Fire safety management method and system for charging site

Abstract

The application relates to the technical field of fire management, in particular to a fire safety management method and system for a charging place, which performs multispectral data collection on the charging place, obtains safety monitoring data of the charging place, performs fire risk prediction on the charging place according to the safety monitoring data, generates fire safety supervision information of the charging place, analyzes the fire safety supervision information, and executes corresponding preset management countermeasures.

Description

Technical Field

[0001] This application relates to the technical field of fire protection management, and in particular to a fire safety management method and system for charging facilities. Background Technology

[0002] With the popularization of electric vehicles, electric bicycles and other electric transportation tools, the number of charging sites has increased dramatically, such as public charging stations and electric vehicle charging sheds in residential areas. These charging sites often concentrate a large number of charging equipment and vehicles waiting to be charged. Once a fire occurs, it may cause serious casualties and property damage. Traditional fire safety monitoring mainly relies on single-type monitoring equipment such as smoke detectors and temperature sensors. These devices can only issue alarms when the fire develops to a certain stage, such as when smoke is produced or the temperature rises significantly, making it difficult to provide accurate early warnings in the early stages when there is a fire hazard. Summary of the Invention

[0003] Therefore, it is necessary to provide a fire safety management method and system for charging sites to address the aforementioned technical problems, which can provide accurate early warnings in the early stages of potential fire hazards.

[0004] Firstly, this application provides a method for fire safety management of charging facilities, the method comprising: Multispectral data was collected from charging sites to obtain safety monitoring data for the charging sites; Based on the aforementioned safety monitoring data, fire risks at charging locations are predicted, and fire safety supervision information for charging locations is generated. The fire safety supervision information is analyzed to implement corresponding preset management measures.

[0005] Secondly, this application also provides a fire safety management system for charging locations, used to implement the fire safety management method for charging locations as described in any one of the first aspects, comprising: The safety monitoring module is used to collect multispectral data from charging sites to obtain safety monitoring data for the charging sites. The risk prediction module is used to predict the fire risk of charging sites based on the safety monitoring data and generate fire safety supervision information for charging sites. The response processing module is used to parse the fire safety supervision information in order to execute corresponding preset management countermeasures.

[0006] The aforementioned fire safety management method for charging sites utilizes multispectral data acquisition to comprehensively obtain safety monitoring data, covering different spectral information, thus improving monitoring accuracy. Based on this, fire risk prediction can be performed, potential hazards can be identified in advance, and reliable fire safety supervision information can be generated. Pre-set management countermeasures can be implemented based on information analysis, and targeted measures can be taken at different risk levels, such as fire extinguishing, shutting down charging piles, and reducing power, to effectively prevent and respond to fires, protect the safety of personnel and property in charging sites, and reduce fire losses. Attached Figure Description

[0007] Figure 1 This is a schematic diagram illustrating the steps of a fire safety management method for a charging location in one embodiment; Figure 2 This is a schematic diagram of the fire safety management system of a charging site in one embodiment. Detailed Implementation

[0008] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0009] The fire safety management method for charging sites provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown: S1: Collect multispectral data from the charging site to obtain safety monitoring data for the charging site; S2: Based on the aforementioned safety monitoring data, predict the fire risk of the charging site and generate fire safety supervision information for the charging site; S3: Analyze the fire safety supervision information to execute corresponding preset management measures.

[0010] Specifically, in step S1 of the embodiment provided in this application, a multispectral AI fire detector pre-installed at the charging location is used to collect visual data of the charging location through multiple spectral channels. These spectral channels include visible light, infrared light, and ultraviolet light. The resulting safety monitoring data includes visible light monitoring data, infrared light monitoring data, and ultraviolet light monitoring data. Data from different spectral channels can reflect the safety status of the charging location from different perspectives, complementing each other and helping to more comprehensively and accurately monitor and assess the fire safety situation of the charging location. A detailed analysis follows: More specifically, the visible light channel of the multispectral AI fire detector collects image and video data of the charging site, forming visible light monitoring data. This data can present an intuitive scene of the charging site, such as the appearance of the equipment and the activities of personnel. Since the human eye mainly observes the world through visible light, the data from the visible light channel is consistent with human intuitive cognition. It can directly observe whether there are obvious abnormalities in the charging site, such as equipment damage, debris accumulation, electrical sparks, and flames. It can provide an important basis for the preliminary judgment of fire risk. For example, if obvious sparks are found in the visible light monitoring data of the charging pile, it can be directly determined that there may be a safety hazard.

[0011] More specifically, infrared light channels are used to collect infrared radiation data emitted by objects in the charging area, generating infrared light monitoring data. This data can reflect the temperature distribution on the surface of objects. Many fire hazards manifest as abnormal local temperature increases in the early stages. Through infrared light monitoring, these potential high-temperature areas can be detected in time, such as internal circuit faults in charging piles and overheating of charging batteries. Infrared light is not limited by lighting conditions and can work normally even at night or in low-light environments. It can monitor the temperature changes of equipment and objects in the charging area in real time and provide early warning of fire risks.

[0012] More specifically, the ultraviolet light channel of the multispectral AI fire detector collects ultraviolet radiation data in the charging area to obtain ultraviolet monitoring data. Electric sparks, electric arcs, etc., emit ultraviolet light when they are generated. By monitoring ultraviolet light, it is possible to detect whether there are abnormal electrical discharge phenomena in the charging area. This is very important for discovering potential faults and fire hazards in electrical equipment. For example, when the electrical connection of the charging pile is loose or short-circuited, it may generate electric sparks. The ultraviolet light channel can capture these abnormal signals, thereby discovering problems in time and taking measures.

[0013] Specifically, in step S2 of the embodiment provided in this application, the electrical sparks and flames of each wireless charging pile and charging electric vehicle in the charging area are identified based on the visible light monitoring data in the safety monitoring data to obtain spark monitoring information. Image recognition algorithms can be used to analyze the images and videos in the visible light monitoring data to identify the characteristics of electrical sparks and flames, and record information such as their location, time, and size. Electrical sparks and flames are direct signs of fire. In the charging area, the electrical systems of wireless charging piles and charging electric vehicles may generate electrical sparks due to short circuits, poor contact, etc. If an electrical spark ignites flammable materials, it will form a flame, which can then cause a fire. By identifying electrical sparks and flames, early signs of fire can be detected in a timely manner, providing an important basis for subsequent fire risk prediction and response.

[0014] More specifically, the model information of wireless charging piles and electric vehicles is determined based on visible light monitoring data. Structural design data of wireless charging piles and electric vehicles are then retrieved from the database. The internal structure of wireless charging piles and electric vehicles is digitally simulated to obtain an internal structure model. The infrared and ultraviolet monitoring data in the safety monitoring data are used to analyze the infrared and ultraviolet intensity of the internal structure model. Based on the analysis results, temperature data is labeled on the internal structure model to obtain the temperature monitoring information of each wireless charging pile and each electric vehicle.

[0015] More specifically, many fires are preceded by an abnormal rise in temperature. Different models of wireless charging stations and electric vehicles have different internal structures and thermal characteristics. By establishing an internal structural model through digital simulation, the internal temperature distribution can be analyzed more accurately. Infrared and ultraviolet rays are closely related to the temperature of an object. By analyzing the intensity of the infrared and ultraviolet bands, temperature data can be annotated onto the internal structural model, thereby enabling real-time monitoring of temperature changes inside wireless charging stations and electric vehicles and timely detection of potential fire hazards.

[0016] More specifically, the spark monitoring information is analyzed to determine the severity and frequency of spark occurrence, yielding the first risk characteristic. For example, the severity of a spark is judged based on its size and duration, and the frequency of spark occurrence is calculated by counting the number of sparks per unit time. Similarly, abnormally high temperatures and abnormal temperature rises are analyzed in the temperature monitoring information to obtain the second risk characteristic. Normal temperature ranges and temperature rise rate thresholds are set; when the temperature exceeds the threshold or the temperature rise rate is too rapid, it is considered abnormal.

[0017] More specifically, by combining the first risk feature and the second risk feature, the fire risk of wireless charging piles and electric vehicles can be predicted, and the fire risk prediction information of each wireless charging pile and electric vehicle can be obtained. Machine learning algorithms or expert systems can be used to calculate the probability of fire occurrence based on the first risk feature and the second risk feature.

[0018] More specifically, the severity and frequency of sparks reflect the severity and stability of electrical system faults. Frequent and severe sparks indicate significant safety hazards in the electrical system and a higher probability of fire. Abnormally high temperatures and abnormal temperature rises are important precursors to fires. When equipment temperature rises abnormally, it may be due to internal short circuits, overloads, or other reasons, which can easily lead to fires. By comprehensively considering the risk characteristics of both sparks and temperature, the fire risk of charging sites can be predicted more comprehensively and accurately, generating reliable fire safety supervision information.

[0019] More specifically, by pre-installing sensor arrays within wireless charging stations at charging locations, data on the actual operation and performance of the wireless charging stations are collected, generating equipment operation monitoring data to supplement fire risk prediction for wireless charging stations. The sensor arrays can include current sensors, voltage sensors, power sensors, etc., collecting operating parameters such as current, voltage, and power of the charging stations. The operating status and performance of wireless charging stations are closely related to fire risk. For example, excessive current or unstable voltage may cause the charging station to overheat, increasing the risk of fire. By collecting equipment operation monitoring data, a more comprehensive understanding of the operation of wireless charging stations can be obtained, providing more data support for fire risk prediction and improving the accuracy of predictions.

[0020] Specifically, in step S3 of the embodiment provided in this application, the fire safety supervision information is analyzed, and the risk level of the charging site is determined according to the preset risk assessment standards. These standards may comprehensively consider factors such as spark monitoring information (e.g., the severity and frequency of sparks), temperature monitoring information (abnormal high temperature and abnormal temperature rise), and equipment operation monitoring data. For example, if obvious flames and a sharp increase in temperature are found in the fire safety supervision information, it can be determined that a fire has occurred; if frequent and severe electric sparks occur, and the temperature is close to the danger threshold, it is determined that a fire is about to occur; if there are only slight temperature abnormalities or occasional small sparks, it is determined that a fire is predicted to occur in the future. Different risk situations require different countermeasures. Accurately judging the risk level is the basis for formulating reasonable management countermeasures. Only by clarifying the risk level can targeted actions be taken to avoid overreaction or underreaction, thereby effectively ensuring the safety of the charging site.

[0021] More specifically, if the risk level indicates that a fire has already occurred, the pre-set fire response devices will be activated to extinguish the fire. These devices include drones and robots equipped with fire-fighting modules. Drones can quickly reach the fire scene to monitor and extinguish the fire from the air, making them suitable for large-scale fires or hard-to-reach areas. Robots can perform fire-fighting operations on the ground, penetrating deep into the fire scene to precisely target the fire source. When a fire has already occurred, it is necessary to take effective fire-fighting measures quickly to reduce the losses caused by the fire. Drones and robots equipped with fire-fighting modules are characterized by rapid response and flexibility, enabling them to reach the fire scene and extinguish the fire in a short time, improving fire-fighting efficiency and reducing the risk of casualties and property damage.

[0022] More specifically, if the risk level is an impending fire, the corresponding wireless charging station will be shut down and an early warning message will be sent. The power supply to the charging station will be cut off through the charging station's control system to stop charging operations. At the same time, early warning messages will be sent to relevant personnel (such as charging site managers and maintenance personnel) via SMS, email, and audible and visual alarms to inform them of the safety hazards present in the charging site. When a fire is about to occur, timely shutdown of the charging station can prevent the fire from spreading further and reduce the source of ignition. Sending early warning messages allows relevant personnel to understand the situation in a timely manner and take appropriate measures, such as conducting on-site inspections and repairs, to prevent the fire from occurring.

[0023] More specifically, if the risk level is a predicted future fire, the operating efficiency of the corresponding wireless charging station will be reduced, and an early warning message will be sent. This can be achieved by adjusting the output power of the charging station, reducing the charging current, and decreasing the heat generated by the equipment. At the same time, early warning messages will also be sent to relevant personnel via SMS, email, and audible and visual alarms to remind them to pay attention to the safety of the charging site. When a fire is predicted to occur, reducing the operating efficiency of the charging station can reduce the heat generated by the equipment and decrease the likelihood of a fire. Sending early warning messages allows relevant personnel to prepare in advance, inspect and maintain the charging site, and eliminate potential safety hazards.

[0024] More specifically, while issuing early warning information, key information from fire safety supervision data is identified to generate reference information for maintenance personnel. This key information may include the charging pile number where the anomaly occurred, the specific details of the anomaly (such as the location of sparks or abnormal temperatures), etc. This information is compiled into a report or document and sent to maintenance personnel. After receiving the early warning information, maintenance personnel need to understand the specific fault situation in order to carry out effective repairs. The generated reference information provides maintenance personnel with detailed fault information, helping them to quickly locate the problem, improve repair efficiency, and thus eliminate safety hazards as soon as possible, ensuring the normal operation of the charging site.

[0025] This application provides a fire safety management method for charging sites, which has the following beneficial effects: The aforementioned fire safety management method for charging sites utilizes multispectral data acquisition to comprehensively obtain safety monitoring data, covering different spectral information, thus improving monitoring accuracy. Based on this, fire risk prediction can be performed, potential hazards can be identified in advance, and reliable fire safety supervision information can be generated. Pre-set management countermeasures can be implemented based on information analysis, and targeted measures can be taken at different risk levels, such as fire extinguishing, shutting down charging piles, and reducing power, to effectively prevent and respond to fires, protect the safety of personnel and property in charging sites, and reduce fire losses.

[0026] In one embodiment, the step of acquiring multispectral data of a charging location to obtain safety monitoring data of the charging location includes: By using a multispectral AI fire detector pre-installed in the charging location, visual data from multiple spectral channels is collected from the charging location, resulting in safety monitoring data composed of safety monitoring data from each spectral channel. Among them, the spectral channels include visible light channels, infrared light channels, and ultraviolet light channels; The security monitoring data includes visible light monitoring data, infrared light monitoring data, and ultraviolet light monitoring data.

[0027] In one embodiment, the step of predicting the fire risk of the charging site based on the safety monitoring data and generating fire safety supervision information for the charging site includes: S21: Based on the visible light monitoring data in the safety monitoring data, identify the electric sparks and flames of each wireless charging pile and electric vehicle in the charging area to obtain spark monitoring information. S22: Based on the infrared monitoring data and ultraviolet monitoring data in the safety monitoring data, identify the internal structural temperature of each wireless charging pile and electric vehicle in the charging site to obtain temperature monitoring information. S23: Combine the spark monitoring information and the temperature monitoring information to predict the fire risk of wireless charging piles and electric vehicles, and obtain the corresponding fire risk prediction information for each wireless charging pile and electric vehicle. S24: Fire risk prediction information for each wireless charging station and electric vehicle being charged constitutes fire safety supervision information for charging sites.

[0028] Specifically, preprocessing operations such as denoising and contrast enhancement are performed on images or video frames in visible light monitoring data to improve image quality and facilitate subsequent feature extraction and recognition. Computer vision technology is used to extract features representing electric sparks or flames from the images, such as color, shape, and texture. For example, flames usually have specific colors such as red and orange, and their shapes are irregular and their edges flicker. The extracted features are matched with pre-defined patterns of electric sparks and flames. Machine learning classifiers (such as convolutional neural networks CNN) are used to determine whether electric sparks and flames exist, and their location, occurrence time, duration, and other information are recorded to form spark monitoring information.

[0029] More specifically, electrical sparks and flames are obvious and intuitive signs of fire. During the charging process, electrical sparks are generated in the electrical systems of wireless charging piles and electric vehicles due to faults such as short circuits and poor contact, which in turn ignite flames. By identifying them, early and obvious safety hazards of charging facilities can be detected in a timely manner, providing a direct basis for fire risk prediction. Visible light monitoring data is relatively easy to acquire, requiring no complex equipment or special environmental conditions. As long as there is a suitable visible light camera, it can be continuously collected, which is low in cost and highly practical.

[0030] More specifically, based on visible light monitoring data, the models of wireless charging piles and electric vehicles are initially determined. Corresponding structural design data is retrieved from the database to construct an internal structural model. Infrared monitoring data is analyzed to obtain thermal radiation information of the object's surface and determine the surface temperature distribution. Ultraviolet monitoring data is processed and, combined with specific physical models and algorithms, its potential temperature-related information is analyzed. The obtained temperature information is then labeled onto the internal structural model. Multispectral data is fused and processed, taking into account both infrared and ultraviolet data, to more accurately reflect the actual temperature of the internal structure and form temperature monitoring information.

[0031] More specifically, many fires involve a gradual increase in temperature before they occur. By monitoring the temperature of the internal structure of wireless charging stations and electric vehicles, potential overheating hazards, such as aging electrical components or overload-induced heating, can be detected before a fire breaks out. This allows for proactive measures to prevent fires. Infrared data directly reflects the surface temperature of an object, while ultraviolet data can provide additional information related to high temperatures in certain situations, such as ultraviolet radiation generated by electrical discharges. Combining the two provides a more comprehensive and accurate understanding of the internal temperature state of the equipment, overcoming the limitations of single-spectrum data.

[0032] More specifically, the severity of sparks (such as spark size and energy) and frequency of spark occurrence in spark monitoring information are quantitatively analyzed; abnormal high temperature values ​​and abnormal heating rates in temperature monitoring information are statistically analyzed and calculated. The quantified spark and temperature characteristics are then fused to form a comprehensive risk feature vector. Historical data is used to train machine learning or deep learning models (such as support vector machines (SVM) or neural networks). The comprehensive risk feature vector is input into the model, and the corresponding fire risk level or probability is output as fire risk prediction information for each wireless charging pile and electric vehicle.

[0033] More specifically, relying solely on spark monitoring information or temperature monitoring information for fire risk prediction has certain limitations. For example, occasional small sparks may not necessarily cause a fire, and a brief temperature rise may be a normal operational fluctuation. Combining the two for comprehensive analysis can provide a more complete assessment of the safety status of charging equipment and improve the accuracy and reliability of fire risk prediction. In actual fire events, sparks and temperature rises are often interconnected and mutually influential. Sparks may cause local temperature increases, while high temperatures increase the likelihood of sparks, ultimately leading to a fire. Therefore, combining the two for prediction is more in line with the actual mechanism of fire occurrence.

[0034] More specifically, fire risk prediction information for each wireless charging station and electric vehicle being charged is compiled, organized, and stored in a specific format (such as tables or database records) to form fire safety supervision information covering all equipment within the entire charging area. Equipment identifiers and timestamps can be added to each information entry to facilitate subsequent querying, analysis, and management. Since each wireless charging station and electric vehicle in the charging area could potentially be a fire hazard, integrating the fire risk prediction information of each device allows for a comprehensive and systematic assessment of the overall fire safety status of the charging area. This provides a basis for developing targeted fire safety management strategies. Presenting the fire risk information of all equipment centrally facilitates unified management and decision-making by charging area managers, fire departments, and other relevant parties. For example, based on the fire safety supervision information, key monitoring equipment can be identified, and regular inspection and maintenance plans can be arranged, improving the efficiency and effectiveness of fire safety management.

[0035] In one embodiment, the step of identifying the internal structural temperature of each wireless charging pile and electric vehicle in the charging area based on infrared and ultraviolet monitoring data from the safety monitoring data to obtain temperature monitoring information includes: S221: Determine the model information of the wireless charging pile and the electric vehicle based on the visible light monitoring data, retrieve the structural design data of the wireless charging pile and the electric vehicle from the database, and perform digital simulation of the internal structure of the wireless charging pile and the electric vehicle to obtain the internal structure model. S222: Based on the infrared and ultraviolet monitoring data in the safety monitoring data, the infrared and ultraviolet intensity of the internal structure model is analyzed, and the temperature data of the internal structure model is labeled according to the analysis results to obtain the temperature monitoring information of each wireless charging pile and each charging electric vehicle.

[0036] Specifically, image recognition technology is used to analyze the appearance features of wireless charging piles and electric vehicles in visible light monitoring data. For example, the brand logos and model labels of charging piles and electric vehicles can be identified. Deep learning-based target detection algorithms, such as the YOLO (You Only Look Once) series of algorithms, can be used to locate and classify the charging piles and electric vehicles in the images and determine their specific models.

[0037] More specifically, based on the identified model information, the corresponding structural design data of the wireless charging pile and the charging electric vehicle are searched in a pre-established database. The database should contain detailed internal structural information, such as the location, material, and dimensions of each component. Using computer-aided design (CAD) software or professional modeling tools, the internal structure of the wireless charging pile and the charging electric vehicle is digitally simulated based on the retrieved structural design data. During the simulation, the shape, position, and connection relationship of each component must be accurately restored to construct an internal structural model that matches the actual situation.

[0038] More specifically, different models of wireless charging piles and electric vehicles have different internal structures and thermal characteristics. By identifying the model and obtaining the corresponding structural design data, personalized temperature analysis can be performed for specific models, improving the accuracy of temperature identification. Accurate structural design data is the foundation for building an internal structural model. Only by conducting digital simulation based on real structural information can a model that matches the actual situation be obtained, providing a reliable platform for subsequent temperature analysis. The internal structural model can intuitively display the internal structure of wireless charging piles and electric vehicles, helping to gain a deeper understanding of the layout and interrelationships of various components, and providing necessary support for accurately identifying the internal structural temperature.

[0039] More specifically, infrared monitoring data is processed to analyze the intensity distribution of the infrared band. The intensity of infrared radiation is closely related to the temperature of an object. A specific algorithm is used to convert the intensity of the infrared band into a temperature value. Ultraviolet monitoring data is also analyzed. Although the direct relationship between ultraviolet radiation and temperature is not as obvious as that between infrared radiation and temperature, in some cases, abnormal ultraviolet radiation may be related to high temperature or electrical discharge. Relevant physical models and empirical data can be combined to analyze the temperature-related information in the ultraviolet monitoring data.

[0040] More specifically, the parsed temperature data is labeled onto the corresponding locations in the internal structural model. Color coding can be used, with different colors representing different temperature ranges, to intuitively display the temperature distribution of the internal structure. For each labeled temperature data, its corresponding location, time, and other information are recorded to form complete temperature monitoring information.

[0041] More specifically, infrared and ultraviolet (UV) monitoring data have different characteristics and advantages. Infrared light directly reflects the thermal radiation of an object, providing relatively accurate temperature information; UV light, on the other hand, can provide additional information related to high temperatures or electrical faults. Combining the two allows for a more comprehensive and accurate identification of the internal structure's temperature. By annotating the temperature data onto the internal structural model, the temperature distribution inside wireless charging piles and charging electric vehicles can be visually displayed. This helps managers quickly identify areas with abnormal temperatures, take timely measures, and prevent safety accidents such as fires. Complete temperature monitoring information records the temperature data and time information for each location, facilitating subsequent data analysis and trend prediction. By analyzing historical temperature data, the operating status and thermal characteristics of the equipment can be understood, providing a basis for equipment maintenance and management.

[0042] In one embodiment, the step of combining the spark monitoring information and the temperature monitoring information to predict the fire risk of wireless charging piles and electric vehicles, and obtaining the corresponding fire risk prediction information for each wireless charging pile and electric vehicle, includes: S231: Analyze the spark monitoring information for spark severity and spark frequency to obtain the first risk characteristic; S232: Analyze the temperature monitoring information for abnormal high temperature and abnormal temperature rise to obtain the second risk characteristic; S233: Combining the first risk characteristics and the second risk characteristics, fire risk prediction is performed on wireless charging piles and charging electric vehicles to obtain fire risk prediction information.

[0043] Specifically, image information of sparks is extracted from visible light monitoring data, and the size, brightness, color, and other characteristics of the sparks are analyzed. Generally speaking, larger, brighter sparks with a bluish-white color have higher energy and are relatively more serious; while smaller, darker sparks with an orange-red color are relatively less serious. If there are relevant sensors that can measure the energy parameters of the sparks, the severity of the sparks can be directly assessed based on the energy level. The greater the energy, the more serious the spark, and the higher the possibility of causing a fire.

[0044] More specifically, a statistical time period is set, such as one hour or one day, and the number of times sparks occur within that time period is counted. The more times sparks occur, the worse the electrical stability of the equipment is and the higher the risk of fire. In addition to counting the absolute frequency, the trend of spark frequency over time can also be analyzed. If the frequency is on the rise, it indicates that the equipment malfunction is worsening and requires more attention.

[0045] More specifically, the severity of sparks directly reflects the severity of electrical system faults. Severe sparks indicate a significant short circuit or leakage problem, which can easily cause a fire. The frequency of sparks reflects the stability of electrical faults. Frequent sparks indicate that the electrical system of the equipment is in an unstable state and may cause a fire at any time, while occasional sparks may be momentary interference with relatively low risk.

[0046] More specifically, based on the normal operating temperature range of wireless charging piles and electric vehicles, an abnormal high temperature threshold is set. For example, for some charging piles, the normal operating temperature is between 20 and 50°C. When the temperature exceeds 70°C, it is judged as an abnormal high temperature. The temperature of each part of the equipment is monitored in real time, and the monitored temperature is compared with the set threshold. Once the threshold is exceeded, the location and duration of the abnormal high temperature are recorded.

[0047] More specifically, within a certain time interval, the ratio of the temperature rise of the equipment to the time is calculated to obtain the heating rate. For example, if the temperature rises by 10°C in 10 minutes, the heating rate is 1°C / minute. A normal heating rate threshold is set, and when the actual heating rate exceeds this threshold, it is judged as an abnormal heating.

[0048] More specifically, abnormally high temperatures are one of the important precursors to fires. When the temperature of equipment rises abnormally, it may be due to internal short circuits, overloads, or other reasons, which can easily lead to a fire. By monitoring abnormally high temperatures, potential fire hazards can be detected in advance. Abnormal temperature rise reflects the rapid changes in equipment temperature. Even if the current temperature has not yet reached the abnormally high temperature threshold, if the rate of temperature rise is too fast, it may mean that there is a serious malfunction in the equipment, which is about to cause a fire.

[0049] More specifically, the first risk characteristic (severity and frequency of sparks) and the second risk characteristic (abnormal high temperature and abnormal temperature rise) are fused to form a comprehensive risk characteristic vector. For example, these characteristic values ​​can be arranged in a certain order to form a multi-dimensional vector.

[0050] More specifically, pre-trained machine learning models, such as decision trees, support vector machines, and neural networks, are used to take the comprehensive risk feature vector as input. The model outputs the corresponding fire risk level or probability. Based on expert experience and rules, the comprehensive risk features are judged. For example, if the severity of sparks is high and the frequency of occurrence is high, and it is accompanied by abnormal high temperature and abnormal temperature rise, it is judged as a high fire risk.

[0051] More specifically, relying solely on spark monitoring information or temperature monitoring information for fire risk prediction has certain limitations. For example, occasional small sparks may not necessarily cause a fire, and a brief temperature rise may be a normal operating fluctuation. Combining the two for comprehensive analysis can provide a more complete assessment of the safety status of charging equipment and improve the accuracy and reliability of fire risk prediction.

[0052] More specifically, in actual fires, sparks and temperature increases are often interconnected and mutually influential. Sparks may cause local temperature increases, while high temperatures may increase the likelihood of sparks, ultimately leading to a fire. Therefore, combining both factors for prediction is more consistent with the actual mechanism of fire occurrence.

[0053] In one embodiment, a sensor array pre-installed within the wireless charging station at the charging location is used to collect data on the actual operation and performance of the wireless charging station, generating equipment operation monitoring data to supplement the fire risk prediction of the wireless charging station.

[0054] Specifically, current sensors are typically installed in the main circuit of a charging station to monitor the magnitude of the charging current in real time. They measure the current value by sensing changes in the magnetic field around the conductor, and have high accuracy and minimal impact on the circuit.

[0055] More specifically, the voltage sensor is connected to the input and output terminals of the charging pile, and obtains voltage data by means of direct measurement or isolated measurement to ensure that the charging pile operates within a suitable voltage range and avoids malfunctions caused by abnormal voltage.

[0056] More specifically, power sensors are generally located near the power module. They can indirectly obtain power by measuring current and voltage and calculating their product. There are also dedicated power measurement chips that can directly measure power. Temperature sensors are distributed on the surfaces of key heat-generating components of the charging pile, such as power modules and transformers. Different types of temperature sensors, such as thermocouples and thermistors, can be used to sense temperature changes in various parts in real time.

[0057] More specifically, the humidity sensor is installed in a well-ventilated location inside the charging station to accurately measure the relative humidity of the environment inside the charging station; the smoke sensor is generally installed on the top of the charging station or in a place where smoke is likely to accumulate, and will send a signal once the smoke concentration reaches a certain threshold.

[0058] More specifically, the sensor array collects data at a certain sampling frequency. For example, current, voltage, and power sensors can collect data once per second to capture instantaneous current fluctuations and power changes; temperature sensors can collect data every 5-10 minutes because temperature changes are relatively slow; humidity and smoke sensors are set to appropriate sampling frequencies according to actual needs. The data acquisition process is usually completed by the data acquisition module inside the charging pile. This module is connected to the sensor array, converts the analog signals output by the sensors into digital signals, and performs preliminary processing and storage.

[0059] More specifically, the collected data is transmitted to the backend server via wired or wireless communication. Wired communication methods can include Ethernet, RS-485, etc., while wireless communication methods can include Wi-Fi, LoRa, 4G / 5G, etc., ensuring the stability and timeliness of data transmission. After receiving the data, the backend server stores it in a database. The database can be a relational database (such as MySQL) or a non-relational database (such as MongoDB) to enable effective data management and retrieval.

[0060] More specifically, the collected equipment operation monitoring data is integrated with previous spark monitoring and temperature monitoring information. For example, current, voltage, and power data at the same time point are correlated with corresponding temperature and spark data to form a more comprehensive dataset. Data analysis techniques, such as machine learning algorithms and data mining techniques, are used to analyze the integrated data. By building models, hidden patterns and features in the data are discovered to more accurately predict the fire risk of wireless charging piles.

[0061] More specifically, current, voltage, and power are key parameters that reflect whether the charging process of a charging pile is normal. Excessive charging current can cause the circuit to overheat, abnormal voltage can cause damage to electronic components, and unstable power can mean that there is a fault in the power module of the charging pile. By monitoring these parameters in real time, abnormal situations in the charging process can be detected in time, providing an important basis for fire risk prediction.

[0062] More specifically, although the overall temperature of the charging pile was previously monitored through multispectral data acquisition, the temperature sensors in the sensor group can more accurately measure the temperature of key heat-generating components. The temperature changes of different components have different impacts on fire risk. For example, excessively high power module temperature can cause short circuits and fires. By monitoring the temperature of key components individually, fire risks can be assessed more specifically.

[0063] More specifically, environmental factors such as humidity and smoke have a significant impact on the safe operation of charging piles. Excessive humidity leads to a decline in electrical insulation performance, causing leakage and short circuits; the appearance of smoke is an early sign of combustion inside the equipment. By monitoring environmental factors, potential safety hazards can be detected in advance, supplementing the data needed for fire risk prediction.

[0064] More specifically, relying solely on spark and temperature information for fire risk prediction is insufficient. The actual operation and performance of charging piles involve multiple aspects. Monitoring data collected by sensor arrays can provide more information dimensions, complementing and verifying spark and temperature information, thereby constructing a more complete fire risk prediction model and improving the accuracy and reliability of predictions.

[0065] In one embodiment, the step of parsing the fire safety supervision information to execute corresponding preset management measures includes: S31: Analyze the fire safety supervision information to determine the risk level of the charging site; S32: If the risk level is that a fire has occurred, then the preset fire response device will be activated to extinguish the fire. S33: If the risk level is that a fire is about to occur, shut down the corresponding wireless charging station and send a warning message; S34: If the risk level is a prediction that a fire will occur in the future, reduce the working efficiency of the corresponding wireless charging pile and send a warning message.

[0066] In one embodiment, the fire response device includes a drone equipped with a fire-fighting module and a robot equipped with a fire-fighting module.

[0067] In one embodiment, while issuing the warning information, key information of the fire safety supervision information is identified to generate reference information for maintenance personnel to refer to.

[0068] Specifically, fire safety supervision information is organized, and key features related to fire risk are extracted, such as the severity and frequency of sparks in spark monitoring information, abnormal high temperature values ​​and abnormal heating rates in temperature monitoring information, and abnormal current and voltage in equipment operation monitoring data. A pre-established risk assessment model is then used, inputting the extracted features into the model. This model can be built based on machine learning algorithms (such as decision trees and neural networks) or on rule-based models developed based on expert experience. The model calculates and judges based on the input features, outputting the risk level of the charging site. Generally, it can be divided into several levels, such as fire already occurred, fire imminent, predicted future fire, and safe. A comprehensive judgment is made by combining multiple features and the model's output. Simultaneously, the judgment results are verified, for example, by comparing them with historical data and actual conditions, to ensure the accuracy of the risk level assessment.

[0069] More specifically, different risk levels require different management strategies. Accurately judging the risk level is the basis for formulating reasonable response measures. Only by clarifying the current risk status of the charging site can targeted actions be taken to avoid overreaction or underreaction. By accurately judging the risk level, fire protection resources and management forces can be allocated reasonably. For high-risk situations, more resources can be invested in a timely manner, while for low-risk situations, relatively mild measures can be taken to improve resource utilization efficiency.

[0070] More specifically, when the risk level is determined to be a fire, the control system immediately sends an activation command to the preset fire response devices (drones and robots equipped with fire-fighting modules). The drones use their onboard positioning systems (such as GPS) and image recognition technology to quickly locate the fire and plan a flight path to the fire scene. The robots, on the other hand, use their internal navigation systems (such as lidar and visual navigation) to autonomously navigate to the fire location within the charging area.

[0071] More specifically, upon arriving at the fire scene, drones and robots select appropriate firefighting methods based on the type and scale of the fire. For example, dry powder fire extinguishers can be used to extinguish electrical fires, while foam fire extinguishers can be used for liquid fires. They spray extinguishing agents onto the fire source through nozzles or spray guns on their firefighting modules. During the firefighting process, drones and robots monitor the fire situation in real time using their own sensors (such as temperature sensors and smoke sensors) and feed the data back to the control system. The control system adjusts the firefighting strategy and the amount of extinguishing agent sprayed based on the feedback information to ensure the firefighting effect.

[0072] More specifically, time is a critical factor after a fire breaks out. Drones and robots, with their rapid response and flexible movement, can reach the fire scene quickly and begin firefighting operations, effectively controlling the spread of fire and reducing losses. In charging stations, fires pose risks such as electrical equipment leakage and explosions. Using drones and robots for firefighting avoids direct contact with hazardous environments, ensuring personnel safety. Drones and robots can precisely locate and operate firefighting equipment based on the fire situation, making more effective use of extinguishing agents and improving firefighting efficiency.

[0073] More specifically, when the risk level is determined to be an impending fire, the control system sends a shutdown command to the corresponding wireless charging station via the communication interface. Upon receiving the command, the charging station immediately cuts off the charging circuit and stops charging. Simultaneously, the control system sends warning information to relevant personnel (such as charging site managers and maintenance personnel) via SMS, email, and audible and visual alarms. The warning information should include key information such as the charging station number, location, and possible causes of the impending fire.

[0074] More specifically, in the event of an impending fire, promptly shutting down the charging station can prevent the fire from spreading further and reduce the likelihood of a fire. Cutting off the charging circuit can prevent electrical sparks and overheating caused by electrical faults, thus reducing the risk of fire. Sending early warning information allows relevant personnel to be informed of the situation in a timely manner and take appropriate measures. Managers can organize personnel to conduct on-site inspections and handle the situation, and maintenance personnel can arrive at the scene as soon as possible to troubleshoot and repair the fault, preventing the fire from occurring.

[0075] More specifically, when the risk level is determined to be a predicted future fire, the control system reduces the charging pile's efficiency by adjusting its output power and lowering the charging current. For example, the output power of the charging pile may be reduced from 100% to 50%. Similarly, warning information is sent to relevant personnel via SMS, email, and audible and visual alarms. In addition to basic information such as the charging pile number and location, the warning information should also remind relevant personnel to pay attention to the safety status of the charging site and to conduct timely inspections and maintenance.

[0076] More specifically, when a fire is predicted to occur, reducing the operating efficiency of charging stations can decrease equipment heat generation, lowering the likelihood of a fire. Reducing charging current can lessen the burden on electrical equipment, preventing fires caused by overload or other reasons. Sending early warning information allows relevant personnel to prepare in advance, inspect and maintain charging sites, and eliminate potential safety hazards. Timely action can prevent fires and ensure the safe operation of charging facilities. Simultaneously with issuing early warning information, key information from the fire safety supervision data is identified to generate reference information for maintenance personnel. More specifically, a thorough analysis of fire safety supervision information is conducted to extract key information related to the cause of malfunctions and maintenance, such as the location of sparks, areas of abnormally high temperatures, and abnormal values ​​of equipment operating parameters. This extracted key information is then organized and formatted into reference information (e.g., reports, tables). The reference information should be concise, clear, and highlight key points to facilitate quick understanding of the problem by maintenance personnel. This generated reference information, along with early warning information, is sent to maintenance personnel. Access can be facilitated through SMS messages with links or email attachments to ensure timely receipt of the reference information by maintenance personnel.

[0077] More specifically, after receiving an early warning message, maintenance personnel need to understand the specific fault situation in order to carry out effective repairs. Reference information can provide maintenance personnel with detailed fault clues, helping them to quickly locate the problem, reduce on-site troubleshooting time, and improve repair efficiency. Through reference information, maintenance personnel can understand the possible causes and severity of the fault in advance, prepare the corresponding tools and materials, and carry out targeted repairs, thereby solving the problem more accurately and ensuring the safe operation of the charging site.

[0078] In one embodiment, such as Figure 2 As shown, a fire safety management system for charging facilities is provided, used to implement the fire safety management method for charging facilities as described in any one of the first aspects, comprising: The safety monitoring module is used to collect multispectral data from charging sites to obtain safety monitoring data for the charging sites. The risk prediction module is used to predict the fire risk of charging sites based on the safety monitoring data and generate fire safety supervision information for charging sites. The response processing module is used to parse the fire safety supervision information in order to execute corresponding preset management countermeasures.

[0079] In this embodiment, the specific implementation of each module in the above system embodiment is described in the above method embodiment, and will not be repeated here.

[0080] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0081] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0082] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0083] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for fire safety management of charging locations, characterized in that, include: Multispectral data was collected from charging sites to obtain safety monitoring data for the charging sites; Based on the aforementioned safety monitoring data, fire risks at charging locations are predicted, and fire safety supervision information for charging locations is generated. The fire safety supervision information is analyzed to implement corresponding preset management measures.

2. The fire safety management method for charging sites as described in claim 1, characterized in that, The steps for acquiring multispectral data from charging locations to obtain safety monitoring data include: By using a multispectral AI fire detector pre-installed in the charging location, visual data from multiple spectral channels is collected from the charging location, resulting in safety monitoring data composed of safety monitoring data from each spectral channel. Among them, the spectral channels include visible light channels, infrared light channels, and ultraviolet light channels; The security monitoring data includes visible light monitoring data, infrared light monitoring data, and ultraviolet light monitoring data.

3. The fire safety management method for charging sites as described in claim 2, characterized in that, The steps for predicting fire risks at charging locations based on the aforementioned safety monitoring data and generating fire safety supervision information for charging locations include: Based on the visible light monitoring data in the aforementioned safety monitoring data, the electrical sparks and flames of each wireless charging pile and electric vehicle in the charging area are identified to obtain spark monitoring information. Based on the infrared and ultraviolet monitoring data in the aforementioned safety monitoring data, the internal structural temperature of each wireless charging pile and electric vehicle in the charging area is identified to obtain temperature monitoring information. By combining the spark monitoring information and the temperature monitoring information, fire risk prediction is made for wireless charging piles and electric vehicles, resulting in fire risk prediction information for each wireless charging pile and electric vehicle. The fire risk prediction information of each wireless charging station and electric vehicle being charged constitutes the fire safety supervision information for charging sites.

4. The fire safety management method for charging sites as described in claim 3, characterized in that, The steps for identifying the internal structural temperature of each wireless charging pile and electric vehicle in the charging area based on the infrared and ultraviolet monitoring data in the aforementioned safety monitoring data to obtain temperature monitoring information include: Based on the visible light monitoring data, the model information of the wireless charging pile and the electric vehicle is determined, and the structural design data of the wireless charging pile and the electric vehicle is retrieved from the database to perform digital simulation of the internal structure of the wireless charging pile and the electric vehicle, thereby obtaining the internal structure model. Based on the infrared and ultraviolet monitoring data in the safety monitoring data, the infrared and ultraviolet intensity of the internal structure model is analyzed, and the temperature data of the internal structure model is labeled according to the analysis results, so as to obtain the temperature monitoring information of each wireless charging pile and each charging electric vehicle.

5. The fire safety management method for charging sites as described in claim 3, characterized in that, The steps for predicting the fire risk of wireless charging piles and electric vehicles by combining the spark monitoring information and the temperature monitoring information, and obtaining the corresponding fire risk prediction information for each wireless charging pile and electric vehicle, include: The spark monitoring information is analyzed for spark severity and spark frequency to obtain the first risk characteristic; The temperature monitoring information is analyzed for abnormal high temperatures and abnormal temperature rises to obtain a second risk characteristic; By combining the first risk feature and the second risk feature, fire risk prediction information is obtained for wireless charging piles and electric vehicles.

6. The fire safety management method for charging sites as described in claim 5, characterized in that, By using sensor arrays pre-installed in wireless charging stations, data on the actual operation and performance of the wireless charging stations are collected, generating equipment operation monitoring data to supplement the data for fire risk prediction of wireless charging stations.

7. The fire safety management method for charging locations as described in claim 1, characterized in that, The steps for parsing the fire safety supervision information to implement corresponding preset management measures include: The fire safety supervision information is analyzed to determine the risk level of the charging location; If the risk level is that a fire has occurred, the preset fire response device will be activated to extinguish the fire. If the risk level is that a fire is about to occur, the corresponding wireless charging station will be shut down and a warning message will be sent. If the risk level is a prediction that a fire will occur in the future, the working efficiency of the corresponding wireless charging station will be reduced and a warning message will be sent.

8. The fire safety management method for charging sites as described in claim 7, characterized in that, Fire response equipment includes drones and robots equipped with fire-fighting modules.

9. The fire safety management method for charging locations as described in claim 7, characterized in that, While issuing the warning information, the key information of the fire safety supervision information is identified to generate reference information for maintenance personnel.

10. A fire safety management system for charging locations, characterized in that, A method for implementing a fire safety management system for a charging facility as described in any one of claims 1-9, comprising: The safety monitoring module is used to collect multispectral data from charging sites to obtain safety monitoring data for the charging sites. The risk prediction module is used to predict the fire risk of charging sites based on the safety monitoring data and generate fire safety supervision information for charging sites. The response processing module is used to parse the fire safety supervision information in order to execute corresponding preset management countermeasures.

Images