Parking lot electric vehicle fire control system using artificial intelligence and thermal imaging sensor

The electric vehicle fire control system uses AI and thermal imaging sensors to monitor and predict fires in parking lots, enabling real-time fire control and minimizing damage and casualties by tracking vehicle locations and automating responses to prevent and extinguish fires.

WO2024232727A1PCT designated stage expired Publication Date: 2024-11-14DAEYOUNG IOT INC

Patent Information

Application Number
PCT/KR2024/006397
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-05-10
Filing Date
2024-05-10
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

Electric vehicle fires in parking lots are difficult to predict and control, leading to extensive damage and casualties due to the inability to effectively monitor and respond to potential fires, especially in underground locations where fires can spread rapidly and are challenging to extinguish with conventional methods.

Method used

A parking lot electric vehicle fire control system utilizing artificial intelligence and thermal imaging sensors that continuously monitors parked electric vehicles, predicts potential fires using machine learning algorithms, and performs fire control by tracking vehicle locations, broadcasting evacuations, and automatically reporting and extinguishing fires.

Benefits of technology

The system effectively minimizes human casualties and prevents large-scale fires by predicting and controlling electric vehicle fires, reducing damage to surrounding vehicles and buildings through real-time monitoring and automated responses.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure KR2024006397_14112024_PF_FP_ABST
    Figure KR2024006397_14112024_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to a parking lot electric vehicle fire control system using artificial intelligence (AI) and a thermal imaging sensor, wherein the system continuously monitors parked electric vehicles and analyzes same using a machine learning-based artificial intelligence algorithm, and minimizes casualties and prevents large-scale fires in advance through fire control when it is predicted that there is a possibility of a fire occurring. The parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor comprises: a fire information collection unit for collecting electric vehicle information as fire information; an environment information collection unit for collecting environment information about a parking lot; an electric vehicle fire sensing unit for predicting or sensing a fire in an electric vehicle by analyzing the collected fire information and environment information using a machine learning-based artificial intelligence algorithm, generating fire analysis information when a fire of the electric vehicle is predicted or sensed, and transmitting the fire analysis information to parking control and fire control; a fire control unit for performing fire control when the fire analysis information is generated; and a parking control unit for recognizing the electric vehicle and tracking the parked location, and performing parking control when the fire analysis information is generated.
Need to check novelty before this filing date? Find Prior Art

Description

Parking lot electric vehicle fire control system using artificial intelligence and thermal imaging sensors

[0001] The present invention relates to a fire control system for electric vehicles in a parking lot using artificial intelligence and a thermal imaging sensor, and more particularly, to a fire control system for electric vehicles in a parking lot using artificial intelligence and a thermal imaging sensor, which recognizes and tracks electric vehicles entering a parking lot, continuously monitors parked electric vehicles, analyzes them using a machine learning-based artificial intelligence (AI) algorithm, and, if a possibility of fire is predicted, performs fire control to minimize casualties and prevent large-scale fires in advance.

[0002] In general, an electric vehicle (hereinafter referred to as an "electric vehicle") is a vehicle that generates driving force by supplying electric energy from a high-voltage battery to an electric motor. Since it does not use fossil fuels, it is an environmentally friendly vehicle with almost no exhaust gas emissions or noise.

[0003] The batteries mainly used in electric vehicles are lithium-ion batteries. If the separator that separates the positive and negative electrodes, which are the main components of the battery, is damaged and the two poles come into contact, excessive current can flow and heat can be generated, which can lead to a fire.

[0004] For example, fires and explosions caused by electric vehicle charging in parking lots are on the rise, with internal short circuits between electric vehicle batteries and their connections, as well as problems with chargers. Electric vehicle fires can cause extensive damage to surrounding vehicles, facilities, and buildings, necessitating significant efforts and urgent improvements to protect precious lives and property.

[0005] In particular, electric vehicle fires are nearly impossible to extinguish with fire extinguishers because the batteries explode, generating ultra-high temperature heat energy, flames (smoke), and toxic gases, which increases the damage caused by the spread of the fire.

[0006] Additionally, if an electric vehicle fire breaks out in an underground parking lot, there is a problem that not only does it damage surrounding vehicles, but the fire can also spread to the building, resulting in a large-scale fire.

[0007] Therefore, electric vehicle fire control technology is required that continuously monitors the status of electric vehicles in parking lots in real time to predict the possibility of fire, responds in real time to the predicted possibility of fire to prevent fires in advance, and reduces casualties and minimizes damage to facilities through optimal fire control in the event of a fire in an electric vehicle.

[0008] Accordingly, the present invention has been proposed to solve the problem of not being able to predict electric vehicle fires in advance in underground and above-ground parking lots and thus prevent electric vehicle fires in advance, and the problem of not being able to provide an optimal fire control system that can minimize casualties and building damage in the event of an electric vehicle fire. The purpose of the present invention is to provide a parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor that recognizes and tracks electric vehicles entering a parking lot, continuously monitors parked electric vehicles, analyzes them using a machine learning-based artificial intelligence (AI) algorithm, and, if a possibility of fire is predicted, performs fire control to minimize casualties and prevent large-scale fires in advance.

[0009] In order to achieve the above-mentioned purpose, the "parking lot electric vehicle fire control system using artificial intelligence and thermal imaging sensor" according to the present invention,

[0010] A fire information collection unit that collects electric vehicle information as fire information for fire analysis and prediction of parked electric vehicles;

[0011] An environmental information collection unit that collects environmental information of the parking lot where electric vehicles are parked;

[0012] An electric vehicle fire detection unit that analyzes fire information and environmental information collected from the above fire information collection unit and environmental information collection unit using a machine learning-based artificial intelligence algorithm to predict or detect a fire in an electric vehicle, and generates fire analysis information when a fire in an electric vehicle is predicted or detected and transmits it to the parking control and fire control side;

[0013] A fire control unit that performs fire control when fire analysis information is generated from the above electric vehicle fire detection unit; and

[0014] It includes a parking control unit that recognizes an electric vehicle entering a parking lot and tracks the parking location, and performs parking control in conjunction with the fire control unit when fire analysis information is generated from the electric vehicle fire detection unit.

[0015] The above parking control unit transmits the tracked electric vehicle parking location information to the fire information collection unit, and the fire information collection unit collects information on the electric vehicle as an image using a thermal imaging sensor based on the electric vehicle parking location information, and transmits the collected image information on the electric vehicle as fire information to the electric vehicle fire detection unit.

[0016] Preferably, the environmental information collection unit is characterized in that it detects at least one of temperature and humidity around an electric vehicle parked in a parking lot as environmental information and transmits the detected information to the electric vehicle fire detection unit.

[0017] Preferably, the parking control unit is characterized by including an electric vehicle recognition unit that recognizes an electric vehicle by recognizing a vehicle entering a parking lot; an electric vehicle location tracking unit that tracks the electric vehicle recognized by the electric vehicle recognition unit in real time to confirm the parking location, and transmits the confirmed parking location information to an electric vehicle fire detection unit.

[0018] More preferably, the parking control unit is characterized by including an access control unit that controls vehicle entry and exit of the parking lot based on fire analysis information generated from the electric vehicle fire detection unit; and a vehicle contact unit that transmits a fire prediction status or a fire occurrence status to a vehicle terminal based on vehicle contact information included in the fire analysis information generated from the electric vehicle fire detection unit.

[0019] More preferably, the vehicle owner contact unit is characterized in that it transmits fire prediction status information or fire occurrence status information in text format based on the electric vehicle owner contact information included in the fire analysis information or the vehicle owner contact information of a general vehicle parked around the electric vehicle.

[0020] Preferably, the fire control unit is characterized by including a fire evacuation broadcast unit that performs a fire evacuation broadcast when receiving fire analysis information from the electric vehicle fire detection unit; a fire reporting unit that analyzes the fire analysis information when the fire analysis information is generated and transmits the fire information, including location information, to a relevant agency to report the fire; and a fire extinguishing system operating unit that operates an installed fire extinguishing system based on the fire location information included in the fire analysis information to extinguish the fire.

[0021] Preferably, the electric vehicle fire detection unit is characterized by including: an information receiving unit that receives fire information collected by the fire information collecting unit and environmental information collected by the environmental information collecting unit; an environmental temperature detection unit that detects the environmental temperature around the electric vehicle parked from the environmental information received by the information receiving unit; and an electric vehicle information extraction unit that extracts only electric vehicle information from the image fire information received by the information receiving unit.

[0022] More preferably, the electric vehicle information extraction unit is characterized in that it extracts only the area of ​​the electric vehicle as electric vehicle information through object classification from the received image fire information.

[0023] Preferably, the electric vehicle fire detection unit includes an electric vehicle temperature detection unit that detects the electric vehicle temperature by calculating the number of pixels by color in the electric vehicle area information extracted from the electric vehicle information extraction unit; an abnormal temperature detection unit that corrects the electric vehicle temperature detected by the electric vehicle temperature detection unit based on the environmental temperature detected by the environmental temperature detection unit, compares the corrected electric vehicle temperature information with preset reference electric vehicle temperature information, extracts a difference, determines that the extracted difference is out of the allowable temperature range, and provides the difference information as information for fire prediction.

[0024] Preferably, the electric vehicle fire detection unit is characterized by including a fire prediction unit that analyzes information for fire prediction provided by the abnormal temperature detection unit using a machine learning-based artificial intelligence algorithm for a certain period of time to determine fire prediction or fire occurrence in the electric vehicle, and outputs fire prediction or fire occurrence determination information.

[0025] More preferably, the machine learning-based artificial intelligence algorithm is characterized by being a fire inference model that predicts fire or determines fire occurrence based on abnormal temperature information and abnormal temperature change time information.

[0026] More preferably, the fire inference model is characterized by predicting fire and determining fire occurrence through learning that matches abnormal temperature information and abnormal temperature change time information in advance.

[0027] Preferably, the electric vehicle fire detection unit is characterized by including a fire analysis information generation unit that generates fire analysis information based on fire prediction information or fire occurrence information determined by the fire prediction unit and transmits the generated fire analysis information to the parking control unit and the fire control unit.

[0028]

[0029] According to the present invention, electric vehicles entering a parking lot are recognized and tracked, and parked electric vehicles are continuously monitored and analyzed using a machine learning-based artificial intelligence (AI) algorithm. If a fire is predicted to be likely to occur, fire control is performed to minimize casualties and prevent large-scale fires in advance.

[0030] In addition, according to the present invention, there is an advantage in that it can recognize and track electric vehicles entering a parking lot, continuously monitor parked electric vehicles, analyze them using a machine learning-based artificial intelligence (AI) algorithm, and automatically activate a fire evacuation broadcast and fire extinguishing system when a fire is detected, thereby minimizing casualties and building fires caused by fire.

[0031] In addition, according to the present invention, when a fire is detected in an electric vehicle, it automatically reports the fire to a relevant organization (e.g., fire station, police station, government office, etc.), thereby having the effect of helping to quickly extinguish the fire.

[0032] Figure 1 is a schematic diagram of a parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor according to the present invention.

[0033] Figure 2 is a configuration diagram of an embodiment of the parking control unit of Figure 1,

[0034] Figure 3 is a configuration diagram of an embodiment of the electric vehicle fire detection unit of Figure 1.

[0035] Fig. 4 is a configuration diagram of an embodiment of the fire control unit of Fig. 1.

[0036] Hereinafter, a parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor according to a preferred embodiment of the present invention will be described in detail with reference to the attached drawings.

[0037] The terms and words used in the present invention described below should not be interpreted as limited to their usual or dictionary meanings, but should be interpreted as meanings and concepts that conform to the technical idea of ​​the present invention based on the principle that the inventor can appropriately define the concept of the term in order to explain his or her own invention in the best way.

[0038] Therefore, the embodiments described in this specification and the configurations illustrated in the drawings are only preferred embodiments of the present invention and do not represent all of the technical ideas of the present invention. Therefore, it should be understood that there may be various equivalents and modified examples that can replace them at the time of filing this application.

[0039] FIG. 1 is a block diagram of a parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor according to a preferred embodiment of the present invention, which may include a parking control unit (100), a fire information collection unit (200), an environmental information collection unit (300), an electric vehicle fire detection unit (400), and a fire control unit (500). Since the present invention has the same configuration and performs the same function in above-ground or underground parking lots, even if the parking lot is not specified and is simply referred to as a parking lot, it can be recognized as an underground or above-ground parking lot.

[0040] The fire information collection unit (200) collects electric vehicle information as fire information for analyzing and predicting fires in electric vehicles parked in a parking lot.

[0041] This fire information collection unit (200) can collect information about the electric vehicle as an image using a thermal imaging sensor based on the electric vehicle parking location information, and transmit the collected image information about the electric vehicle as fire information to the electric vehicle fire detection unit (400).

[0042] The environmental information collection unit (300) collects environmental information of the parking lot where the electric vehicle is parked.

[0043] This environmental information collection unit (300) can detect at least one of the temperature and humidity around an electric vehicle parked in a parking lot as environmental information and transmit it to the electric vehicle fire detection unit (400).

[0044] The electric vehicle fire detection unit (400) analyzes fire information and environmental information collected from the fire information collection unit (200) and environmental information collection unit (300) using a machine learning-based artificial intelligence algorithm to predict or detect a fire in an electric vehicle, and when a fire in an electric vehicle is predicted or detected, it generates fire analysis information and transmits it to the parking control and fire control sides.

[0045] As shown in FIG. 3, the electric vehicle fire detection unit (400) may include an information receiving unit (401) that receives fire information collected by the fire information collecting unit (200) and environmental information collected by the environmental information collecting unit (300), an environmental temperature detecting unit (402) that detects the environmental temperature around the electric vehicle parked in the environmental information received by the information receiving unit (401), and an electric vehicle information extracting unit (403) that extracts only electric vehicle information from the image fire information received by the information receiving unit (401).

[0046] Here, the electric vehicle information extraction unit (403) can extract only the electric vehicle area as electric vehicle information through object classification from the received video fire information.

[0047] Preferably, the electric vehicle fire detection unit (400) may include an electric vehicle temperature detection unit (404) that detects the electric vehicle temperature by calculating the number of pixels by color in the electric vehicle area information extracted by the electric vehicle information extraction unit (403), an abnormal temperature detection unit (405) that corrects the electric vehicle temperature detected by the electric vehicle temperature detection unit (404) based on the environmental temperature detected by the environmental temperature detection unit (402), compares the corrected electric vehicle temperature information with preset reference electric vehicle temperature information, extracts a difference, determines that the temperature is abnormal when the extracted difference deviates from the allowable temperature range, and provides the difference information as information for fire prediction.

[0048] Preferably, the electric vehicle fire detection unit (400) may include a fire prediction unit (406) that analyzes information for fire prediction provided by the abnormal temperature detection unit (405) using a machine learning-based artificial intelligence algorithm for a certain period of time to determine fire prediction or fire occurrence in the electric vehicle and outputs fire prediction or fire occurrence determination information.

[0049] Here, the machine learning-based AI algorithm can be a fire inference model that predicts or determines fire occurrence based on abnormal temperature information and the time of change in that temperature. Furthermore, the fire inference model predicts fire occurrence and determines fire occurrence through pre-training to match abnormal temperature information with the time of change in that temperature.

[0050] Preferably, the electric vehicle fire detection unit (400) may include a fire analysis information generation unit (407) that generates fire analysis information based on fire prediction information or fire occurrence information determined by the fire prediction unit (406) and transmits the generated fire analysis information to the parking control unit (100) and the fire control unit (500).

[0051] The fire control unit (500) performs fire control when fire analysis information is generated from the electric vehicle fire detection unit (400).

[0052] As illustrated in FIG. 4, the fire control unit (500) may include a fire evacuation broadcast unit (501) that performs a fire evacuation broadcast when receiving fire analysis information from the electric vehicle fire detection unit (400), a fire reporting unit (502) that analyzes the fire analysis information when the fire analysis information is generated and transmits the fire information including location information to the relevant agency when a fire occurs to report the fire, and a fire extinguishing system operating unit (503) that operates the installed fire extinguishing system based on the fire location information included in the fire analysis information to extinguish the fire.

[0053] The parking control unit (100) recognizes electric vehicles entering the parking lot and tracks their parking locations, and when fire analysis information is generated from the electric vehicle fire detection unit (400), it performs parking control in conjunction with the fire control unit (500).

[0054] This parking control unit (100) can transmit the tracked electric vehicle parking location information to the fire information collection unit (200).

[0055] Preferably, the parking control unit (100) may include, as illustrated in FIG. 2, an electric vehicle recognition unit (101) that recognizes an electric vehicle by recognizing a vehicle entering a parking lot, an electric vehicle location tracking unit (102) that tracks the electric vehicle recognized by the electric vehicle recognition unit (101) in real time to confirm the parking location, and transmits the confirmed parking location information to an electric vehicle fire detection unit (400).

[0056] In addition, the parking control unit (100) may include an access control unit (103) that controls vehicle entry and exit to the parking lot based on fire analysis information generated from the electric vehicle fire detection unit (400), and a vehicle owner contact unit (104) that transmits a fire prediction status or fire occurrence status to the vehicle owner terminal based on vehicle owner contact information included in the fire analysis information generated from the electric vehicle fire detection unit (400).

[0057] Here, the vehicle owner contact information (104) can transmit fire prediction status information or fire occurrence status information as a text message based on the electric vehicle owner contact information included in the fire analysis information or the vehicle owner contact information of a general vehicle parked around the electric vehicle.

[0058] The operation of the parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor according to a preferred embodiment of the present invention configured as described above is described in detail as follows.

[0059] Here, it is assumed that vehicles parked in the parking lot have previously registered their vehicle information (vehicle number, owner contact information, etc.) with the electric vehicle fire detection unit (400). In other words, vehicle information can be provided when making a reservation for the parking lot.

[0060] When a user driving an electric vehicle (possibly the owner) enters a parking lot to park, the electric vehicle recognition unit (101) installed at the entrance of the parking lot recognizes the electric vehicle entering the parking lot. Here, the electric vehicle can be recognized by photographing the vehicle license plate with a camera and recognizing the vehicle license plate. For example, in the case of an electric vehicle, the license plate is plated in a specific color (e.g., blue), so the license plate area is extracted from the photographed license plate, and the license plate color is recognized through pixel-by-pixel color recognition to recognize the electric vehicle.

[0061] When a vehicle entering a parking lot is recognized as an electric vehicle, the parking control unit (100) provides information on the recognized electric vehicle to the electric vehicle fire detection unit (400) and at the same time provides information on the electric vehicle location tracking unit (102).

[0062] The above electric vehicle location tracking unit (102) tracks the recognized electric vehicle in real time to confirm the parking location, and provides the confirmed parking location information to the electric vehicle fire detection unit (400) and fire information collection unit (200).

[0063] For example, the electric vehicle location tracking unit (102) can track the location using an omnidirectional camera. The omnidirectional camera is installed throughout the parking lot (in all locations where the parking location can be recognized) and can record the entire movement path of a vehicle moving through the parking lot. The omnidirectional video may include recording time information. The omnidirectional camera is normally in an idle state, and when electric vehicle recognition information is transmitted from the electric vehicle recognition unit (101), it performs a recording operation and tracks the electric vehicle's location based on the captured omnidirectional video. The location information of the omnidirectional cameras installed in the parking lot is stored in advance, so that the location of the omnidirectional video captured by a specific omnidirectional camera can be easily identified within the parking lot. When an electric vehicle being tracked through the video tracking of the omnidirectional camera is parked in a specific parking space, parking information is recorded by the omnidirectional camera, and parking location information of the electric vehicle can be generated based on the installation location information of the omnidirectional camera. Preferably, the parking location of the electric vehicle can be accurately identified using pillars, the shape of the parking lot, the number on the floor, etc. surrounding the parking space where the vehicle is parked.

[0064] The above fire information collection unit (200) collects thermal image images by capturing information about the electric vehicle using a thermal imaging sensor (thermal imaging camera) based on the electric vehicle parking location information, and transmits the collected image information about the electric vehicle as fire information to the electric vehicle fire detection unit (400).

[0065] In addition, the environmental information collection unit (300) collects environmental information of the parking lot where the electric vehicle is parked and transmits it to the electric vehicle fire detection unit (400).

[0066] For example, the environmental information collection unit (300) detects at least one of the temperature and humidity around an electric vehicle parked in a parking lot as environmental information using environmental sensors such as a temperature sensor and a humidity sensor, and transmits the detected information to the electric vehicle fire detection unit (400). In the present invention, the use of temperature information using a temperature sensor as environmental temperature is described as an example.

[0067] The above electric vehicle fire detection unit (400) analyzes fire information and environmental information collected from the fire information collection unit (200) and environmental information collection unit (300) using an artificial intelligence algorithm based on machine learning (artificial intelligence machine learning or deep learning) to predict or detect a fire in an electric vehicle, and when a fire in an electric vehicle is predicted or detected, fire analysis information is generated and transmitted to the parking control and fire control sides.

[0068] For example, the information receiving unit (401) of the electric vehicle fire detection unit (400) receives fire information collected by the fire information collecting unit (200) and environmental information collected by the environmental information collecting unit (300).

[0069] In addition, the environmental temperature detection unit (402) detects the environmental temperature around the electric vehicle parked from the environmental information received from the information receiving unit (401) and transmits it to the abnormal temperature detection unit (405).

[0070] In addition, the electric vehicle information extraction unit (403) extracts only electric vehicle information from the thermal image fire information received from the information receiving unit (401) and transmits it to the electric vehicle temperature detection unit (404).

[0071] Here, the electric vehicle information extraction unit (403) can extract only the electric vehicle area as electric vehicle information from the received thermal image fire information through object classification. Object classification can be performed by directly adopting well-known edge detection or object detection and classification techniques in image processing techniques to classify and detect only the electric vehicle area from the thermal image.

[0072] Next, the electric vehicle temperature detection unit (404) detects the electric vehicle temperature by calculating the number of pixels by color in the electric vehicle area information extracted from the electric vehicle information extraction unit (403).

[0073] Here, the thermal image, which is the electric vehicle area information, contains black and white shading (dark and light) or color information for each pixel. Temperature can be detected by calculating the color information extracted for each pixel. The technology for extracting temperature from images captured by a thermal imaging camera is also well-known, so a detailed description thereof will be omitted. The detected electric vehicle temperature information is transmitted to the abnormal temperature detection unit (405).

[0074] The above abnormal temperature detection unit (405) corrects the electric vehicle temperature detected by the electric vehicle temperature detection unit (404) based on the environmental temperature detected by the environmental temperature detection unit (402).

[0075] In other words, the detected electric vehicle temperature may differ depending on the actual surrounding temperature. For example, the detected electric vehicle temperature may differ depending on whether the surrounding temperature is low or high. Therefore, in order to compensate for this difference and accurately detect the electric vehicle temperature, the detected electric vehicle temperature is compensated using the measured environmental temperature information. The compensation value for the electric vehicle temperature compensation is generated in advance through experiments, etc. according to the surrounding environmental temperature, and then stored in memory in the form of a table. When only the surrounding environmental temperature is detected, the compensation value is extracted from the table in real time to compensate for the electric vehicle temperature. For example, when the surrounding environmental temperature is high, the compensation value threshold can be set to be lowered so that the electric vehicle temperature value is slightly lower, and when the surrounding environmental temperature is low, the compensation value threshold can be set to be higher so that the electric vehicle temperature value is slightly higher.

[0076] Through this process, the temperature information of the electric vehicle is accurately detected, and the detected electric vehicle temperature information is compared with the preset reference electric vehicle temperature information to determine whether the temperature of the electric vehicle is abnormal, and the difference is extracted. Next, the extracted difference is compared with the allowable temperature range, which is the temperature range for determining whether the temperature of the electric vehicle is normal, and if the temperature of the electric vehicle deviates from the allowable temperature range, it is determined as an abnormal temperature. If the temperature is determined to be abnormal, the difference temperature between the temperature of the electric vehicle and the allowable temperature range is extracted and transmitted to the fire prediction unit (406).

[0077] The above fire prediction unit (406) analyzes the information (differential temperature) for fire prediction provided by the above abnormal temperature detection unit (405) using a machine learning-based artificial intelligence algorithm for a certain period of time to predict a fire or determine the occurrence of a fire in an electric vehicle, and outputs fire prediction or fire occurrence determination information.

[0078] Here, the machine learning-based AI algorithm can be a fire inference model that predicts or determines fire occurrence based on abnormal temperature information and the change time of that abnormal temperature. A fire inference model is a model that predicts and determines fire occurrence through learning by matching abnormal temperature information with the change time of that abnormal temperature in advance. In other words, the fire inference model can be used to track abnormal temperatures (differential temperatures) based on time information for a certain period of time, thereby predicting fire potential or detecting fire to determine whether a fire has actually occurred in an electric vehicle.

[0079] Here, the battery separator melting temperature of electric vehicles is 130 to 150℃, and the battery cathode material thermal decomposition temperature is 240℃. Therefore, if the allowable temperature range is set based on this, it is possible to predict fire in electric vehicles or determine the state of fire occurrence.

[0080] When a fire is determined through fire prediction or fire detection, the fire determination information is transmitted to the fire analysis information generation unit (407).

[0081] The above fire analysis information generation unit (407) generates fire analysis information based on fire prediction information or fire occurrence information, which is fire determination information determined by the fire prediction unit (406), and transmits the generated fire analysis information to the parking control unit (100) and the fire control unit (500).

[0082] The above parking control unit (100) performs fire response actions based on the above fire analysis information.

[0083] Here, fire response actions can be taken by controlling vehicle entry and exit from the parking lot through the access control unit (103). In particular, entry into the parking lot can be restricted, while exit can be permitted, allowing vehicles to quickly leave the parking lot.

[0084] In addition, the vehicle owner contact information (104) can transmit the fire prediction status or fire occurrence status to the vehicle owner terminal based on the vehicle owner contact information included in the fire analysis information.

[0085] Here, the electric vehicle fire detection unit (400) can provide the registered vehicle information of the electric vehicle and the registered vehicle information of general vehicles parked around the electric vehicle to the parking control unit (100) when a fire is predicted or detected for a specific electric vehicle.

[0086] The above-mentioned vehicle owner contact section (104) transmits fire prediction status information or fire occurrence status information as a text message based on the electric vehicle owner contact information included in the fire analysis information or the contact information of owners of general vehicles parked around the electric vehicle, thereby promptly allowing the owners of the electric vehicle and the owners of general vehicles parked around the electric vehicle to evacuate the vehicles.

[0087] By delivering fire prediction information to the owner of an electric vehicle or delivering information about the possibility of fire in surrounding electric vehicles to the owners of regular vehicles around the electric vehicle, the owner of the electric vehicle or regular vehicle can quickly move the vehicle or respond by checking its status, thereby preventing fires in electric vehicles in advance. In addition, in the event of a fire in an electric vehicle, the surrounding regular vehicles can also be prevented from starting fires.

[0088] In particular, if the parking lot is located underground in a building, fires in electric vehicles can be prevented in advance using fire prediction information as described above, and large-scale fires in buildings caused by electric vehicle fires can also be prevented in advance.

[0089] Meanwhile, the fire control unit (500) quickly controls the fire by using the location tracking information of the electric vehicle in which a fire has occurred or a fire is predicted when fire analysis information is generated from the electric vehicle fire detection unit (400).

[0090] For example, when the fire evacuation broadcast unit (501) receives fire analysis information from the electric vehicle fire detection unit (400), a fire evacuation broadcast is performed to notify people around the electric vehicle of this and promptly evacuate, thereby preventing casualties in advance. Similarly, in the case of an underground parking lot, an evacuation broadcast is also performed in the building, prompting people in the building to promptly evacuate, thereby preventing casualties in the building in advance.

[0091] In addition, the fire reporting unit (502) analyzes the fire analysis information when the above fire analysis information occurs, and automatically reports the fire by transmitting the fire information, including location information, to the relevant organizations (fire department, police station, etc.). Here, fire reports can be made automatically via a communication network by registering the phone numbers and other information of the relevant fire-related organizations in advance. The relevant organizations that receive the fire report can respond quickly to the fire and minimize damage caused by the fire.

[0092] In addition, the fire extinguishing system operating unit (503) operates the installed fire extinguishing system based on the fire location information included in the fire analysis information to extinguish the fire. Fire extinguishing can be simply accomplished by automatically operating the fire extinguishing equipment. In particular, the fire control unit (500) is linked with the parking control unit (100) in advance to know the location information of the electric vehicle in which the fire has occurred or the fire has occurred, so that the fire extinguishing equipment in the corresponding parking lot area can be operated quickly and accurately to enable a quick initial response in the event of a fire, thereby preventing the fire from spreading.

[0093] According to the present invention described above, electric vehicles entering a parking lot are recognized and tracked, parked electric vehicles are continuously monitored, and if a possibility of fire is predicted by analyzing the electric vehicles using a machine learning-based artificial intelligence (AI) algorithm, casualties can be minimized and large-scale fires can be prevented in advance through fire control.

[0094] In addition, according to the present invention, electric vehicles entering a parking lot are recognized and tracked, and parked electric vehicles are continuously monitored and analyzed using a machine learning-based artificial intelligence (AI) algorithm. When a fire is detected, the fire evacuation broadcast and fire extinguishing system are automatically activated, thereby minimizing casualties and building fires caused by fire.

[0095] In addition, according to the present invention, when a fire is detected in an electric vehicle, a fire report can be automatically made to a relevant organization (e.g., fire station, police station, government office, etc.), thereby helping to quickly extinguish the fire.

[0096] Although the invention made by the present inventor has been specifically described according to the above embodiments, it is obvious to a person having ordinary skill in the art that the present invention is not limited to the above embodiments and can be modified in various ways without departing from the spirit thereof.

[0097] The present invention is applied to a technology for predicting a fire in an electric vehicle in a parking lot or for fire control when a fire occurs.

Claims

1. A fire information collection unit that collects electric vehicle information as fire information for fire analysis and prediction of parked electric vehicles; An environmental information collection unit that collects environmental information of the parking lot where electric vehicles are parked; An electric vehicle fire detection unit that analyzes fire information and environmental information collected from the above fire information collection unit and environmental information collection unit using a machine learning-based artificial intelligence algorithm to predict or detect a fire in an electric vehicle, and generates fire analysis information when a fire in an electric vehicle is predicted or detected and transmits it to the parking control and fire control side; A fire control unit that performs fire control when fire analysis information is generated from the above electric vehicle fire detection unit; and It includes a parking control unit that recognizes an electric vehicle entering a parking lot, tracks the parking location, and performs parking control in conjunction with the fire control unit when fire analysis information is generated from the electric vehicle fire detection unit. A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that the parking control unit transmits the tracked electric vehicle parking location information to the fire information collection unit, and the fire information collection unit collects information on the electric vehicle as an image using a thermal imaging sensor based on the electric vehicle parking location information, and transmits the collected image information on the electric vehicle as fire information to the electric vehicle fire detection unit.

2. In claim 1, the environmental information collection unit, A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it detects at least one of the temperature and humidity around an electric vehicle parked in a parking lot as environmental information and transmits the detected information to the electric vehicle fire detection unit.

3. In claim 1, the parking control unit, An electric vehicle recognition unit that recognizes electric vehicles by recognizing vehicles entering a parking lot; A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized by including an electric vehicle location tracking unit that tracks an electric vehicle recognized by the electric vehicle recognition unit in real time to confirm a parking location and transmits the confirmed parking location information to an electric vehicle fire detection unit.

4. In claim 3, the parking control unit, An access control unit that controls vehicle entry into a parking lot based on fire analysis information generated from the above electric vehicle fire detection unit; A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it includes a vehicle owner contact unit that transmits a fire prediction status or fire occurrence status to a vehicle owner terminal based on vehicle owner contact information included in fire analysis information generated from the electric vehicle fire detection unit.

5. In claim 4, the above-mentioned contact part, A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it transmits fire prediction status information or fire occurrence status information as a text message based on the contact information of the electric vehicle owner included in the above fire analysis information or the contact information of the owners of general vehicles parked around the electric vehicle.

6. In claim 1, the fire control unit, A fire evacuation broadcast unit that performs a fire evacuation broadcast when receiving fire analysis information from the above electric vehicle fire detection unit; A fire reporting unit that analyzes the fire analysis information when the above fire analysis information occurs and transmits the fire information, including location information, to the relevant authorities to report the fire; and A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it includes a fire extinguishing system operating unit that operates the installed fire extinguishing system based on fire location information included in the above fire analysis information to extinguish a fire.

7. In claim 1, the electric vehicle fire detection unit, An information receiving unit that receives fire information collected from the fire information collection unit and environmental information collected from the environmental information collection unit; An environmental temperature detection unit that detects the environmental temperature around the electric vehicle parked from the environmental information received from the above information receiving unit; A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it includes an electric vehicle information extraction unit that extracts only electric vehicle information from the image fire information received from the above information receiving unit.

8. In claim 7, the electric vehicle information extraction unit, A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized by extracting only the electric vehicle area as electric vehicle information through object classification from received video fire information.

9. In claim 7, the electric vehicle fire detection unit, An electric vehicle temperature detection unit that detects the electric vehicle temperature by calculating the number of pixels by color in the electric vehicle area information extracted from the electric vehicle information extraction unit; A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it includes an abnormal temperature detection unit that corrects the electric vehicle temperature detected by the electric vehicle temperature detection unit based on the environmental temperature detected by the environmental temperature detection unit, compares the corrected electric vehicle temperature information with preset reference electric vehicle temperature information, extracts a difference, determines that the extracted difference is an abnormal temperature if it deviates from the allowable temperature range, and provides the difference information as information for fire prediction.

10. In claim 9, the electric vehicle fire detection unit, A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it includes a fire prediction unit that analyzes information for fire prediction provided by the above-mentioned abnormal temperature detection unit using a machine learning-based artificial intelligence algorithm for a certain period of time to predict a fire or determine the occurrence of a fire in an electric vehicle, and outputs fire prediction or fire occurrence determination information.

11. In claim 10, the machine learning-based artificial intelligence algorithm, A fire control system for electric vehicles in a parking lot using artificial intelligence and a thermal imaging sensor, characterized by a fire inference model that predicts fire or determines fire occurrence based on abnormal temperature information and abnormal temperature change time information, and predicts fire and determines fire occurrence through learning by matching abnormal temperature information and abnormal temperature change time information in advance.

12. In claim 10, the electric vehicle fire detection unit, A parking lot electric vehicle fire control system using artificial intelligence and a thermal imaging sensor, characterized in that it includes a fire analysis information generation unit that generates fire analysis information based on fire prediction information or fire occurrence information determined by the above fire prediction unit and transmits the generated fire analysis information to the parking control unit and the fire control unit.

Citation Information

Patent Citations

  • Smart parking management system

    KR101624991B1

  • The method for operating a parking management system according to recognition of electric vehicle and Hi-pass

    KR101819568B1

  • Vehicle control system using mobile network

    KR1020160072971A

  • Fire fighting system for electric vehicle and its operation method

    KR102372971B1

  • KR20200088736A

Cited By

  • Fire early warning method and system for new energy vehicle in tunnel

    CN120199009A