A safety detection method and system suitable for narrow space power construction

Abstract

The application discloses a safety detection method and system suitable for narrow space electric power construction, relates to the field of electric power construction, and is innovative in adopting a multilevel sensor network for different risk level areas, establishing a physical special line life channel completely independent of a conventional data network for a highest danger signal, realizing a microsecond level closed loop from sensing to execution, ensuring absolute reliability of the highest level safety signal, reducing false positives and false negatives caused by sensor drift by deploying a calibration column integrated with high precision reference on site and setting a mechanism of automatically triggering calibration after periodic and event, setting a threshold and a phased response strategy, trying to automatically eliminate risks in priority under the premise of ensuring safety, avoiding unnecessary construction interruption caused by slight abnormalities or transient interference, and improving construction efficiency and intelligent level of emergency response.

Description

Technical Field

[0001] This invention relates to the field of power construction, and in particular to a method and system for safety inspection during power construction in confined spaces. Background Technology

[0002] When carrying out construction and maintenance of power equipment in confined spaces such as substations, underground utility tunnels, and cable layers, the working environment is complex, space is limited, and high-risk equipment is densely packed.

[0003] Current common practices mainly rely on manual inspections and the installation of simple fixed sensors and independent fire or gas alarm devices, which have the following drawbacks: Existing sensor networks focus on only a single parameter and lack the ability to perceive risks introduced by construction dynamics in real time. Alarm signals mostly rely on general industrial buses or wireless networks for transmission, which cannot cope with the deadly dangers that develop at the microsecond level, such as electric arc light.

[0004] Sensors are prone to drift and aging after long-term operation, which can lead to distorted measurement data and potential false alarms or missed alarms. Currently, there is a lack of automatic online calibration methods for harsh industrial environments.

[0005] Most existing systems only set fixed thresholds for a single monitoring parameter to trigger over-limit alarms. They cannot integrate and analyze multi-dimensional information such as temperature, partial discharge, gas concentration, and harmonics to make trend judgments. They can only trigger alarms at the critical point of an accident or after it has occurred, thus missing the best opportunity for intervention.

[0006] Therefore, we propose a safety inspection method and system for electrical construction in confined spaces to solve the above problems. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for safety inspection during power construction in confined spaces, in order to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method and system for safety inspection during electrical construction in confined spaces, comprising the following steps: S1. Divide the monitoring space into safe areas, critical areas and high-risk areas, and deploy low-power wireless sensor networks, high-precision wired sensor networks and hard-wired sensors based on physical dedicated lines accordingly. S2. Perform periodic or event-triggered online calibration of the sensor, and acquire and update the sensor data correction module; S3. Receive hard-wired sensor signals from the high-risk area, process them in microseconds through the dedicated fast channel of the edge gateway, and drive the local actuator to perform corresponding fast protection actions according to the signal type. S4. Receive sensor data from safe and critical areas, perform data feature fusion and risk value calculation, and execute corresponding early warning or control operations based on the risk level threshold of the comprehensive risk value.

[0009] Preferably, step S1 includes: Wireless sensors are deployed in a safe area. By default, they are in a low-power, timed monitoring mode. When a person's location tag is detected, the sensor is activated and enters a continuous monitoring mode. Deploy high-precision wired sensors on key equipment in key areas, and use industrial Ethernet for continuous monitoring and real-time data reporting; Hardwired sensors are deployed in extremely high-risk locations within high-risk areas. The sensor signals are directly connected to the edge gateway's fast channel via physical cables, forming a dedicated physical line independent of the conventional data network.

[0010] Preferably, step S2 includes: A calibration column is set up in the monitoring space. The calibration column integrates a high-precision reference sensor and a standard gas concentration simulator. Set calibration trigger conditions, which include reaching a preset time interval or the system completing a single alarm event; When any trigger condition is met, the calibration column is controlled by the edge gateway to release calibration gas or simulated reference signal, and data from the corresponding surrounding sensors is read. Calculate the deviation between the sensor-reported data and the calibration column reference value, and update the calibration coefficients to correct the original sensor data; If the sensor data deviation continues to exceed the maximum allowable threshold, the sensor is marked as requiring maintenance and a maintenance work order is generated.

[0011] Preferably, in step S3, the local actuator is driven to perform the corresponding fast protection action according to the signal type, specifically including: Upon receiving an arc light signal, execute a millisecond-level trip command and trigger an audible and visual alarm; If the received gas concentration data reaches the first threshold, the forced ventilation will be activated. If the concentration continues to rise to a higher second threshold, an audible and visual alarm and a shutdown command will be triggered. Upon receiving the emergency stop button signal, immediately cease construction operations in the area.

[0012] Preferably, after the forced exhaust is started, the rate of decrease in gas concentration is used as feedback. If the rate of decrease does not meet the requirements, the shutdown command is executed.

[0013] Preferably, step S4 involves data feature fusion and risk value calculation, specifically including: Multiple risk features were extracted from the sensor data, including: temperature and heating rate, three-phase imbalance, total harmonic distortion, UHF partial discharge pulse count, and gas concentration change rate. Map and merge the various risk characteristic values ​​into a single comprehensive risk value; The overall risk value is compared with preset thresholds for high, medium, low, and no risk levels to determine the current risk level.

[0014] Preferably, in step S4, the corresponding early warning or control operation is performed, specifically as follows: Determined to be high-risk, work was forced to stop and personnel were evacuated; If the risk level is determined to be medium, triggering warning lights or dashboard alarms, it is recommended to suspend operations. If the risk level is determined to be low, a broadcast announcement or signboard notification will be made. If the risk is deemed to be no, periodic data monitoring and reporting will be conducted.

[0015] Preferably, the power construction safety inspection method further includes an incident handling step: Reported data is categorized into alarm events and abnormal events; Alarm events are recorded, data is recorded, preset operations are executed, diagnostic conclusions and maintenance work orders are generated, and sent to mobile terminals. For abnormal events, implement corresponding measures and continuously monitor them. Based on the monitoring feedback, decide whether to lift the alarm or escalate it to generate early warnings and handling suggestions.

[0016] A safety monitoring system for electrical construction in confined spaces, comprising: The partitioned sensor network module includes low-power wireless sensor clusters deployed in secure areas, high-precision wired sensors deployed in critical equipment, and edge hardwired sensors deployed in high-risk equipment and directly connected via physical dedicated lines. The edge gateway includes a fast channel and a fast rule engine for microsecond-level processing of hardwired sensor signals, as well as a computing unit for processing regular sensor data. The online calibration module includes a calibration column installed on-site and a calibration logic unit for controlling the calibration process and calculating and storing calibration coefficients; The data fusion and risk assessment module is used to extract and fuse features from sensor data from safe and critical areas, calculate the comprehensive risk value, and determine the risk level. The execution and early warning output module is used to drive the corresponding local executor or output early warning information based on the instructions of the fast rule engine or the risk level determined by the risk assessment module.

[0017] Preferably, the edge gateway's fast channel and fast rule engine are physically and logically independent of the computing unit and network that process regular data, and the signals from hardwired sensors are directly connected to the digital input interface of the fast channel via physical cables.

[0018] The technical effects and advantages of this invention are as follows: 1. For areas with different risk levels, a multi-level sensor network is innovatively adopted to establish a dedicated physical lifeline channel for the highest risk signals that is completely independent of the conventional data network. This achieves a microsecond-level closed loop from perception to execution, ensuring the absolute reliability of the highest level of safety signals.

[0019] 2. By deploying calibration columns with integrated high-precision benchmarks on-site and setting up mechanisms for periodic and event-triggered automatic calibration, the online accuracy verification and coefficient correction of key sensors are automatically completed during construction, improving the long-term reliability of monitoring data and reducing false alarms and missed alarms caused by sensor drift.

[0020] 3. By extracting and fusing features from multi-dimensional sensor data such as temperature, partial discharge, gas, and electrical parameters, a dynamic comprehensive risk value is generated to identify the evolution trend of potential hazards and issue low-level warnings in the early stages of risk accumulation, thereby transforming passive response into proactive prevention.

[0021] 4. By setting threshold-based and phased response strategies, and prioritizing the automatic elimination of risks while ensuring safety, unnecessary construction interruptions caused by minor anomalies or momentary disturbances are avoided, thereby improving construction efficiency and emergency response levels. Attached Figure Description

[0022] Figure 1 A schematic diagram of the overall process of invention; Figure 2 Flowchart for invention area division and sensor deployment; Figure 3 Flowchart for dividing the invention area; Figure 4 A flowchart for the invention of sensor calibration; Figure 5 To develop a flowchart for handling high-risk areas; Figure 6 A flowchart for data fusion and risk assessment of safe and critical areas was developed. Figure 7 To develop an abnormal event handling flowchart; Figure 8 The flowchart for the invention system. Detailed Implementation

[0023] 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.

[0024] This invention provides, for example Figures 1-8 The method and system for safety inspection during electrical construction in confined spaces, as shown, include the following steps: S1. Divide the monitoring space into safe areas, critical areas and high-risk areas, and deploy low-power wireless sensor networks (i.e., the first sensor network), high-precision wired sensor networks (i.e., the second sensor network), and hard-wired sensors based on physical dedicated lines (i.e., the third sensor network) accordingly. S2. Perform periodic or event-triggered online calibration on all the above sensors, and acquire and update the sensor data correction module; S3. Receive hard-wired sensor signals from the high-risk area, process them in microseconds through the dedicated fast channel of the edge gateway, and drive the local actuator to perform corresponding fast protection actions according to the signal type. S4. Receive sensor data from the safe area and the critical area, perform data feature fusion and risk value calculation, and execute corresponding early warning or control operations based on the risk level threshold of the comprehensive risk value.

[0025] Based on step S1, including: Wireless sensors are deployed in a safe area. By default, they are in a low-power, timed monitoring mode. When a person's location tag is detected, the sensor is activated and enters a continuous monitoring mode. Deploy high-precision wired sensors on key equipment in key areas, and use industrial Ethernet for continuous monitoring and real-time data reporting; Hardwired sensors are deployed in extremely high-risk locations within high-risk areas. The sensor signals are directly connected to the edge gateway's fast channel via physical cables, forming a dedicated physical line independent of the conventional data network.

[0026] Step S1 specifically involves: initially monitoring multiple areas and dividing the monitored area into three zones, namely: a safety zone, which is a personnel activity area such as cable layer passages and equipment room corridors, and an area without directly high-risk equipment; and a wireless sensor network, which can be a wireless temperature, humidity, and smoke sensor network. The wireless sensors initially perform low-power timed monitoring by default and are normally in a low-power sleep state, but can be woken up periodically to report. During use, it is determined whether a personnel positioning tag has entered the area. When a construction worker wearing a positioning tag enters the area, it triggers the surrounding sensor cluster to wake up, exits the low-power sleep state, and enters continuous monitoring mode, increasing its detection frequency. Thus, when a construction worker enters the area, the environment is effectively monitored. When no construction worker enters the area, it remains in a low-power sleep state and reports periodically, avoiding energy waste. Critical Area: This area serves as the placement area for critical equipment involved in power construction. High-precision wired sensors are installed on the critical equipment involved in construction. The critical equipment can be switchgear connectors and main transformer bushings. Real-time monitoring of the critical equipment ensures the safety of construction personnel during construction and detects the status of the critical equipment. The high-precision wired sensors can be set as temperature sensors and UHF partial discharge sensors. The high-precision wired sensors provide continuous monitoring throughout the construction period, and the monitored data is continuously monitored and reported in real time via industrial Ethernet.

[0027] High-risk areas: These areas are where extremely high-risk equipment is placed during power construction. Hardwired sensors are deployed at extremely high-risk locations at the construction site. Their signals are directly connected to the fast digital input channel of the edge gateway. The signals are transmitted directly through physical cables without any network protocol encapsulation, routing, or software queue processing, achieving microsecond-level delay from sensing to gateway reception. This enables the monitoring and reporting of rapidly changing, potentially fatal, dangers such as electric arcs and instantaneous gas concentration spikes. The hardwired sensors deployed can be arc light sensors and gas sensors. Arc light sensors are installed at the construction point and near the busbar in the switch cabinet, while gas sensors are installed under the cable layer and GIS gas chamber. Gas sensors can be methane gas sensors and carbon monoxide gas sensors, or gas sensors can be installed according to the gas monitoring needs of confined spaces. The sensor signals are directly connected to the edge gateway fast channel, forming a physical dedicated line independent of the conventional data network. When the gas concentration detected by the gas sensor reaches a relatively high proportion of the lower explosive limit, the non-explosion-proof power supply is immediately cut off and the strong exhaust is started to exchange air and reduce the gas concentration.

[0028] Specifically, for the same risk of gas concentration, hardwired sensors can be deployed at the same location with a higher threshold, while smart network sensors can be deployed to monitor concentration change trends with a lower threshold for early warning, forming a double protection.

[0029] Furthermore, in high-risk areas, an emergency stop button can be directly connected via a dry contact signal. This button can be installed at the construction site in a prominent location, allowing construction personnel to quickly start and stop the machine.

[0030] It is particularly emphasized that by establishing a physical dedicated line independent of the conventional network, information reporting can be avoided in the event of congestion, interference or interruption of the on-site Wi-Fi and bus network. The hard-wired lifeline ensures uninterrupted information reporting and ensures that the highest level alarm signals are not lost. The edge gateway's fast rule engine processes rules in firmware or dedicated logic circuits, which is not affected by busy or crashing processes of the upper-layer operating system.

[0031] Step S2 includes: A calibration column is set up in the monitoring space. The calibration column integrates a high-precision reference sensor and a standard gas concentration simulator. Set calibration trigger conditions, which include reaching a preset time interval or the system completing a single alarm event; When any trigger condition is met, the calibration column is controlled by the edge gateway to release calibration gas or simulated reference signal, and data from the corresponding surrounding sensors is read. Calculate the deviation between the sensor-reported data and the calibration column reference value, and update the calibration coefficients to correct the original sensor data; If the sensor data deviation continues to exceed the maximum allowable threshold, the sensor is marked as requiring maintenance and a maintenance work order is generated.

[0032] Step S2 specifically involves setting calibration columns in the switch chamber and capacitor chamber, with the calibration columns corresponding to the sensor settings. A high-precision reference temperature and humidity sensor and a standard gas concentration simulator are integrated on the calibration column, which can release a trace amount of calibration gas to help the sensor calibrate its accuracy. Furthermore, during the use of the sensor, it is necessary to trigger the calibration of the sensor. During calibration, the calibration column is activated to release the labeling gas. At this time, the corresponding sensor collects data, reads data from surrounding sensors, and updates the sensor data correction module. Sensors deployed in safe areas, critical areas, and high-risk areas detect data and generate raw data. The raw data is then corrected by applying calibration coefficients to the sensor data correction module structure, so that the sensor outputs reliable sensing data. Specifically, regarding the sensor calibration process, the triggering conditions can be set in two ways: The first method involves calibration after the preset calibration time interval is reached. Before each day's construction, the calibration program is automatically triggered. The edge gateway instructs the calibration column to release calibration gas and reads data from all surrounding gas sensors. The deviation value is calculated and stored as a calibration coefficient. This is then used in conjunction with subsequent data reporting for automatic correction to ensure the monitoring accuracy of the sensors. The second method involves calibration after an alarm is triggered. Whenever the system processes a gas concentration exceeding the limit alarm, it automatically triggers a calibration of the sensor at that location during a safe period to verify whether the sensor status has been affected by the event.

[0033] It is particularly important to emphasize that if neither of the two triggering conditions is met, calibration will be performed after the next trigger. If either condition is met, a calibration command can be initiated through the edge gateway, starting the calibration column and releasing the labeling gas. At this time, the corresponding sensor will collect data, read data from surrounding sensors, calculate the deviation value and calibration coefficient, and then update the sensor data correction module. The original data and the sensor data correction module structure will be adjusted using the calibration coefficient to correct the data, so that the sensor outputs reliable sensing data, thus completing the data acquisition.

[0034] It is particularly important to note that if the deviation between the data reported by the corresponding sensor and the calibration column reference value continues to exceed the set maximum threshold, the system will determine that the sensor needs maintenance, mark the device as having suspicious data, generate a work order, and carry out manual repair.

[0035] In step S3, the local actuator is driven to perform the corresponding fast protection action according to the signal type, specifically including: Upon receiving an arc light signal, execute a millisecond-level trip command and trigger an audible and visual alarm; If the received gas concentration data reaches the first threshold, the forced ventilation will be activated. If the concentration continues to rise to a higher second threshold, an audible and visual alarm and a shutdown command will be triggered. Upon receiving the emergency stop button signal, immediately cease construction operations in the area.

[0036] In step S3, the local actuator is driven to perform the corresponding fast protection action according to the signal type. Specifically, when receiving hard-wired sensor data from a high-risk area, the signal is input into the fast channel of the edge gateway to stably receive the sensor monitoring data, which is divided into the following data information: Upon receiving arc light data, a millisecond-level trip command is executed and an audible and visual alarm is triggered, ensuring that the transmission path of arc light and gas explosion signals is the shortest and most direct, completely independent of the conventional data network. The corresponding local actuator is driven through the hardware interface to execute immediately, cutting off the accident chain within microseconds to milliseconds and preventing disaster from occurring. The system receives gas concentration data. When the gas concentration reaches the first set threshold, it activates forced ventilation and continues to receive gas concentration data. By activating forced ventilation, it attempts to automatically eliminate risks without triggering a panic alarm on site, thus avoiding unnecessary construction interruptions due to false alarms or minor leaks. While continuously receiving gas concentration data, it makes judgments. If the gas concentration decreases, no alarm is triggered. If the gas concentration reaches the second threshold, an audible and visual alarm is triggered. That is, the audible and visual alarm is only triggered when the measures are ineffective and the concentration continues to rise to a more dangerous level, notifying personnel to pay emergency attention or evacuate. The system drives the local actuator through the corresponding hardware interface to execute immediately, cutting off the accident chain within microseconds to milliseconds to prevent disasters. Specifically, after the forced exhaust is started, the rate of decrease in gas concentration is used as feedback. If the rate of decrease does not meet the requirements, the operation is upgraded to execute a shutdown order.

[0037] Specifically, when forced ventilation is activated due to high gas concentration, the resulting execution results, such as the rate of concentration decrease, are used as feedback data and uploaded to determine the effectiveness of the treatment. If an increase in concentration is detected in the area, ventilation can be pre-activated to prevent it from reaching the point of triggering a shutdown. If the concentration continues to rise when pre-activation of ventilation is performed, and the rate of decrease cannot be guaranteed to be within a safe concentration range, then a shutdown will be implemented to ensure personnel safety.

[0038] Upon receiving data from pressing the emergency stop button, operations in the construction area are immediately halted. This means that when a risk is manually assessed and personnel initiate an emergency stop, the corresponding local actuator is driven through the hardware interface to execute the action immediately, breaking the chain of events within microseconds to milliseconds and preventing a disaster from occurring. Upon receiving data indicating excessively high temperatures, operations in the construction area can be halted. Furthermore, trend analysis can predict when temperatures may exceed the limit in the future, providing early warnings to stop construction, avoid dangers caused by excessively high temperatures, and prevent disasters.

[0039] High-risk areas will be handled using reliable and rapid dedicated channels, forming a physical defense line with rapid feedback and execution to protect the lives of personnel and the safety of core equipment.

[0040] Step S4 involves data feature fusion and risk value calculation, specifically including: Multiple risk features were extracted from the sensor data, including: temperature and heating rate, three-phase imbalance, total harmonic distortion, UHF partial discharge pulse count, and gas concentration change rate. Map and merge the various risk characteristic values ​​into a single comprehensive risk value; The overall risk value is compared with preset thresholds for high, medium, low, and no risk levels to determine the current risk level.

[0041] The data feature fusion and risk value calculation in step S4 are as follows: when receiving sensor data from safe and critical areas, data feature fusion is performed and processed. The data features are converted into risk values ​​to obtain a comprehensive risk value. The risk level is determined according to the actual situation. Its function is to provide early warning of risk states that will not cause disasters instantly but will accumulate over time and eventually lead to accidents. These factors may include slow temperature rise due to poor contact of equipment joints, gradual deterioration of insulation materials in humid environments, and slow accumulation of harmful gases in the construction area. It receives raw data streams from various sensors, aligns data from different sensors at the same time based on timestamps, and performs specific risk assessments. Specific risk characteristics include: continuous high load is a major cause of overheating; three-phase imbalance leads to neutral point shift, increasing equipment stress; total harmonic distortion (THD) is calculated as the effective percentage of harmonic content in the current waveform, as harmonics cause additional losses and heat generation; the number of discharge pulses captured by UHF sensors within one second is statistically analyzed to reflect the activity level of insulation defects; the rate of temperature rise reflects the acceleration process of a fault better than absolute temperature, and a sudden increase in the rate is a serious warning signal; and the rate of change in gas concentration is crucial, as slow leaks may be ignored by absolute concentration thresholds, but the rate of change can detect them early.

[0042] Each characteristic value is mapped to a unified risk index for quantification and combination. The current risk level is judged to determine the level of cooperation, and corresponding actions are taken. The risk index can be specially set and adjusted according to the actual types of risks on site and the risk index ratio of each type of risk. In step S4, the corresponding early warning or control operation is performed, specifically as follows: Determined to be high-risk, work was forced to stop and personnel were evacuated; If the risk level is determined to be medium, triggering warning lights or dashboard alarms, it is recommended to suspend operations. If the risk level is determined to be low, a broadcast announcement or signboard notification will be made. If the risk is deemed to be no, periodic data monitoring and reporting will be conducted.

[0043] Step S4 specifically includes the following: When the overall risk value is at the high-risk threshold, work is forcibly stopped and personnel are evacuated; when the overall risk value is at the medium-risk threshold, warning lights or dashboard alarms are activated, and work is advised to be suspended; when the overall risk value is at the low-risk threshold, broadcast voice and dashboard prompts are given; when the overall risk value is at the no-risk threshold, normal monitoring is carried out, and data is reported periodically; after assessing the risk threshold, corresponding responses are taken to prevent the occurrence of risks in advance. This method for safety inspection during power construction also includes incident handling procedures: Reported data is categorized into alarm events and abnormal events; Alarm events are recorded, data is recorded, preset operations are executed, diagnostic conclusions and maintenance work orders are generated, and sent to mobile terminals. For abnormal events, implement corresponding measures and continuously monitor them. Based on the monitoring feedback, decide whether to lift the alarm or escalate it to generate early warnings and handling suggestions.

[0044] Specifically, during the monitoring process using sensor data, the reported data is received and the corresponding events are classified into alarm events and abnormal events. In the event of an alarm, the alarm data is recorded. Based on the alarm data, the corresponding operation is executed according to the settings, the corresponding equipment and environment are associated, diagnostic conclusions and handling suggestions are generated, a maintenance work order is formed, and the decision is sent to the mobile terminal. The data received by the staff's public mobile terminal is processed and the current status data is updated. For abnormal events, corresponding abnormal event handling is performed and continuous monitoring is conducted. Based on the monitoring feedback, it is determined whether the abnormality has improved. When the abnormality improves, the abnormal alarm is lifted, the abnormal event is recorded, and the current status data is updated. If the abnormality does not improve or the rate of improvement does not meet the set threshold, an early warning and handling suggestions are generated, the decision is sent to the mobile terminal, and the current status data is updated.

[0045] A safety monitoring system for electrical construction in confined spaces, comprising: The partitioned sensor network module includes low-power wireless sensor clusters deployed in secure areas, high-precision wired sensors deployed in critical equipment, and edge hardwired sensors deployed in high-risk equipment and directly connected via physical dedicated lines. The edge gateway includes a fast channel and a fast rule engine for microsecond-level processing of hardwired sensor signals, as well as a computing unit for processing regular sensor data. The online calibration module includes a calibration column installed on-site and a calibration logic unit for controlling the calibration process and calculating and storing calibration coefficients; The data fusion and risk assessment module is used to extract and fuse features from sensor data from safe and critical areas, calculate the comprehensive risk value, and determine the risk level. The execution and early warning output module is used to drive the corresponding local executor or output early warning information based on the instructions of the fast rule engine or the risk level determined by the risk assessment module.

[0046] The edge gateway’s fast lane and fast rule engine are physically and logically independent of the computing units and networks that process regular data. The signals from hard-wired sensors are directly connected to the digital input interface of the fast lane via physical cables.

[0047] Specifically, the system integrates a personnel positioning system and defines the status of each zone, which can be divided into disabled, monitored construction, and free movement states. When authorized personnel enter a high-risk area for construction, the system can automatically upgrade the monitoring level to a critical area and activate additional video surveillance or environmental monitoring in that area. At the same time, unauthorized personnel are prohibited from entering high-risk areas and / or from performing unsupervised solo work in critical areas. All status changes are recorded and reported synchronously with personnel entry and exit records.

[0048] For hard-wired lifelines, dual-redundancy configurations should be considered for sensors or lines carrying critical signals. These critical signal sensors include arc light sensors, methane gas sensors, carbon monoxide gas sensors, and hard-wired emergency stop button signal inputs connected to different fast channels of the edge gateway. The fast rule engine makes judgments, and the edge gateway itself has a watchdog mechanism and dual power supply redundancy. In case of a fault, it can send its own fault alarm to the upper-level monitoring center. The wireless sensor network has self-organizing network and routing redundancy capabilities. When some nodes fail, data can be uploaded through other paths, and the system should periodically perform connectivity self-checks on all communication links, including physical leased lines.

[0049] The system generates and encrypts a complete, tamper-proof audit log, recording all timestamps and key data from raw sensor data, calibration events, risk calculation processes, rule engine judgment results to the final executor actions.

[0050] For each alarm, the log should be associated with the location information of personnel in the relevant area at the time of the trigger, as well as the sequence of all protective actions performed.

[0051] Establish an electronic workflow from construction permit to construction supervision and finally to construction completion. After construction is completed, the person in charge needs to trigger the construction completion process through a mobile terminal. At this time, the system should automatically check the relevant areas: whether the comprehensive risk value has returned to the risk-free level, whether all mandatory actions have been reset according to the procedures, and whether the gas concentration has dropped to a safe level. Only when all the checks are passed will the system confirm the completion of construction and may restore the status of high-risk areas to disabled, while switching the wireless sensor network back to low power mode.

[0052] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for safety inspection during electrical construction in confined spaces, characterized in that, Includes the following steps: S1. Divide the monitoring space into safe areas, critical areas and high-risk areas, and deploy a first network consisting of low-power wireless sensors, a second network consisting of high-precision wired sensors and a third network consisting of hard-wired sensors based on physical dedicated lines accordingly; deploy the first network in the safe area, which is in low-power timed monitoring mode by default. When a personnel positioning tag is detected, it is woken up and enters continuous monitoring mode. Deploy a second network, Industrial Ethernet, on key equipment in key areas for continuous monitoring and real-time data reporting; In extremely high-risk locations within high-risk areas, a third network is deployed. The signals from the third network sensors are directly connected to the edge gateway's fast channel via physical cables, forming a dedicated physical line independent of the conventional data network. S2. Perform periodic or event-triggered online calibration on the sensors in the first network, the second network, and the third network, and acquire and update the sensor data correction module; S3. Receive hard-wired sensor signals from the high-risk area, process them in microseconds through the dedicated fast channel of the edge gateway, and drive the local actuator to perform corresponding fast protection actions according to the signal type. S4. Receive sensor data from safe and critical areas, perform data feature fusion and risk value calculation, obtain a comprehensive risk value from the risk value, and execute corresponding early warning or control operations based on the risk level threshold of the comprehensive risk value.

2. The method for safety inspection of electrical construction in confined spaces according to claim 1, characterized in that, Step S1 specifically includes: The safe zone is the area where personnel move around in the cable layer passage and equipment room corridor, and there are no directly high-risk equipment. The first network deployed includes a cluster consisting of wireless temperature and humidity sensors and wireless smoke sensors. The critical area is the area where key equipment involved in power construction is placed. The high-precision wired sensors are installed on the switchgear joints and main transformer bushings involved in the construction. The second network includes a cluster composed of temperature sensors and ultra-high frequency partial discharge sensors. The high-risk area refers to the area where extremely high-risk equipment is placed during power construction. The third network deployed at the extremely high-risk location of the construction site includes a cluster of arc light sensors and gas sensors. The arc light sensors are installed in the switch cabinet near the construction site and busbars, and the gas sensors are installed below the cable layer and GIS gas chamber. The gas sensors include methane gas sensors and carbon monoxide gas sensors. The sensor signal of the third network is directly connected to the fast channel of the edge gateway, forming a physical dedicated line independent of the conventional data network. When the gas sensor detects that the concentration reaches a preset ratio of the lower explosive limit, the non-explosion-proof power supply is immediately cut off and the forced ventilation is started.

3. The method for safety inspection of electrical construction in confined spaces according to claim 1, characterized in that, Step S2 includes: A calibration column is set up in the monitoring space. The calibration column integrates a high-precision reference sensor and a standard gas concentration simulator. Set calibration trigger conditions, which include reaching a preset time interval or the system completing a single alarm event; When any trigger condition is met, the calibration column is controlled by the edge gateway to release calibration gas or simulated reference signal, and data from the corresponding surrounding sensors is read. Calculate the deviation between all sensor-reported data and the calibration column reference value, and update the calibration coefficients to correct the original sensor data; If the data deviation of any sensor continues to exceed the maximum allowable threshold, the sensor is marked as requiring maintenance and a maintenance work order is generated.

4. The method for safety inspection of electrical construction in confined spaces according to claim 1, characterized in that, In step S3, the local actuator is driven to perform the corresponding fast protection action according to the signal type, specifically including: When the arc light sensor receives an arc light signal, it executes a millisecond-level trip command and triggers an audible and visual alarm. When the gas concentration data received by the gas sensor reaches the first threshold, the forced ventilation is activated. If the concentration continues to rise to a higher second threshold, an audible and visual alarm and a shutdown command are triggered. When the edge gateway receives the emergency stop button signal, it immediately stops the construction operation in that area.

5. A method for safety inspection during electrical construction in confined spaces according to claim 1, characterized in that, After the forced exhaust is started, the rate of decrease in gas concentration is used as feedback. If the rate of decrease does not meet the requirements, the shutdown command is executed.

6. A method for safety inspection during electrical construction in confined spaces according to claim 1, characterized in that, Step S4 involves data feature fusion and risk value calculation, specifically including: Multiple risk features are extracted from sensor data from safe and critical areas. These multiple risk features include: temperature and heating rate, three-phase imbalance, total harmonic distortion, UHF partial discharge pulse count, and gas concentration change rate. Multiple risk characteristics are mapped and merged into a single comprehensive risk value; The overall risk value is compared with preset thresholds for high, medium, low, and no risk levels to determine the current risk level.

7. A method for safety inspection during electrical construction in confined spaces according to claim 6, characterized in that, In step S4, the corresponding early warning or control operation is performed, specifically as follows: When the overall risk value is determined to be high risk, work must be stopped and personnel evacuated. When the overall risk value is determined to be medium risk, a warning light or dashboard alarm will be triggered, and it is recommended to suspend operations. When the overall risk value is determined to be low risk, a broadcast voice or signboard prompt will be made. When the comprehensive risk value is determined to be risk-free, periodic data monitoring and reporting will be carried out.

8. A method for safety inspection during electrical construction in confined spaces according to claim 1, characterized in that, This method for safety inspection during power construction also includes incident handling procedures: The reported data is divided into alarm events and abnormal events; For alarm events, record data, execute preset operations, generate diagnostic conclusions and maintenance work orders, and send them to mobile terminals; For abnormal events, corresponding measures are taken and continuous monitoring is carried out. Based on the monitoring feedback, a decision is made on whether to lift the alarm or escalate it to generate early warning and handling suggestions.

9. A safety monitoring system for electrical construction in confined spaces, used to implement the method described in any one of claims 1-8, characterized in that, include: The partitioned sensor network module includes low-power wireless sensor clusters deployed in secure areas, high-precision wired sensors deployed in critical equipment, and edge hardwired sensors deployed in high-risk equipment and directly connected via physical dedicated lines. The edge gateway includes a fast channel and a fast rule engine for microsecond-level processing of hardwired sensor signals, as well as a computing unit for processing regular sensor data. The online calibration module includes a calibration column installed on-site and a calibration logic unit for controlling the calibration process and calculating and storing calibration coefficients; The data fusion and risk assessment module is used to extract and fuse features from sensor data from safe and critical areas, calculate the comprehensive risk value, and determine the risk level. The execution and early warning output module is used to drive the corresponding local executor or output early warning information based on the instructions of the fast rule engine or the risk level determined by the risk assessment module.

10. A safety monitoring system for electrical construction in confined spaces according to claim 9, characterized in that, The edge gateway’s fast lane and fast rule engine are physically and logically independent of the computing units and networks that process regular data. The signals from hard-wired sensors are directly connected to the digital input interface of the fast lane via physical cables.

Images