An online monitoring system for safety guarantee of temporary power supply at construction sites

Abstract

The application discloses a kind of based on construction site temporary power supply safety guarantee online monitoring system, it is related to electrical safety monitoring technical field, including intelligent circuit breaker array, multi-source perception unit, edge computing gateway, adaptive topology modeling module, multi-physical field risk assessment module, edge consensus decision module and hierarchical intervention execution module;The intelligent circuit breaker array is deployed in each temporary power supply branch of construction site, for real-time acquisition current signal, voltage signal and residual current signal, and executes local operation of breaking;The multi-source perception unit is cooperatively laid with the intelligent circuit breaker array, for obtaining environmental temperature and humidity data, ground water state, real-time position information and work behavior characteristics of personnel;The adaptive topology modeling module is integrated in the edge computing gateway, for extracting harmonic fingerprint features and transient response waveform when each power equipment is accessed based on the current signal collected by the intelligent circuit breaker array.

Description

Technical Field

[0001] This invention relates to the field of electrical safety monitoring technology, and in particular to an online monitoring system for temporary power supply safety at construction sites. Background Technology

[0002] Electrical safety monitoring technology refers to a comprehensive technology that uses sensors, data acquisition devices, communication networks, and intelligent analysis algorithms to collect, transmit, and evaluate key parameters such as voltage, current, leakage current, temperature, and insulation status in power systems in real time. This allows for the identification of potential electrical faults or safety hazards, and timely warnings or automatic protective measures. Its core objective is to prevent safety accidents such as electric shock, electrical fires, and equipment damage. It is widely used in industrial, construction, rail transportation, civil construction, and temporary power supply scenarios, and is a crucial technical means to ensure personal safety and the reliable operation of power systems. In emerging fields such as modern smart grids and smart construction sites, electrical safety monitoring technology is continuously evolving towards multi-source sensing fusion, edge intelligent decision-making, and proactive risk prevention and control.

[0003] Construction sites frequently dismantle and modify power distribution lines, and equipment is plug-and-play. However, existing monitoring systems generally rely on manual input or fixed topology models, which cannot reflect the actual wiring relationships in real time. This leads to frequent fault location errors, false alarms, and missed alarms. Furthermore, traditional systems only trigger protection based on single thresholds such as residual current, voltage, or temperature, and cannot identify hidden high-risk conditions formed by the coupling of multiple factors such as rain, leakage, and personnel proximity. They also lack effective early warning capabilities for the main causes of actual electric shock accidents. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an online monitoring system for temporary power supply safety at construction sites to address the frequent dismantling and alteration of power distribution lines and the plug-and-play nature of equipment at construction sites. Existing monitoring systems generally rely on manual input or fixed topology models, which cannot reflect the actual wiring relationships in real time, resulting in frequent fault location errors, false alarms, and missed alarms. Furthermore, traditional systems only trigger protection based on single thresholds such as residual current, voltage, or temperature, and cannot identify hidden high-risk states formed by the coupling of multiple factors such as rain, leakage, and personnel proximity, thus lacking effective early warning capabilities for the main causes of actual electric shock accidents.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an online monitoring system for temporary power supply safety at construction sites, comprising: The system includes an intelligent circuit breaker array, a multi-source sensing unit, an edge computing gateway, an adaptive topology modeling module, a multi-physics risk assessment module, an edge consensus decision-making module, and a hierarchical intervention execution module. The intelligent circuit breaker array is deployed in each temporary power branch of the construction site to collect current signals, voltage signals and residual current signals in real time, and to perform disconnection operations locally. The multi-source sensing unit is deployed in conjunction with the smart circuit breaker array to acquire environmental temperature and humidity data, ground water status, real-time personnel location information, and operational behavior characteristics. The adaptive topology modeling module is integrated into the edge computing gateway. It is used to extract the harmonic fingerprint features and transient response waveforms of each electrical device when it is connected based on the current signal collected by the smart circuit breaker array, construct an initial electrical connection diagram through a lightweight clustering algorithm, and dynamically update the electrical connection diagram when a new device is detected or the original circuit is changed, and output real-time topology data. The multiphysics risk assessment module is integrated into the edge computing gateway. It is used to receive the real-time topology data, the residual current signal, the environmental temperature and humidity data, the ground water status and the real-time location information of the personnel, construct a spatiotemporal risk map with the power consumption nodes as vertices and the environmental and personnel status as edge attributes, and calculate the composite risk index corresponding to each power consumption node based on the graph neural network model. The edge consensus decision module is integrated into the edge computing gateway. When the composite risk index exceeds the first preset threshold or the residual current signal exceeds the safety limit, it generates a fault feature message and broadcasts it to the neighboring smart circuit breaker node through the local wireless communication protocol. After receiving feature consistency confirmations returned by no less than a preset number of neighboring nodes, it generates a local power-off command. The graded intervention execution module is used to execute differentiated response actions according to the range in which the composite risk index is located: when the composite risk index is in the medium-risk range, it triggers audible and visual warnings and enhanced human-machine interaction prompts; when the composite risk index is in the high-risk range or when the local power outage command is received, it drives the corresponding smart circuit breaker to perform an emergency disconnection operation and simultaneously uploads the event log to the remote monitoring platform.

[0007] As a preferred embodiment of the online monitoring system for temporary power supply safety at construction sites described in this invention, the adaptive topology modeling module extracts harmonic fingerprint features and transient response waveforms from current signals and constructs an electrical connection diagram. The specific steps are as follows: When any smart circuit breaker detects that the branch current jumps from zero to non-zero, the device access identification process is initiated. Data collection before power-on Instantaneous current signal within milliseconds ,in The preset time window length; right Perform a fast Fourier transform to obtain the amplitude sequences of the fundamental and higher harmonic components. , as a harmonic fingerprint feature; Simultaneously calculate three transient parameters of the current rising edge: rising slope overshoot Stable time ; The rising slope Defined as the average rate of change of current during the rise from 10% to 90% of its steady-state value; oversurge. Defined as the difference between the peak current and the steady-state current; settling time Defined as the time from the moment of power-on until the current fluctuation amplitude is less than the preset tolerance range; Harmonic fingerprint features With transient parameters splicing to form device access feature vector ; Will Input a lightweight clustering model, output device category labels and node identifiers. ; Add a vertex to the electrical connection diagram. Then, a mapping edge is established between the vertex and the corresponding physical address of the smart circuit breaker to complete the topology update.

[0008] As a preferred embodiment of the online monitoring system for temporary power supply safety at construction sites described in this invention, the lightweight clustering model employs an improved K-means++ initialization strategy, specifically including: Let the current set of cluster centers be... ; The first cluster center was selected as a random sample. For each new center thereafter, calculate the value of each unselected sample. Square distance to the nearest existing center ; With probability Select the next cluster center; When the new input feature vector The minimum distance to all existing cluster centers is greater than a preset threshold. At that time, Joining as a new cluster center and assign a unique node identifier. ; The preset threshold The system dynamically adjusts the feature vectors of historical devices based on the maximum intra-class distance to ensure that new devices can be effectively identified without being mistakenly merged into existing categories.

[0009] As a preferred embodiment of the online monitoring system for temporary power supply safety at construction sites described in this invention, the multi-physics risk assessment module constructs a spatiotemporal risk map and calculates a composite risk index, specifically through the following steps: By each power consumption node As a vertex of the graph, the vertex attribute vector is defined as a triple consisting of the residual current value, the temperature rise rate and the device type code; Based on the environmental and personnel data output by the multi-source sensing unit, edge attributes are constructed, including: For any two nodes, if their spatial distance is less than the preset neighborhood radius... Then, graph edges are created, and edge attributes include the distance between the node and the nearest water accumulation area, the relative humidity of the environment, and the spatial distance between the nearest worker and the node; Generate an adjacency matrix based on real-time topology data. ,in If and only if node and Electrically directly connected or spatially separated by less than ; Vertex attribute matrix Edge attribute matrix and adjacency matrix Input a three-layer graph convolutional neural network; The node representation update for layer 11 uses the following formula: ; in, For the normalized adjacency matrix, For a degree matrix, its diagonal elements ; For the first Layer node embedding matrix; For the first Layer-learnable weight matrix; A learnable fusion coefficient matrix that matches the edge attribute dimension; This is the aggregated representation of the edge attribute tensor along the node dimension; It is the ReLU activation function; Take the output of the last layer Each row is mapped to the Sigmoid function through a fully connected layer to obtain the composite risk index corresponding to each node. .

[0010] As a preferred embodiment of the online monitoring system for temporary power supply safety at construction sites described in this invention, wherein: the composite risk index The rules for dividing the numerical intervals are as follows: when A value <0.3 is considered low risk and no intervention is triggered; when 0.3 ≤ A value less than 0.6 is considered medium risk; when... A value ≥0.6 is considered high risk; the results of the medium and high risk assessments serve as input conditions for the tiered intervention execution module; the composite risk index... It comprehensively reflects the triple coupling effect of electrical fault tendency, environmental hazard level and personnel exposure level.

[0011] As a preferred embodiment of the online monitoring system for temporary power supply safety at construction sites described in this invention, the specific process by which the edge consensus decision module generates a local power outage command is as follows: When any node satisfies the composite risk index ≥0.6 or the residual current value exceeds the safety limit. At that time, construct a fault characteristic message. ,in For timestamps, For Center, radius Local topological subgraphs within; Through local wireless communication protocol Broadcast to all nearby smart circuit breaker nodes; Each neighboring node received Then, compare it with the residual current value it collected itself. With the message If satisfied Furthermore, the composite risk index at this node If the value is ≥0.6, a consistency confirmation message is returned; Let the total number of neighboring nodes within the broadcast range be . The minimum number of confirmations required is ,and ≥2; If within the preset time limit Received no less than If a consistency confirmation message is received, the edge consensus decision module will generate a local power-off command. ; The consistency judgment introduces the following risk synergy enhancement factor. Used to quantify the cumulative effect of risks across multiple nodes: ; in, Represents a node The set of neighboring nodes, The risk sensitivity coefficient is taken as a positive value; when Exceeding the preset collaboration threshold At that time, even if local Even if the risk level is not high, a consensus voting process can be triggered to achieve proactive protection driven by group risk perception.

[0012] As a preferred embodiment of the online monitoring system for temporary power supply safety at construction sites described in this invention, wherein: the preset time limit Fixed at three seconds; if in If the required number of consistent confirmation messages are not received within the specified timeframe, the distributed consensus process is abandoned, and the system is instead asked to determine whether the hard rule conditions are met. ; If the conditions are met, a local power-off command will be generated directly. Otherwise, maintain power supply.

[0013] As a preferred embodiment of the online monitoring system for temporary power supply safety at construction sites described in this invention, the specific process for the graded intervention execution module to perform differentiated response actions is as follows: When a medium-risk assessment result is received, the sound and light alarm is activated to issue intermittent warnings, and the location of the risk node, a 3D animation of the risk cause, and a safe operation guide text are overlaid in the field of vision of the operator through the augmented reality terminal. When a high-risk assessment result or a local power outage command is received When the power supply to the corresponding intelligent circuit breaker is interrupted, a disconnection control signal is sent to the circuit breaker to forcibly cut off the power supply to that branch. Synchronously generate event logs ,in For ambient relative humidity, For nodes Distance from the nearest flooded area For the most recent workers and nodes Spatial distance, This is a snapshot of the local topology at the moment of power failure; right Calculate hash digest and will and After combining and encrypting, the data is uploaded to the remote monitoring platform. The integrity of the event log is ensured by the following verification formula: ; in, This indicates a field concatenation operation. This indicates that the adjacency matrix of the topological snapshot is expanded row-wise into a one-dimensional vector; this hash digest Used to verify that logs have not been tampered with during transmission and storage, supporting incident accountability and security auditing.

[0014] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements any step of the online monitoring system for temporary power supply safety assurance at construction sites as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the online monitoring system for temporary power supply safety assurance at construction sites as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By constructing an online monitoring system integrating adaptive topology modeling, multi-physics risk fusion assessment, and edge consensus power outage, it effectively solves the core problems of inaccurate safety monitoring, delayed response, and protection failure caused by dynamic wiring without drawings, complex and changeable environment, non-professional operation by personnel, and easy communication interruption in temporary power supply scenarios at construction sites. The system can achieve plug-and-play automatic circuit modeling without relying on manual topology input. Based on graph neural networks, it quantifies the coupled risks of electrical, environmental, and human behavior, breaking through the limitations of traditional single threshold alarms. Even when the backbone network is interrupted, it can still reliably execute emergency power outages through a lightweight consensus mechanism, improving the early warning capability of electric shock accidents, the reliability of emergency response, and the overall safety protection level. At the same time, the hierarchical intervention and event log traceability mechanism take into account both the continuity of operations and the needs of post-event auditing, providing a highly robust, intelligent, and engineering-practical technical solution for the safety management of temporary power supply at construction sites. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0018] Figure 1 This is a schematic diagram of the online monitoring system for temporary power supply safety at construction sites in Example 1. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Example, refer to Figure 1 As one embodiment of the present invention, this embodiment provides an online monitoring system for temporary power supply safety at construction sites, comprising: The system includes an intelligent circuit breaker array, a multi-source sensing unit, an edge computing gateway, an adaptive topology modeling module, a multi-physics risk assessment module, an edge consensus decision-making module, and a hierarchical intervention execution module.

[0023] The intelligent circuit breaker array is deployed in each temporary power branch of the construction site to collect current signals, voltage signals and residual current signals in real time, and to perform disconnection operations locally.

[0024] Furthermore, each circuit breaker unit in the intelligent circuit breaker array incorporates a high-precision current transformer, voltage sampling circuit, and residual current detection module. It synchronously acquires the instantaneous values ​​of three-phase or single-phase current, the effective values ​​of phase voltage, and the amplitude of residual current at a sampling frequency of no less than 10 kHz. Each circuit breaker unit is equipped with an independent microcontroller for real-time calculation of the effective current value, power factor, harmonic distortion rate, and leakage current action criteria. When the residual current amplitude is detected to continuously exceed 30 mA for 30 milliseconds, or the phase current mutation rate exceeds a preset slope threshold, the microcontroller immediately drives a magnetic latching relay to cut off the power supply to the corresponding branch, with a disconnection response time of no more than 20 milliseconds. All circuit breaker units are installed in a temporary distribution box via standard DIN rails and establish a communication connection with an edge computing gateway via RS485 or power line carrier interface, periodically uploading electrical parameters and event status.

[0025] It should be noted that the intelligent circuit breaker array, through its built-in high-precision sensing and local high-speed processing capabilities, can independently identify and disconnect leakage current, overcurrent, and transient anomalies without relying on external systems, with a response time controlled within 20 milliseconds, meeting the timeliness requirements for personal electric shock protection. At the same time, its periodic uploading of electrical characteristic data provides the original basis for upper-level topology modeling and risk assessment, ensuring the bottom-level safety baseline and supporting system-level intelligent analysis, forming a closed-loop linkage from execution to perception.

[0026] Multi-source sensing units are deployed in conjunction with intelligent circuit breaker arrays to acquire environmental temperature and humidity data, ground water accumulation status, real-time personnel location information, and operational behavior characteristics.

[0027] Furthermore, the multi-source sensing unit includes distributed environmental sensing nodes, millimeter-wave radar positioning base stations, video analytics cameras, and wearable positioning tags. The environmental sensing nodes integrate temperature and humidity sensors and capacitive water accumulation detection probes, fixed at the bottom of the distribution box and in low-lying areas along cable laying paths, to monitor the local relative humidity and ground water depth in real time. The millimeter-wave radar positioning base stations are installed at the four corners of the construction area, using multipath reflection signals to analyze the spatial coordinates of UWB or Bluetooth positioning tags carried by workers, achieving a positioning accuracy better than 0.5 meters. The video analytics cameras are deployed above high-risk electrical areas, using a lightweight YOLOv5s model to identify in real time whether personnel are wearing insulated gloves, crossing warning lines, or approaching exposed wiring terminals. All sensing data is aggregated to the edge computing gateway via a local LoRa or Wi-Fi 6 network, and after timestamp alignment, it is spatiotemporally correlated with electrical parameters, providing an input basis for multiphysics risk assessment.

[0028] It should be noted that the multi-source sensing unit achieves simultaneous acquisition of temperature and humidity, water accumulation, personnel location and operation behavior by deploying environmental, positioning and vision sensors around the power distribution box and in the work area; the combination of millimeter-wave radar and lightweight image recognition model can still stably output personnel spatial coordinates and safety behavior status in complex lighting and dust environments, providing the risk assessment module with real, reliable, spatiotemporally aligned non-electrical dimension input, significantly improving the system's perception coverage of typical high-risk scenarios on the construction site.

[0029] The adaptive topology modeling module is integrated into the edge computing gateway. It is used to extract the harmonic fingerprint features and transient response waveforms of each electrical device when it is connected based on the current signal collected by the smart circuit breaker array. It constructs an initial electrical connection diagram through a lightweight clustering algorithm and dynamically updates the electrical connection diagram when a new device is detected or the original circuit is changed, and outputs real-time topology data.

[0030] Furthermore, when any smart circuit breaker detects that the branch current has changed from zero to non-zero, it initiates the device access identification process. Data collection before power-on Instantaneous current signal within milliseconds ,in The preset time window length; right Perform a fast Fourier transform to obtain the amplitude sequences of the fundamental and higher harmonic components. , as a harmonic fingerprint feature; Simultaneously calculate three transient parameters of the current rising edge: rising slope overshoot Stable time ; The rising slope Defined as the average rate of change of current during the rise from 10% to 90% of its steady-state value; oversurge. Defined as the difference between the peak current and the steady-state current; settling time Defined as the time from the moment of power-on until the current fluctuation amplitude is less than the preset tolerance range; Harmonic fingerprint features With transient parameters splicing to form device access feature vector ; Will Input a lightweight clustering model, output device category labels and node identifiers. ; Add a vertex to the electrical connection diagram. And establish a mapping edge between the vertex and the corresponding smart circuit breaker physical address to complete the topology update; The lightweight clustering model employs an improved K-means++ initialization strategy, specifically including: Let the current set of cluster centers be... ; The first cluster center was selected as a random sample. For each new center thereafter, calculate the value of each unselected sample. Square distance to the nearest existing center ; With probability Select the next cluster center; When the new input feature vector The minimum distance to all existing cluster centers is greater than a preset threshold. At that time, Joining as a new cluster center and assign a unique node identifier. ; Preset threshold The system dynamically adjusts the feature vectors of historical devices based on the maximum intra-class distance to ensure that new devices can be effectively identified without being mistakenly merged into existing categories.

[0031] It should be noted that the adaptive topology modeling module automatically identifies new loads and updates electrical connections by utilizing the current characteristics of the device at the moment of power-on, eliminating the need for manual input of wiring diagrams and effectively addressing the engineering reality of frequent dismantling and modification of temporary power supplies. The improved K-means++ clustering strategy adopted reduces computational overhead while ensuring recognition accuracy, enabling the edge gateway to continuously maintain a dynamic topology consistent with the physical site under resource-constrained conditions, providing a reliable network foundation for subsequent fault location and risk tracing.

[0032] The multiphysics risk assessment module is integrated into the edge computing gateway to receive real-time topology data, residual current signals, environmental temperature and humidity data, ground water status and real-time personnel location information. It constructs a spatiotemporal risk map with power consumption nodes as vertices and environmental and personnel status as edge attributes, and calculates the composite risk index corresponding to each power consumption node based on a graph neural network model.

[0033] Furthermore, by each electricity consumption node As a vertex of the graph, the vertex attribute vector is defined as a triple consisting of the residual current value, the temperature rise rate and the device type code; Based on the environmental and personnel data output by the multi-source sensing unit, edge attributes are constructed, including: For any two nodes, if their spatial distance is less than the preset neighborhood radius... Then, graph edges are created, and edge attributes include the distance between the node and the nearest water accumulation area, the relative humidity of the environment, and the spatial distance between the nearest worker and the node; Generate an adjacency matrix based on real-time topology data. ,in If and only if node and Electrically directly connected or spatially separated by less than ; Vertex attribute matrix Edge attribute matrix and adjacency matrix Input a three-layer graph convolutional neural network; No. The node representation of a layer is updated using the following formula: ; in, For the normalized adjacency matrix, For a degree matrix, its diagonal elements ; For the first Layer node embedding matrix; For the first Layer-learnable weight matrix; A learnable fusion coefficient matrix that matches the edge attribute dimension; This is the aggregated representation of the edge attribute tensor along the node dimension; It is the ReLU activation function; Take the output of the last layer Each row is mapped to the Sigmoid function through a fully connected layer to obtain the composite risk index corresponding to each node. .

[0034] It should be noted that the multiphysics risk assessment module models electrical parameters, environmental conditions, and personnel locations into a unified graph structure, and uses graph convolutional neural networks to mine the risk propagation and superposition effects between nodes, thereby quantifying the degree of composite danger that cannot be reflected by a single parameter. This method can identify hidden high-risk combinations such as minor leakage + damp ground + personnel proximity, realizing the transformation from passive alarm to proactive prediction, and significantly improving the system's ability to match the causes of real construction accidents.

[0035] The edge consensus decision module is integrated into the edge computing gateway. When the composite risk index exceeds the first preset threshold or the residual current signal exceeds the safety limit, it generates a fault feature message and broadcasts it to the neighboring smart circuit breaker node through the local wireless communication protocol. After receiving feature consistency confirmations from no less than a preset number of neighboring nodes, it generates a local power-off command.

[0036] Furthermore, the compound risk index The rules for dividing the numerical intervals are as follows: when A value <0.3 is considered low risk and no intervention is triggered; when 0.3 ≤ A value less than 0.6 is considered medium risk; when... A value ≥0.6 is considered high risk; the results of medium and high risk assessments serve as input conditions for the tiered intervention implementation module; composite risk index. It comprehensively reflects the triple coupling effect of electrical fault tendency, environmental hazard level and personnel exposure level; The specific process by which the edge consensus decision module generates a local power-off command is as follows: When any node satisfies the composite risk index ≥0.6 or the residual current value exceeds the safety limit. At that time, construct a fault characteristic message. ,in For timestamps, For Center, radius Local topological subgraphs within; Through local wireless communication protocol Broadcast to all nearby smart circuit breaker nodes; Each neighboring node received Then, compare it with the residual current value it collected itself. With the message If satisfied Furthermore, the composite risk index at this node If the value is ≥0.6, a consistency confirmation message is returned; Let the total number of neighboring nodes within the broadcast range be . The minimum number of confirmations required is ,and ≥2; If within the preset time limit Received no less than If a consistency confirmation message is received, the edge consensus decision module will generate a local power-off command. ; Consistency judgment introduces the following risk synergy enhancement factors Used to quantify the cumulative effect of risks across multiple nodes: ; in, Represents a node The set of neighboring nodes, The risk sensitivity coefficient is taken as a positive value; when Exceeding the preset collaboration threshold At that time, even if local Even if the risk level is not high, the consensus voting process can still be triggered to achieve proactive protection driven by the group's risk perception. Preset time limit Fixed at three seconds; if in If the required number of consistent confirmation messages are not received within the specified timeframe, the distributed consensus process is abandoned, and the system is instead asked to determine whether the hard rule conditions are met. ; If the conditions are met, a local power-off command will be generated directly. Otherwise, maintain power supply.

[0037] It should be noted that the edge consensus decision-making module, while retaining the traditional residual current hard protection mechanism, introduces a neighboring node collaborative verification mechanism. By broadcasting fault characteristics and collecting consensus confirmations, it avoids unnecessary power outages caused by single-point misjudgments. At the same time, a risk collaboration enhancement factor is set up so that the system can still initiate early warning or intervention when the risk of multiple nodes accumulates but does not reach the single-point threshold, which enhances the sensitivity to progressive hidden dangers and takes into account both safety and power supply continuity.

[0038] The graded intervention execution module is used to execute differentiated response actions based on the range of the composite risk index: when the composite risk index is in the medium risk range, it triggers audible and visual warnings and enhanced human-machine interaction prompts; when the composite risk index is in the high risk range or a local power outage command is received, it drives the corresponding smart circuit breaker to perform an emergency disconnection operation and simultaneously uploads the event log to the remote monitoring platform.

[0039] Furthermore, when a medium-risk assessment result is received, the system activates an intermittent audible and visual alarm and displays the location of the risk node, a 3D animation of the risk cause, and safe operation instructions text overlaid on the operator's field of vision via an augmented reality terminal. When a high-risk assessment result or a local power outage command is received When the power supply to the corresponding intelligent circuit breaker is interrupted, a disconnection control signal is sent to the circuit breaker to forcibly cut off the power supply to that branch. Synchronously generate event logs ,in For ambient relative humidity, For nodes Distance from the nearest flooded area For the most recent workers and nodes Spatial distance, This is a snapshot of the local topology at the moment of power failure; right Calculate hash digest and will and After combining and encrypting, the data is uploaded to the remote monitoring platform. The integrity of the event log is ensured by the following verification formula: ; in, This indicates a field concatenation operation. This indicates that the adjacency matrix of the topological snapshot is expanded row-wise into a one-dimensional vector; this hash digest Used to verify that logs have not been tampered with during transmission and storage, supporting incident accountability and security auditing.

[0040] It should be noted that the graded intervention execution module implements differentiated responses based on the risk level. In the case of medium risk, only warnings and visual guidance are triggered to reduce interference with normal operations. Only in the case of high risk or confirmed faults is a forced power outage executed, reflecting the concept of refined safety management. The event log contains complete contextual information at the time of the power outage and ensures data integrity through hash digests, providing traceable and tamper-proof technical evidence for post-event analysis and supporting the continuous improvement of the construction company's safety management system.

[0041] This embodiment also provides a computer device applicable to the online monitoring system for temporary power supply safety at construction sites, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the online monitoring system for temporary power supply safety at construction sites as proposed in the above embodiment.

[0042] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0043] This embodiment also provides a storage medium on which a computer program is stored. When executed by a processor, the program implements the online monitoring system for temporary power supply safety at construction sites as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0044] In summary, this invention effectively solves the core challenges of inaccurate safety monitoring, delayed response, and protection failure in temporary power supply scenarios at construction sites by constructing an online monitoring system that integrates adaptive topology modeling, multi-physics risk fusion assessment, and edge consensus power outage. These challenges stem from factors such as the lack of wiring diagrams, complex and changing environments, non-professional personnel operation, and easy communication interruptions. The system achieves plug-and-play automatic circuit modeling without relying on manual topology input. Based on graph neural networks, it quantifies the coupled risks of electrical, environmental, and human behavior, breaking through the limitations of traditional single-threshold alarms. Even when the backbone network is interrupted, it can still reliably execute emergency power outages through a lightweight consensus mechanism, improving the early warning capability, emergency response reliability, and overall safety protection level for electric shock accidents. At the same time, the hierarchical intervention and event log traceability mechanism take into account both operational continuity and post-event auditing needs, providing a highly robust, intelligent, and engineering-practical technical solution for the safety management of temporary power supply at construction sites.

[0045] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An online monitoring system for temporary power supply safety at construction sites, characterized in that: include: The system includes an intelligent circuit breaker array, a multi-source sensing unit, an edge computing gateway, an adaptive topology modeling module, a multi-physics risk assessment module, an edge consensus decision-making module, and a hierarchical intervention execution module. The intelligent circuit breaker array is deployed in each temporary power branch of the construction site to collect current signals, voltage signals and residual current signals in real time, and to perform disconnection operations locally. The multi-source sensing unit is deployed in conjunction with the smart circuit breaker array to acquire environmental temperature and humidity data, ground water status, real-time personnel location information, and operational behavior characteristics. The adaptive topology modeling module is integrated into the edge computing gateway. It is used to extract the harmonic fingerprint features and transient response waveforms of each electrical device when it is connected based on the current signal collected by the smart circuit breaker array, construct an initial electrical connection diagram through a lightweight clustering algorithm, and dynamically update the electrical connection diagram when a new device is detected or the original circuit is changed, and output real-time topology data. The multiphysics risk assessment module is integrated into the edge computing gateway. It is used to receive the real-time topology data, the residual current signal, the environmental temperature and humidity data, the ground water status and the real-time location information of the personnel, construct a spatiotemporal risk map with the power consumption nodes as vertices and the environmental and personnel status as edge attributes, and calculate the composite risk index corresponding to each power consumption node based on the graph neural network model. The edge consensus decision module is integrated into the edge computing gateway. When the composite risk index exceeds the first preset threshold or the residual current signal exceeds the safety limit, it generates a fault feature message and broadcasts it to the neighboring smart circuit breaker node through the local wireless communication protocol. After receiving feature consistency confirmations returned by no less than a preset number of neighboring nodes, it generates a local power-off command. The graded intervention execution module is used to execute differentiated response actions according to the range in which the composite risk index is located: when the composite risk index is in the medium-risk range, it triggers audible and visual warnings and enhanced human-machine interaction prompts; when the composite risk index is in the high-risk range or when the local power outage command is received, it drives the corresponding smart circuit breaker to perform an emergency disconnection operation and simultaneously uploads the event log to the remote monitoring platform.

2. The online monitoring system for temporary power supply safety at construction sites as described in claim 1, characterized in that: The adaptive topology modeling module extracts harmonic fingerprint features and transient response waveforms from current signals and constructs an electrical connection diagram. The specific steps are as follows: When any smart circuit breaker detects that the branch current jumps from zero to non-zero, the device access identification process is initiated. Data collection before power-on Instantaneous current signal within milliseconds ,in The preset time window length; right Perform a fast Fourier transform to obtain the amplitude sequences of the fundamental and higher harmonic components. , as a harmonic fingerprint feature; Simultaneously calculate three transient parameters of the current rising edge: rising slope overshoot Stable time ; The rising slope Defined as the average rate of change of current during the rise from 10% to 90% of its steady-state value; oversurge. Defined as the difference between the peak current and the steady-state current; settling time Defined as the time from the moment of power-on until the current fluctuation amplitude is less than the preset tolerance range; Harmonic fingerprint features With transient parameters splicing to form device access feature vector ; Will Input a lightweight clustering model, output device category labels and node identifiers. ; Add a vertex to the electrical connection diagram. Then, a mapping edge is established between the vertex and the corresponding physical address of the smart circuit breaker to complete the topology update.

3. The online monitoring system for temporary power supply safety at construction sites as described in claim 2, characterized in that: The lightweight clustering model employs an improved K-means++ initialization strategy, specifically including: Let the current set of cluster centers be... ; The first cluster center was selected as a random sample. For each new center thereafter, calculate the value of each unselected sample. Square distance to the nearest existing center ; With probability Select the next cluster center; When the new input feature vector The minimum distance to all existing cluster centers is greater than a preset threshold. At that time, Joining as a new cluster center and assign a unique node identifier. ; The preset threshold The system dynamically adjusts the feature vectors of historical devices based on the maximum intra-class distance to ensure that new devices can be effectively identified without being mistakenly merged into existing categories.

4. The online monitoring system for temporary power supply safety at construction sites as described in claim 3, characterized in that: The multiphysics risk assessment module constructs a spatiotemporal risk map and calculates a composite risk index. The specific steps are as follows: By each power consumption node As a vertex of the graph, the vertex attribute vector is defined as a triple consisting of the residual current value, the temperature rise rate and the device type code; Based on the environmental and personnel data output by the multi-source sensing unit, edge attributes are constructed, including: For any two nodes, if their spatial distance is less than the preset neighborhood radius... Then, graph edges are created, and edge attributes include the distance between the node and the nearest water accumulation area, the relative humidity of the environment, and the spatial distance between the nearest worker and the node; Generate an adjacency matrix based on real-time topology data. ,in If and only if node and Electrically directly connected or spatially separated by less than ; Vertex attribute matrix Edge attribute matrix and adjacency matrix Input a three-layer graph convolutional neural network; The node representation update for layer 11 uses the following formula: ; in, For the normalized adjacency matrix, For a degree matrix, its diagonal elements ; For the first Layer node embedding matrix; For the first Layer-learnable weight matrix; A learnable fusion coefficient matrix that matches the edge attribute dimension; This is the aggregated representation of the edge attribute tensor along the node dimension; It is the ReLU activation function; Take the output of the last layer Each row is mapped to the Sigmoid function through a fully connected layer to obtain the composite risk index corresponding to each node. .

5. The online monitoring system for temporary power supply safety at construction sites as described in claim 4, characterized in that: The composite risk index The rules for dividing the numerical intervals are as follows: when A value <0.3 is considered low risk and no intervention is triggered; when 0.3 ≤ A value less than 0.6 is considered medium risk; when... A value ≥0.6 is considered high risk; the results of the medium and high risk determinations serve as input conditions for the graded intervention execution module. The composite risk index It comprehensively reflects the triple coupling effect of electrical fault tendency, environmental hazard level and personnel exposure level.

6. The online monitoring system for temporary power supply safety at construction sites as described in claim 5, characterized in that: The specific process by which the edge consensus decision module generates a local power-off command is as follows: When any node satisfies the composite risk index ≥0.6 or residual current value exceeds safety limit At that time, construct a fault characteristic message. ,in For timestamps, For Center, radius Local topological subgraphs within; Through local wireless communication protocol Broadcast to all nearby smart circuit breaker nodes; Each neighboring node received Then, compare it with the residual current value it collected itself. With the message If satisfied Furthermore, the composite risk index at this node If the value is ≥0.6, a consistency confirmation message is returned; Let the total number of neighboring nodes within the broadcast range be . The minimum number of confirmations required is ,and ≥2; If within the preset time limit Received no less than If a consistency confirmation message is received, the edge consensus decision module will generate a local power-off command. ; The consistency judgment introduces the following risk synergy enhancement factor. Used to quantify the cumulative effect of risks across multiple nodes: ; in, Represents a node The set of neighboring nodes, The risk sensitivity coefficient is taken as a positive value; when Exceeding the preset collaboration threshold At that time, even if local Even if the risk level is not high, a consensus voting process can be triggered to achieve proactive protection driven by group risk perception.

7. The online monitoring system for temporary power supply safety at construction sites as described in claim 6, characterized in that: The preset time limit Fixed at three seconds; if in If the required number of consistent confirmation messages are not received within the specified timeframe, the distributed consensus process is abandoned, and the system is instead asked to determine whether the hard rule conditions are met. ; If the conditions are met, a local power-off command will be generated directly. Otherwise, maintain power supply.

8. The online monitoring system for temporary power supply safety at construction sites as described in claim 7, characterized in that: The specific process for the tiered intervention execution module to execute differentiated response actions is as follows: When a medium-risk assessment result is received, the sound and light alarm is activated to issue intermittent warnings, and the location of the risk node, a 3D animation of the risk cause, and a safe operation guide text are overlaid in the field of vision of the operator through the augmented reality terminal. When a high-risk assessment result or a local power outage command is received When the power supply to the corresponding intelligent circuit breaker is interrupted, a disconnection control signal is sent to the circuit breaker to forcibly cut off the power supply to that branch. Synchronously generate event logs ,in For ambient relative humidity, For nodes Distance from the nearest flooded area For the most recent workers and nodes Spatial distance, This is a snapshot of the local topology at the moment of power failure; right Calculate hash digest and will and After combining and encrypting, the data is uploaded to the remote monitoring platform. The integrity of the event log is ensured by the following verification formula: ; in, This indicates a field concatenation operation. This indicates that the adjacency matrix of the topological snapshot is expanded row-wise into a one-dimensional vector; this hash digest Used to verify that logs have not been tampered with during transmission and storage, supporting incident accountability and security auditing.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the online monitoring system for temporary power supply safety assurance at construction sites as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the online monitoring system for temporary power supply safety assurance at construction sites as described in any one of claims 1 to 8.

Images