follows: When the platform sends data back to the "Two-Ticket, Three-System" management system, it simultaneously sends a compliance verification warning signal. This signal includes the risk level, a violation summary, and a warning timestamp. Upon receiving this warning signal, the "Two-Ticket, Three-System" management system automatically marks the status of the ticket (e.g., marked as compliance abnormal) and locks its termination process. At this point, even if the user clicks to terminate or close the ticket in the ticket system, the system will refuse to execute due to the failure of compliance verification and will display a prompt: "A significant / major risk event occurred during the execution of this ticket, and it can only be terminated after handling and review." After completing the risk handling (e.g., educating the violators and implementing corrective measures), on-site management personnel (such as the work supervisor and safety officer) must submit a handling completion report on the platform or the "Two-Ticket" system. The authorized project manager or safety supervision department will retrieve the monitoring evidence chain through the system for online review. After confirming that the risk has been completely eliminated, they will click "Review Passed." At this point, the "Two-Ticket, Three-System" management system will unlock the ticket, allowing it to enter the termination process normally. If the review fails, it will remain locked and the alarm will be escalated.

[0109] Based on the archived safety monitoring and analysis results, a compliance traceability report for the corresponding operation is generated, completing the closed loop of operation process control.

[0110] After the final archiving of tickets, the platform automatically performs data aggregation and report generation operations to support subsequent regulatory audits. The specific implementation steps are as follows: The platform aggregates all monitoring and analysis results generated during the entire period from the effective date of the ticket to its termination (including normal compliance records, risk events at all levels, handling processes, and review records).

[0111] Based on aggregated data, the platform uses a preset report template to automatically generate a "Compliance Traceability Report for the Entire Workflow." This report typically includes the following sections: Basic information about the task: ticket number, task content, time, location, and personnel.

[0112] Compliance Overview: Overall security status assessment.

[0113] Risk event details: List the analysis results, screenshots of evidence of violations, and corresponding handling instructions for each risk event in a timeline.

[0114] Handling and review records: handling measures for major / significant risk events, reviewers and review times.

[0115] Attachments to the complete chain of evidence: key video clips and location tracking maps of all violations.

[0116] Archiving and For Reference: The generated reports are automatically stored in the archive of the Two Tickets and Three Systems Management System, and are connected to the Enterprise Resource Planning System or the government regulatory data platform to meet the audit requirements of regulatory departments such as the National Energy Administration for compliance traceability of the entire process of power construction safety, thus solving the defect of the disconnect between early warning results and compliance traceability in the existing technology.

[0117] Preferred options also include: The update module is used to re-trigger the first generation module, adjustment module, and analysis module to perform corresponding operations when changes are detected in the dynamic data of the two-ticket three-system process, so as to update the dynamic monitoring benchmark model of personnel behavior and the full-dimensional monitoring rules of personnel behavior in real time.

[0118] Specifically, the update module establishes a real-time data synchronization channel with the two-ticket, three-system management system based on message middleware or database log capture technology. This module runs continuously as a daemon process, monitoring the dynamic data of the entire two-ticket, three-system process in real time. The types of change events monitored include, but are not limited to: Changes in ticket status: such as a work ticket changing from effective to in progress, or from in progress to postponement, as well as the progress of the operation ticket execution steps, the termination or invalidation of the ticket.

[0119] Changes to basic permit data: such as coordinate shifts or range scaling due to adjustments in site conditions, changes in personnel IDs due to temporary additions or replacements, or approved extensions of work time due to weather or schedule reasons.

[0120] Changes in safety management data: such as the equipment status changing from operation to maintenance or vice versa after the power outage and restoration operation is completed; changes in the arrangement of safety measures, such as the grounding wire being connected or removed, or the barrier being moved.

[0121] When any of the above data changes, the two-ticket three-system management system pushes a real-time event notification to the update module, which includes the change type, change content, and associated ticket number.

[0122] Specifically, upon receiving a change event notification, the update module immediately initiates the linkage triggering logic. This logic follows the principle of "change triggers, triggers update," and specifically performs the following operations: First, the change event is verified to confirm that the associated ticket is still within its valid lifecycle at the current moment (i.e., not in an invalid state). For changes to expired or invalid tickets, the update module does not process them to avoid invalid triggers.

[0123] The update module sends a regeneration command to the first generation module, carrying the updated dynamic data of the two-ticket, three-system process. Based on this, the first generation module re-executes the baseline model generation action, that is, re-parses the execution status of the current ticket, re-extracts the basic data of the work permit and the work safety control data, and generates a brand-new baseline model for dynamic monitoring of personnel behavior that reflects the changed work conditions.

[0124] Once the new baseline model is generated, the update module automatically triggers the adjustment module. The adjustment module takes the new baseline model as input and re-executes the rule adaptive adjustment action, that is, it comprehensively refreshes the rules for monitoring personnel violation permissions, fall risk, collision risk, and electric shock risk based on the new compliance permission boundaries, risk levels, and security control requirements.

[0125] After the adjustment module completes the rule update, the update module notifies the analysis module to perform a hot switch of the monitoring rules. The analysis module smoothly transitions the running old version of the monitoring rules to the new version, ensuring that the monitoring process is uninterrupted and does not require a restart, achieving a seamless update. Thereafter, all subsequent real-time monitoring and analysis are executed based on the updated rules.

[0126] The following example illustrates the complete workflow of the update module: Suppose that the work order for a substation maintenance operation was originally scheduled for 09:00-17:00, and the personnel list included Zhang San and Li Si. At 10:00 AM, due to weather conditions, the work supervisor applied through the work order system to extend the work time to 18:00 and added Wang Wu to the personnel list.

[0127] Change capture: After the two-ticket system approves the extension and updates the personnel list, the update module monitors the changes in work time and personnel list in real time.

[0128] Trigger update: The update module sends a regeneration command to the first generation module.

[0129] Baseline Model Update: The first generation module generates a new baseline model based on the new work time window and personnel list, in which the compliance permission boundary has been updated to allow Zhang San, Li Si, and Wang Wu to enter the work area between 09:00 and 18:00.

[0130] Rule Adjustment: The adjustment module is based on the new benchmark model. The permission verification logic in the personnel violation permission monitoring rules will be updated synchronously. Wang Wu's ID will be added to the whitelist of legitimate personnel, and the job completion time threshold will be adjusted from 17:00 to 18:00.

[0131] Real-time monitoring: The analysis module completes hot-switching of rules. When Wang Wu enters the work area in the afternoon, the platform determines that he is compliant based on the updated rules and no longer issues false alarms; at 17:10, if Zhang San is still in the work area and has not triggered other violations, the platform determines that his behavior is normal based on the updated 18:00 threshold and will not trigger an overtime stay alarm.

[0132] Through the above implementation methods, the update module ensures that the platform can be fully synchronized with the management rhythm of the "two-ticket, three-system" approach. Every change in ticket and every change in equipment status can instantly drive the self-evolution of the monitoring benchmark, ensuring that the monitoring system always operates in a state that is perfectly in line with the legal requirements on site. This fundamentally solves the root cause defect of the disconnect between static rules and dynamic operating conditions in existing technologies.

[0133] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A remote safety coordination platform for power construction sites, characterized in that, include: The acquisition module is used to acquire dynamic data of the entire process of the two-ticket and three-system management system output by the two-ticket and three-system management system, as well as the construction site personnel behavior data collected by the intelligent monitoring network; The first generation module is used to dynamically generate a benchmark model for dynamic monitoring of personnel behavior that is adapted to the current working conditions based on the dynamic data of the entire process of the two-ticket three-system. The adjustment module is used to adaptively adjust the full-dimensional monitoring rules for personnel behavior based on the aforementioned dynamic monitoring benchmark model for personnel behavior. The analysis module is used to perform real-time monitoring and analysis of the personnel behavior data at the construction site based on the adjusted full-dimensional monitoring rules for personnel behavior, and to determine the safety monitoring and analysis results. The second generation module is used to generate corresponding safety emergency control instructions based on the safety monitoring and analysis results and send them to the target terminal, and synchronously transmit the safety monitoring and analysis results back to the two-ticket three-system management system to form a closed loop for the entire operation process control.

2. The remote safety coordination platform for power construction sites according to claim 1, characterized in that, The dynamic data for the entire process of the "two-ticket, three-system" includes the full lifecycle status data of work tickets and operation tickets, basic data of work permits, and data on work safety management. Among them, the full lifecycle status data includes the status of the ticket as pending approval, effective, in execution, extended, terminated, and invalid; the basic data of work permits includes the scope of permitted work, the list of operators, the type of work, the work time, and the voltage level of the equipment; the work safety management data includes the power outage and restoration status of equipment, the layout of safety measures, and the distribution information of live equipment.

3. The remote safety coordination platform for power construction sites according to claim 1, characterized in that, The step of dynamically generating a dynamic monitoring benchmark model for personnel behavior adapted to the current work conditions based on the dynamic data of the entire process of the two-ticket, three-system system includes: Based on the full lifecycle status data in the dynamic data of the two-ticket three-system process, the execution status of the current ticket is determined. When the ticket is in the effective or in execution state, the benchmark model generation action is triggered. Extract the basic data of work permits and the data of work safety management from the dynamic data of the two-ticket three-system process to determine the compliance authority boundaries, risk level and safety management requirements of the current work. Based on the aforementioned compliance authority boundaries, risk levels, and security control requirements, a dynamic monitoring benchmark model for personnel behavior adapted to the current work conditions is generated.

4. The remote safety coordination platform for power construction sites according to claim 1, characterized in that, The comprehensive monitoring rules for personnel behavior include: rules for monitoring personnel violations of permissions, rules for monitoring personnel fall risks, rules for monitoring personnel and equipment collisions, and rules for monitoring personnel electric shock risks.

5. The remote safety coordination platform for power construction sites according to claim 4, characterized in that, The adaptive adjustment of the multi-dimensional monitoring rules for personnel behavior based on the dynamic monitoring benchmark model includes: Based on the compliance permission boundaries in the aforementioned personnel behavior dynamic monitoring benchmark model, the power construction permission zoning map and the corresponding compliance standards for each zoning are updated in real time, and the permission verification logic and violation judgment threshold in the personnel violation permission monitoring rules are adaptively adjusted.

6. The remote safety coordination platform for power construction sites according to claim 4, characterized in that, The adaptive adjustment of the multi-dimensional monitoring rules for personnel behavior based on the dynamic monitoring benchmark model for personnel behavior also includes: Based on the risk level and safety control requirements in the dynamic monitoring benchmark model of personnel behavior, the posture recognition model and fall risk threshold in the personnel fall risk monitoring rule, the safety distance threshold in the personnel and equipment collision monitoring rule, and the boundary of the live danger zone and safety protection threshold in the personnel electric shock risk monitoring rule are adaptively adjusted respectively.

7. The remote safety coordination platform for power construction sites according to claim 1, characterized in that, The process involves real-time monitoring and analysis of the construction site personnel behavior data based on the adjusted multi-dimensional monitoring rules, to determine the safety monitoring and analysis results, including: The on-site personnel behavior data is compared in real time with the adjusted all-dimensional monitoring rules for personnel behavior. Based on the classification standards for violation risk levels, the risk level of the safety monitoring analysis results is determined. The risk levels include normal compliance, general risk, significant risk, and major risk.

8. The remote safety coordination platform for power construction sites according to claim 7, characterized in that, The step of generating corresponding security emergency control instructions based on the security monitoring and analysis results and sending them to the target terminal includes: When the risk level of the safety monitoring and analysis results is general risk, a voice warning message is generated on-site and sent to the smart terminal worn by the operator and the on-site safety officer's terminal. When the risk level of the safety monitoring and analysis results is relatively high, a violation prevention instruction and on-site audible and visual alarm information are generated and sent to the on-site control terminal, safety officer terminal and work supervisor terminal. When the risk level of the safety monitoring and analysis results is major risk, an emergency equipment lockout command, a site work stoppage and evacuation command, and the highest level alarm information are generated and sent to the control terminal of the corresponding work equipment, the site emergency broadcast terminal, the project manager's terminal, and the enterprise safety management department's terminal.

9. The remote safety coordination platform for power construction sites according to claim 1, characterized in that, The process of synchronously transmitting the safety monitoring and analysis results back to the two-ticket, three-system management system to form a closed loop for the entire operation process control includes: The safety monitoring and analysis results, the corresponding details of violations, and the data on the handling process are synchronously transmitted back to the two-ticket three-system management system and bound and archived with the corresponding work ticket or operation ticket file; When the security monitoring analysis results indicate a significant or major risk, a compliance verification warning is triggered for the corresponding ticket in the two-ticket three-system management system. After the risk is handled and reviewed, the ticket is allowed to enter the final process. Based on the archived safety monitoring and analysis results, a compliance traceability report for the corresponding operation is generated, completing the closed loop of operation process control.

10. The remote safety coordination platform for power construction sites according to claim 1, characterized in that, Also includes: The update module is used to re-trigger the first generation module, adjustment module, and analysis module to perform corresponding operations when changes are detected in the dynamic data of the two-ticket three-system process, so as to update the dynamic monitoring benchmark model of personnel behavior and the full-dimensional monitoring rules of personnel behavior in real time.

Images