Electric Vehicle Fire Prevention AI Receiver System

Abstract

The present invention relates to an electric vehicle fire prevention AI receiver system in which fire detectors are attached to a charger, a charging connector, a battery cell of an electric vehicle, etc., to detect fires, an analyzer performs time-series analysis of the detection signals from the fire detectors based on artificial intelligence (AI) technology, and a controller operates in an intensive monitoring mode, a warning alarm mode, or a fire prevention mode according to the abnormal pattern values ​​of the analysis results to accurately determine fire risks without false alarms or malfunctions, and prevents fires by cutting off the electrical connection between the charger and the electric vehicle or stopping the operation of the charger if a fire is expected to occur. The electric vehicle fire prevention AI receiver system according to the present invention comprises: a source data unit for collecting data by cause of fire according to the environment; a data labeling unit for classifying the data collected in the source data unit into normal state data and malfunction data by cause of fire; and an AI modeling unit for learning an AI model that applies the data separated by the data labeling unit; a charger for supplying charging power; a charging connector connected to the charger and connected to the electric vehicle; and a fire detector provided in one or more of the battery cells of the electric vehicle; an analyzer for performing time-series analysis of a detection signal received from the fire detector using an AI model learned in the AI ​​modeling unit; and a controller that operates in an intensive monitoring mode, a warning alarm mode, and a fire prevention mode according to the analysis result of the analyzer.

Description

Technology Field

[0001] The present invention relates to an electric vehicle fire prevention AI receiver system that prevents the occurrence of fire during electric vehicle charging. More specifically, the invention relates to an electric vehicle fire prevention AI receiver system in which fire detectors are attached to the charger, charging connector, and battery cell of the electric vehicle to detect fires, an analyzer performs time-series analysis of the detection signals from the fire detectors based on artificial intelligence (AI), and a controller operates in a concentrated monitoring mode, a warning alarm mode, or a fire prevention mode according to the abnormal pattern values ​​of the analysis results to accurately determine fire risk without false alarms or malfunctions, and prevents fires by disconnecting the electrical connection between the charger and the electric vehicle or stopping the operation of the charger if a fire is expected to occur. Background Technology

[0003] Recently, with the proliferation of electric vehicles, charging stations are increasing, and along with this, electric vehicle fires are also on the rise.

[0004] In the case of electric vehicle fires, not only is it difficult to extinguish, but even if the fire is extinguished in the first instance, there is a risk of a secondary fire until the energy stored in the battery is depleted.

[0005] Therefore, when an electric vehicle fire occurs, significant damage to life and property often results from the spread of the fire. Furthermore, extinguishing an electric vehicle fire consumes a considerable amount of time, manpower, and resources.

[0007] Due to the risk of electric vehicle fires, it is important to effectively prevent fires starting from the charging stage.

[0009] For reference, prior art regarding the prevention of electric vehicle fires includes published patent 10-2024-0162979 "Prevention and extinguishing of fire during electric vehicle charging" and registered patent 10-2672274 "Electric vehicle charging station fire monitoring system and method".

[0010] Conventional technology primarily uses surveillance cameras to monitor the charging status of electric vehicles. However, this method has limitations in identifying critical information regarding fire risks, such as temperature changes during charging, and the occurrence of a fire is only recognized after it has already happened, which is inconvenient as it requires continuous monitoring by a supervisor. The problem to be solved

[0012] The present invention is developed to solve the problems of the prior art as described above, and aims to provide an electric vehicle fire prevention AI receiver system that analyzes the detection signal of a fire detector in a time series based on artificial intelligence technology and operates in an intensive monitoring mode, a warning alarm mode, and a fire prevention mode according to the numerical values ​​of the analyzed abnormal patterns, thereby reliably preventing the occurrence of fire and ensuring that no false alarms or malfunctions occur during the process.

[0013] In addition, another objective is to provide an electric vehicle fire prevention AI receiver system in which a fire detector is attached so as to be in close contact with the charging connector, thereby enabling more accurate detection of the overheating temperature of the charging connector. means of solving the problem

[0015] The electric vehicle fire prevention AI receiver system according to the present invention

[0016] An AI server comprising: a source data unit for collecting data by cause of fire according to the environment; a data labeling unit for classifying the data collected in the source data unit into normal state data and malfunction data by cause of fire; and an AI modeling unit for training an AI model that applies the data separated by the data labeling unit;

[0017] A charger that supplies charging power, a charging connector connected to the charger and connected to an electric vehicle, and a fire detector provided in one or more of the battery cells of the electric vehicle;

[0018] An analyzer that performs time-series analysis of a detection signal received from the fire detector using an AI model learned in the AI ​​modeling unit;

[0019] It comprises a controller that operates in intensive monitoring mode if the abnormal pattern in the analysis result of the analyzer is 20% or more, in warning alarm mode if it is 30% or more, and in fire prevention mode if it is 40% or more.

[0021] And the above controller is characterized by cutting off the electrical connection between the charger and the electric vehicle or stopping the operation of the charger in a fire prevention mode, and

[0023] A detachable member comprising a holder that encloses the fire detector and a band that is connected to the holder and elastically fastened to the fire detector, thereby detachably coupling the fire detector to the outer surface of the charging connector; further comprising

[0024] The band is characterized by having a male projection with a catch formed on the end side, and a plurality of pairs of female plates with a hook formed to engage with the catch formed on the opposite side.

[0026] The above pair of female plates are characterized by having an insertion hole formed therein, and the male projection is provided with a locking pin that is inserted into and engages with the insertion hole. Effects of the invention

[0028] The electric vehicle fire prevention AI receiver system according to the present invention is a highly useful invention for industrial development, as it analyzes detection signals from fire detectors attached to the charging connector, charger, and battery cells of the electric vehicle—which have a high probability of ignition during charging—using a time-series AI solution based on artificial intelligence technology to accurately determine the risk of fire occurrence; if the risk of fire is high or a fire occurs, it immediately stops the operation of the charger or disconnects the electrical connection between the charger and the electric vehicle to prevent the occurrence or spread of fire; it eliminates issues regarding false alarms or malfunctions regarding fire occurrence; and the fire detector is closely attached to the charging connector via a detachable member to precisely detect heat generation temperatures. Brief explanation of the drawing

[0030] FIG. 1 is a configuration diagram of an electric vehicle fire prevention AI receiver system according to the present invention. FIG. 2 is a drawing illustrating a detachable member according to the present invention. FIG. 3 is a drawing illustrating a fastening member for fastening a band of a detachable member according to the present invention. Specific details for implementing the invention

[0031] Hereinafter, an electric vehicle fire prevention AI receiver system according to the present invention will be described in more detail with reference to the drawings.

[0033] Before describing the electric vehicle fire prevention AI receiver system according to the present invention in more detail,

[0034] The present invention is capable of various modifications and may take various forms, and embodiments (aspects or examples) are to be described in detail in the text. However, this is not intended to limit the present invention to the specific disclosed forms, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention.

[0036] As shown in FIG. 1, the electric vehicle fire prevention AI receiver system according to the present invention comprises a fire detector (10, 20, 30), an analyzer (40), a controller (50), and an AI server (60).

[0038] The above fire detector (10, 20, 30) is attached to one or more of the charging connector (1) connected to the electric vehicle, the charger (2) that supplies charging power to the charging connector (1), and the battery cell of the electric vehicle to which the charging connector (1) is connected, and detects the temperature by coming into contact with them.

[0039] Here, the fire detector provided in the charging connector (1) is referred to as the first fire detector (10), the fire detector provided in the charger (2) is referred to as the second fire detector (20), and the fire detector provided in the battery cell is referred to as the third fire detector (30).

[0040] The first fire detector (10) is attached so as to be in contact with the outer surface of the charging connector (1), the second fire detector (20) is attached so as to be in contact with the power supply or / and control board inside the charger (2), and the third fire detector (30) is attached directly or indirectly to the battery cell (3) in the electric vehicle.

[0041] The first fire detector (10), second fire detector (20), and third fire detector (30) transmit detection signals in real time to the analyzer (40). The transmission of detection signals can be done wirelessly or via a wired connection, but wireless transmission may be preferable.

[0042] The above fire detectors (10, 20, 30) can detect minute heat fluctuations by using contact-type heat detectors with 16-bit resolution.

[0044] The detection signal regarding the temperature detected by the fire detectors (10, 20, 30) can be transmitted to the analyzer (40) in real time and stored in memory in real time, and the detection signal stored in memory can be transmitted to the analyzer (40).

[0046] The analyzer (40) analyzes the detection signals transmitted by the first fire detector (10), second fire detector (20), and third fire detector (30), namely the temperature of the charging connector (1), charger (2), and battery cell (3), using artificial intelligence (AI) technology.

[0047] The analyzer (40) above performs time-series analysis of the detection signal using a time-series AI solution. The time-series AI solution learns the input detection signal in real time using an RNN, which is a time-series AI model, and determines whether an abnormal pattern occurs in the detection signal (i.e., temperature) through time-series analysis. The SAX / PAA algorithm may also be applied to the time-series analysis of the detection signal.

[0048] When charging of an electric vehicle begins, the temperature of the charging connector (1), charger (2), battery cell (3), etc., rises gradually, and once it enters a stable phase, it generally maintains a constant temperature without significant fluctuation.

[0049] The analyzer (40) checks the amount of change over time of the temperature detected by the first fire detector (10), the second fire detector (20), and the third fire detector (30) to determine whether a rapid change in temperature deviating from the normal temperature rise rate or a temperature rise deviating from the constant range occurs, that is, whether there is an abnormal signal (abnormal pattern).

[0050] The analyzer (40) can be configured to separate the first fire detector (10), the second fire detector (20), and the third fire detector (30) so that even if an abnormal signal is detected in any one of them, it can be configured to enter the intensive monitoring mode, the warning alarm mode, and the fire prevention mode. Alternatively, it can be configured to enter the monitoring mode, the warning alarm mode, and the fire prevention mode only when it is detected that an abnormal signal is detected in two or more or all of the first fire detector (10), the second fire detector (20), and the third fire detector (30), thereby preventing the occurrence of problems such as false alarms or malfunctions.

[0052] The AI ​​server (60) provides the AI ​​model that the analyzer (40) learns to analyze the detection signals of the fire detectors (10, 20, 30) in a time series.

[0053] The above AI server (60) includes a source data section (61), a data labeling section (62), and an AI modeling section (63).

[0054] The above source data unit (61) creates various experimental environments or installs fire detectors (10, 20, 30) at actual electric vehicle charging stations and collects source data while charging electric vehicles.

[0055] The source data classifies detection signal data from fire detectors (10, 20, 30) according to each cause of fire. The source data mainly consists of time-series data, and collects and classifies data before and after based on hourly, daily, and weekly intervals.

[0056] The above data labeling unit (62) classifies source data into normal state data, malfunction data, and data by cause of actual fire using crowdsourcing techniques, etc.

[0057] The above AI modeling unit (63) learns an AI model that analyzes the detection signals of the fire detectors (10, 20, 30) in a time series.

[0058] AI modeling can apply RNN-based artificial intelligence algorithms for handling time-series data. To classify specific fire patterns, algorithms such as SAX / PAA can be applied to extract features from the time-series data. The AI ​​model is trained by applying fire cause-specific data obtained from data labeling.

[0059] We measure the accuracy of the AI ​​modeling algorithm using pre-classified test data, and also establish an experimental environment to test whether the algorithm actually works.

[0060] The accuracy of the AI ​​modeling is enhanced by repeatedly performing the process of securing additional training data and rebuilding the AI ​​model until the desired level of classification results regarding fire detection and cause is obtained.

[0062] The controller (50) operates by entering an intensive monitoring mode when the abnormal pattern of the detection signal analyzed by the analyzer (40) is 20% or more, a warning alarm mode when it is 30% or more, and a fire prevention mode when it is 40% or more.

[0063] When the above controller (50) enters the intensive monitoring mode, it notifies the manager of the charging station via text message, etc., and when it enters the warning alarm mode, it sounds a warning so that the manager of the charging station can check. When it enters the fire prevention mode, it stops the operation of the above charger (2) so that charging power is no longer supplied, and disconnects the electrical connection between the above charging connector (1) and the electric vehicle charging port to cut off the connection with the electric vehicle battery cell (3), thereby preventing the occurrence of a fire or minimizing the spread of a fire.

[0064] The charging connector (1) is equipped with an automatic release device so that the connection between the charging connector (1) and the electric vehicle is released by the control of the controller (50).

[0066] Referring to FIG. 2, the first fire detector (10), which is connected to the charging connector (1) and detects the temperature of the charging connector (1), includes a casing (11) in which a sensing board and a communication module are built, and a sensing plate (13) that is exposed on the lower surface of the casing (11) and connected to the sensing board.

[0067] The above sensing plate (13) is in contact with the outer surface of the charging connector (1) to detect the temperature of the charging connector (1).

[0068] The sensing plate (13) may use a metal plate whose resistance value changes according to temperature. It is preferable that the sensing plate (13) be provided protruding from the lower surface of the casing (11) and elastically supported inside the casing (11) so as to be in close contact with the outer surface of the connector (1).

[0069] The above casing (11) is detachably coupled to the charging connector (1) through a detachable member (70).

[0070] The above detachable member (70) is fastened to the charging connector (1) so that the sensing plate (13) of the first fire detector (10) is in close contact with the surface of the charging connector (1).

[0071] The above detachable member (70) comprises a holder (71) that wraps around the upper surface and side edges of the upper side of the casing (11), and a band (73) that is connected to the holder (71) and detachably coupled to the holder (71). The band (73) has elasticity and pulls the casing (11) coupled to the holder (71) toward the connector (1), thereby causing the sensing plate (13) attached to the casing (11) to adhere to the outer surface of the connector (1).

[0072] One side of the band (73) is connected to one side of the holder (71), and the other side of the holder (71) is provided with a loop (75) through which the band (73) passes. After passing through the loop (75), the band (73) is fastened through a fastening member.

[0073] The above fastening member includes a male projection (81) protruding from the end side of the band (73) and a plurality of a pair of female plates (83) protruding from the opposite side of the band (73).

[0074] A catch (811) is formed on both sides of the male projection (81), and a hook (831) is formed on the inner surface of the pair of female plates (83), so that when the male projection (81) is inserted between the pair of female plates (83), the catch (811) is engaged with the hook (831).

[0075] When the connection and separation between the male projection (81) and the pair of female plates (83) are repeated, the pair of female plates (83) may separate, and then the locking connection between the locking jaw (811) and the hook jaw (831) weakens, so that when an external impact is applied, the locking connection is released and the male projection (81) can be separated from the pair of female plates (83).

[0076] To prevent this, a 'C'-shaped locking pin (85) capable of moving back and forth inward and outward is provided at the front and rear of the male projection (81), respectively, and a spring (86) is interposed between the front and rear locking plates to push the two locking pins (85) outward and advance them. Additionally, an insertion hole (835) into which the locking pin (85) is inserted and caught is formed on the inner surface of the pair of female plates (83). When the 'C'-shaped locking pin (85) is inserted into the insertion hole (835), it holds the pair of female plates (83) so that they do not spread apart, and also prevents the male projection (81) from being lifted and separated from the pair of female plates (83).

[0077] The male projection (81) is provided with a 'U'-shaped push pin (87) capable of moving back and forth in the vertical direction, and the locking pin (85) is provided with an inclined surface (851) on which the end of the push pin (87) rests. When a user attaches or detaches the male projection (81) to or from the pair of female plates (83), the user presses the push pin (87) with a finger, and the push pin (87) moves downward and forward. As it moves forward, the end of the push pin (87) travels along the inclined surface (851) and moves the locking pin (85) inward.

[0078] When the male projection (81) is coupled to the pair of female plates (83), the push pin (87) is first pressed by the user's finger and advanced downward, and the advancing push pin (87) moves along the inclined surface (851) so that the locking pin (85) is retracted inward, and then the locking jaw (811) and the hook jaw (831) are locked together, and then when the user removes their finger, the locking pin (85) is advanced outward by the restoring force of the spring (86) so that the insertion hole (835) is inserted.

[0079] When the male projection (81) is separated from the pair of female plates (83), the push pin (87) is pressed by the user's finger and advanced downward, and the advancing push pin (87) moves along the inclined surface (851) and the locking pin (85) is retracted inward, and then the locking pin (85) is disengaged from the insertion hole (835), and then when the user pulls the band they were holding, the locking connection between the hook jaw (831) and the locking jaw (811) is forcibly released, and the male projection (81) is separated from the pair of female plates (83).

[0081] In describing the present invention above, an electric vehicle fire prevention AI receiver system having a specific shape and structure has been described with reference to the attached drawings; however, the present invention is susceptible to various modifications and changes by those skilled in the art, and such modifications and changes should be interpreted as falling within the scope of protection of the present invention. Explanation of the symbols

[0083] 10 : 1st fire detector 20 : 2nd fire detector 30: Third fire detector 40: Analyzer 50 : Controller 60 : AI Server 61: Source Data Department 62: Data Labeling Department 63 : AI Modeling Section 70 : Detachable Member 71 : Holder 73 : Band 81 : Male projection 83 : Female plate 85 : Locking pin 86 : Spring 87 : Push pin

Claims

Claim 1 An electric vehicle comprising: an AI server (60) including a source data unit (61) for collecting data by cause of fire according to the environment, a data labeling unit (62) for classifying the data collected in the source data unit (61) into normal state data and malfunction data by cause of fire, and an AI modeling unit (63) for training an AI model by applying the data separated by the data labeling unit (62); a charger (2) for supplying charging power, a charging connector (1) connected to the charger and connected to the electric vehicle, and a fire detector (10, 20, 30) provided in one or more of the battery cells (3) of the electric vehicle; an analyzer (40) for time-series analysis of detection signals received from the fire detectors (10, 20, 30) using an AI model trained in the AI ​​modeling unit (63); and a controller (50) that operates in an intensive monitoring mode if the abnormal pattern in the analysis result of the analyzer (40) is 20% or more, a warning alarm mode if it is 30% or more, and a fire prevention mode if it is 40% or more. In a fire prevention AI receiver system, the system further comprises a detachable member (70) that detachably connects the fire detector (10) to the outer surface of the charging connector (1); the detachable member (70) includes a holder (71) that wraps around the casing (11) of the fire detector (10) and is provided with a ring (75), and a band (73) that is connected to the holder (71) and is fastened through a fastening member after passing through the ring (75); the fastening member includes a male projection (81) protruding from the end side of the band (73) and a plurality of a pair of female plates (83) protruding from the opposite side of the band (73); a catch (811) is formed on both sides of the male projection (81), and a hook (831) is formed on the inner surface of the pair of female plates (83), so that the male projection (81) is connected to the pair of When inserted between the female plate (83), the above-mentioned catch (811) is engaged with the above-mentioned hook (831), and a 'C'-shaped catch pin (85) capable of moving back and forth inward and outward is provided on the front and rear of the above-mentioned male projection (81), respectively.A spring (86) is interposed between the front and rear locking pins (85) to push the two locking pins (85) outward and advance them, and an insertion hole (835) is formed on the inner surface of the pair of female plates (83) into which the locking pin (85) is inserted and caught, so that when the 'C'-shaped locking pin (85) is inserted into the insertion hole (835), the pair of female plates (83) are held so that they do not spread apart and the male projection (81) is lifted and separated from the pair of female plates (83), and a 'C'-shaped push pin (87) is provided on the male projection (81) so that it can move back and forth in the up and down direction, and the locking pin (85) is provided with an inclined surface (851) on which the end of the push pin (87) is placed, so that when the push pin (87) is pressed and advanced downward, the advancing push pin (87) An electric vehicle fire prevention AI receiver system characterized by the following: moving along the inclined surface (851) so that the locking pin (85) is retracted inward, and subsequently the locking projection (811) and the hook projection (831) are engaged; subsequently, when the push pin (87) is released, the locking pin (85) is advanced outward by the restoring force of the spring (86) so that the insertion hole (835) is inserted; and when the push pin (87) is pressed again and advanced downward, the advancing push pin (87) moves along the inclined surface (851) so that the locking pin (85) is retracted inward, and then the locking pin (85) is disengaged from the insertion hole (835). Claim 2 An electric vehicle fire prevention AI receiver system according to claim 1, wherein the controller (50) cuts off the electrical connection between the charger (2) and the electric vehicle or stops the operation of the charger (2) in a fire prevention mode. Claim 3 delete Claim 4 delete

Images