An electrical fire real-time monitoring system based on wireless sensor network

Abstract

The application discloses a kind of electrical fire real-time monitoring system based on wireless sensor network, it is related to electrical fire monitoring technical field, including cloud monitoring platform, the electrical data acquisition module, electrical fire anomaly analysis module, abnormal early warning module, node management configuration module and visual monitoring module are connected with communication to cloud monitoring platform, wherein, electrical signal connection between each module;The electrical data acquisition module is used to utilize the monitoring data of electrical node of deployed wireless sensor acquisition, and monitoring data is transmitted to cloud monitoring platform by wireless communication technology, wherein, monitoring data includes current, voltage, temperature, humidity and leakage current.The wireless sensor of the present application is deployed in the key node of electrical equipment, and the fusion monitoring of multiple parameters is carried out, compared with traditional wired sensor, wireless sensor does not need wiring, installation is flexible, can more comprehensively capture early signal of electrical fire, significantly improve the accuracy of fire warning.

Description

Technical Field

[0001] This invention relates to the field of electrical fire monitoring technology, and specifically to a real-time electrical fire monitoring system based on a wireless sensor network. Background Technology

[0002] With the rapid development of industrialization and urbanization, and the widespread use of various electrical equipment and lines, the risk of electrical fires has also increased. Electrical fires are usually caused by aging electrical equipment, short circuits, overloads, poor contact, etc. The characteristics of electrical fires are that the initial fire is often hidden and difficult to detect, and if it is not detected and extinguished in time, the fire may spread rapidly, causing great casualties and property losses.

[0003] In existing technologies, traditional electrical fire monitoring systems rely on wired connections, which are not only complex to install and easily limited by the installation environment, but also have poor scalability, making it difficult to adapt to changes in building functions and updates to electrical equipment. This results in limited monitoring coverage and significant expansion difficulties. To address these issues, a real-time electrical fire monitoring system based on a wireless sensor network is proposed. By combining wireless networking and a wireless sensor network, the system improves the level of electrical fire monitoring, enabling real-time monitoring and early warning of electrical fires, thereby solving the aforementioned problems. Summary of the Invention

[0004] The purpose of this invention is to provide a real-time electrical fire monitoring system based on a wireless sensor network to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A real-time electrical fire monitoring system based on a wireless sensor network includes a cloud monitoring platform. The cloud monitoring platform is communicatively connected to an electrical data acquisition module, an electrical fire anomaly analysis module, an anomaly early warning module, a node management and configuration module, and a visualization monitoring module. The modules are connected by electrical signals.

[0007] The electrical data acquisition module is used to collect monitoring data of electrical nodes using deployed wireless sensors and transmit the monitoring data to the cloud monitoring platform through wireless communication technology. The monitoring data includes current, voltage, temperature, humidity and leakage current.

[0008] The electrical fire anomaly analysis module is used to extract monitoring data from the cloud monitoring platform, combine it with the constructed electrical fire identification model to analyze fire risk trends, and identify potential hazards.

[0009] The abnormal early warning module is used to combine the electrical fire identification model with the output and preset early warning rules to trigger graded early warning measures;

[0010] The node management and configuration module is used for remote management and optimization of the wireless sensor network;

[0011] The visualization monitoring module is used to display the electrical system topology diagram, real-time monitoring parameters and early warning records through a mobile APP, and provides historical curve comparison to assist in fault diagnosis.

[0012] A further improvement of the technical solution of the present invention is that: the electrical data acquisition module includes a wireless sensor node unit and a wireless communication unit;

[0013] The wireless sensor node unit is used to collect monitoring data including current, voltage, temperature, humidity and leakage current through wireless sensors deployed at key nodes of electrical equipment.

[0014] The wireless communication unit is used to transmit data between the wireless sensor node unit and the cloud monitoring platform through wireless networking technology. It has a wide coverage, adapts to complex environments, supports multi-hop relay, and solves the signal obstruction problem.

[0015] A further improvement to the technical solution of the present invention is that the wireless sensor node unit specifically includes:

[0016] Based on the layout of electrical equipment and monitoring requirements, the key node locations for wireless sensors are determined. The wireless sensors will be installed at the key node locations of electrical equipment (distribution boxes, cable joints, etc.) by magnetic attraction, adhesive, or clipping. After deployment, the wireless sensors at each node will automatically start the initialization program, complete the hardware self-test (sensor calibration, communication module activation), establish a connection with the gateway through the wireless protocol, and then synchronously configure the monitoring parameters (sampling frequency, threshold range) to ensure that the node enters the standby state.

[0017] The wireless sensors at each node synchronously collect current, voltage, temperature, humidity and leakage current data according to the monitoring parameters. The acquisition process adopts time-division multiplexing technology to avoid signal interference, and converts analog signals into digital signals through the built-in ADC (analog-to-digital converter).

[0018] The wireless sensor node unit packages the processed data, adds timestamps and node number information to form a standardized data packet. The timestamp ensures that the data sequence is traceable, and the node number enables the location of the specific device. It then establishes a connection with the wireless communication unit, enters the transmission queue, and waits for the transmission command.

[0019] A further improvement to the technical solution of the present invention is that the wireless communication unit specifically includes:

[0020] After the wireless communication unit is powered on, it performs hardware initialization, scans the surrounding wireless channels, automatically discovers connectable gateways and nearby nodes, and constructs a network topology and determines the optimal communication path by combining the preset wireless networking protocol (LoRaWAN or Zigbee).

[0021] The wireless sensor node unit sends the encapsulated data packet (including timestamp and node number) to the wireless communication unit. The wireless communication unit uses its built-in caching mechanism to store the data packet according to the first-in-first-out (FIFO) principle to avoid loss due to excessive instantaneous data volume. Then, it waits for an appropriate time window to transmit the data packet to the cloud monitoring platform.

[0022] Before data transmission, the wireless communication unit encrypts the data packet. After transmission is completed, the wireless communication unit returns to sleep mode, waiting for the next trigger command.

[0023] A further improvement to the technical solution of the present invention is that the electrical fire anomaly analysis module specifically includes:

[0024] Monitoring data is extracted from the cloud monitoring platform, linked and integrated according to timestamp and node number, and the data quality is improved through data cleaning (noise removal and missing value filling) and normalization (data scaling to a uniform dimension). Feature analysis is performed on the monitoring data to extract risk features related to electrical fires, including abnormal current features, abnormal voltage features, abnormal temperature features, abnormal humidity features, and leakage current features.

[0025] The extracted risk features are analyzed to determine the sub-indicators of each risk feature. Based on electrical safety standards, the corresponding safety thresholds of the sub-indicators of each risk feature are set to form a risk feature library.

[0026] The extracted risk feature data is input into a pre-built electrical fire identification model to analyze the anomalies of each risk feature and identify the sub-indicators of the risk features that are abnormal. Specifically, an electrical fire identification model is built based on a neural network model using historical monitoring data and known electrical fire cases. The feature data extracted from the historical monitoring data is input into the neural network model for training to obtain the final electrical fire identification model.

[0027] By analyzing the abnormal situations of sub-indicators of each risk characteristic and combining them with preset safety thresholds, the fire risk trend value is calculated comprehensively, the fire risk trend is analyzed, and potential electrical safety hazards are identified.

[0028] A further improvement to the technical solution of this invention lies in the fact that the sub-indicators of each risk characteristic are specifically as follows:

[0029] The sub-indicators of the current anomaly characteristics are overload current, current mutation, and harmonic distortion rate.

[0030] The sub-indicators of the voltage anomaly characteristics are voltage deviation, voltage fluctuation, and three-phase imbalance.

[0031] The sub-indicators of the temperature anomaly characteristics are the rate of temperature rise, absolute temperature threshold, and temperature gradient difference.

[0032] The sub-indicators of the humidity anomaly characteristics are relative humidity and humidity abrupt change;

[0033] The sub-indicators of the leakage current characteristics are leakage current threshold, leakage current fluctuation, and leakage duration.

[0034] A further improvement to the technical solution of the present invention is that the anomaly warning module specifically includes:

[0035] Receive the output results of the electrical fire identification model from the electrical fire anomaly analysis module, and analyze the fire risk trend value and related anomaly sub-indicator information;

[0036] Based on the preset early warning rules and fire risk trend values, different risk levels are divided into low risk level, medium risk level and high risk level, and corresponding early warning thresholds and early warning measures are matched for each risk level.

[0037] Based on the determined risk level, select corresponding measures from the pre-set early warning scheme library;

[0038] After an alert is triggered, the processing progress is continuously tracked, on-site feedback results are received (hazards have been eliminated, false alarms have been confirmed), and processing records are uploaded to the cloud monitoring platform.

[0039] A further improvement of the technical solution of the present invention is that the node management configuration module includes a node self-diagnosis unit and a node optimization unit;

[0040] The node self-diagnostic unit is used to detect the status of the wireless sensor node, including its working status, communication status, detection accuracy, and battery level, to troubleshoot faults or abnormal information of the wireless sensor, and to issue maintenance notifications.

[0041] The node optimization unit is used to temporarily remove wireless sensor nodes based on the self-diagnosis results of the nodes and dynamically adjust the network topology of the wireless sensor network.

[0042] A further improvement to the technical solution of this invention is that the node management configuration module specifically includes:

[0043] The node self-diagnostic unit periodically collects key status data of the wireless sensor nodes, including sensor operating status (sampling frequency, data output stability), communication status (signal strength, packet loss rate), detection accuracy (deviation from the reference value), and battery power (remaining capacity, battery discharge rate). Through the built-in protocol parsing module, the raw data is converted into structured information and associated with timestamps and node numbers to form a complete status log.

[0044] Fault features are extracted from the status log, including sampling frequency deviation, number of communication interruptions, abnormal signal strength, abnormal packet loss rate, detection value drift rate, abnormal remaining capacity, and battery discharge rate. The extracted fault features are compared with preset fault feature benchmark thresholds to identify potential abnormal patterns of wireless sensors and troubleshoot existing wireless sensor faults.

[0045] The system will push the identified wireless sensor faults or anomalies to the maintenance personnel, issue maintenance notices, and mark the wireless sensors with faults or anomalies.

[0046] A further improvement to the technical solution of the present invention is that the node optimization unit specifically includes:

[0047] The node optimization unit obtains the self-diagnosis results of the wireless sensor nodes from the node self-diagnosis unit, parses the node number and fault / abnormal information, locates the position of the faulty node in the topology by associating with the network topology database, identifies its directly connected parent node, child node and associated links, and generates a list of faulty nodes.

[0048] Based on the list of faulty nodes, the topology of the current wireless sensor network is analyzed to identify faulty or abnormal nodes, thereby determining the areas that need to be adjusted.

[0049] Based on the analysis results, assess the impact of removing faulty or abnormal wireless sensor nodes on the network topology, conduct a rehearsal of removing faulty or abnormal wireless sensor nodes, calculate the coverage of the remaining links after removal, analyze the load changes and redundancy of parent nodes, identify whether it affects key monitoring areas, and if the network still meets the criteria of coverage > 95% and load < 80% after removal, generate a removal instruction and plan a topology reconstruction scheme.

[0050] The system executes removal commands to temporarily isolate faulty or abnormal wireless sensor nodes, dynamically adjusts the network topology, reattaches the child nodes of faulty or abnormal wireless sensor nodes to nearby healthy nodes, enables backup communication paths, and reallocates data traffic using Dijkstra's shortest path algorithm. After adjustment, it verifies network connectivity and data accuracy to ensure topology stability.

[0051] Due to the adoption of the above technical solution, the technical progress achieved by this invention compared to the prior art is as follows:

[0052] 1. This invention provides a real-time electrical fire monitoring system based on a wireless sensor network. By deploying wireless sensors at key nodes of electrical equipment, multi-parameter fusion monitoring is performed. Compared with traditional wired sensors, wireless sensors do not require wiring, are flexible in installation, and can more comprehensively capture early signals of electrical fires, significantly improving the accuracy of fire early warning.

[0053] 2. This invention provides a real-time electrical fire monitoring system based on a wireless sensor network. It adopts wireless networking technology, supports multi-hop relay, solves the signal obstruction problem, and enables the monitoring system to cover a wider area and adapt to complex environments. At the same time, the wireless sensor network is easy to expand, and the sensor nodes can be flexibly adjusted according to the layout of electrical equipment and monitoring needs, so as to meet the needs of building function adjustment and electrical equipment update, thereby improving the scalability and flexibility of the system.

[0054] 3. This invention provides a real-time electrical fire monitoring system based on a wireless sensor network. By combining an electrical fire identification model and preset early warning rules, the system analyzes and processes the monitoring data to achieve intelligent early warning. Based on the fire risk trend value, different risk levels can be classified and corresponding early warning measures can be matched to achieve graded response, which can more effectively deal with electrical fire risks. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0056] Figure 1 This is a schematic diagram of the functional modules of the system of the present invention;

[0057] Figure 2 This is a schematic diagram of the workflow of the electrical fire anomaly analysis module of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0059] Example 1, as Figure 1 , Figure 2 As shown, the present invention provides a real-time electrical fire monitoring system based on a wireless sensor network, including a cloud monitoring platform. The cloud monitoring platform is communicatively connected to an electrical data acquisition module, an electrical fire anomaly analysis module, an anomaly early warning module, a node management and configuration module, and a visualization monitoring module, wherein the modules are connected by electrical signals.

[0060] The electrical data acquisition module is used to collect monitoring data of electrical nodes using deployed wireless sensors and transmit the monitoring data to the cloud monitoring platform through wireless communication technology. The monitoring data includes current, voltage, temperature, humidity and leakage current. The electrical data acquisition module includes a wireless sensor node unit and a wireless communication unit.

[0061] The wireless sensor node unit is used to collect monitoring data including current, voltage, temperature, humidity, and leakage current through wireless sensors deployed at key nodes of electrical equipment. It replaces traditional wired sensors, eliminating the need for wiring, offering flexible installation, supporting multi-parameter fusion monitoring, and improving the accuracy of fire early warning. Based on the layout of the electrical equipment and monitoring requirements, the key node locations for the wireless sensors are determined, and the sensors are installed at these locations (distribution boxes, cable joints, etc.) using magnetic, adhesive, or clip-on methods. After deployment, each node's wireless sensor automatically initiates an initialization program, completing hardware self-tests (sensor calibration, communication module activation), establishing a connection with the gateway via wireless protocol, and synchronously configuring monitoring parameters (sampling frequency, threshold range) to ensure the node enters a standby state. The wireless sensor at the point synchronously collects current, voltage, temperature, humidity, and leakage current data according to the monitoring parameters. The acquisition process adopts time-division multiplexing technology to avoid signal interference. The analog signal is converted into a digital signal through the built-in ADC (analog-to-digital converter). The wireless sensor node unit packages the processed data, adds timestamp and node number information to form a standardized data packet. The timestamp ensures the data time sequence is traceable, and the node number enables the location of the specific device. It then establishes a connection with the wireless communication unit, enters the transmission queue, and waits for the transmission command. During non-acquisition cycles, the wireless sensor node unit enters a low-power sleep mode, retaining only the timed wake-up function. The sleep current is less than 10μA, extending battery life (up to 5-10 years). After waking up, the node quickly resumes working state and performs a new round of data acquisition and transmission.

[0062] The wireless communication unit is used for data transmission between wireless sensor node units and the cloud monitoring platform via wireless networking technology. It boasts wide coverage, adaptability to complex environments, and supports multi-hop relay to resolve signal obstruction issues. Upon power-up, the wireless communication unit performs hardware initialization, scans surrounding wireless channels, automatically discovers connectable gateways and neighboring nodes, and constructs a network topology based on a preset wireless networking protocol (LoRaWAN or Zigbee), determining the optimal communication path. It also supports a multi-hop relay mechanism, extending network coverage through inter-node data forwarding. The wireless sensor node unit sends the encapsulated data packet (including timestamp and node number) to the wireless communication unit. The wireless communication unit utilizes its built-in buffering mechanism to store data packets according to the first-in-first-out (FIFO) principle, avoiding data loss due to excessive instantaneous data volume. It then waits for an appropriate time window to transmit the data packets to the cloud monitoring platform. Before data transmission, the wireless communication unit encrypts the data packets to prevent data leakage or tampering. The transmission process adopts a low-power mode, reducing energy consumption through duty cycle optimization (1% duty cycle). A two-way confirmation mechanism is established between the wireless communication unit and the cloud monitoring platform. After the receiving end successfully parses the data, it sends back an ACK signal, triggering the sending end to clear the buffered data. After the transmission is completed, the wireless communication unit returns to the sleep state, waiting for the next trigger command.

[0063] The electrical fire anomaly analysis module is used to extract monitoring data from the cloud monitoring platform, combine it with the constructed electrical fire identification model to analyze fire risk trends, identify potential hazards, and integrate monitoring data extracted from the cloud monitoring platform by timestamp and node number. Data quality is improved through data cleaning (noise removal, missing value filling) and normalization (scaling data to a uniform dimension). Feature analysis is performed on the monitoring data to extract risk features related to electrical fires, including abnormal current, voltage, temperature, humidity, and leakage current features. Each extracted risk feature is analyzed to determine its sub-indicators, and these are then combined with electrical safety standards (GB). (50016-2014) Set corresponding safety thresholds for sub-indicators of each risk feature to form a risk feature library. Input the extracted risk feature-related data into a pre-built electrical fire identification model to analyze the abnormality of each risk feature and identify the sub-indicators of the risk feature with abnormality. Among them, using historical monitoring data and known electrical fire cases, an electrical fire identification model is built based on a neural network model. The feature data extracted from the historical monitoring data is input into the neural network model for training. By adjusting the parameters of the model, normal and abnormal electrical operating states are identified. The trained model is verified and optimized to ensure the accuracy and generalization ability of the model. The model performance is evaluated using cross-validation method. Based on the verification results, the model is further optimized and adjusted to improve its ability to identify electrical fires, resulting in the final electrical fire identification model. Analyze the abnormality of the sub-indicators of each risk feature, combine the preset safety thresholds, comprehensively calculate the fire risk trend value, analyze the fire risk trend, and identify potential electrical safety hazards.

[0064] The specific sub-indicators for each risk characteristic are as follows:

[0065] Sub-indicators of abnormal current characteristics include overload current, current mutation, and harmonic distortion rate. Overload current is defined as a continuous current exceeding 1.2 times the rated current, which may be caused by equipment overload or short circuit leading to fire. Current mutation is defined as a current fluctuation exceeding 30% in a short period of time, which may be caused by poor contact or insulation aging leading to local overheating. Harmonic distortion rate is defined as the proportion of the third harmonic >15%, indicating that nonlinear load causes additional heat loss and accelerates insulation aging. Sub-indicators of abnormal voltage characteristics include voltage deviation, voltage fluctuation, and three-phase imbalance. Voltage deviation is defined as a continuous deviation from the rated voltage ±10%, which may be caused by power supply abnormalities or equipment failure leading to overload. Voltage fluctuation is defined as a frequency >5 times / minute and an amplitude >5%, which may be caused by poor contact or arc discharge leading to sparks. Three-phase imbalance is defined as a three-phase current / voltage deviation >15%, leading to local overheating. Sub-indicators of abnormal temperature characteristics include temperature rise rate, absolute temperature threshold, and temperature gradient difference. Temperature rise rate is defined as a temperature rise rate within 10 minutes. An internal temperature rise of >10℃ may be due to localized overheating caused by a short circuit or poor contact. The absolute temperature threshold is a critical node temperature >80℃, exceeding the insulation material's tolerance limit. A temperature gradient difference of >20℃ between different parts of the same equipment indicates an internal fault. Sub-indicators of abnormal humidity characteristics are relative humidity and humidity mutation. Relative humidity >85% indicates that a long-term high humidity environment accelerates the corrosion of metal parts and reduces insulation performance. Humidity mutation is a humidity change of >30% in a short period of time, which may be due to condensation caused by equipment sealing failure, leading to a short circuit. Sub-indicators of leakage current characteristics are leakage current threshold, leakage current fluctuation, and leakage duration. A leakage current threshold of >30mA may be due to electric shock or fire caused by insulation damage. Leakage current fluctuation is >±10%, indicating unstable insulation condition. Leakage duration is >30 minutes, which may be due to insulation failure caused by equipment aging or moisture.

[0066] The expression for the fire risk tendency value is:

[0067] ;

[0068] In the formula, This is a fire risk directional value, a comprehensive indicator reflecting the overall fire risk of the electrical system. This represents the category of risk characteristics, with values ​​ranging from 1 to 5. For the first The number of sub-indicators of risk-like characteristics, For the first Sub-indicator indexes under risk-like characteristics, with values ​​ranging from 1 to... , For the first The weights of risk characteristics reflect the relative importance of that type of risk characteristic in the overall fire risk assessment; the sum of all weights is 1. For the first Under the risk characteristics of the first Monitoring values ​​of individual sub-indicators, For the first Under the risk characteristics of the first The safety threshold for each sub-indicator, when Exceed When this occurs, it indicates that the sub-indicator is abnormal. For the first Under the risk characteristics of the first The sensitivity coefficient of each sub-indicator reflects the degree of influence of that sub-indicator on fire risk, and its value ranges from 0 to 1. The higher the sensitivity coefficient, the greater the impact of that sub-indicator on fire risk. The range of values ​​should be ,, It is a relatively large positive number, the specific value of which depends on the situation of each sub-indicator and the parameter settings. When more sub-indicators show abnormalities (i.e. Exceed When the number of cases increases, It will increase; conversely, when the monitored values ​​of each sub-indicator are all below the safety threshold and the deviation is not significant, A value that is small or even close to 0 indicates a low fire risk. The sensitivity coefficients for overload current, current surge, harmonic distortion rate, voltage anomaly, voltage fluctuation, three-phase imbalance, temperature rise rate, absolute temperature threshold, temperature gradient difference, relative humidity, humidity surge, leakage current threshold, leakage current fluctuation, and leakage duration are all set at 0.6.

[0069] The anomaly warning module is used to trigger tiered warning measures by combining the electrical fire identification model with its output and preset warning rules. It receives the output results of the electrical fire identification model from the electrical fire anomaly analysis module, analyzes the fire risk trend value and related anomaly sub-indicator information, and classifies different risk levels—low risk, medium risk, and high risk—based on preset warning rules and fire risk trend values. It also matches corresponding warning thresholds and warning measures to each risk level. Based on the determined risk level, it selects corresponding measures from a preset warning scheme library. For low-risk levels, abnormal electrical nodes are included in 15-minute polling monitoring, and warning information is pushed to the on-duty personnel's terminal. For medium-risk levels, the maintenance team is notified to conduct on-site inspections, remotely cut off the power to non-critical equipment, and initiate area cooling / dehumidification programs. For high-risk levels, the circuit power to the abnormal electrical node and related electrical nodes is immediately cut off. Multiple methods, including SMS, APP notifications, and audible and visual alarms, are used to simultaneously trigger warnings, ensuring timely information transmission. After a warning is triggered, the module continuously tracks the processing progress, receives on-site feedback (hazard eliminated, false alarm confirmed), and uploads processing records to the cloud monitoring platform.

[0070] Multiple risk levels correspond one-to-one with multiple warning thresholds, and the specific correspondence is as follows:

[0071] The warning threshold for low-risk levels is: ;

[0072] The warning threshold for medium-risk level is: ;

[0073] The warning threshold for high-risk levels is: ;

[0074] in, This represents the fire risk trend value. These are the upper threshold for low-risk levels and the lower threshold for medium-risk levels. These are the upper threshold for medium-risk levels and the lower threshold for high-risk levels;

[0075] The node management and configuration module is used for remote management and optimization of wireless sensor networks, including dynamic adjustment of network topology, addition and removal of nodes, fault diagnosis and recovery of the network, maintaining network stability and efficiency, improving the reliability and real-time performance of data transmission, and adapting to the expansion and changes in system scale.

[0076] The visualization monitoring module is used to display the electrical system topology diagram, real-time monitoring parameters and early warning records through a mobile APP, and provides historical curve comparison to assist in fault diagnosis.

[0077] Example 2, as Figure 1 , Figure 2As shown, based on Embodiment 1, the present invention provides a technical solution: preferably, the node management configuration module includes a node self-diagnosis unit and a node optimization unit;

[0078] The node self-diagnostic unit is used to detect the status of wireless sensor nodes, including operating status, communication status, detection accuracy, and battery level. It identifies faults or anomalies in the wireless sensors and issues maintenance notifications, improving system reliability and maintainability. Through automatic diagnosis, it quickly alerts users before or during faults, preventing serious operational problems. The node self-diagnostic unit periodically collects key status data from the wireless sensor nodes, including sensor operating status (sampling frequency, data output stability), communication status (signal strength, packet loss rate), detection accuracy (deviation from reference value), and battery level (remaining capacity, battery discharge rate). Through a built-in protocol parsing module, it converts the raw data into structured information and associates it with timestamps and node numbers to form a complete status log. Fault characteristics, including sampling frequency deviation and communication... The system tracks and analyzes fault characteristics such as the number of interruptions, abnormal signal strength, abnormal packet loss rate, abnormal detection value drift rate, abnormal remaining capacity, and battery discharge rate. It compares these extracted fault features with preset fault characteristic thresholds to identify potential abnormal patterns in wireless sensors and pinpoint existing faults. Specifically, abnormal sampling frequency is defined as the actual value deviating from the set value by more than 10%; abnormal communication interruptions are defined as three consecutive communication failures; abnormal signal strength is defined as RSSI < -90dBm for 5 minutes; abnormal packet loss rate is defined as a packet loss rate > 10% for 10 minutes; abnormal detection value drift rate is defined as a deviation > 3% of the baseline value per hour; abnormal remaining capacity is defined as capacity < 15% and voltage < 3.6V; and abnormal battery discharge rate is defined as a rate > 10% / hour. The system then pushes the identified wireless sensor fault or abnormal information to maintenance personnel, issues maintenance notices, and marks the faulty or abnormal wireless sensors.

[0079] The node optimization unit, based on the self-diagnostic results of the nodes, temporarily removes wireless sensor nodes and dynamically adjusts the network topology of the wireless sensor network. The unit obtains the self-diagnostic results of the wireless sensor nodes from the node self-diagnostic unit, parses the node number and fault / abnormal information, locates the faulty node's position in the topology by associating it with the network topology database, identifies its directly connected parent nodes, child nodes, and associated links, and generates a list of faulty nodes. Based on this list, it analyzes the current wireless sensor network topology, identifies faulty or abnormal nodes, determines the areas requiring adjustment, assesses the impact of removing faulty or abnormal wireless sensor nodes on the network topology based on the analysis results, performs a rehearsal of the removal process, and calculates the coverage of the remaining links after removal. Coverage is assessed by analyzing parent node load changes and redundancy to identify whether they affect key monitoring areas. If the network still meets the criteria of coverage > 95% and load < 80% after removal, a removal command is generated, a topology reconstruction scheme is planned, and the removal command is executed. Faulty or abnormal wireless sensor nodes are temporarily isolated, and the network topology is dynamically adjusted. The child nodes of faulty or abnormal wireless sensor nodes are reattached to nearby healthy nodes, prioritizing nodes with signal strength > -70dBm. Backup communication paths are enabled to ensure data link redundancy ≥ 2. Data traffic is redistributed using Dijkstra's shortest path algorithm to avoid single-point load > 80%. After adjustment, network connectivity (Ping test success rate > 99%) and data accuracy (checksum matching rate 100%) are verified to ensure topology stability.

[0080] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A wireless sensor network based electrical fire real-time monitoring system comprising of a cloud monitoring platform, characterized in that: The cloud monitoring platform is connected with an electrical data acquisition module, an electrical fire anomaly analysis module, an anomaly early warning module, a node management configuration module and a visual monitoring module, wherein the modules are connected by electrical signals; The electrical data acquisition module is configured to collect monitoring data of electrical nodes by using the deployed wireless sensors and transmit the monitoring data to the cloud monitoring platform by wireless communication technology, wherein the monitoring data includes current, voltage, temperature, humidity and leakage current; The electrical fire anomaly analysis module is configured to extract the monitoring data from the cloud monitoring platform, analyze the fire risk trend in combination with the constructed electrical fire identification model, and identify potential hazards; The anomaly early warning module is configured to trigger graded early warning measures in combination with the electrical fire identification model and the output and preset early warning rules; The node management configuration module is configured to remotely manage and optimize the wireless sensor network, and includes a node self-diagnosis unit and a node optimization unit; The node self-diagnosis unit is configured to detect the state of the wireless sensor node, including the working condition, the communication state, the detection accuracy and the battery capacity, troubleshoot the faults or abnormal information of the wireless sensor, and send a maintenance notification; The node optimization unit is configured to temporarily exclude the wireless sensor node in combination with the self-diagnosis result of the node, dynamically adjust the network topology of the wireless sensor network, and specifically includes: The node optimization unit obtains the self-diagnosis result of the wireless sensor node from the node self-diagnosis unit, analyzes the node number and fault / abnormal information, locates the position of the fault node in the topology by associating with the network topology database, identifies the directly connected parent node, child node and associated link, and generates a fault node list; Based on the fault node list, the topology structure of the current wireless sensor network is analyzed to identify the nodes with faults or abnormalities, and then the area to be adjusted is determined; According to the analysis result, the influence of the excluded wireless sensor node with faults or abnormalities on the network topology is evaluated, a pre-performance of the excluded wireless sensor node with faults or abnormalities is performed, the coverage rate of the remaining link after exclusion is calculated, the parent node load change and redundancy are analyzed, and whether the key monitoring area is affected is identified, if the network after exclusion still meets the standards of coverage rate > 95% and load < 80%, an exclusion instruction is generated, and a topology reconstruction scheme is planned; The exclusion instruction is executed to temporarily isolate the wireless sensor node with faults or abnormalities, and dynamically adjust the network topology, the child node of the wireless sensor node with faults or abnormalities is re-mounted to the adjacent healthy node, the standby communication path is enabled, and the data flow is redistributed by using the Dijkstra shortest path algorithm, after the adjustment is completed, the network connectivity and data accuracy are verified; The visual monitoring module is configured to display the electrical system topology graph, real-time monitoring parameters and early warning records through a mobile APP.

2. The electrical fire real-time monitoring system based on wireless sensor network according to claim 1, characterized in that: The electrical data acquisition module includes a wireless sensor node unit and a wireless communication unit; The wireless sensor node unit is configured to collect monitoring data including current, voltage, temperature, humidity and leakage current by deploying wireless sensors at key nodes of electrical equipment. The wireless communication unit is configured to perform data transmission between the wireless sensor node unit and the cloud monitoring platform through a wireless networking technology.

3. The electrical fire real-time monitoring system based on wireless sensor network according to claim 2, characterized in that: The wireless sensor node unit specifically comprises: According to the electrical equipment layout and monitoring requirements, the key node positions of the wireless sensor are determined, and the wireless sensor is installed at the key node positions of the electrical equipment by magnetic attraction, adhesion or buckling. After deployment, the wireless sensor of each node automatically starts the initialization program, completes the hardware self-checking, establishes a connection with the gateway through the wireless protocol, and then synchronizes the monitoring parameters; The wireless sensor of each node synchronously collects current, voltage, temperature, humidity and leakage current data according to the monitoring parameters, and converts the analog signals into digital signals through the built-in ADC. The wireless sensor node unit packages the processed data, adds a timestamp and node number information, forms a standardized data packet, establishes a connection with the wireless communication unit, enters a transmission queue, and waits for a sending instruction.

4. The electrical fire real-time monitoring system based on wireless sensor network according to claim 2, characterized in that: The wireless communication unit specifically comprises: After the wireless communication unit is powered on, hardware initialization is performed, the surrounding wireless channels are scanned, the connectable gateway and adjacent nodes are automatically discovered, and the network topology is constructed in combination with the preset wireless networking protocol to determine the optimal communication path; The wireless sensor node unit sends the packaged data packet to the wireless communication unit, stores the data packet in the built-in buffer mechanism of the wireless communication unit according to the first-in-first-out principle, and then waits for an appropriate time window to transmit the data packet to the cloud monitoring platform; Before data transmission, the wireless communication unit performs encryption processing on the data packet, and after the transmission is completed, the wireless communication unit returns to a dormant state and waits for the next trigger instruction.

5. The electrical fire real-time monitoring system based on wireless sensor network according to claim 2, characterized in that: The electrical fire anomaly analysis module specifically comprises: The monitoring data is extracted from the cloud monitoring platform, associated and integrated according to the timestamp and node number, and the monitoring data is analyzed to extract risk features related to electrical fires, including current anomaly features, voltage anomaly features, temperature anomaly features, humidity anomaly features and leakage current features; The extracted risk features are analyzed to determine the sub-indices of the risk features, and the corresponding safety thresholds of the sub-indices of the risk features are set in combination with the electrical safety standards to form a risk feature library; The extracted risk feature related data is input into the pre-constructed electrical fire identification model to analyze the abnormal conditions of the risk features, identify the sub-indices of the risk features that are abnormal, wherein the electrical fire identification model is constructed based on a neural network model using historical monitoring data and known electrical fire cases, the feature data extracted from the historical monitoring data is input into the neural network model for training, and the final electrical fire identification model is obtained; The abnormal conditions of the sub-indices of the risk features are analyzed, the safety thresholds are combined, the fire risk trend value is calculated comprehensively, the fire risk trend is analyzed, and potential electrical safety hazards are identified.

6. The electrical fire real-time monitoring system based on wireless sensor network according to claim 5, characterized in that: The sub-indices of the risk features are specifically as follows: The sub-indices of the current anomaly features are overload current, current mutation and harmonic distortion rate; The sub-indices of the voltage anomaly features are voltage deviation, voltage fluctuation and three-phase unbalance degree; The sub-indices of the temperature anomaly feature are temperature rise rate, absolute temperature threshold and temperature gradient difference; The sub-indices of the humidity anomaly feature are relative humidity and humidity mutation; The sub-indices of the electric leakage current feature are electric leakage current threshold, electric leakage current fluctuation and electric leakage duration.

7. The electrical fire real-time monitoring system based on wireless sensor network according to claim 5, characterized in that: The abnormal early warning module specifically comprises: receiving the output result of the electrical fire identification model from the electrical fire anomaly analysis module, analyzing the fire risk trend value and related abnormal sub-index information; combining the preset early warning rules and the fire risk trend value, dividing different risk levels, respectively low risk level, medium risk level and high risk level, and matching corresponding early warning thresholds and early warning measures for each risk level; according to the determined risk level, selecting the corresponding measures from the preset early warning scheme library; after the early warning is triggered, continuously tracking the processing progress, receiving the field feedback result, and uploading the processing record to the cloud monitoring platform.

8. The electrical fire real-time monitoring system based on wireless sensor network according to claim 1, characterized in that: The node management configuration module specifically comprises: The node self-diagnosis unit periodically collects the key state data of the wireless sensor node, including the sensor working state, the communication state, the detection accuracy and the battery capacity, converts the original data into structured information through the built-in protocol analysis module, associates the time stamp and the node number, and forms a complete state log; extracting fault features from the state log, including sampling frequency deviation, communication interruption times, signal strength anomaly, packet loss rate anomaly, detection value drift rate, residual capacity anomaly and battery discharge rate, and comparing the extracted fault features with the preset fault feature benchmark threshold to identify the potential abnormal mode of the wireless sensor and troubleshoot the existing wireless sensor fault; pushing the troubleshooted wireless sensor fault or abnormal information to the operation and maintenance personnel, issuing a maintenance notice, and marking the wireless sensor with fault or abnormality.

Images